This is another volume in the new Cambridge series of com-
panions to major philosophers. Each volume will contain
specially commissioned essays by an international team of
scholars, together with a substantial bibliography, and will
serve as a reference work for students and nonspecialists.
One aim of the series is to dispel the intimidation such
readers often feel when faced with the work of a difficult and
challenging thinker.
Plato stands at the head of our philosophical tradition, be-
ing the first Western thinker to produce a body of writing that
touches upon a wide range of topics still discussed by philoso-
phers today. In a sense he invented philosophy as a distinct
subject, for although many of these topics were discussed by
his intellectual predecessors and contemporaries, he was the
first to bring them together by giving them a unitary treat-
ment. He conceives of philosophy as a discipline with a dis-
tinctive intellectual method, and he makes radical claims for
its position in human life and the political community. This
volume contains fifteen new essays discussing Plato's views
about knowledge, reality, mathematics, politics, ethics, love,
poetry, and religion. There are also analyses of the intellec-
tual and social background of his thought, the development of
his philosophy throughout his career, the range of alternative
approaches to his work, and the stylometry of his writing.
New readers and nonspecialists will find this the most
convenient, accessible guide to Plato currently available. Ad-
vanced students and specialists will find a conspectus of
recent developments in the interpretation of Plato.
THE CAMBRIDGE COMPANION TO
PLATO
OTHER VOLUMES IN THIS SERIES OF CAMBRIDGE
COMPANIONS:
AQUINAS Edited by NORMAN KRETZMAN and
ELEONORE STUMP
ARISTOTLE Edited by JONATHAN BARNES
DESCARTES Edited by JOHN COTTINGHAM
FOUCAULT Edited by GARY GUTTING
FREUD Edited by JEROME NEU
HEGEL Edited by FREDERICK BEISER
HEIDEGGER Edited by CHARLES GUIGNON
HOBBES Edited by TOM SORRELL
HUME Edited by DAVID FATE NORTON
HUSSERL Edited by BARRY SMITH and DAVID
WOODRUFF SMITH
KANT Edited by PAUL GUYER
LEIBNIZ Edited by NICHOLAS JOLLEY
LOCKE Edited by VERE CHAPPELL
MARX Edited by TERRELL CARVER
MILL Edited by JOHN SKORUPSKI
NIETZSCHE Edited by BERND MAGNUS
SARTRE Edited by CHRISTINA HOWELLS
SPINOZA Edited by DON GARRETT
The Cambridge Companion to
PLATO
Edited by Richard Kraut
University of Illinois at Chicago
CAMBRIDGE
UNIVERSITY PRESS
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 2RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/0521430186
© Cambridge University Press 1992
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.
First published 1992
Reprinted 1993 (twice), 1995, 1996 (twice), 1997,1999 (twice)
A catalogue record for this publication is available from the British Library
ISBN-10 0-521-43018-6 hardback
ISBN-10 0-521-43610-9 paperback
Transferred to digital printing 2005
FOR GREGORY VLASTOS
1907-1991
CONTENTS
List of contributors page ix
Chronology xii
Abbreviations xiii
1 Introduction to the study of Plato i
RICHARD KRAUT
2 Plato: The intellectual background 51
T. H. IRWIN
3 Stylometry and chronology 90
LEONARD BRANDWOOD
4 Socrates and the early dialogues 121
TERRY PENNER
5 Mathematical method and philosophical truth 170
IAN MUELLER
6 Inquiry in the Meno 200
GAIL FINE
7 Plato and Greek religion 227
MICHAEL L. MORGAN
Vll
viii Contents
8 Platonic love 248
G. R. F. FERRARI
9 Plato's metaphysical epistemology 277
NICHOLAS P. WHITE
10 The defense of justice in Plato's Republic 311
RICHARD KRAUT
11 Plato on poetic creativity 338
ELIZABETH ASMIS
12 Good-bye to the Third Man 365
CONSTANCE C. MEINWALD
13 Plato's Sophist on false statements 397
MICHAEL FREDE
14 Disintegration and restoration: Pleasure and pain
in Plato's Philebus 425
DOROTHEA FREDE
15 Plato's later political thought 464
TREVOR J. SAUNDERS
Bibliography 493
Index of names and subjects 531
Index of passages 541
CONTRIBUTORS
ELIZABETH ASMIS is Associate Professor of Classics at the Univer-
sity of Chicago. The author of Epicurus' Scientific Method (Cornell
University Press, 1984) and numerous articles on Hellenistic philoso-
phy, she is currently working on Greek views of poetry from Plato to
the Neoplatonists.
LEONARD BRANDWOOD is Lecturer in the Department of Greek
and Latin at Manchester University. He is the author of A Word
Index to Plato (W. S. Maney & Son, 1976) and The Chronology of
Plato's Dialogues (Cambridge University Press, 1990).
G. R. F. FERRARI is Associate Professor of Classics at the University
of California, Berkeley. He is the author of Listening to the Cicadas:
A Study of Plato's Phaedrus (Cambridge University Press, 1987) and
of articles on Plato, the pre-Socratics, and archaic Greek culture.
GAIL FINE is Professor of Philosophy at Cornell University. She is the
author of numerous articles on the metaphysics and epistemology of
Plato and Aristotle. Her book On Ideas: Aristotle's Criticism of
Plato's Theory of Forms is to be published by Oxford University Press.
DOROTHEA FREDE is Professor of Philosophy at the University of
Hamburg. She is the author of Aristoteles und die Seeschlacht
(Vandenhoeck & Ruprecht, 1970) and has written numerous articles
on Plato, Aristotle, later Greek philosophy, and the philosophy of
Martin Heidegger. Her translation of Plato's Philebus is forthcoming
(Hackett Publishing Company).
MICHAEL FREDE is Professor of the History of Philosophy at Oxford
University and Fellow of Keble College. He is the author of Prddika-
IX
x Contributors
tion und Existenzaussage (Vandenhoeck & Ruprecht, 1967), Die
stoische Logik (Vandenhoeck & Ruprecht, 1974), and (with Giinther
Patzig) a translation of and commentary on Aristotle's Metaphysics
Z (C.H. Beck, 1988). Some of his many papers on Plato, Aristotle,
Stoicism, Skepticism, ancient medicine, and ancient grammatical
theories have been collected in Essays in Ancient Philosophy (Uni-
versity of Minnesota Press and Oxford University Press, 1987).
T. H. 1 RWIN is Professor of Philosophy at Cornell University. He is
the author of Plato's Moral Theory (Clarendon Press, 1977), Aris-
totle's First Principles (Clarendon Press, 1988), Classical Thought
(Oxford University Press, 1989), and translations of and commentar-
ies on Plato's Gorgias (Clarendon Press, 1979) and Aristotle's Nico-
machean Ethics (Hackett Publishing Company, 1985), as well as
numerous articles on Greek philosophy.
RICHARD KRAUT is Professor of Philosophy at the University of
Illinois at Chicago. He is the author of Socrates and the State (Prince-
ton University Press, 1984) and Aristotle on the Human Good
(Princeton University Press, 1989) and is currently writing a transla-
tion of and commentary on Aristotle's Politics: Books VII and VIII.
CONSTANCE c. MEINwALD is Assistant Professor of Philosophy at
the University of Illinois at Chicago. The author of Plato's Parmen-
ides (Oxford University Press, 1991), she is currently working on
Plato's late metaphysics.
MICHAEL L. MORGAN is Professor of Philosophy at Indiana Univer-
sity, Bloomington. He is the author of Platonic Piety: Philosophy
and Ritual in Fourth-Century Athens (Yale University Press, 1990)
and has written numerous articles on Plato as well as on Jewish
thought.
IAN MUELLER is Professor of Philosophy at the University of Chi-
cago. He is the author of Philosophy of Mathematics and Deductive
Structure in Euclid's Elements (MIT Press, 1981), as well as numer-
ous articles on ancient Greek philosophy, science, and mathematics.
He is currently preparing a translation of Alexander of Aphrodisias's
commentary on Aristotle's Prior Analytics.
TERRY PENNER is Professor of Philosophy at the University of Wis-
consin, Madison. He is the author of The Ascent from Nominalism:
Contributors xi
Some Existence Arguments in Plato's Middle Dialogues (D. Reidel
Publishing Company, 1987) and is currently at work on a sequel to
that volume, Plato and the Philosophers of Language. He is also
preparing a study of the philosophy of Socrates.
TREVOR j . SAUNDERS is Professor of Greek at the University of
Newcastle upon Tyne. He has produced three volumes in the Pen-
guin Classics series: a translation of Plato's Laws (1970), a revision
of T. A. Sinclair's translation of Aristotle's Politics (1981), and (as
contributing editor) Plato, Early Socratic Dialogues (1987). He has
written numerous articles on the political philosophy of Plato and
Aristotle, and his latest book is Plato's Penal Code (Clarendon Press,
1991).
NICHOLAS P. WHITE is Professor of Philosophy at the University of
Michigan. He is the author of Plato on Knowledge and Reality
(Hackett Publishing Company, 1976), A Companion to Plato's Re-
public (Hackett Publishing Company, 1979), and numerous articles
on Plato, Aristotle, and Stoicism. He is also the translator of
Epictetus's Handbook (Hackett Publishing Company, 1983), and his
translation of Plato's Sophist is forthcoming (Hackett Publishing
Company).
CHRONOLOGY
Plato's life*
427: born
Plato's writings*
c- 399~c- 387: composes
early dialogues: Ap.,
Chrm., Cri., Euphr., H.
Mi., Ion, La., Pit.-, Euthd,
Gig., H. Ma., Lys.,
Menex., Rep. I.
c. 387-c. 367: composes
middle dialogues: Meno,
Cra., Phd., Smp., Rep. II-
X, Phdr., Prm., Tht.
Other events
431-404: Pelopponesian
War
399: death of Socrates
384: birth of Aristotle
c- 365-347: composes
late dialogues: TL, Criti.,
Sph., Pol, Phil, Laws
367: Aristotle joins
Academy
387: first visit to Sicily;
makes contact with
Pythagorean
philosophers; founds
Academy upon his return
to Athens
367-365: second visit to
Sicily, upon death of
Dionysius I of Syracuse;
involvement in
Syracusan politics,
described in Seventh
Letter
361: third visit to Sicily,
described in Seventh
Letter
347: dies
*For further information, see Chapter i, notes i, 3, 24, and 25.
+For further information, see Chapter i, notes 16-18, 20, 21, 25, 39, 57, and 61.
X l l
ABBREVIATIONS
I. ANCIENT AUTHORS
ARISTOPHANES
Acharn. Acharnians
ARISTOTLE
Ath. Pol. Constitution of the Athenians
De An. De Anima
Met. Metaphysics
N.E. Nicomachean Ethics
Poet. Poetics
Soph. El. De Sophisticis Elenchis
Top. Topics
ISOCRATES
Antid. Antidosis
Panath. Panathenaicus
OLYMPIODORUS
Prol. Anonymous Prolegomena to the Philosophy of
Plato
PLATO
Ale.
Ap.
Chrm.
Cleit.
Cra.
Cri.
Criti.
Alcibiades
Apology
Charmides
Cleitophon
Cratylus
Crito
Critias
Xlll
XIV Abbreviations
Epin.
Epist.
Euphr.
Euthd.
Grg.
H. Ma.
H. Mi.
La.
Lys.
Menex.
Phd.
Phdr.
Phil.
Pol.
Prm.
Pit.
Rep.
Smp.
Sph.
Theag.
Tht.
Ti.
Epinomis
Epistles (Letters)
Euthyphro
Euthydemus
Gorgias
Hippias Major
Hippias Minor
Laches
Lysis
Menexenus
Phaedo
Phaedrus
Philebus
Politicus (Statesman)
Parmenides
Protagoras
Republic
Symposium
Sophist
Theages
Theaetetus
Timaeus
SEXTUS EMPIRICUS
A.M. Adversus Mathematicos
II. MODERN TEXTS
D.K.
O.C.T.
H. Diels and W. Kranz, Die Fragmente der
Vorsokratiker, Seventh edition, 1954
Oxford Classical Texts
Cambridge Companions Online © Cambridge University Press, 2006
RICHARD KRAUT
1 Introduction to the study of
Plato
Plato (427-347 B.C.)1 stands at the head of our philosophical tradi-
tion, being the first Western thinker to produce a body of writing
that touches upon the wide range of topics that are still discussed by
philosophers today under such headings as metaphysics, epistemol-
ogy, ethics, political theory, language, art, love, mathematics, sci-
ence, and religion. He may in this sense be said to have invented
philosophy as a distinct subject, for although all of these topics were,
of course, discussed by his intellectual predecessors and contempo-
raries, he was the first to bring them together by giving them a
unitary treatment. He conceives of philosophy as a discipline with a
distinctive intellectual method, and he makes radical claims for its
position in human life and the political community. Because philoso-
phy scrutinizes assumptions that other studies merely take for
granted, it alone can give us genuine understanding; since it discov-
ers a realm of objects inaccessible to the senses and yields an orga-
nized system of truths that go far beyond and in some cases under-
mine common sense, it should lead to a transformation in the way
we live our lives and arrange our political affairs. It is an autono-
mous subject and not the instrument of any other discipline, power,
or creed; on the contrary, because it alone can grasp what is most
important in human life, all other human endeavors should be subor-
dinate to it.2
This conception of philosophy and the substantive philosophical
theories that support it were controversial from the very start; al-
I am most grateful to Terence Irwin, Constance Meinwald, and Ian Mueller for their
helpful comments on an earlier draft of this essay.
Cambridge Companions Online © Cambridge University Press, 2006
2 THE CAMBRIDGE COMPANION TO PLATO
though there have been long periods during which some form of
Platonism flourished,* there have always been at the same time
various forms of opposition to Plato's astonishingly ambitious con-
ception of the subject.* For this reason he can be considered not only
the originator of philosophy but the most controversial figure in its
historical development. For one cannot argue that philosophy must
limit its ambitions without understanding the almost limitless
hopes that gave birth to the subject and explaining why these - all of
them or some - are misguided or unachievable. If we are forced to
retreat from his ideal of a comprehensive and unitary understanding
that transforms our lives and society, we must decide what alterna-
tive intellectual goal to put in its place. For this reason, Plato pro-
vides us with an invaluable test case and standard of comparison:
Our conception of what philosophy should be (and whether there
should be any such thing) must be developed in agreement with or
opposition to alternatives provided by the history of the subject, and
so inevitably we must ask whether the ambitions of the subject's
inventor are worthy and capable of fulfillment.
Although Plato invented philosophy as a unified and comprehen-
sive discipline, he of course could not have created it from nothing,
and so to understand how he arrived at his views we must take
account of the currents of his time. His attitudes toward political
developments in Athens and Sparta and his reaction to the intellec-
tual issues raised by the science, speculation, and poetry of the fifth
and fourth centuries decisively shaped his philosophical develop-
ment. The sophistic movement, the mathematical work of the Py-
thagoreans, the theory of flux advocated by Heraclitus and Cratylus,
the unchanging and unitary being postulated by Parmenides - all of
these played an important role in his thinking. 5 But the intellectual
influence that was paramount was Socrates, a man who wrote noth-
ing but whose personality and ideas were so powerful that no one
who came into contact with him could react to him with indiffer-
ence. For Socrates, to philosophize was to engage in adversarial con-
versation about how one's life should be lived; because the ideas he
expressed and the questions he raised were perceived as threatening,
he was tried and convicted on the charge of refusing to recognize the
gods of the city, introducing other new divinities, and corrupting the
young.6 While Socrates was alive, Plato was one of many young
people who fell under his spell, and so great was his influence that
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 3
Plato made him the central figure in many of his works, most or all
of which were composed after the death of Socrates in 399 B.C.?
Plato's writings are almost without exception in dialogue form,8 and
frequently the figure who takes the leading role in these conversa-
tions is Socrates.9 Plato did not write a part for himself in these
dialogues; rather, when they advance philosophical positions, it is
often the character named "Socrates" who expounds them.10 And so
newcomers to these works naturally raise the question how any
distinction can be made between the philosophy of Socrates and that
of Plato. How can we distinguish them, since in many dialogues
they have the same lines, the former doing all the talking, the latter
all the writing? Could we not say with equal justice that it was
Socrates (and not just Plato) who invented philosophy?
We could not, for Plato's works themselves provide us with good
evidence that Socrates focused his investigative skills on one ques-
tion only - How should one live one's life? - and was not similarly
preoccupied with the broader range of issues that absorbed their au-
thor. We have in Plato's Apology an account of the speech Socrates
gave in his defense, and he says here that although his whole life has
been devoted to the discussion of virtue, he has, despite his best
efforts, not been able to acquire any wisdom about it - except for the
wisdom that consists in knowing that he does not know. Knowledge
of such matters, Socrates thinks, is possessed by the gods alone; the
best we humans can do is to imitate his own example and recognize
the severe limitations in our moral understanding. This profession of
ignorance is a feature of several other Platonic dialogues: In the
Laches, the Charmides, the Hippias Major, and the Euthyphro, for
example, Socrates searches for an understanding of virtue and moral-
ity, but each dialogue ends with a confession that such understanding
still eludes him. By contrast, when we turn to the Republic, we find
the interlocutor called "Socrates" giving definitions of justice, cour-
age, temperance, and wisdom; and in addition he puts forward an
ambitious program of study, ranging over arithmetic, geometry, har-
monics, and astronomy, that will take us away from the unreal world
of sensible objects and eventually culminate in understanding the
Form of the Good and the unification of all branches of knowledge.
How can Socrates be so opposed to himself: a seeker who professes
ignorance about the one subject that absorbs him - the human
good - and yet (in the Republic and elsewhere) a confident theoreti-
Cambridge Companions Online © Cambridge University Press, 2006
4 THE CAMBRIDGE COMPANION TO PLATO
cian who speculates at length not only about morality but also about
knowledge, reality, politics, and the human soul? The most plausi-
ble answer, one that is now widely accepted by many scholars, is
this: In the Apology and in several other works that search for ethi-
cal definitions but show no deep interest in mathematics and make
no inquiry into metaphysics, we have a portrait of the historical
Socrates,11 but then, as Plato continued to develop in his thinking,
he retained Socrates as the main interlocutor of his dialogues, even
though the doctrines of this more ambitious ''Socrates" go well be-
yond anything dreamt of by that philosopher.12 And this interpreta-
tion of Plato's development accords with the distinction Aristotle
makes between the real Socrates and the Socrates who is a mouth-
piece for Plato: the former, he says, professed ignorance and inquired
about ethical matters but not "the whole of nature",-^ to the latter
he attributes no such limitations, but instead regards him as a
thinker who speculated about a wide range of issues and fell into
utter confusion when he posited a realm of separately existing
Forms and made the Form of the Good central to ethical theory.z*
Evidently, Aristotle reads the Republic as a presentation of the phi-
losophy of Plato and not of Socrates.^ We can assume, then, that for
some time Plato continued to use "Socrates" as the name of his
principal interlocutor because he wanted to emphasize the continu-
ity between himself and his teacher. Socrates gave to Plato the funda-
mental idea that it is vital to our well-being to discover the single
uniting factor in our diverse use of moral terms; and Plato also
inherited from Socrates the method of seeking the truth by exposing
our beliefs to the systematic cross-examination of interlocutors.
When Plato used that method to make the discoveries that Socrates
also sought to make, he paid tribute to his teacher by letting him
continue in his role as the main interlocutor.
This way of distinguishing between the philosophies of Socrates
and Plato has been given further support by studies of Plato's style of
composition that have been undertaken since the nineteenth cen-
tury.16 There has now emerged a broad consensus that we can say, at
least in many cases, which of Plato's works were written during
which periods of his life; for it is widely accepted that he wrote the
Laws in his later years,x? and we can determine which dialogues are
stylistically closest and which farthest from this late work. And so,
partly because of these stylistic studies, and partly because of Aris-
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 5
totle's distinction between Socrates and Plato, it has now become
common to divide Plato's writings into three periods: early, middle,
and late.18
The early dialogues are the ones in which he is most fully under
the influence of Socrates (hence these are often called Socratic dia-
logues), and among them are the works which unsuccessfully seek
definitions of moral properties. During this period, Plato wrote the
Apology, Charmides, Crito, Euthyphro, Gorgias, Hippias Minor,
Ion, Laches, and Protagoras.1* These have been listed in alphabetical
order, for although there may be good reason for saying of some of
them that they were written later than others in this group, scholars
are very far from a consensus about such issues. It would be safe to
say, however, that the Gorgias is one of the latest in this group - and
probably the latest - for it contains a number of features that link it
with dialogues that do not belong to this early period.20 Other works
besides those just mentioned are widely accepted as early, but since
they have a greater stylistic similarity to the middle dialogues than
do the ones listed above, there is some basis for thinking that, like
the Gorgias, they were composed after the earliest of the early dia-
logues, but prior to the middle dialogues. These are (in alphabetical
order) the Euthydemus, Hippias Major, Lysis, Menexenus, and Re-
public Book I.21
Although many of these works portray Socrates as someone who
raises questions that neither he nor his interlocutors are able to
answer, it would be a serious mistake to regard him as a purely
negative thinker who had no convictions of his own. On the con-
trary, he passionately defends a number of theses that are radically at
odds with the common sense of his time (and ours). For example, he
holds that human well-being does not consist in wealth, power, or
fame, but in virtue; that so long as one remains a good person one is
immune to misfortune; that to possess the virtues is to have intellec-
tual mastery over a distinct subject matter,- that this mastery can be
acquired only by means of a successful investigation of what the
virtues are; and that if one leaves these questions unexamined, one's
life is not worth living.22 Although Plato no doubt accepted these
doctrines when he was under the spell of Socrates and wrote his
earliest dialogues, he was eventually to modify them in important
ways. For example, one of his most significant departures is his
belief that Socrates had overlooked a nonrational aspect of human
Cambridge Companions Online © Cambridge University Press, 2006
6 THE CAMBRIDGE COMPANION TO PLATO
motivation; as he argues in the Republic, a training in virtue in-
volves both an appeal to reason and an education of the emotions
and appetites, and is not a purely intellectual matter, as Socrates had
thought.
But although Plato was eventually to disagree with some of the
views of his teacher, the greatest contrast between them, as I have
been emphasizing, lies in the different scope of their intellectual
interests: Socrates does not conduct investigations into matters out-
side of ethics, whereas Plato explores in detail a much wider range of
issues. We can see this broadening of intellectual interests when we
turn to the Meno, for here for the first time the interlocutor named
"Socrates" devotes considerable attention to an issue outside the
realm of moral philosophy; although he begins with a typically So-
cratic question - What is virtue? — and finds no adequate answer, he
soon faces an unprecedented question about the legitimacy of his
method of inquiry, a question that challenges our ability to move
from a state of ignorance and acquire knowledge. "Socrates" re-
sponds to this challenge by proposing a radical theory of knowledge
according to which the human soul is born with the ability to recol-
lect what it once learned in a previous existence,- and he defends this
theory by conducting an experiment in which it is shown that a
slave can make significant progress toward an understanding of ge-
ometry, if he is asked the right sorts of questions.^ It is widely
believed that when this doctrine of recollection is introduced, Socra-
tes the interlocutor is entering a terrain that had not been investi-
gated by the historical Socrates, a terrain that was explored by Plato
for the first time in his middle period. This shift is signaled by the
fact that an experiment about the learning of a geometrical theorem
is chosen to support the doctrine of recollection, for Plato's deep
interest in mathematics is evident throughout the middle and late
dialogues, but is not yet present in his early works.2* The Meno is
therefore commonly regarded as a "transitional" dialogue, one that
comes between Plato's early and middle periods, and contains ele-
ments of both. It is here that we can most clearly see the transforma-
tion of "Socrates" into Plato. It is thought to have been written at
some point between 386 and 382, when Plato was in his early or
middle forties and Socrates had been dead for at least thirteen
years.25
The separation of body and soul is a theme more fully investigated
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 7
in the Phaedo, and although this dialogue portrays the last conversa-
tion and death of Socrates - and therefore forms a dramatic unity
with the Euthyphro (Socrates on his way to court), Apology (Socrates
on trial), and Crito (Socrates refusing to escape from prison) - it is
widely agreed that unlike these works it belongs to Plato's middle
period. It refers (72e-73a) to the conversation held in the Meno and
therefore can safely be assumed to have been written after it; and in
the Phaedo we see how Plato's interest in questions Socrates had left
unexplored or undeveloped continues to grow. Here for the first time
in Plato's writings we find a decisive declaration and argument26
that there exists a realm of objects utterly different from the ones
with which we are familiar: These objects are changeless (78d),2?
revealed to us by thought rather than sensation (79a), different from
both body and soul (79b-c), and everlasting (79d).28 Equality2* is one
of Plato's foremost examples of such an object; and his attention to
this mathematical property, like his discussion of geometrical learn-
ing in the Meno, reflects the fact that he is moving beyond the
exclusively ethical explorations of Socrates. He argues (74b-d) that
Equality itself cannot be identical to equal sticks or any other observ-
ably equal objects; for it is possible to make a mistake about two
equal objects and to believe them unequal, but no one can make a
comparable mistake about Equality itself, and take it to be unequal.
The equal sticks are in some way inferior to Equality, though when
Plato makes this assertion (74d-e) he does not say explicitly what it
is about them that is defective, and why Equality does not share this
deficiency.3° The defectively equal sticks "participate in" Equality
and are therefore "called after" that Form, but they are not Equality
itself. 3 *
The singular Greek terms Plato often uses to designate the new
kind of object he discovered - eidos and idea - are conventionally
translated "Form" and "Idea," though the latter term must be used
with caution, for he clearly thinks that these entities are not
thoughts or any other creations of a mind;^2 they are by nature
uncreated, and their existence is not dependent on being known or
thought. He takes these to be the objects that we are trying to under-
stand when we ask the sorts of questions Socrates asked, and in this
way he sees his own philosophy as continuous with that of Socrates.
Although Socrates asked such questions as What is virtue? - and,
according to Aristotle, was the first to engage in this sort of
Cambridge Companions Online © Cambridge University Press, 2006
8 THE CAMBRIDGE COMPANION TO PLATO
inquiry" - he showed no interest in a further series of second-order
questions that can be asked about the objects of the very question he
was asking: Is what we are looking for something that exists indepen-
dently of human beings? Is it something that can be detected by
means of the senses? Can it change or perish? How is it possible for
us to learn about it? What is its relationship with people and acts
that are rightly considered virtuous but are not identical to what
virtue is? How are these objects of thought related to each other?
Plato's attempt to answer these questions by positing the exis-
tence of a separate realm of abstract objectss* called "Forms" and
exploring their special properties is often called his theory of Forms,
but this phrase should not mislead us into assuming that after writ-
ing the early dialogues he quickly developed a dogmatic system that
gave decisive and unambiguous answers to all of the important ques-
tions that can be asked about the Forms. On the contrary, it is more
reasonable to take him to be developing and perhaps even revising
the theory as he continues to explore the nature of these objects.35
For example, Plato makes no attempt in the Phaedo to say which
Forms there are; he clearly believes that there are Forms of Equality,
Beauty, Goodness, Justice, and Piety; but we are not told how to
decide what else is on this list.*6 By the time he writes the Republic
(a middle dialogue), he gives a general, but imprecise, answer: When-
ever a name is applied to many different things, there is a Form
corresponding to the name (596a).^ And he posits the existence of
Forms corresponding to commonplace artifacts (beds, for example),
no less than Forms for moral and mathematical properties. Does this
mean that whenever we invent a word and apply it to a plurality of
objects, we must posit a corresponding Form, even when we have no
justification for introducing the new word and the objects to which
it applies ought not to be classifed together? In the Statesman (a
dialogue belonging to the late period) Plato makes it clear that in his
opinion there are no Forms corresponding to names that are not
supported by a justified classification of reality into groups. For ex-
ample, he says, it is arbitrary to divide the peoples of the world into
tv& groups - Greeks and non-Greeks - because there is nothing that
unifies the latter group into a genuine whole (262c-e). And so there
is no Form corresponding to "barbarian," even though the Greek
term - which simply designated anyone other than a Greek - was a
name of long standing and familiarity. And so we can see some
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 9
progression or development in Plato's articulation of the theory of
Forms: At first he does not try to delimit their range,- then he gives
their range in a general way; and finally he refines or modifies his
criterion.*8
11
In the Meno and Phaedo Plato's thought had not yet fully matured
into a comprehensive philosophy, and had he died at this early stage in
his development he would have been recognized as a thinker who had
to some degree broken away from Socrates, but he would merely be a
minor star in the philosophical firmament. His claim to greatness
rests principally on later works of the middle period - the Sympo-
sium, Republic, Parmenides, Theaetetus, and Phaedrus (to list them
in one plausible chronological order);39 and on the late dialogues-
the Timaeus (to which the Critias is a brief sequel), Sophist, States-
man (Politicus), Philebus, and LawsA° And although the Phaedo is
continuous with the later works of Plato's middle period in its deci-
sive positing of a realm of Forms, there are also important ways in
which Plato's later work in the middle period moves away from his
initial formulations in the Phaedo. In this dialogue, all of the impedi-
ments to philosophical progress are located in the body's incessant
demands; and because the body blocks any attempts we make to
achieve a pure understanding of the Forms, the strongest wish of the
true philosopher is to die and be freed from corporeal imprisonment
(65e-67b). By contrast, in the Republic Plato argues that the soul is
divided into three components, and this allows him to locate the
impediments to philosophical progress in the soul itself, rather than
the body. More important, the longing for death and the pessimism
about our prospects for understanding that pervade the Phaedo give
way in the Republic to a confidence that if one pursues the right
course of study one can, while one is still alive, achieve a full under-
standing of the realm of the Forms, including the Good, the Form that
Plato declares to be of greatest importance.*1 Those who grasp this
Form will not only attain supreme happiness for themselves, but
their contact with an otherwordly realm, so far from making them
unfit for political life, will enable them to be of immense value to
other members of the community.*2 Perhaps the most important dif-
ference between the Phaedo and the Republic is that although in the
Cambridge Companions Online © Cambridge University Press, 2006
IO THE CAMBRIDGE COMPANION TO PLATO
later dialogue Plato once again argues (in Book X) for the immortality
of the soul, it is principally devoted to the project of demonstrating
the great advantage of leading a philosophical and fully virtuous life
quite apart from the posthumous rewards that those who lead such a
life will eventually receive in another world. Even though the soul is
immortal, we can dispense with this thesis and still show that for the
sake of one's own happiness there is overwhelming reason to lead a
virtuous and philosophical life.
In the Republic we see how Plato intends to solve the problems
that had preoccupied Socrates. In Book I, we have a typically So-
cratic attempt to define a central moral property - justice - and here
too it ends in failure. Although Socrates argues that justice and self-
interest coincide, his interlocutors ask him to make a fresh attempt
to defend this thesis in Book II; and this reflects the fact that Plato is
about to use new materials to defend one of the principal ethical
beliefs of his teacher. What follows is a unified metaphysical, episte-
mological, ethical, political, and psychological theory that goes far
beyond the doctrines of the early dialogues.** The Republic is in one
sense the centerpiece of Plato's philosophy, for no other single work
of his attempts to treat all of these topics so fully,- but at the same
time it gives us only a small glimpse of his mature thought, because
nearly all of the dialogue's components are developed further and in
many cases treated more fully in works that he wrote afterward.
Metaphysics is more thoroughly explored in the Parmenides and
Sophists epistemological issues are treated with greater profundity
and thoroughness in the Theaetetus; our prospects for understand-
ing the sensible world through the Forms are most fully explored in
the Timaeus; the question of how best to lead one's life is reopened
in the Philebus;^ feasible institutions for a good political commu-
nity are most fully described in the Laws; and his thoughts about
human psychology continue to be developed in the Phaedrus,
Philebus, and Laws. So the Republic gives us only a partial, though
an indispensable, picture of Plato's philosophy.
It is in Book VII of the Republic that we find Plato's well-known
and powerful image of the human condition: Ordinary human be-
ings, untouched by philosophical education, are likened to prisoners
in a cave who are forced to gaze on shadows created by artificial light
and cast by artifacts paraded by unseen manipulators (5i4a-5i9a).
Their conception of what exists and of what is worth having is so
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 11
severely limited and the deception by which they are victimized is
so systematic that they cannot even recognize that they are con-
fined, and would not immediately regard an interruption in their
routine ways of thought as a liberation. Here Plato is of course think-
ing of the psychological resistance Socrates encountered to his ques-
tioning (517a); but he is also making a far more audacious claim,
because he means somehow to downgrade the reality of the ordinary
world of sensible objects. The shadows cast on the wall of the cave
are less real than the objects of which they are images (5 isa-e); and
in the same way, when the prisoners make progress, leave the cave,
and learn to understand the Forms, they recognize the existence of a
realm of objects that are more real than anything they saw in the
cave (518c). Similarly, in Book X of the Republic, Plato distinguishes
three types of thing to which the word "bed" can be given - a paint-
ing of a bed, a bed created by a carpenter, and the Form - and he
holds that they constitute a series of increasing reality. The painter
does not make a real bed, but only an image of a bed, and the product
of the carpenter is not completely real either. It is only the Form that
is really real (596e-597a).*6
It would be a mistake to think that in these passages Plato is
trying to cast doubt on the existence of the sensible world. After all,
in saying that the painter's image of a bed is not a true bed, he is not
expressing doubts about the existence of the painting; rather, he is
trying to express the point that the painter's image is in some way
derivative from or dependent on the functional object he is represent-
ing. It is this same relationship of dependency that he thinks exists
between the visible bed of the carpenter and the Form. When we
point to an image in a painting and call it a bed, what we say is in a
way correct, if our claim is taken in the right way. What makes it
something that is properly called a bed is its having the right relation-
ship to the functional bed, even though it has radically different
properties from the latter. In this case, the right relationship consists
in visual similarity. In the same way, Plato is suggesting, there is a
dependency of ordinary sensible objects like beds upon the Forms
after which they are named. When we point to the carpenter's cre-
ation and call it a bed, our assertion is acceptable only if it is con-
strued in the right way. That physical object is not precisely what it
is to be a bed - and if this is what we mean when we call it a bed
then we are confused and not simply mistaken. Rather, what we say
Cambridge Companions Online © Cambridge University Press, 2006
12 THE CAMBRIDGE COMPANION TO PLATO
when we call it a bed is acceptable only if the sensible object has the
right relationship to the Form, even though the Form has radically
different properties from physical beds. In this case, of course, the
relationship is not one of visual resemblance, since nothing can be
an image of a Form in that way. Instead, the point is that whether an
observable object can properly be called a bed depends on what it is
for something to be a bed; it depends, in other words, on the Form,
since this is the object one succeeds in understanding when one has
a proper conception of what a bed is. So, when Plato says that Forms
are completely real, he means that they are at the top of a scale that
ranks objects according to their degree of ontological dependency:
Just as images of sensible objects receive their names because of
their relationship with something with radically different proper-
ties, so the sensible objects also receive their names because of the
kind of relationship they have to Forms; but the Forms are not in
turn dependent for their names on anything else.*?
The analogy of the cave also brings out a pervasive feature of
Plato's political philosophy: Those who are limited in their concep-
tion of what exists and what is worthwhile are not the best judges of
their own interests, and they can be expected to resist initial efforts
to improve their lives. What makes a political system a good one is
not the consent of the governed, for if false values are prevalent
people may willingly accept only those political systems that per-
petuate their confinement. A good political community, Plato as-
sumes, must be one that promotes the well-being of all the citizens;
and if the citizens fail to understand where their own good lies, then
it is the proper task of political leaders to educate them. Although
Plato is therefore in favor of giving extraordinary powers to rulers
who themselves have a philosophical understanding of the human
good, he is not unconcerned about the possibility that such power
might be misused or arouse resentment. It is partly for this reason
that private wealth and the family are abolished in the ruling class:
These powerful sources of political corruption and favoritism must
be eliminated in order to give reasonable assurances to those who
are ruled that they will not be exploited by those who are more
powerful. One of the driving forces behind Plato's depiction of an
ideal society is that in such a society there must be a deep feeling of
community among all the citizens, in spite of the fact that they
cannot all share an equal understanding of the human good. No one
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 13
individual or group is to be favored at the expense of any other. The
ideal city is not one designed for the maximal happiness of the
philosophers or any other group; instead, institutions must be de-
signed so that there is a fair pattern of benefits for all (4i9a-42ia).
Although there is a great deal in the political philosophy of the
Republic that we rightly reject/8 it also contains elements that
many still find appealing. Plato has an attractive vision of a certain
kind of community: one in which no one is favored by traditional
privileges of wealth, birth, or gender/9 one in which no one's well-
being is ignored and no one is allowed to be indifferent to others; one
in which every member of the community leads a life that is to some
degree objectively worthwhile.*0
Another remarkable feature of the Republic is that Plato takes
there to be a kinship between the kind of community that human
beings are capable of forming and the more rarified community he
thinks exists among the Forms. The Forms are not a mere aggregate
of abstract objects, but are in some way connected to each other:
They form an ordered kosmos (500c) and therefore each must be
studied not in isolation from others but as a member of a unified
whole. What we must strive for in our souls and as members of a
political order - the unification of diverse elements into a harmo-
nious whole - is something that the Forms already possess by their
very nature. But Plato devotes very little attention in the Republic
to the articulation of the structure possessed by the Forms; what he
insists upon is that there is one Form that is central to the being and
knowability of all the others: the Form of the Good (5O5a-5O9c).
The project of studying the relationships that exist among the Forms
is taken up by a number of dialogues written after the Republic. In
the Phaedrus he assigns to the dialectician the tasks of finding unity
in a diversity of Forms and diversity in a unity, and uses his concep-
tion of love as one kind of divine madness as an illustration of such a
structure. In the Parmenides, there is a complex treatment of the
relationship between Unity and such other Forms as Sameness and
Difference, Motion and Rest, Limited and Unlimited. And the explo-
ration of such relationships is a recurring feature of much of Plato's
late work, playing an especially important role in the Statesman,
Sophist, and Philebus. If Plato equates Goodness and Unity - and
there are some reasons to believe that he does*1 - then the elaborate
treatment of Unity found in the Parmenides might be read as a
Cambridge Companions Online © Cambridge University Press, 2006
14 THE CAMBRIDGE COMPANION TO PLATO
continuation of Plato's preoccupation with the Good. And in the
Timaeus the entire sensible world is viewed as an expression of the
goodness of a divine craftsman who looks to the pattern of the Forms
and shapes the recalcitrant and disorderly material at his disposal
into a good (29a—30b) though far from perfect (46d-e) series of struc-
tures. So some of the metaphysical ideas briefly suggested in the
Republic — that the Forms constitute a structured whole, that the
Good is foremost among them, that the goodness of a complex group
of objects consists in their unification - continue to guide Plato's
development in his later works.
in
Thus far, we have seen how Plato's thought underwent a transition
in the Meno, and, beginning with the Phaedo, developed into a com-
prehensive philosophy based on the theory of Forms. And I have just
been discussing some themes that connect the Republic and dia-
logues that postdate it. But it should not be thought that the course
of Plato's development after the Republic is a matter about which
scholars have achieved consensus. On the contrary, even among
scholars who believe that Plato continued to develop throughout his
life, there are deep divisions about the overall shape of the changes
he made. At one extreme, some hold that he entirely abandoned the
theory of Forms, having seen that it is not necessary to posit such
objects.5* A somewhat less radical proposal is that Plato faced an
emergency as he became increasingly aware of deficiencies in the
theory of Forms, but that he did not alter the theory because he
could not locate the source of his difficulties.53 Others believe that
while he did retain his belief in the existence of these abstract ob-
jects, he recognized that his conception of them during his middle
period went badly astray, and he therefore developed a new under-
standing of their nature.54 Still another view is that Plato changed
the theory of Forms not so much by rejecting certain components of
it as by adding new elements that make his conception of Forms
more suble and less vulnerable to misunderstandings and objec-
tions.55 In addition to these disagreements about whether Plato
changed, and how, there are related disagreements about the chrono-
logical sequence of his works.
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 15
To understand these controversies, it is best to look more care-
fully at the division of Plato's works into three phases - early, mid-
dle, and late - and to be more specific about which works are gener-
ally thought to belong to the latter two groups. As I said earlier,
studies of Plato's style initiated in the nineteenth century and con-
tinuing to the present have begun with the point, about which there
is universal consensus, that the Laws is a late work. A good deal of
cumulative evidence has pointed to the conclusion that there are
five other works that are closely related to the Laws as measured by
a variety of stylistic features. These are (to list them alphabetically)
the Critias, Philebus, Sophist, Statesman, and Timaeus. Within this
group, we can confidently say that the Statesman was written after
the Sophist, since it refers back to it several times,*6 and that the
Timaeus precedes the Critias, since the latter's depiction of the lost
island of Atlantis is obviously a sequel to the account of this subject
initiated in the former work. Any further attempt to order their
composition is more conjectural; for example, though Diogenes
Laertius says that the Laws is Plato's last work, the Critias is obvi-
ously incomplete and is therefore a competing candidate for this
position. But stylistic studies suggest that they were composed in
the following order: Timaeus, Critias, Sophist, Statesman, Philebus,
Laws."
The greatest source of contention regarding the late dialogues is
the Timaeus, for it is not universally agreed that it does belong to his
late period. It was argued by G. E. L. Owen some forty years ago that
the stylometric studies that placed this dialogue in the late period
were mistaken; furthermore, Owen argued, the philosophical con-
tent of the dialogue and any coherent account of Plato's develop-
ment requires us to place the Timaeus in the middle period, after the
composition of the Republic, and prior to the major shift in Plato's
thought that Owen sought to locate in the Parmenides and Theae-
tetus.^ It might be thought that Owen's thesis is confirmed by the
opening page of the Timaeus, for it refers to a conversation, held on
the previous day, in which Socrates was describing the institutions
of the best city, and those institutions are the very ones we find in
the Republic. From this it might be inferred that the Timaeus was
the dialogue Plato composed immediately after completing the Re-
public, but such an inference would be unsound, since Plato may
Cambridge Companions Online © Cambridge University Press, 2006
16 THE CAMBRIDGE COMPANION TO PLATO
have had literary or dramatic reasons for drawing a close connection
between these two dialogues. Their dramatic proximity is not a
reliable guide to the closeness of their dates of composition.
Before giving further consideration to the place of the Timaeus, it
will be helpful to call attention to a conspicuous and puzzling fea-
ture of five of the six dialogues that are often classified as late - the
Timaeus, Critias, Sophist, Statesman, Philebus, and Laws: With the
exception of the Philebus, Socrates either plays a minor role (in the
Timaeus, Critias, Sophist, and Statesman) or is completely absent
(in the Laws). Now, it might be thought that this feature can be of no
significance precisely because Socrates is restored to a major role in
the Philebus. But there is an explanation for Plato's deviation in this
one case from a general rule he had decided to adopt: The Philebus is
devoted to an examination of the place of pleasure in the best human
life, and so it is understandable that Plato should bring back his
teacher as the dialogue's main interlocutor. It is unlikely, then, that
Plato's decision about whether to let Socrates be the chief interlocu-
tor is unmotivated and arbitrary; he has a special reason for giving
him the leading role in the Philebus. And so it is plausible to assume
that he also has a good reason for giving Socrates a small role or none
at all in these other late dialogues. But what is his reason? One
hypothesis that leaps to mind is that Plato is consciously rejecting at
least some of the major tenets of his middle period, and that he
therefore signals this divergence by giving Socrates a smaller role to
play. 59 But of course this explanation is merely an initial hypothesis
and must be substantiated by a careful examination of the content of
these later works.
It is important to be aware of the fact that there are two other
dialogues outside this group of six that also give Socrates an unusual
role to play: the Parmenides and Theaetetus. It is widely agreed that
these belong to the middle period, but it is extremely likely that
they were written after the Republic, because each subjects to criti-
cal scrutiny a doctrine that is regarded in the Republic as unprob-
lematic: In the first part of the Parmenides (i26a-i35d) the theory of
Forms is exposed to criticism, and in the Theaetetus we find an
unsuccessful search for an adequate definition of knowledge and a
critical discussion of the conception of knowledge taken for granted
in the Republic and earlier works.60 In the Parmenides, the difficul-
ties in the theory of Forms are pressed by Parmenides, and Socrates
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 17
is too young to have developed successful answers (i35d); in the
Theaetetus, Socrates takes a leading role in asking questions and
raising difficulties, but, like a barren midwife who can only help
others give birth, he says he cannot produce positive views on his
own (i48e-i5id). Both of these roles-the object of unanswered
criticism, the intellectual midwife - are new departures in the dra-
matic structure of the middle dialogues.61
Now, the fact that these two critical works give an unusual role to
Socrates is easily explained: If, for the first time since he began to
develop the metaphysics and epistemology associated with Forms,
Plato is subjecting that theory to hard questioning, then he would
naturally assign a more passive role to the chief expounder of that
theory. And if we accept this hypothesis, then we can also propose a
conservative explanation of why Socrates continues to play so small
a role throughout the remainder of Plato's work, with the exception
of the Philebus: having become accustomed in the Parmenides to
writing a large part for someone other than Socrates, and having
continued in the Theaetetus to use Socrates in a role other than that
of a mouthpiece for positive doctrine, perhaps Plato sees no reason
to revert to his earlier habit of making his teacher the main ex-
pounder of doctrine, except when the subject matter specifically
calls for such a change. In other words, there may be no more in-
volved in Plato's frequent practice of minimizing the role of Socrates
in the late dialogues than a certain kind of conservatism. Having
good reason for altering the role of Socrates in the Parmenides and
Theaetetus, Plato simply saw no reason (except in the Philebus) to
make yet another change.
If we find that the late dialogues contain nothing that rejects posi-
tions adopted in the middle works, and that Plato has survived the
self-criticism of the Parmenides and the Theaetetus without alter-
ing what he had earlier believed, then we can account for the smaller
role of Socrates in this way. But it is also open to us to combine
several ways of accounting for this change in the conversational
roles of the later works. If we find that Plato altered his views in
moderate rather than radical ways after writing the Parmenides and
Theaetetus, then, although these moderate changes might not on
their own have been a sufficient reason for demoting Socrates as a
speaker, they may have been large enough to discourage him from
reverting to the earlier pattern of Socratic leadership in conversa-
Cambridge Companions Online © Cambridge University Press, 2006
18 THE CAMBRIDGE COMPANION TO PLATO
tion. So the change Plato makes in the role assigned to Socrates by
itself tells us nothing about the pattern of his own intellectual devel-
opment. To see whether Plato's later works are continuous with his
middle dialogues or whether there are mild or sharp breaks, there is
no substitute for looking at the actual content of these late dia-
logues; our interpretation of that content will tell us what to make
of the smaller role assigned to Socrates, and not vice versa.
And when we look at the content of the late dialogues [Timaeus,
Critias, Sophist, Statesman, Philebus, and Laws) and the two middle
dialogues that assign Socrates an unusual role (Parmenides and
Theaetetus), what do we find? A radical break, smooth continuity, or
something in between? There are really many questions here, for it is
possible that on some issues Plato saw no need for modifications
whereas on others he made significant revisions. On such questions,
scholars have been deeply divided. One of the crucial issues in this
debate concerns a series of objections presented in the opening pages
of the Parmenides against the theory of Forms. These objections re-
ceive no explicit answer in this or any other dialogue. Aristotle
thought that one of them was fatal to some ways of arguing for the
existence of Forms, and sought to avoid a similar problem for his
conception of universals.62 As expressed in the Parmenides, this ob-
jection holds that if there is a reason for positing a single Form of
Largeness then there is equally a reason for positing an unlimited
number of Forms of this same kind. The reason for positing the "first"
Form of Largeness is that whenever a number of things are large, there
must be a single Form by virtue of which they are large; but now when
we consider that Form of Largeness together with those large things,
there must be a further Form of Largeness, by virtue of which the large
things and Largeness Itself are large; the process by which we recog-
nize "new" Forms of Largeness can be repeated indefinitely, and so
there is no one Form of Largeness, contrary to our initial hypothesis.
It has seemed to many scholars that this argument depends crucially
on the highly questionable assumption that a Form of Largeness is
itself a large thing. But does Plato's theory of Forms, as developed in
such middle dialogues as the Phaedo, Symposium, Republic, and
Phaedrus, commit him to such an assumption? Is this the assump-
tion that he meant to single out for further examination in the
Parmenides7. Did he modify his theory of Forms, or did he perhaps
entirely abandon his belief in the existence of these objects, in the
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 19
light of the objections recorded in the Parmenides! These are among
the main questions scholars have raised in their debates about the
pattern of Plato's later development.6*
The evidence of stylometric studies plays an especially important
role in this debate, because if their results are accepted they place the
Timaeus among the late dialogues and therefore assign it a position
after the writing of the Parmenides. In the Timaeus, Plato clearly
continues to uphold some of the views about the Forms that play a
large role in the middle dialogues. He holds that these objects alone
are changeless, and contrasts their invulnerability to alteration with
the constant fluctuation that characterizes objects in the world of
sensation; because of these radical differences, the Forms are capable
of being known, whereas objects of sensation are not.6* Furthermore,
the Forms are described in the Timaeus as paradigms6* - objects to
which the divine craftsman looks in creating the sensible world, and
to which we must look in order to acquire knowledge - and this too is
a doctrine that is central to Plato's philosophy in the middle dia-
logues.66 So, if we give serious consideration to stylometric studies
and accept the finding of so many of them that the Timaeus is among
Plato's latest works, then we must conclude that nothing in the
Parmenides or Theaetetus led him to abandon some of the central
doctrines of his middle period. Of course, this does not mean that
there are no new developments in the late dialogues; obviously there
are. And it may also be that Plato modified his conception of abstract
objects in significant ways. But if the Timaeus is late, then there is
also a good deal of continuity in Plato's thought.
Any consideration of Plato's development beyond his middle pe-
riod must also pay careful attention to the further work he did in
political and moral theory after writing the Republic. It is remark-
able that after giving, in this dialogue, a rather elaborate blueprint
for an ideal society, he took up a similar project near the end of his
life and devoted his longest work - the Laws - to the development
of a complex political system and legal code. Some of the main
doctrines of the Republic are preserved intact here: Moral education
is the principal business of the political community, and there is no
toleration for those who put forward doctrines that would under-
mine the virtue of the citizens. But there are also striking differences
between the ideal community of the Republic and the new Utopia
depicted in the Laws: No specialized training in mathematics or
Cambridge Companions Online © Cambridge University Press, 2006
2O THE CAMBRIDGE COMPANION TO PLATO
dialectic is prescribed for an elite group of citizens, and instead of
assigning total responsibility and power to one small group of deci-
sion makers, Plato widely distributes the functions of government
and establishes an elaborate system of safeguards against the abuse
of power. Although power is unevenly divided, no citizen is com-
pletely deprived of a legislative or judicial role. Does this mean that
in his later period Plato came to be less opposed to democratic ideas
than he once had been? Perhaps. But it may also be that in the Laws
he accepts limited democratic features and envisages a smaller role
for philosophers because in this work he is merely describing a
second-best political community (739a-74oa); if that is the proper
explanation, then he might have continued to believe that ideally
philosophers should have absolute control over political matters.6?
IV
Thus far, we have been focusing our attention exclusively on what
we can learn about Plato from his own writings, but something
should be added now about whether there are any other important
sources for our understanding of his philosophy. Of course, we are
remarkably fortunate to have so much from Plato's own hand; in
fact, we possess every philosophical work he ever composed, in the
form of copies made during the medieval period, which derive ulti-
mately from the original sheets of papyrus on which Plato wrote.68
(By contrast, most of Greek tragedy and comedy, and a good deal of
early Greek and Hellenistic philosophy, are lost to us.) But in addi-
tion to the abundance of material we have from Plato himself, we
also have reports from Aristotle and later philosophers of antiquity
about Plato's teaching in the Academy. The value of these reports for
our understanding of Plato is, however, a matter of considerable
debate among scholars.
Before turning to these reports, we should take note of Plato's
recognition, at Phaedrus 274b-278b, of the limitations of the writ-
ten word and his insistence upon the superiority of speech as an
instrument of teaching and learning. For some scholars believe that,
in view of Plato's low opinion of writing, it is a matter of urgency
that we try to interpret the reports we have about his oral teaching.^
He points out in this dialogue that when one discusses philosophy
with another person, one has an opportunity to respond to questions
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 21
and defend one's assertions. In addition, what one says to one person
may be different from what one says to another; and to some one
should say nothing at all - presumably because some listeners will
be less sympathetic or prepared than others, and will therefore raise
different challenges or obstacles. Written philosophy lacks this flexi-
bility; it says the same thing to everyone, and leaves the questions of
its audience unanswered [2j$c-2j6a). Furthermore, the existence of
philosophical books can lead to a deterioration of memory, if they
are used as a substitute for understanding; and they entice students
into thinking that reading by itself creates wisdom (275a-b). They
are no substitute for the give-and-take of dialogue, for this alone, and
not the mere spouting of doctrine, can give rise to understanding and
wisdom.
Of course, these assertions of the supremacy of speech and reserva-
tions about the value of philosophical writing do not lead Plato to
reject the written word completely. As we have seen, he did a great
deal of writing after the Phaedrus, and so we cannot take this dia-
logue as a farewell to the written word or a repudiationn of the value
of writing philosophy. After all, Plato does say in the Phaedrus that
writing, when properly used, can come to the aid of a memory weak-
ening with age, and can also be helpful to the students with whom
one discusses philosophy (276d). The point then, is that written
works can serve a purpose, but only so long as they are accompanied
by philosophical dialogue. It is no mystery, then, that Plato wrote
voluminously, and continued to do so even after expressing his reser-
vations about the written word. So the Phaedrus gives us no good
reason to think that Plato decided not to put into writing his sincere
views about philosophical matters; nor does it give us evidence that
he deliberately refrained from putting some of his convictions into
writing. This is a significant and controversial point, because some
scholars do believe that Plato refused to write down the most impor-
tant points of his philosophy, and that we can recover these ideas
only through reports of his oral teaching.
Strong misgivings about the writing of philosophy are also ex-
pressed briefly in the second (3i4b-c) and more fully in the seventh
(34ib-345a) of Plato's Letters - though the authenticity of these
works, it should be recalled, is a matter of controversy. In the Sev-
enth Letter, the author writes that he, Plato, is greatly annoyed
because he has heard that Dionysius, the tyrant of Syracuse, recently
Cambridge Companions Online © Cambridge University Press, 2006
22 THE CAMBRIDGE COMPANION TO PLATO
composed a work based on philosophical discussions they had had.
Plato is eager to dissociate himself from anything Dionysius may
have written, and to do so he announces his objections to putting the
most serious matters into writing. The matters he discussed with
Dionysius are ones he never has and never will commit to writing
(341b). Why not? One reason is that this written work will be of no
help to the many, and will be used by some as a substitute for
wisdom; on the other hand, the few who are capable of understand-
ing his views will be able to discover the truth without relying on a
written exposition (34id-e). Some of these misgivings about writing
correspond to those expressed in the Phaedrus, but Plato's repudia-
tion of expressing certain matters in written form seems to go be-
yond anything said in the Phaedrus. The Seventh Letter tells us that
there are certain thoughts that Plato refuses to put into writing,
whereas the Phaedrus expresses no such self-imposed limitation. In
fact, certain portions of the Seventh Letter appear to say that certain
thoughts are not to be expressed either orally or in written form,
because words themselves are matters of convention and this makes
them ill-suited instruments for grasping true being (341c, 342e-
343c). If Plato is saying that certain truths cannot or should not be
captured in language, he is again going far beyond the reservations
expressed in the Phaedrus about writing. And he is saying that we
should not look to either his oral or his written teachings for an
expression of the deepest truths.
However, if we take the Seventh Letter to be saying that Plato will
orally divulge his most important thoughts to his students, but will
not put them in writing, and if we take this work to be authentic,
then it becomes a matter of urgency to see whether we can discover
what it is that he said but refused to write. Now, it is a certainty that
Plato expressed some philosophical views that he did not put into
writing; not only is that a natural assumption to make about any
philosopher, especially one who sees so much value in dialogue, but
in addition we have a passage from Aristotle's Physics in which a
distinction is made between what Plato says about place in the
Timaeus and what he says in "so-called unwritten opinions" (IV.2
2O9bi4-i5). But significantly, Aristotle gives no special weight to
the latter; that is, he does not suggest that we should downgrade
Plato's views in the Timaeus because this is merely a written compo-
sition or that we should attach priority to the unwritten opinion of
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 23
Plato precisely because it was unwritten. In fact, Aristotle fre-
quently looks to Plato's dialogues for his information about what
Plato thought; he never suggests that because of Plato's views about
the defects of writing he communicated his deepest philosophical
thoughts only in speech and that it is therefore to the "so-called
unwritten opinions" that we must turn whenever we can. So Aris-
totle's way of treating the written works and the unwritten opinions
counts heavily against the suggestion we might have drawn from the
Seventh Letter that to understand Plato we must give greatest
weight to what he said but did not write.
Elsewhere, Aristotle attributes to Plato certain views without as-
signing them to any particular dialogue, but also without saying
explicitly that these opinions were unwritten. For example, in the
Metaphysics he says that according to Plato there are, in some way
between sensible objects and Forms, mathematical objects which
differ from sensible objects in that they are eternal and unchange-
able and from Forms in that they are many and alike (1.6 987^4-18).
In addition, he attributes to Plato the doctrine that the elements of
the Forms are the great and the small - which constitute the mate-
rial element - and unity - which is substance (987b!8-21). This lat-
ter passage is especially significant, for it indicates that in Plato's
opinion the Forms are not the most basic entities, but are in some
way derived from something else. Here we have a view that might be
called Plato's deepest thought about reality, since it posits some-
thing more basic than even the Forms.
But on what basis does Aristotle attribute this view to Plato? And
at what point in Plato's career does he think this view was adopted?
He does not answer either of these questions. And so a number of
different options are open to us:
1. We might think that Aristotle attributes this view to Plato
(whether justifiably or not) on the basis of what he reads in
some dialogue or group of dialogues. In this case, Aristotle is
not providing us with information about Plato's philosophy
that is not already available in the dialogues.?0
2. We might think that Plato did not express this view in writ-
ing, and that it was a fresh idea that did not occur to him
until very late in his philosophical career. In this case, it
could be a mistake to take anything in the dialogues to have
Cambridge Companions Online © Cambridge University Press, 2006
24 THE CAMBRIDGE COMPANION TO PLATO
been shaped by unspoken assumptions revealed to us only
by later reports of his oral teaching. Alternatively, even if
Plato's unwritten opinions occurred to him at the same time
that he was writing some of his latest work, it is possible
that he did not develop these opinions about the generation
of the Forms into a dialogue because he thought that these
new ideas were too tentative and undeveloped to merit their
preservation for future generations.?x In this case, any re-
ports we have of opinions that do not find expression in the
dialogues should be given less emphasis in our attempt to
understand the dialogues than the evidence of the dialogues
themselves.
3. We might think that Plato held this view about the genera-
tion of the Forms at an earlier point in his philosophical
career, and that he refused to write it down for reasons given
in the Seventh Letter. This third option, of course, is open to
us only if the latter work was written by Plato, and only if it
is taken to mean that he reveals his deepest convictions in
conversations with a few students and refuses to divulge
them in written form.
Which of these options is most credible is a matter of ongoing
debate among scholars. The central question in the debate is whether
reports of Plato's teaching give us greater insight into what we find in
the dialogues. If his works can be understood well enough without
recourse to these later reports about his philosophy, then the later
material may supplement our knowledge of Plato, but they should
not play a controlling role in our interpretation of the dialogues. On
the other hand, if we find that the only way we can make sense of
Plato is by letting the later reports guide our reading of the dialogues,
then of course those reports will have proven to be of great worth. At
present, it is fair to say that only a small number of scholars believe
that Plato's written works are a mystery that can be solved only by
recovering oral doctrines that he refused to commit to writing be-
cause of their great importance. But it should not be concluded that
the reports of Plato's unwritten opinions deserve no attention whatso-
ever. They may give us helpful clues about the tendency of Plato's
latest thoughts, and perhaps they can help us solve some interpretive
difficulties about his last dialogues.
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 25
There is another kind of disagreement among scholars about how
Plato's dialogues are to be read, in addition to the question of the
relationship between his writing and his oral opinions. This second
controversy stems from the fact that so much of his writing (almost
all of it, if we exclude the Letters) takes the form of a dialogue
between several interlocutors, and that Plato never gives a speaking
part to himself. A question can therefore be raised about how we can
tell, on the basis of what the interlocutors of the dialogues say, what
Plato himself believes. After all, when we read a play of Sophocles or
Euripides, we all recognize that what the characters say need not
represent the beliefs of the author. And so it is reasonable to ask why
we should make a different assumption when we read a Platonic
dialogue. Why assume that some one figure in these works presents
the convictions of Plato himself? Some scholars, using this analogy
between a dramatic work and a Platonic dialogue, hold that Plato's
thought is no more contained in the words of any one interlocutor
than the beliefs of dramatists are revealed by the words of any of
their characters.?2
But the comparison between Plato's dialogues and dramatic works
is misleading in a number of ways, in spite of the fact that in each
genre there is dialogue among two or more characters. To begin with
what is most obvious: Plato's works were not written to be entered
into competition and performed at civic religious festivals, as were
the plays of the Greek tragedians and comedians. Plato is not assign-
ing lines to his speakers in order to win a competition or to compose
a work that will be considered beautiful or emotionally satisfying by
official judges or an immense audience. The dramatist does have
this aim, and if it suits his purpose to have his main characters
express views that differ from his own, he will do so. But if Plato's
aim in writing is to create an instrument that can, if properly used,
guide others to the truth and the improvement of their souls, then it
may serve his purpose to create a leading speaker who represents the
sincere convictions of Plato himself. The point is that, if Plato's
aims differ from those of a dramatist, then he will have a reason that
the dramatist lacks for using his main speakers as a mouthpiece for
his own convictions.
Furthermore, many of Plato's dialogues in his middle and later
Cambridge Companions Online © Cambridge University Press, 2006
l6 THE CAMBRIDGE COMPANION TO PLATO
periods exhibit a high degree of collaboration among the interlocu-
tors: Although they involve questions and answers among several
speakers, these characters cooperate in developing and refuting philo-
sophical theories. In the Phaedo, Socrates listens to the attempts of
his interlocutors to undermine his conception of the soul, but he
argues against them and has little trouble in securing their agree-
ment. Similarly, in the Republic Socrates accepts questions and chal-
lenges from Glaucon and Adeimantus, but soon persuades them to
accept his answers. In the Theaetetus Socrates and the dialogue's
eponymous interlocutor cooperate in undermining the various con-
ceptions of knowledge that are discussed. In the Philebus Socrates
completely wins over an initially recalcitrant Protarchus, and the
latter becomes a docile yes-man whose role is primarily to seek
clarification. So, Plato's dialogues cannot have been intended merely
to dramatize conflict between opposing characters and to give expres-
sion to competing philosophical ideas. Nor can they have been de-
signed merely to give mental exercise to the reader, for that purpose
would have been much better served by simply recording as many
arguments as possible on opposite sides of a question.
When the dialogues are read in their entirety, they take on the
shape that we would expect of works that record the intellectual
development of a single individual who is struggling to express and
argue for the truth as he best understands it. There is development
and perhaps there are even reversals, but there is at the same time
the kind of continuity that indicates that Plato is using his main
speaker to express his own views. And so although the dialogue form
might be used by a philosopher in order to reveal the deficiencies of
the views expressed by all of the interlocutors, we have strong rea-
son for thinking that this is not in fact what Plato is doing. The
dialogue form of his works should not keep us from saying that they
are vehicles for the articulation and defense of certain theses and the
defeat of others. Though they are not philosophical treatises, many
of them share these purposes with philosophical treatises.
But why, then, did he not simply write philosophical treatises,
rather than dialogues?73 Relying on some points that were made ear-
lier, we can answer as follows:74 Plato begins his career as a writer in
order to give expression to the philosophy and way of life of Socrates.
His purpose in doing so is not purely historical; rather, he regards
Socrates as a model of wisdom and insight, and he sets down his
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 27
portrait of Socrates so that he and others will have an enduring re-
minder of this remarkable man. Since Socrates is above all someone
who enters into dialogue with others, and not a propounder of system-
atic doctrine, the dialogue form is the perfect medium for the expres-
sion of his life and thought. But Plato is himself a philosopher and not
merely a follower of Socrates, and when he develops views that go
beyond those of his teacher, he continues to use the dialogue form for
their expression. His decision to do so need not be regarded as a
mystery. The dialogue form provides a natural way to air challenges
the reader might be expected to make to the theories under discus-
sion; assigning an objection to a speaker is a vivid way of clarifying
and defending the views being presented. Furthermore, the misgiv-
ings expressed in the Phaedrus about writing may have added to
Plato's reasons for retaining the dialogue form. Oral exchange is the
essential tool of philosophy, yet reading books can entice one into
thinking that this encounter with the written word is by itself suffi-
cient for wisdom. Therefore, it is entirely appropriate to put into one's
writing something that reminds the reader that insight comes
through discussion with others and not through mere reading. What
better way to give expression to this warning against the misuse of
books than to make each of one's works a dialogue? Even when the
dialogue conveys no drama and no real opposition of viewpoints - as
in such earlier works as the Protagoras and Gorgias - it continues to
serve this further purpose.
But my answer to the question Why did Plato write dialogues?
would not be accepted by all scholars. Some students of Plato take
him to be less than fully forthcoming and straightforward in the
expression of his philosophical views, and they take the dialogue
form as a device by which he avoids telling us everything he be-
lieves. According to this interpretation, one of Plato's aims in writ-
ing is to get his readers to think for themselves; and to accomplish
this goal, he deliberately inserts fallacies, ambiguities, and other
deficiencies into his works. He of course does have his own convic-
tions, but he has reasons for making his readers do a considerable
amount of work before they discover his true views. And so, in order
to see what Plato is driving at, we must go beyond what any of the
dialogues actually says, using the speeches of the interlocutors as
signs of a concealed message. This way of approaching Plato need
not regard his unwritten opinions as the truths to which he is trying
Cambridge Companions Online © Cambridge University Press, 2006
28 THE CAMBRIDGE COMPANION TO PLATO
to point in the dialogues. Rather, the idea is that each dialogue
contains within itself all the materials we need for its understand-
ing, but that these materials are like an anagram that must be un-
scrambled before its meaning can be revealed. The explicit state-
ments of the interlocutors are the materials from which we must
construct Plato's hidden message.?*
But this is a hazardous way of approaching Plato. First, even if one
locates what one takes to be a deficiency in the reasoning of an
interlocutor, it is a further matter to say that Plato himself regarded
it as a deficiency and meant it to be recognized as such by the reader.
To show that Plato is engaged in this kind of project one must
uncover a pattern of egregious error in the dialogues, for only in this
case could we legitimately infer that these deficiencies are meant to
convey something to us. And it is difficult to establish that such a
pattern exists.
Second, to defend this approach, we would have to suggest a mo-
tive for Plato's deceptive practices as a writer, and, unless we wish to
appeal to the Seventh Letter, it is not easy to find such a motive.
Might Plato have feared political persecution, had he said what he
really believed? Plato's objections to Athenian democracy are de-
clared openly in the Republic (555b-565e); he obviously is not writ-
ing in order to curry favor with the masses.?6 Did he want to conceal
his message because he thought his readers should be forced to think
for themselves rather than take him as an authority? That is un-
likely, because we can see from reports by Aristotle and other an-
cient writers about the Platonists that they were internally divided,
and Aristotle provides in his own person evidence that at the Acad-
emy there were many opponents of the theory of Forms.?? In the
midst of all this controversy, Plato could hardly have believed that
his writings would be treated by many as authoritative pronounce-
ments to be accepted without question.
Third, if we think that the manifest meaning of a dialogue must be
set aside and its hidden message revealed, we leave ourselves little
or nothing to serve as evidence. For example, suppose we think that
Plato has deliberately given bad arguments in the Republic for the
thesis that justice is more advantageous than injustice. Even if we
are convinced that the deficiencies of the argument can only have
been deliberate, and even if we can plausibly arrive at Plato's motive
for this deception, we are still left with the problem of what hidden
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 29
message we should bring away from this dialogue. Is Plato trying to
show that in fact justice is less advantageous than injustice? Is he
trying to show that although justice is more advantageous, we must
not try to establish this thesis by means of the arguments used in the
Republic, but must seek new arguments of our own? Or is he trying
to show that when we try to present arguments about what is advan-
tageous, we inevitably fall into error, and that therefore this is an
area of human life inaccessible to human reason? And of course
there are many other possibilities. None of them can be discounted
on the grounds that it cannot be found in any text of Plato, because
the whole point of this approach to reading Plato is that we cannot
expect to find what he believes (as opposed to what his interlocutors
say) in the text. But, having abandoned the text on the grounds that
it does not contain what Plato believes, we have no way of support-
ing one suggestion as opposed to another regarding what he does
believe.
Our best chance of understanding Plato is therefore to begin with
the assumption that in each dialogue he uses his principal interlocu-
tor to support or oppose certain conclusions by means of certain
arguments because he, Plato, supports or opposes those conclusions
for those reasons. In reading him this way, we need make no hazar-
dous assumptions about why he wrote, and why he wrote in dia-
logue form. And of course we are always free to question our work-
ing hypothesis when certain passages or even whole dialogues resist
this approach, and plausible motives for deception are suggested by
the text itself. It is fair to say that this is the approach adopted by a
great many scholars, and that it has considerably enhanced our un-
derstanding of the dialogues. This methodological principle is not an
a priori assumption about how Plato must be read, but is rather a
successful working hypothesis suggested by an intelligent reading of
the text and confirmed by its fruitfulness.
Reading Plato in this way allows us to make use of whatever
material we have in the dialogues to contribute to our understanding
of them: If the scene-setting that occurs at various points of the
dialogue helps us understand the argument, or if the characteriza-
tion of the interlocutors gives us clues as to why the argument takes
the course it does, so much the better for our interpretation.?8 The
fundamental idea is that unless we have good evidence to the con-
trary, we should take Plato to be using the content of his interlocu-
Cambridge Companions Online © Cambridge University Press, 2006
3O THE CAMBRIDGE COMPANION TO PLATO
tors' speeches, the circumstances of their meeting, and whatever
other material he has at his disposal, to state conclusions he believes
for reasons he accepts.
NOTES
1 Modern accounts of Plato's life and thought give different dates for his
birth, reflecting a discrepancy among the ancient sources. I have fol-
lowed W. K. C. Guthrie, A History of Greek Philosophy, vol. 4 (Cam-
bridge: Cambridge University Press, 1975), 10. R. Robinson and J. D.
Denniston, in their entry on Plato in N. G. L. Hammond and H. H.
Scullard, The Oxford Classical Dictionary, id ed. (Oxford: Clarendon
Press, 1970), give 429. W. D. Ross, Plato's Theory of Ideas (Oxford:
Clarendon Press, 1951), 10, gives 429-7. For references to further discus-
sion, see Guthrie, History of Greek Philosophy, 4: 10 n. 2. For methods
used in the ancient world for keeping track of the year, see Alan E.
Samuel, "Calendars and Time-Telling/' in Civilization of the Ancient
Mediterranean, vol. 1, ed. Michael Grant and Rachel Kitzinger (New
York: Scribner's, 1988), 389-95.
2 Plato's name for the discipline described in this paragraph is dialektike
("dialectic"), and he conceives of the philosopher as someone who is
studying or has mastered this subject. The principal texts I have used in
my account of dialectic are Republic 5O9d-5iid and 53id-534e. The
Greek noun dialektike is derived from a verb, dialegesthai, which
means "to engage in conversation" (cf. "dialogue"); accordingly, a dialec-
tician is, among other things, someone who is trained at asking and
answering questions and in this way defending a position against criti-
cism J534b-d). Other characterizations of dialectic, which add consider-
ably to the Republic, are found at Phaedrus 265d-266b and Philebus 16-
17a, 57C-59C. The equation of philosophy and dialectic is not accepted
by Aristotle (see for example Metaphysics ioo4b25-6), though it is not
immediately clear whether this merely reflects a verbal difference from
Plato or whether Aristotle, in making philosophy superior to dialectic,
intends to find fault with dialectic as Plato conceives it. For the latter
interpretation, see T. H. Irwin, Aristotle's First Principles (Oxford: Cla-
rendon Press, 1988), 137-8.
The references made in this note, and throughout this volume, to
passages in Plato's works cite "Stephanus pages," and the reader might
like a brief explanation of this common practice. To maintain a uniform
system of reference, the outer margins of modern editions and transla-
tions of Plato's works supply the pagination of the edition of Plato pub-
lished in 1578 by Henri Estienne (c.1528/31-1598). These are called
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 31
Stephanus pages after the Latinized form of "Estienne." A brief account
of his contributions to scholarship can be found in John Edwin Sandys, A
History of Classical Scholarship, vol. 2 (New York: Hafner, 1958), 175-
7. His edition of Plato was standard for over two centuries. Unless other-
wise indicated, it should be assumed that the Greek text of Plato used in
this volume is that of John Burnet, Platonis Opera, 5 vols. (Oxford:
Clarendon Press, 1900-7) (often referred to as the Oxford Classical Text
or O.C.T.). Similarly, all modern editions and translations of the works
of Aristotle supply the pagination of the edition of the Greek text pro-
duced in 1831 by Immanuel Bekker.
3 The school Plato founded (c. 387 B.C.), called the Academy after a park
located on the outskirts of Athens and sacred to the hero Academus, was
in continuous existence for many centuries, but those who in turn suc-
ceeded Plato as leaders of the Academy - Speusippus, Xenocrates, Pole-
mon, and Crates - deviated from his philosophy in important ways.
When Crates died in c. 276 B.C., his successor, Arcesilaus, heavily influ-
enced by certain Platonic dialogues that seem to show that philosophy
can achieve no positive results, made the Academy a center of skepti-
cism; but a watered-down, Stoic version of Platonism was revived when
Antiochus of Ascalon abandoned skepticism and founded the Old Acad-
emy (c. 87 B.C.). At this time, and continuing for two centuries, there
begins a Platonist movement in Athens and Alexandria, called Middle
Platonism by some scholars, to distinguish it from both Plato himself and
the Neo-Platonist school later founded by Plotinus; the representative of
Middle Platonism best known today is Plutarch (c. A.D. 45-125). For a
valuable treatment of this movement, see John Dillon, The Middle
Platonists (Ithaca, N.Y.: Cornell University Press, 1977). Plotinus (c. A.D.
205-70) founded his own school at Rome in 244, and although he is
heavily influenced by Middle Platonism as well as other philosophical
currents, he places greatest weight on Plato's dialogues. His writings have
as much influence as do Plato's on the development of later forms of
Platonism. A revived and syncretic form of Platonism was the ascendant
philosophical trend in the pagan world from the time of Plotinus until the
closing of the pagan schools by Justinian in 529. Other Neo-Platonic
thinkers of this period who had a decisive influence on medieval thought
were Plotinus's student, Porphyry (232-c. 305), Iamblichus (c. 250-c.
325), and Proclus (4io/4i2?-485). On the influence of Platonism, see
Raymond Klibansky, The Continuity of the Platonic Tradition during the
Middle Ages (London: Warburg Institute, 1939); Ernst Cassirer, The Pla-
tonic Renaissance in England (Edinburgh: Nelson, 1953). Plato's influ-
ence on Renaissance thought is discussed in many of the contributions to
Charles B. Schmitt, ed., The Cambridge History of Renaissance Philoso-
Cambridge Companions Online © Cambridge University Press, 2006
32 THE CAMBRIDGE COMPANION TO PLATO
phy (Cambridge: Cambridge University Press, 1988). A brief overview of
Platonism from antiquity to the twentieth century is presented by D. A.
Rees, "Platonism and the Platonic Tradition," in The Encyclopedia of
Philosophy, ed. Paul Edwards (New York: Macmillan and Free Press,
1967), 6:33-341. On the influence of Plato in Victorian Britain, see Frank
M. Turner, The Greek Heritage in Victorian Britain (New Haven: Yale
University Press, 1980), chap. 8. Other works on later Platonism are listed
in the bibliography to this volume under the heading Platonism after
Plato.
4 This feature of Plato's relation to the philosophical tradition is distorted
by the widely known tribute paid to him by Alfred North Whitehead:
"The safest general characterization of the European philosophical tradi-
tion is that it consists of a series of footnotes to Plato." See Process and
Reality: An Essay in Cosmology corrected ed., ed. D. R. Griffin and D. W.
Sherburne (New York: Free Press, 1978), 39. Since footnotes are mere
supplements and are not intended to contradict the body of the text,
Whitehead's statement misleadingly suggests not only that subsequent
philosophers were less significant than Plato but also that Plato's work
was the universally accepted starting point for all later philosophy. In fact,
the context of Whitehead's remark shows that he did not mean to say that
Plato's thought was uncontroversial; rather, he was alluding to "the
wealth of general ideas scattered throughout" Plato's writings (ibid). This
is certainly a remarkable feature of the Platonic corpus, but it is also a fair
characterization of other major figures in the history of philosophy.
5 The essay of T. H. Irwin in this volume (Chapter 2) discusses these
historical influences, as well as the political and moral climate of Plato's
time, and tries to show how these materials are transformed by Plato as
he develops his principal positions in metaphysics, epistemology, and
political philosophy. Irwin also addresses the question of why Plato's
writings take the form of dialogues - an issue I briefly consider later in
this essay. Plato's relationship to the religous currents of the fifth and
fourth centuries, briefly treated by Irwin, are more fully discussed by
Michael L. Morgan in his contribution to this volume (Chapter 7). He
focuses on the influence of mystery religions on Plato's thought, with
their emphasis on an ecstatic experience in which one achieves a kind of
union with what is divine.
6 For two excellent accounts of the trial of Socrates, see Thomas C.
Brickhouse and Nicholas D. Smith, Socrates on Trial (Princeton: Prince-
ton University Press, 1989); and C. D. C. Reeve, Socrates in the Apology
(Indianapolis: Hackett, 1989). A more controversial account, and one
more hostile to Socrates, can be found in I. F. Stone, The Trial of Socrates
(Boston: Little, Brown, 1988). For detailed criticism of Stone's interpreta-
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 33
tion, see T. H. Irwin, "Socrates and Athenian Democracy/' Philosophy
and Public Affairs 18 (1989): 184-205.
7 Diogenes Laertius, Lives of Eminent Philosophers, 3.35, reports that
Socrates heard Plato give a reading of the Lysis, and German scholars of
the nineteenth century, relying on this testimony, assumed that some of
Plato's writings were written before the death of Socrates. This assump-
tion was attacked by George Grote in Plato and the Other Companions
of Socrates, and later scholars have generally accepted his arguments.
For a brief account of this debate and a survey of scholarly opinion, see
Guthrie, History of Greek Philosophy, 4: 54-6. Some of those who be-
lieve that Plato did not begin to write until after the death of Socrates
also conjecture that the Apology was his first work, since he would have
wished to preserve the memory of his teacher while the trial was still
fresh in his memory. See ibid., 4: 71-2.
8 The exceptions are the Apology and the Letters, but there is consider-
able disagreement among scholars about whether the latter were actu-
ally written by Plato. (Conflicting views about the authenticity of the
Seventh Letter sue found in this volume: contrast the essays of T. H.
Irwin and Terry Penner.) The Apology consists almost entirely of a court-
room speech and is not happily classified as a dialogue, although there is
some interchange when Socrates examines one of his accusers. Con-
versely, in a number of the works that begin with an interchange be-
tween characters, and that are therefore classified as dialogues, there is
little or no alteration of speakers after the introductory section. These
are the Menexenus, Timaeus, and Critias.
9 We will later return to the question of what significance should be
attributed to the fact that Socrates plays a much smaller role in some
dialogues than in others, and is entirely absent from some of them.
10 Some scholars question the assumption I make in this sentence that
Plato endorses the views expounded by the main interlocutor (often
Socrates). Instead, they would say that Plato refuses to insert himself in
the dialogue and address the reader directly precisely because he does
not endorse the views of any one interlocutor - even Socrates-and
wants to leave the truth unstated. I will discuss this approach more fully
and express my doubts about it later in this essay.
11 This does not mean that for a certain portion of his life Plato was merely
trying to write an accurate historical account of what Socrates said, and
that he decided to postpone the project of expressing his own philosophi-
cal ideas. It is more plausible to assume that for a period of time Plato
accepted the philosophy of Socrates and therefore did not distinguish the
historical and the philosophical aspect of his works: To write a histori-
cally faithful account of a conversation between Socrates and other inter-
Cambridge Companions Online © Cambridge University Press, 2006
34 THE CAMBRIDGE COMPANION TO PLATO
locutors was to write a philosophical work that expresses the truth as
he, Plato, saw it.
12 For a fuller presentation of this general line of interpretation, see the
essay by Terry Penner in this volume (Chapter 4), as well as Gregory
Vlastos, Socrates: Ironist and Moral Philosopher (Cambridge: Cam-
bridge University Press, 1991), Chaps. 2-4. The thesis that we can and
should distinguish the philosophies of Socrates and Plato is widely ac-
cepted, but the further claim of Vlastos that they are "so diverse in
content and method that they contrast as sharply with one another as
with any third philosophy you care to mention" (p. 46) is more controver-
sial. Vlastos holds that Socrates7 ontological beliefs do not advance be-
yond common sense (pp. 53-66) and that his ethical inquiries produce
"no spill-over into epistemology" (p. 67). To this it can be objected that
Socrates did take a step beyond common sense by explicitly assuming
that a single factor unites whatever is virtuous and by requiring a wise
person to know this single factor.
13 See De Sophisticis Elenchis i83b7-8: "Socrates asked questions but did
not answer them; for he agreed that he did not know"; Metaphysics 1.6
987b 1-4: "Socrates was concerned with ethical matters and not with
the whole of nature, and in ethical matters he sought the universal and
fixed thought for the first time on definitions."
14 For Aristotle's view that Plato arrived at the theory of Forms by "separat-
ing" the universal, which Socrates had not separated, see Metaphysics
1.6 987bi-io, XIII.9 108602-5. Some of Aristotle's main objections to
the Forms are presented in Metaphysics I.9 and XIII.4-5. His complaints
about the Form of the Good are presented in Nicomachean Ethics 1.6.
15 A further indication of this is the statement in Magna Moralia Li
n82ai5-28 (written either by Aristotle or one of his followers) that
Socrates neglects the irrational part of the soul, and that Plato corrects
this error. Here the Republic is taken to present the thought of Plato
rather than that of Socrates. For a fuller presentation of the way in which
Aristotle distinguishes between the historical Socrates and the character
who presents Platonic doctrine, see W. D. Ross, Aristotle's Metaphysics
vol. 1 (Oxford: Clarendon Press, 1924), xxxiii-xlv.
16 For a critical and historical review of the widely varying measures used
to study Plato's stylistic development, see Leonard Brandwood's essay in
this volume (Chapter 3). For a more detailed historical survey of stylo-
metric studies, see, by the same author, The Chronology of Plato's Dia-
logues (Cambridge: Cambridge University Press, 1990). A brief and acces-
sible introduction to stylometry can be found in Ian Mueller, "Joan
Kung's Reading of Plato's Timaeus," in Nature, Knowledge and Virtue:
Essays in Memory of Joan Rung, ed. Terry Penner and Richard Kraut
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 35
(Edmonton, Alberta: Academic Printing and Publishing, 1989), 1-27.
Prior to the stylistic studies initiated in the nineteenth century, the
dialogues were arranged in groups according to their content. For exam-
ple, Diogenes Laertius, Lives of Eminent Philosophers, 3.57-61, records
the arrangement into tetralogies made by Thrasyllus (d. A.D. 36). For a
comprehensive list of chronological studies and their results, ranging
from 1792 through 1981, see Holger Thesleff, "Studies in Platonic Chro-
nology," in Commentationes Humanarum Litterarum (Helsinki: Scien-
tarum Fennica) 70 (1982): 8-17.
17 Some of the evidence for the lateness of the Laws: Aristotle says in the
Politics (126^26) that it was written after the Republic; Plutarch [De
Iside et Osiride 37off.) says that Plato wrote it when he was an old man;
a battle referred to at Laws 638b is often identified as one that took place
in 356 B.C. (nine years before Plato died). Diogenes Laertius, Lives of
Eminent Philosophers, 3.37, implies that work on the Laws was not
entirely finished when Plato died, but in the same paragraph he reports a
story that the Phaedrus was Plato's first dialogue, so his chronological
information does not inspire confidence. For further references, see
Guthrie, History of Greek Philosophy, 5:322.
18 In addition to the works widely accepted as Plato's there are several
others attributed to him in antiquity by Diogenes Laertius and included
among his works in medieval manuscripts, but whose Platonic author-
ship is either a matter of current controversy or widely rejected. Some of
the Letters (there are thirteen of them) are widely assumed to be spuri-
ous, and some scholars take all of them to be unauthentic; but in most
cases there is no consensus about whether Plato was their author. If they
are genuine, they were written during his later period. References to
scholarly treatments of their authenticity can be found in Guthrie, His-
tory of Greek Philosophy, 5:399-401. Other works that were attributed
to Plato in antiquity, but whose authenticity is now a matter of debate,
include: Alcibiades I, Alcibiades II, Cleitophon, Epinomis, Hipparchus,
Minos, Rivals (sometimes the alternative title Lovers is used), and
Theages. The Epinomis is late, and is thought by some to be the work of
Plato's student, Philip of Opus; this attribution is also reported by
Diogenes Laertius, Lives of Eminent Philosophers, 3.37. The other dia-
logues just listed have affinities to Plato's early works. In addition, a
brief list of succinct Definitions is included in the medieval manuscripts
of Plato's works, but it is universally considered spurious. Two of the
doubtful or spurious works - the Letters and Epinomis - are widely
available in English, but translations of all those mentioned above ex-
cept the Definitions can be found in the Loeb Classical Library. See IA
and IB of the bibliography to this volume for further information on
Cambridge Companions Online © Cambridge University Press, 2006
36 THE CAMBRIDGE COMPANION TO PLATO
texts and translations. For doubts about the authenticity of the Hippias
Major — now widely accepted as genuine - see Charles Kahn, "The Beau-
tiful and the Genuine," Oxford Studies in Ancient Philosophy 3 (1985):
261-87. This is a review of a work that defends the authenticity of the
dialogue: Paul Woodruff, Plato: Hippias Major, trans, with commentary
and essay (Indianapolis: Hackett, 1982).
19 The essay of Terry Penner in this volume (Chapter 4) discusses some of
the main elements of the moral philosophy contained in these dia-
logues, emphasizing their mutual dependence and coherence, and stress-
ing the egoism of Socrates, his rejection of relativism, his profession of
ignorance, and his method of education. Penner does not share the view I
just expressed in the text that in these early works Socrates is unsuccess-
ful, and takes himself to be unsuccessful, in his search for definitions. I
defend this interpretation in Socrates and the State (Princeton: Prince-
ton University Press, 1984), chap. VIII.
20 See E. R. Dodds, Plato: Gorgias, A Revised Text with Introduction and
Commentary (Oxford: Clarendon Press, 1959), 18-24; Terence Irwin,
Plato: Gorgias, translated with notes (Oxford: Clarendon Press, 1979),
5-8. For an alternative proposal about the place of the Gorgias within
the early dialogues, see Charles H. Kahn, "Did Plato Write Socratic
Dialogues?" Classical Quarterly 31 (19811:305-20, esp. 308-11. He ar-
gues that the early dialogues in which Socrates seeks definitions (e.g.,
Laches, Charmides, Euthyphro) are designed to orient the reader toward
the doctrines of the middle period, and he therefore places them later
than the Gorgias. Kahn holds (pp. 307-8) that although the Apology
might give us a reliable historical portrait of Socrates, all of the other
early works of Plato involve significant departures from Socrates. He
therefore doubts (p. 310 n. 13) the value of Aristotle's testimony regard-
ing the differences between Socrates and Plato.
21 There are reasons other than those having to do with style for assuming
that Euthydemus, Lysis, and Hippias Major were written at the end of
the early period. See Vlastos, "The Socratic Elenchus,;/ Oxford Studies
in Ancient Philosophy 1 (19831:57-8; and Vlastos, Socrates, chap. 4. It
should also be noted that if one pays no attention whatsoever to philo-
sophical content and looks only for stylistic similarities, then one would
place the Cratylus, Meno, Phaedo, and Symposium among the later
dialogues of the early period. See Leonard Brandwood, A Word Index to
Plato (Leeds: Maney & Son, 1976), xvii. But there are sound reasons
based on philosophical content to place three of these - the Cratylus,
Phaedo, and Symposium - in the middle period, and that is a widely
held opinion (though the position of the Cratylus is disputed; see n. 39).
Similarly, because of its philosophical content, the Meno is often labeled
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 37
a "transitional" dialogue that cannot be happily classified as early or
middle.
22 The first two claims are made at Gorgias 47oe and Apology 3oc-d respec-
tively; the last at Apology 38a. However, it is not always clear whether
Socrates means to identify virtue and happiness, or whether he thinks
there is a weaker relation between them. For discussion, see Vlastos,
Socrates, chap. 8. The Apology makes it clear that according to Socrates
a virtuous person must be able to pass intellectual tests, and many of the
early dialogues take it for granted that since being virtuous is a matter of
being an expert on a certain subject, one who has developed this exper-
tise is able to explain the subject to others.
23 On the way in which this geometrical experiment provides an answer to
the methodological challenge of the dialogue, see the essay by Gail Fine
in this volume (Chapter 6). She argues that it is the distinction between
knowledge and true belief, rather than the theory of recollection, that
plays the crucial role in Plato's attempt to show that lacking knowledge
is not a bar to acquiring it.
24 For this contrast, see Vlastos, Socrates, chap. 4. Plato's developing inter-
est in mathematics can also be found in the Gorgias, a dialogue often
thought to be somewhat earlier than the Meno. See Vlastos, Socrates,
128-9; Irwin, Plato: Gorgias, 7-8. Plato's association with such leading
mathematicians of his time as Archytas and Timaeus is cited in the
Seventh Letter (35oa-b) and Cicero's Academica I.10.16. For Plato's high
degree of involvement in the mathematical research of his time, see the
essay by Ian Mueller in this volume (Chapter 5). He discusses Plato's
fundamental idea, pursued in the Meno, Phaedo, and Republic Books VI-
VII, that progress can be made in philosophy if it adopts the "method of
hypothesis" that had been working so smoothly in mathematics. This
aspect of Plato's methodology is central to his philosophy, since he re-
gards the existence of Forms as a "hypothesis" (see Phd. 99d-iosb).
25 See R. S. Bluck, Plato's Meno (Cambridge: Cambridge University Press,
1961), 108-20, for a discussion of the dialogue's date,- he locates it in the
vicinity of 386/5. Platonic scholarship would have a much easier task if
each dialogue had made reference to some historical event to which we
could confidently assign a precise date. Unfortunately, there is at best
meager evidence of this sort, and assigning dates to the dialogues is a
highly conjectural matter. Because of its reference to historical events,
we know that the Menexenus - a work likely to have been written
around the same time as the Gorgias-was composed after 386; see
Dodds, Plato: Gorgias, 24. The Theaetetus (i42a-b) refers to the fatal
wounds of its eponymous interlocutor after a battle at Corinth, and this
is generally assumed to have taken place in 369. Book I of the Laws
Cambridge Companions Online © Cambridge University Press, 2006
38 THE CAMBRIDGE COMPANION TO PLATO
(638b) refers to the defeat of the Locrians by the Syracusans, and this has
been taken as a reference to events of 356. One other method of finding
approximate dates for the dialogues involves using historical sources
other than the dialogues to determine dates for certain events in Plato's
life. (These sources include the Letters, which may give us accurate
information about Plato's life even if they were not composed by Plato
himself.) The Seventh Letter says that Plato was forty when he made his
first visit to Sicily (324a, 326b); later writers say that the motive of the
trip was to enter discussions with Pythagoreans living there. Pythago-
rean influences can be seen in the Gorgias and Meno, and so assigning
these works a date not far from 387 is plausible. He returned within the
year and is said to have established the Academy, a center for philosophi-
cal study, soon afterward. He made a second visit to Syracuse after the
death of its tyrant, Dionysius I, in 367, and returned to Athens in 365; a
third and final visit occurred in 361. The purpose of these two further
visits, according to the Seventh Letter, was political: At the urging of his
friend, Dion, Plato was attempting to mold the politics of Syracuse by
giving a philosophical education to its new tyrant, Dionysius II. These
later dates, of course, give us an earliest time of composition for the
Seventh Letter, whether it is by Plato or not. And they are sometimes
used as a general framework for speculating about the dates of composi-
tion of the dialogues.
26 There is, however, disagreement among scholars about whether Plato
gives arguments for the existence of Forms, or whether he merely takes
their existence for granted. The latter view can be found in R. Hackforth,
Plato's Phaedo (Indianapolis: Bobbs-Merrill, 1955), 50; David Gallop,
Plato's Phaedo (Oxford: Clarendon Press, 1975), 95,- William J. Prior,
Unity and Development of Plato's Metaphysics (London: Croom Helm,
1985), 10. But it may be that this interpretation is based on too narrow a
conception of what constitutes a proof of or an argument for their exis-
tence. If positing the existence of the Forms provides solutions to prob-
lems not otherwise solvable, or explains phenomena not otherwise ex-
plainable, then that would constitute an argument on their behalf. This
latter approach to the theory of Forms is defended by H. F. Cherniss, "The
Philosophical Economy of the Theory of Ideas/' in Studies in Plato's
Metaphysics, ed. R. E. Allen (London: Routledge & Kegan Paul, 1965), 1-
12. Aristotle (see note 14 above) takes the Platonists to have arguments
(bad ones, of course) for the existence of Forms. For a recent account of
what those arguments are, with special attention to the argument of
Phaedo 74b-e that Equality is not identical to any observable equal ob-
jects, see Terry Penner, The Ascent from Nominalism: Some Existence
Arguments in Plato's Middle Dialogues (Dordrecht: D. Reidel, 1987).
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 39
27 It is a frequent refrain of Plato's thought that by contrast with the Forms
the objects we can sense are always becoming and never remain the
same. See Phd. 78e; Cra. 439b-d; Smp. 2 i ia-b; Rep. 479a-e; Ti. iyt-
28a, 37e-38b, 49d— 5od, 52a. But the proper interpretation of this
thought is a difficult matter. For discussion, see Nicholas White's contri-
bution to this volume (Chapter 9). He argues that for Plato the flux of
sensible objects is part of their general predicament, namely that they do
not have their properties in a way that is independent of their circum-
stances or the viewpoints from which they are perceived. Aristotle tells
us that because Plato came under the influence of Heraclitus and
Cratylus, he took all sensible objects to be in constant flux and therefore
unsuitable as objects of knowledge. See Metaphysics 1.6 987a32-b7,
XIII.4 io78bi2-i7, XIII.9 io86a32-bi3. It is a controversial matter
whether the flux attributed to all things by Heraclitus and all sensible
objects by Plato is a matter of change as we normally understand that
notion; instead, they may have a broader notion of flux that includes not
only alteration of quality over time but the presence of opposite quali-
ties at the same time. For discussion, see Charles H. Kahn, The Art and
Thought of Heraclitus (Cambridge: Cambridge University Press, 1979);
Jonathan Barnes, The Presocratic Philosophers, vol. 1 (London: Rout-
ledge & Kegan Paul, 1979), chap. 4; G. S. Kirk, f. E. Raven, and M.
Schofield, The Presocratic Philosophers, id ed. (Cambridge: Cambridge
University Press, 1983), chap. 6.
28 Some scholars believe that in the Timaeus Plato distinguishes between
mere everlastingness and eternity, and attributes the latter property to
the Forms. What is everlasting (the term "sempiternal" is sometimes
used) endures as long as time itself: There is no time at which it fails to
exist. What is eternal is not in time at all, and so it does not undergo the
passage of time. The body and soul of the created universe are eternal
and indestructible, having come into existence with the creation of
time, whereas the Forms are not in time. The crucial statement in which
this distinction may be found is the one in which Plato calls time a
"moving image of eternity" (Ti. 37d). See too Parmenides I4oe-i42a for
the claim that the One is not in time and has no share of time. For
discussion and reference to scholarly literature, see Richard Sorabji,
Time, Creation and the Continuum (Ithaca, N.Y.: Cornell University
Press, 1983), 108-12; and Richard Patterson, "The Eternality of Platonic
Forms," Archiv fur Geschichte der Philosophie 67 (19851:27-46. In any
case, whether or not Plato made this distinction, it is present in Plotinus
at (among other places) Enneads 3.7.2. See Sorabji, Time, Creation and
the Continuum, 112-14.
29 It has become a widespread practice to use uppercase initial letters (as in
Cambridge Companions Online © Cambridge University Press, 2006
4O THE CAMBRIDGE COMPANION TO PLATO
Form, Idea, Equality, The Good, The Large) to designate either the eternal
and changeless objects posited by Plato or some member of this class, but
it is important to realize that this is our own convention and not Plato's.
The ancient Greeks wrote only in uppercase letters; the lowercase was
introduced much later as a cursive device. See Rachel Kitzinger, "Alpha-
bets and Writing," in Grant and Kitzinger, Civilization of the Ancient
Mediterranean, vol. i, esp. 412; Herbert Weir Smyth, Greek Grammar
(Cambridge, Mass.: Harvard University Press, 1963), 8. Though capitaliz-
ing words that refer to Forms is a helpful and widely used convention, it
unfortunately forces one to use the lowercase for two different types of
occurrence, namely: those in which one is sure that Plato is not referring
to a Form, and those in which one wants to leave it open whether or not
he is referring to a Form.
30 A reasonable conjecture is that Equality cannot fail to be Equality,
whereas the sticks that are in fact equal can fail to be equal sticks. The
deficiency of the equal sticks is that they derive their equality from their
meeting the definition of Equality itself; whereas the latter entity does
not derive its nature from anything else: It is Equality. It is this ontologi-
cal difference that underlies the fact that we can make mistakes about
equal sticks (misperceiving them as unequal) but cannot make an analo-
gous mistake about Equality (judging that it fails to measure up to Equal-
ity itself). This passage in the Phaedo (74d-e) has been taken by some
scholars to depend on the assumption that no two physical objects can
be exactly equal, but this interpretation is not widely accepted. An
influential attack on it was made by Alexander Nehamas, "Plato on the
Imperfection of the Sensible World/' American Philosophical Quarterly
12 (1975): 105-17. For further discussion, see Richard Patterson, Image
and Reality in Plato's Metaphysics (Indianapolis: Hackett, 1985), chap.
5. Nicholas White's contribution to this volume (Chapter 9) also dis-
cusses this passage.
31 SeePhd. iooc, 101c, 102b; Smp. 211b; Rep. 4766.-, Prm. 130b, I3oe-i3ia.
32 It should not be assumed that the mere presence of these words in Plato
indicates that he is referring to the abstract objects posited by his theory
of Forms. Eidos and idea in their ordinary usage can designate any class
or kind of object, without committing the speaker to the eternity of the
class or any of the other special characteristics Plato attributes to the
Forms. Plato frequently uses these terms in this ordinary way. See e.g. Ti.
35a4 and Sph. 219C2. Occurrences of eidos and idea at Euphr. s^4, 6dn,
and 6e3 are a matter of debate,- they are taken by R. E. Allen to indicate
that a theory of Forms is present even in the early dialogues. See Plato's
"Euthyphro" and the Earlier Theory of Forms (London: Routledge &
Kegan Paul, 1970), 28-9. Against this, see Vlastos, Socrates, 56-66.
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 41
33 See Metaphysics 1.6 98yb2-4, XIII.4 1078^7-19, XIII.9 io86b2-5.
34 These terms - ''separate/' "realm," "abstract" - call for comment: (A) At
Prm. 130b it is agreed that the Forms exist separately, and although this
separate existence is not treated as one of their problematic features,
neither is it explained. Aristotle says that Plato, unlike Socrates, sepa-
rated universals and thereby went astray (Metaphysics XIII.4 io78b3O,
XIII.9 io86b4~7), but he too does not explain what is involved in separa-
tion. A likely conjecture is that the separation of the Forms from sensibles
consists in their ontological independence: They exist whether or not
sensible objects participate in them. For discussion, see Gail Fine, "Sepa-
ration," Oxford Studies in Ancient Philosophy 2 (i984):3i-87; Vlastos,
Socrates, 256-65. (B) By the term "realm" I mean to convey the idea that
according to Plato the Forms are united together in a systematic arrange-
ment. See e.g. Rep. 500c; I will briefly comment on this aspect of the
Forms later in this essay. (C) To call the Forms "abstract objects" is to use
a contemporary philosophical phrase that does not correspond to any
equivalent term in Plato's Greek. An abstract object is not merely one
that cannot be detected by means of the senses - otherwise the soul as
Plato conceives it would count as an abstract object, and so would atoms
as we conceive them. In addition to being unobservable, abstract objects
lack spatial location and are incapable of change.
35 This is now a common view, but it has had its detractors. For an oppos-
ing interpretation, see Paul Shorey, The Unity of Plato's Thought (Chi-
cago: University of Chicago Press, i960). He says (p. 88), "Plato on the
whole belongs to the type of thinkers [sic] whose philosophy is fixed in
early maturity (Schopenhauer, Herbert Spencer), rather than to the class
of those who receive a new revelation every decade (Schelling)." But the
two extremes Shorey mentions here are not the only possibilities.
36 See Phd. 75C-d, where Plato connects the existence of Forms with the
question "What is . . . ?" that typically arises in a Socratic interchange.
In the Parmenides (i3ob-e), Socrates is portrayed as someone who is not
yet sufficiently practiced in the theory of Forms, and although he is
confident that there are Forms of Justice, Beauty, and Goodness, he is
less confident about Human, Fire, and Water; then, when he is asked
about such alleged Forms as Hair, Mud, and Dirt, he expresses confi-
dence that they do not exist, but is immediately criticized by Parmeni-
des for shying away from such objects. This suggests that in Plato's
development there was, at least initially, some uncertainty about which
Forms to posit.
37 An alternative reading was proposed many years ago according to which
Plato is merely saying that if there is a Form of X then it is one Form. See
J. A. Smith, "General Relative Clauses in Greek," Classical Review 31
Cambridge Companions Online © Cambridge University Press, 2006
42 THE CAMBRIDGE COMPANION TO PLATO
(19171:69-71. Recent discussion of this alternative can be found in Gail
Fine, "The One Over Many/7 Philosophical Review 89 (19801:213 n. 25;
and Patterson, Image and Reality in Plato's Metaphysics, 203 n. 8.
38 It is not easy to decide whether in the Statesman Plato has modified his
position in the Republic, or whether he is making explicit a point he had
been assuming earlier. If he is taking it for granted in the earlier dialogue
that any word commonly applied to a plurality is a genuine name, then
he has changed his mind by the time he wrote the later dialogue. But if
he is assuming in the Republic that there are narrow restrictions on
what counts as a true name, then there need have been no change of
mind. The Cratylus, which is often thought to be chronologically closer
to the Republic than to the Statesman (though see n. 39), discusses the
idea that name-giving is a skill, and that names in use can fall short of
ideal names. See, e.g., 389d-39oe. But this allows that a bad name is a
name nonetheless - just as a badly crafted bed is still a bed. If "barbar-
ian" is a name, but a bad one, then the criterion given in the Republic
needs modification, and Plato has come to recognize this in his later
period. On the other hand, if "barbarian" is not a true name - that is, if it
is not really a name at all - then Plato has merely become more explicit
in his later work. For futher discussion, see Patterson, Image and Real-
ity in Plato's Metaphysics, 123-8; and Fine, "The One Over Many,"
197-240.
39 I have listed these works in the order presented in Brandwood, Word
Index, xvii. It is tempting, however, to think that the Phaedrus should
be placed after the Republic and prior to the Parmenides and Theaete-
tus, because of philosophical resemblances between the Symposium,
Republic, and Phaedrus. On the other hand, the method of division and
collection advocated by the Phaedrus (265d-266b) connects it with a
great deal of material in the later period. To see the diversity of opinion
about where the Phaedrus should be placed, see Ross, Plato's Theory of
Ideas, 2. Another dialogue commonly assigned to Plato's middle period
is the Cratylus, but there is some difference of opinion about this.
Stylometric tests suggest that it is earlier than the Republic, but some
scholars argue that its philosophical content requires a later date. For a
brief account of the dispute, see Guthrie, History of Greek Philosophy,
5:1-2. Vlastos proposes the following ordering of the middle dialogues:
Cratylus, Phaedo, Symposium, Republic II-X, Phaedrus, Parmenides,
Theaetetus. See Vlastos, Socrates, 47. On the position of the Sympo-
sium, see K. J. Dover, "The Date of Plato's Symposium," Phronesis 10
(19651:2-20.
40 Again, I follow the order presented in Brandwood, Word Index, xvii.
However, the position of the Timaeus is controversial; some locate it
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 43
among the middle rather than the late dialogues. I will return to this
issue.
41 There is an obvious similarity between the lover's ascent to the Form of
Beauty, as described in the Symposium (2ioa-2i2b), and the philoso-
pher's ascent to the Form of the Good in the Republic. These two Forms
need not be identical, but it is plausible to think that according to Plato
one cannot have knowledge of either one without the other. (At Phile-
bus 64d-65a, Plato says that the good consists, at least partially, in
beauty; and it is arguable that this is meant as an account of the Form of
the Good.) Just as it is said in the Symposium that one must not confine
one's affection to some one individual, but must love what all objects of
beauty have in common (2iob-d), so in the Republic the family is abol-
ished among the guardians so that no single individual will be favored in
a way that detracts from the well-being of all (457d-466b). The essay by
G. R. F. Ferrari in this volume (Chapter 8) discusses the way in which
Plato moves in the Symposium from the familiar topic of falling in love
with other people to the love of the good that motivates all human
action. He also explores the connections among the several speeches of
the Symposium, and the relationship between Plato's treatment of love
in this dialogue and in the Phaedrus.
42 This paradoxical idea is anticipated in the Gorgias: Although Socrates
may seem to be the antithesis of a politician because of his indifference
to worldly goods (485a-486d), it is really conventional politicians who
have the least power (466a-e) and it is Socrates who alone practices the
true art of politics (52id).
43 In my contribution to this volume, "The Defense of Justice in Plato's
Republic" (Chapter 10), I try to show how Plato's metaphysics plays a
central role in his attempt to defend the Socratic thesis that justice
coincides with self-interest.
44 In the Sophist Plato finally comes to terms with a problem that had
puzzled him throughout his career: A true statement says what is and a
false statement says what is not; but since all statements must say
something, and that something must be, it is unclear how false state-
ments are possible. See Euthd. 283e-286d and Tht. 187C-200C. To solve
the difficulty in the Sophist, Plato conducts a complex discussion of the
nature of being and not-being. Michael Frede's contribution to this vol-
ume (Chapter 13) offers an interpretation of Plato's solution.
45 This dialogue contains Plato's fullest discussion of pleasure and its role
in the good life - a topic that he had explored earlier in the Gorgias and
Republic Book IX. As in the Republic, he uses metaphysics to solve the
problem of how we should lead our lives, and he makes important dis-
tinctions among the different kinds of pleasures there are. For an over-
Cambridge Companions Online © Cambridge University Press, 2006
44 THE CAMBRIDGE COMPANION TO PLATO
view of the dialogue's complex structure and its relation to Plato's ear-
lier treatments of pleasure, see Dorothea Frede's contribution to this
volume (Chapter 14).
46 Plato's insistence upon the artist's remoteness from true reality in Book
X of the Republic is the central element in his radical critique of the role
of the poet in the political community, but his attitude toward poetry is
far more complex than is often realized, as other dialogues besides the
Republic reveal. See the essay of Elizabeth Asmis in this volume (Chap-
ter 11) for a comprehensive overview of Plato's lifelong quarrel with
poetry.
47 For further discussion of degrees of reality in Plato, see Gregory Vlastos,
Platonic Studies (Princeton: Princeton University Press, 1981), essays 2
and 3.
48 The most obvious objections are directed against Plato's exclusion of
nearly all citizens from political participation and his eagerness to sup-
press unorthodox ideas. In K. R. Popper's phrase, Plato is an enemy of
"the open society." See The Open Society and its Enemies, vol. 1, 4th ed.
(New York: Harper & Row, 1963). Popper attributes to Plato the view
that "the criterion of morality is the interest of the state" (p. 107), but it
is more plausible to take the Republic to contain a theory of what is
good for any individual and to presuppose that an ideal state is one that
best promotes the good of its individual citizens. For a full reply to
Popper, see R. B. Levinson, In Defense of Plato (Cambridge, Mass.: Har-
vard University Press, 1953); also see Renford Bambrough, ed., Plato,
Popper, and Politics (Cambridge: Cambridge University Press, 1967). For
a recent overview of Plato's political theory, see George Klosko, The
Development of Plato's Political Theory (New York: Methuen, 1986).
Two older treatments of this topic that are still worth consulting are E.
Barker, Greek Political Theory: Plato and His Predecessors (London:
Methuen, 1918); and the chapter on Plato in George Sabine, A History of
Political Theory, 4th ed. (Hinsdale, 111.: Dryden Press, 1973). Plato's pro-
posal for controlling poetry is discussed by Iris Murdoch, The Fire and
the Sun: Why Plato Banished the Artists (Oxford: Clarendon Press,
1977).
49 One feature of Plato's political philosophy that has received consider-
able attention recently is his thesis that qualified women of the ideal
city ought to hold high political office. Trevor Saunder's contribution to
this volume (Chapter 15) emphasizes Plato's eagerness, even in a
nonutopian work like the Laws, to enlarge the political role played by
women. For recent discussion, see Julia Annas, "Plato's Republic and
Feminism," Philosophy 51 (19761:307-21; Susan Moller Okin, "Philoso-
pher Queens and Private Wives: Plato on Women and the Family," Phi-
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 45
losopher and Public Affairs 6 (1977):345-69; Nicholas D. Smith, "Plato
and Aristotle on the Nature of Women/' Journal of the History of Phi-
losophy 21 (19831:467-78; Gregory Vlastos, "Was Plato a Feminist?"
Times Literary Supplement, March 17—23, 1989, pp. 276, 288-9; Doro-
thea Wender, "Plato: Misogynist, Paedophile and Feminist/' Arethusa 6
50 However, even if Plato is right in thinking that the state should favor a
definite conception of the good (and many philosophers in the liberal
tradition would dispute this claim), he can still be criticized for proposing
too narrow a conception as the one that should be officially endorsed.
51 The closest he comes to such an identification is at Philebus 65 a, where
Socrates says that even if the good cannot be captured by means of one
characteristic, it can be understood in terms of beauty, measure, and
truth. The first two members of this triad are tied by Plato to some
notion of unity. Beauty and measure result when a limit is placed on
what is unlimited and excessive [Phil. 24a-26b), and so goodness (inso-
far as it involves beauty and measure) is conceptually connected with
unity (insofar as what is limited in thereby unified). Note that at Rep.
462a-b the unification of the city is called its greatest good; certainly
Plato would add that the unification of the soul is also its greatest good.
For further discussion, see my essay on the Republic in this volume
(Chapter 10). Aristotle was aware of the view that the one itself is the
good itself: See Metaphysics XIV.4 1091^13-15, and compare Eudemian
Ethics 1.8 i2i8ai5-28. But he does not in these passages attribute this
view to Plato. A student of Aristotle's, Aristoxenus, reported in his
work, Elementa Harmonica II.30-31, that Plato gave a public lecture on
the good, and it is possible to interpret his report to mean that this
lecture maintained that the Good is Unity - although the statement
may instead mean that there is one Good. Contrast the Revised Oxford
Translation in Jonathan Barnes, ed., The Complete Works of Aristotle
(Princeton: Princeton University Press, 1984), 2397, and that of Hans
Joachim Kramer, Plato and the Foundations of Metaphysics (Albany:
State University of New York Press, 1990), 203.
52 This is the view advocated by Gilbert Ryle in several publications. See
"Plato's Parmenides," in Studies in Plato's Metaphysics, ed. R. E. Allen
(London: Routledge & Kegan Paul, 1965), 97-147; Plato's Progress (Cam-
bridge: Cambridge University Press, 1966); "Plato," in The Encyclope-
dia of Philosophy, ed. Paul Edwards, 6:314-33, esp. 324-5. Ryle's view
has won few supporters, for even after the theory of Forms is severely
criticized in the early pages of the Parmenides, Parmenides maintains
that if one refuses to posit the existence of these objects one destroys the
power of discourse (i35b-c). For discussion of his interpretation, see G.
Cambridge Companions Online © Cambridge University Press, 2006
46 THE CAMBRIDGE C O M P A N I O N TO PLATO
E. L. Owen, "Notes on Ryle's Plato/' in Ryle, ed. O. P. Wood and G.
Pitcher (Garden City, N.Y.: Doubleday, 1970), 341-72.
53 This view, forcefully presented by Gregory Vlastos in "The Third Man
Argument in the Paimenides," in Allen, Studies in Plato's Metaphysics,
231-63, has been at the center of scholarly controversy ever since its
initial publication in 1954.
54 For a highly influential interpretation of this sort see G. E. L. Owen,
"The Place of the Timaeus in Plato's Dialogues," in Allen, Studies in
Plato's Metaphysics, 293-338. Owen holds that Plato gives up the view
that Forms are paradigms to which sensible objects have a resemblance,
as well as the view that being and becoming are mutually exhaustive
categories. It is quite possible to accept Owen's general thesis - that
Plato changed his conception of the nature of the Forms - while differ-
ing with him about the nature of that change. For a recent interpretation
of this type, see Kenneth M. Sayre, Plato's Late Ontology: A Riddle
Resolved (Princeton: Princeton University Press, 1983). He holds that
Plato gave up the view that the existence of Forms is independent of the
existence of sensible objects.
55 This is the approach defended by Prior, Unity and Development in
Plato's Metaphysics: "Without altering . . . the Theory of Forms . . . he
augments and clarifies his metaphysics" (p. 2). Similarly, Constance C.
Meinwald, Plato's Paimenides (New York: Oxford University Press,
1991), 171: "Instead of seeing the middle dialogues as a perfected edifice
that the late works then tear down, we can think of the masterpieces of
the middle period as showing the need for work that the difficult final
dialogues in fact take on."
56 Statesman 257a, 258b, 266d, 284b, 286b.
57 The reader may find it helpful to have an integrated list of the various
chronologies suggested at various points in this essay: (A) early dia-
logues: (1) first group (in alphabetical order): Apology, Charmides, Chto,
Euthyphro, Hippias Minor, Ion, Laches, Protagoras; (2) second group (in
alphabetical order): Euthydemus, Gorgias, Hippias Major, Lysis, Menex-
enus, Republic I; (B) middle dialogues (in chronological order): Meno,
Cratylus, Phaedo, Symposium, Republic II-X, Phaedrus, Parmenides,
Theaetetus; (C) late dialogues (in chronological order): Timaeus, Critias,
Sophist, Statesman, Philebus, Laws. This order differs from the one
found in Vlastos, Socrates, 46-7, in only three minor respects: He puts
the Gorgias and Republic I in Ai and the Meno in A2.
58 See "The Place of the Timaeus in Plato's Dialogues," in Allen, Studies
in Plato's Metaphysics, 313-38. This volume contains an influential
reply to Owen by Cherniss, "The Relation of the Timaeus to Plato's
Later Dialogues," 339-78. For further discussion, see Prior, Unity and
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 47
Development in Plato's Metaphysics, 168-93; Sayre, Plato's Late Ontol-
ogy, 256-67; Mueller, "Joan Kung's Reading of Plato's Timaeus." Muel-
ler says (p. 20): "The stylistic arguments advanced by Owen show at
most that the Timaeus may be the first of the late dialogues. . . . There
are . . . no stylistic grounds for treating the Timaeus as part of an intellec-
tual project abandoned by Plato."
59 This is not intended as a hypothesis that Owen would embrace, since he
takes the Timaeus to be an expression of the theory of Forms of the
middle period, in spite of the fact that Socrates has so small a role to play
in it. It may be that the Timaeus rejects some of the doctrines of the
middle period; if so, this escaped Owen's notice.
60 For the view that possessing knowledge of something involves having
the ability to reason about it or to give an account of it, whereas merely
having a belief is compatible with lacking this ability, see Gorgias 465 a,
5ooe-5oia; Meno 98a; Phaedo 76b; Republic 53ie, 534b, (cf. 475c, 493c,
497c, 5ioc, 533c); Timaeus 5id-e. For Plato's puzzles about what an
account [logos] is, see Theaetetus 2oic-2ioa. An extensive discussion of
these puzzles can be found in Myles Burnyeat, The Theaetetus of Plato
(Indianapolis: Hackett, 1990), 128-241.
61 A stylistic feature of the Parmenides and Theaetetus also indicates that
these come after some of the other dialogues of the middle period: At
Tht. 143c, it is announced that the interlocutors will talk directly to
each other, and thus the use of "he said" and the like is avoided. Such
streamlining is also adopted in the Parmenides (starting at 137c). This
suggests that these works were written after such dialogues as the Re-
public, which makes frequent use of phrases reporting dialogue. See
Brandwood, Chronology of Plato's Dialogues, 1, 251. One other chrono-
logical signal deserves mention: At Tht. 183c, Socrates says that he met
Parmenides when he was young. We have no evidence about the birth or
death of Parmenides that would make such a meeting an impossibility,
but since we are taking the theory of Forms to be a product of Plato's
middle age, the conversation depicted in the Parmenides between its
eponymous speaker and Socrates is of course a fiction cast back in time
for dramatic purposes. The reference at Tht. 183c to a meeting between
Parmenides and Socrates is therefore plausibly treated as an indication
that this dialogue was composed after the Parmenides.
62 See De Sophisticis Elenchis 17933, Metaphysics I.9 99obi7, VII. 13
io39a2. For discussion, see Joan Kung, "Aristotle on Thises, Suches, and
the Third Man Argument," Phronesis 26 (19811:207-47.
63 A crucial issue in the debate is whether the Parmenides itself provides a
way of answering the objections raised against the theory of Forms-
even though the dialogue does not make this answer explicit. In her
Cambridge Companions Online © Cambridge University Press, 2006
48 THE CAMBRIDGE COMPANION TO PLATO
contribution to this volume (Chapter 12), Constance Meinwald argues
that in the remainder of the dialogue Plato makes and exploits a distinc-
tion between two different kinds of predication, and that this distinction
provides him with solutions to the problems presented in the first part of
the dialogue.
64 See Ti. iye-28a, 376-380, 49b-5od, 5ie-52b.
65 See Ti. 29b, 48e-49a, sod, 52a, 53c.
66 See e.g. Rep. sooe, 540a; Prm. i32d.
67 For a comprehensive discussion of the major institutions of the Laws,
and its relationship to the political philosophy of the Republic, see the
contribution of Trevor Saunders to this volume (Chapter 15).
68 On the process by which Greek texts were produced and preserved from
antiquity through the Renaissance, see L. D. Reynolds and N. G. Wilson,
Scribes and Scholars: A Guide to the Transmission of Greek and Latin
Literature, 3d ed. (Oxford: Clarendon Press, 1991). For a brief account of
methods of preserving written works in the classical world, and a full
bibliography, see Susan A. Stephens, "Book Production/7 in Grant and
Kitzinger, Civilization of the Ancient Mediterranean, 1:421-36.
69 For an introduction to the problem and a guide to some of the literature,
see Guthrie, History of Greek Philosophy, 5: chap. 8. For a full defense of
approaching Plato's philosophy via the testimony about his unwritten
doctrines, see Hans Joachim Kramer, Plato and the Foundations of Meta-
physics, ed. and trans. John R. Caton (Albany: State University of New
York Press, 1990). This is the interpretive method used by Giovanni
Reale in A History of Ancient Philosophy, vol. 2, ed. and trans. John R.
Caton (Albany: State University of New York Press, 1990). For criticism
of this approach, see Harold Cherniss, The Riddle of the Early Academy
(Berkeley: University of California Press, 1945); Gregory Vlastos, "On
Plato's Oral Doctrine/7 in Vlastos, Platonic Studies, 379-403.
70 This is the view of Cherniss, The Riddle of the Early Academy. He takes
Aristotle to be giving questionable interpretations of what he reads in
the dialogues rather than accurate reports of what he heard. By contrast,
Sayre, in Plato's Late Ontology, argues that the doctrines Aristotle attri-
butes to Plato are accurate reports of doctrines Plato puts forward in the
Philebus, although Aristotle's reports use a different terminology.
71 This is the suggestion of Vlastos, "On Plato's Oral Doctrine/' 397-8.
72 "In none of his dialogues does Plato ever say anything. Hence we cannot
know from them what Plato thought. If someone quotes a passage from
the dialogues in order to prove that Plato held such and such a view, he
acts about as reasonably as if he were to assert that according to Shake-
speare life is a tale told by an idiot, full of sound and fury, signifying
nothing." Thus Leo Strauss, The City and Man (Chicago: University of
Cambridge Companions Online © Cambridge University Press, 2006
Introduction to the study of Plato 49
Chicago Press, 1964), 50. Citing Plato's attitude toward writing in the
Phaedrus and Xenophon's account of Socrates in the Memorabilia,
Strauss goes on (pp. 5 3-5) to suggest that a Platonic dialogue is contrived
to lead the ordinary reader to "salutary opinions" while revealing the
truth to "men possessing the best natures." Plato's own convictions,
therefore, may be quite different from the ones for which Socrates ar-
gues. For example, Strauss takes Plato to be saying in the Republic that
"the just city is against nature because the equality of the sexes and
absolute communism are against nature" (p. 127). See too Rudolf H.
Weingartner, The Unity of the Platonic Dialogue (Indianapolis: Bobbs-
Merrill, 1973), 1-7, for the view that Plato's purpose in composing his
dialogues is not to endorse the view of any one of his speakers.
73 For a treatment of this issue that differs from the one I put forward here,
see the essays collected by Charles L. Griswold, Jr., in Platonic Writings,
Platonic Readings (New York: Routledge, 1988). Many of the papers are
guided by the premise that (as Griswold puts it) "deficiencies, paradoxes,
tensions, and even fallacies in a Platonic dialogue ought to be taken not as
signaling Plato's inability to reason well but as intentionally designed
invitations to the reader to sort through the topic at hand himself" (p. 5).
Note the assumption implicit in this sentence that Plato's reasoning is
defective in all of these ways, and so we must choose between accusing
him of being a bad philosopher and taking the text not as containing an
argument for a thesis Plato himself held but merely as an invitation to
think for oneself. A survey of different methodologies for reading Plato,
with special emphasis paid to the nineteenth and twentieth centuries,
can be found in E. N. Tigerstedt, Interpreting Plato (Uppsala: Almquist &
Wiksell International, 1977). His overview is marred, in my opinion, by
his assumption that because of internal conflicts in every dialogue we
cannot take Socrates to be a mouthpiece for Plato's views. See esp. pp. 98-
9. No doubt there are difficulties of interpretation in every dialogue and
on nearly every page of Plato, but our own difficulties in understanding
the text should not lead us to believe that it is filled with contradictions.
74 For treatments of this issue that are consonant with my own, see the
essay of Terence Irwin in this volume (Chapter 2) and Vlastos, Socrates,
5i-3.
75 This way of reading Plato is suggested by Leo Strauss in Persecution and
the Art of Writing (Glencoe, 111.: Free Press, 1952), 22-37. For criticism
of Strauss's methodology, see Myles Burnyeat, "Sphinx Without a Se-
cret," New York Review of Books 32 (May 30, 1985), 30-6; later issues
contain replies.
76 To support his view that Plato, like many other thinkers, concealed his
views for fear of persecution, Strauss says: "A glance at the biographies of
Cambridge Companions Online © Cambridge University Press, 2006
5O THE CAMBRIDGE COMPANION TO PLATO
Anaxagoras, Protagoras, Socrates, Plato, Xenophon, Aristotle . . . [there
follows a long list of medieval and modern philosophers] is sufficient to
show that they witnessed or suffered, during at least part of their life-
times, a kind of persecution which was more tangible than social ostra-
cism." See Persecution and the Art of Writing, 33. Strauss is certainly
right that Plato witnessed persecution in the form of the trial of Socrates.
But this does not show that he himself concealed his true views in order to
avoid the same fate. One could by the same method of argument show
that all contemporary writers who have witnessed persecution must be
writing in an esoteric manner.
77 Aristotle's Metaphysics, Books XIII and XIV, are filled with accounts of
the differing opinions about mathematical objects in Plato's Academy.
Plato's first and second successors as head of the Academy, Speusippus
and Xenocrates, departed from Plato's views in significant ways. For an
account of their views, see Guthrie, History of Greek Philosophy,
5:457-83.
78 That these features of the dialogues can help us understand their content
has been argued by many scholars. For references, see Charles L. Gris-
wold, Jr., Self-Knowledge in Plato's Phaedrus (New Haven: Yale Univer-
sity Press, 1986), 244-6 nn. 7-8. See too Michael C. Stokes, Plato's
Socratic Conversations: Drama and Dialectic in Three Dialogues (Balti-
more: Johns Hopkins University Press, 1986), 1-36. Paying attention to
scene-setting and characterization is by no means incompatible with
holding that Socrates or some other interlocutor is a mouthpiece for
Plato's philosophy; but if one denies the latter, then these dramatic
features of the dialogues take on all the more significance, for they may
provide clues to the hidden message of the dialogue. Needless to say, just
as it is possible to misconstrue the content of an interlocutor's speech or
its connection with other speeches, so it is possible to misunderstand
the significance of the dramatic features of a dialogue,- and either kind of
mistake may lead to the other.
Cambridge Companions Online © Cambridge University Press, 2006
T. H. IRWIN
Plato: The intellectual
background
I. INFLUENCES ON PLATO
We lack the materials for a proper biography of Plato.1 He hardly
refers to himself at all in the dialogues.2 The ancient ''Lives" are
infected by gossip, legend, and fiction^ and the ostensibly autobio-
graphical Seventh Letter is probably spurious.* Fortunately, however,
Aristotle provides us with important evidence on Plato's intellectual
development. He says that Plato was first influenced by Cratylus the
Heraclitean, and later by Socrates (Met. 987a32-bio). It is unlikely
that Aristotle derived his claim about Craytlus from reading Plato's
dialogues; 5 he probably had some independent source. And since he
was probably well informed about Socrates and Plato, his statement
deserves to be taken seriously.
Aristotle implies that Plato is influenced both by the older, "pre-
Socratic"6 tradition of the "naturalists" (phusiologoi; cf. Aristotle,
De Caelo 28(^25-9) and by the more recent application of philoso-
phy to moral and political questions. What, then, did Plato find
when he looked at these two movements in Greek philosophy?
II. NATURAL PHILOSOPHY AND RELIGION
The dialogues reveal Plato's interest in many aspects of Greek natural-
ist thinking. He mentions Pythagorean mathematical speculation/
Heraclitus,8 Anaxagoras,* Zeno and Parmenides,10 and Empedocles.11
Though he never mentions Democritus by name, he probably some-
times refers to him.12 He also refers to the medical theories that are
I have benefited from helpful criticisms and suggestions by Gail Fine, Richard Kraut,
and Susan Sauve.
51
Cambridge Companions Online © Cambridge University Press, 2006
52 THE CAMBRIDGE COMPANION TO PLATO
often closely related to pre-Socratic speculations.13 What does he ac-
cept or reject from the naturalists?
Naturalists give arguments (logoi), in contrast to the traditional
stories (muthoi) told about the gods by the poets (Aristotle, Met.
ioooa9-2o). Instead of appealing simply to authority and tradition,
they claim to explain natural processes by some rationally convinc-
ing principle or argument. In Aristotle's view, they appeal to a mate-
rial cause, arguing that if these material elements are combined in
the right way, a given result necessarily follows. The appeal to neces-
sity and natural law leaves no room for the traditional Homeric view
that natural processes are basically irregular and unpredictable, and
that gods can interfere with them or manipulate them as they
please.x*
On this basic issue both Socrates and Plato agree with the pre-
Socratics,- in doing so they challenge widespread and deep-seated
religious assumptions of their contemporaries. For in rejecting the
Homeric picture of the irregular universe the naturalists also reject
the view that we sometimes incur divine punishment by failing to
sacrifice the right number of oxen or by fighting on an ill-omened
day, and that we can sometimes placate the gods by offering the right
sacrifices. Traditional and civic religion - from a farmer's sacrifices
to the local nymphs and heroes to the Panathenaic civic procession
bringing a new robe to Athena in the Parthenon^ - was understood
as a means of securing a god's favor by offering gifts,- and people
regularly assumed that a natural disaster or a defeat in a war must
result from some ritual offense.16
Naturalism does not imply atheism. Anaximander and Heraclitus
(among others) regard the world order as a manifestation of divine
justice; they see divine action in the order itself, not (as the Homeric
view suggests) in capricious interference by the gods. But some pre-
Socratic systems, especially the Atomism of Leucippus and Democri-
tus, clearly tend to eliminate any role for a designing or controlling
intelligence; given the motions of the atoms in the void for infinite
past time, and given the laws of their combination, nothing else is
needed (in the Atomist view) to explain the existence, maintenance,
and eventual dissolution of the world order, w
According to Plato, Socrates was interested in naturalism early in
his career, but was disappointed, because naturalists did not try to
explain how the natural order is ordered for the best by an intelligent
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 5 3
designer (Phd. 96a-99d).18 Plato accepts the belief in an intellectual
designer; he criticizes the pre-Socratics for regarding the natural
order as merely the product of "chance" and "necessity" without
any design or purpose (Laws 889a-89oa; cf. Phil. 28c-3oe). When he
presents his own cosmology in the Timaeus, he recognizes two
causes - intelligent design aiming at the best and the nonteleolo-
gical necessity of the "wandering" cause (47c—48e).1* The wandering
cause marks Plato's agreement with nontheistic naturalism, since
he allows that some tendencies of matter are simply brute facts,
with no explanation showing why it is best for them to be as they
are. But in recognizing intelligence as the cause partly controlling
the matter, Plato affirms a theistic view.
In Plato's view, the gods are entirely just and good, with no anger,
jealousy, spite, or lust (Ti. 29c). They lack the desires, aims, and
caprices that might well seem to be essential to the gods who are the
traditional objects of propitiatory cult and sacrifice.20 Plato recog-
nizes this conflict with the tradition; for in his ideal state he advo-
cates a thorough censorship of the Homeric poems and other sources
of the traditional views (Rep. 377b-392a).
Plato's attitude is not completely alien to Greek tradition. From
Homer onward Zeus leads a double life. He is sometimes a god with
ordinary passions and caprices who just happens to be more power-
ful than the rest; but sometimes he is the controller of the universe,
and his designs are above the normal anthropomorphic level of the
Homeric gods.21 Greeks were familiar with the view that the gods
demand justice and punish injustice (in later generations or in an
afterlife); but they had not succeeded in reconciling this view with
the presuppositions of propitiatory sacrifice, which sought to placate
the gods by material transactions independent of the moral char-
acter of the sacrificer.22
Some of the tensions between different elements of the traditional
views can be seen in Euthyphro, a self-styled expert in piety. In
prosecuting his father for causing the death of a slave, he violates a
traditional bond of filial loyalty (whose influence is strong in, e.g.,
the Aeschylean Orestes).2* On the other hand, failure to act might be
taken to show indifference to the pollution resulting from unpun-
ished homicide,- and Euthyphro himself argues that if an injustice
has been committed, the gods demand punishment for it (Euphr.
4b7-c3, 5d8-6a5, 7b7~9). He tries to reconcile his view of the gods'
Cambridge Companions Online © Cambridge University Press, 2006
54 THE CAMBRIDGE COMPANION TO PLATO
demands with his other moral convictions,- but he has not worked
out a satisfactory connection between his religious and his moral
views. At first, he claims that the approval of the gods by itself
determines what is pious (961-3). He is shown that this claim
makes the gods' moral outlook, and therefore the requirements of
piety, the mere product of their arbitrary will; and then he agrees
that the gods demand piety and justice because of the nature of these
virtues themselves, not simply because the gods happen to approve
of justice and piety (iod-nb). Euthyphro is by no means a thought-
less or unenlightened representative of traditional views,- and Socra-
tes' interrogation of him shows that when the moral component of
traditional views receives its proper emphasis and articulation, it
undermines other elements of traditional views.2*
People were wrong, but not completely wrong, to suppose that
Socrates and Plato were abandoning belief in the gods of Athens and
the gods of the Greeks, and thereby shattering people's conception of
what it meant to be Athenians and Greeks.2* In the Clouds Aris-
tophanes presents Socrates as a believer in nonpersonal cosmic
forces, rather than the gods who support morality;26 his charge is
false and probably malicious, but not entirely baseless.
III. NATURALISM, METAPHYSICS, AND
EPISTEMOLOGY
Naturalism could never have been taken seriously if naturalists had
claimed to rely exclusively on the evidence of naive observation and
appearance,- from this point of view it does not seem that natural
processes are as regular as the naturalists claim they are. The natural-
ist must claim to describe some reality that underlies and explains
the appearances, and so must claim to have some cognitive access to
this reality, beyond what is immediately accessible to the senses;
this cognitive access must come through reason, argument, and
theory.
By relying on reason against the senses, we discover the non-
apparent facts underlying the apparent, and in doing so we discover
(according to the naturalists) "nature" (phusis) as opposed to mere
"convention" [nomos)^ "Convention" consists of the beliefs that
rest on mere appearances and that have no basis in "nature," which
is the reality discovered by reason. Democritus states the contrast
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 5 5
sharply. He argues that all the properties recognized by the senses
are mere matters of convention, and have no basis in reality: "By
convention there is sweet, bitter, hot, cold, color, but in reality at-
oms and void;/ (Sextus Empiricus, A.M. VII 135).
While naturalists agree in drawing this contrast between nature
and convention, they do not agree about the character of the reality
that reason discovers. Plato explores some of the questions raised by
two accounts of reality: Heraclitus's claim that there is far more
change and instability than the senses reveal to us, and Parmenides'
claim that there is no change at all.
Aristotle testifies to Plato's early interest in the Heraclitean doc-
trine of flux. According to Plato himself, "Heraclitus says some-
where that everything passes away and nothing remains, and in
likening beings to the flow of a river says that you could not step
into the same river twice" [Cra. 402a).28 Elsewhere he ascribes to
Heraclitus the view that everything "is always being drawn together
in being drawn apart" (Sph. 24262-3). Plato thereby implies that the
doctrine of flux includes two claims.2?
The first claim is about succession of properties in the same sub-
ject over time. Heraclitus argues that there is more change over time
than we suppose there is. The river has been replaced by a different
one when we step into "it" for the second time; for, since it has
different waters, it violates our assumption^0 that X is the same from
time t1 to time t2 if and only if X has the same components at t2 that
Xhad at tx. The same assumption implies that trees, rocks, and other
apparently stable things go out of existence during the time when
we suppose they are stable (since everything is always having some
of its matter replaced).
The second claim about flux is about compresence of opposite
properties in the same subject at the same time. We suppose that
things have stable, fixed, and unqualified properties; for we suppose
that some things are straight and other things are crooked, some
good and others bad, some just and others unjust. In fact, however,
things lack this sort of stability; they are both "drawn together" and
"drawn apart" at the same time (not just at different times, as in the
first kind of flux), and in general opposites are compresent in them.
One and the same letter at the same time is both straight (if it has a
straight stroke) and crooked (if it has a crooked stroke), sea water is
good (for fish) and bad (for human beings), and striking a blow is just
Cambridge Companions Online © Cambridge University Press, 2006
56 THE CAMBRIDGE COMPANION TO PLATO
(if done by an official exacting a punishment) and unjust (if done by
an individual in a private feud).
For these reasons Heraclitus believes in universal flux and instabil-
ity. His claims provoke an extreme reaction in Parmenides, who
rejects the possibility of change altogether (D.K., 28 B 5.7-9).
Parmenides affirms that we cannot speak of, think of, or know, what
is not; but any true cosmology requires the existence of change and
requires us to be able to speak and think of what is not (since change
requires something to become what it previously was not); hence no
cosmology is true.31
Parmenides draws these startling conclusions from premises that
seem self-evident, even trivial. He argues:
1. We cannot think (say, know) and think nothing (since think-
ing nothing is not thinking at all).
2. But what is not (or "not being") is nothing.
3. Hence we cannot think (say, know) what is not.
Parmenides assumes that thinking, saying, and knowing are analo-
gous to other activities referred to by transitive verbs. For to kick or
grasp what is not is to kick or grasp nothing, and so is not to kick or
grasp at all; similarly, it seems obvious that to think or say what is
not is to think or say nothing, and so not to think or say at all; and to
know what is not would be to know what is false, and so not to
know anything at all. To Parmenides' naturalist successors, his argu-
ment seemed largely convincing; and they tried to show in various
ways that their cosmological principles did not require the sort of
reference to "what is not" that Parmenides had challenged.3*
Heraclitus and Parmenides reach their conclusions because they
reject the senses in favor of reason. Parmenides goes further than
Heraclitus; he rejects the evidence of the senses altogether, since
they seem to present a world that includes change. The belief in
change is simply a result of human convention (D.K., 28 B 8.38-9).
Parmenides, however, does not try to abolish cosmology,- instead he
confines it to "belief" or "seeming" (doxa)} he seeks to give the best
account he can of how things appear, while denying that this appear-
ance corresponds to any reality. Parmenides, like Heraclitus, evi-
dently does not intend this skepticism about the senses to extend to
reason as well.
Democritus follows Heraclitus and Parmenides in relying on rea-
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 5 7
son against the senses; but he develops a skeptical argument, from
conflicting and equipollent sensory appearances. If the same water
appears cold to you and warm to me, there is no reason (Democritus
claims) to prefer either your appearance or mine (the two appear-
ances are equipollent); but they cannot both be true (since they are
contradictory), and so they must both be false (Aristotle, Met.
ioo9a38-bi2; cf. Plato, Tht. i52b-c). The same form of argument
applies to all colors, sounds, smells, tastes, temperatures,- and so
things cannot really have any of these properties.
In contrast to ordinary sensible things, the atoms that constitute
reality have, in Democritus's view, only weight, shape, size, and
motion. But this restriction on their properties does not seem to
protect them against skeptical argument. For Democritus claims
that the characteristics of the atoms explain the appearances of sensi-
ble things, because, for instance, sharp atoms produce bitter tastes;
but if sensory evidence is totally unreliable, the sensory evidence
and analogies that support the claim about sharp atoms are appar-
ently worthless. And so Democritus's skeptical argument seems to
undermine his own theory. As he makes the senses say in a conversa-
tion with reason, "Wretched mind, do you take your proofs from us
and then overthrow us? Our overthrow is your downfall" (D.K., 68 B
125; cf. Aristotle, Met. ioo9bn-i2).
The sophist Protagoras reacts to the skeptical argument not by
supposing (as Democritus does) that there is some objective, mind-
independent world that we cannot claim to know, but by rejecting the
basic naturalist contrast between reality and appearance. He claims
that "a man is the measure of all things, of those that are, of how they
are, and of those that are not, of how they are not," and that "as things
appear to each of us, so they are" (Plato, Tht. 152a). According to
Protagoras we should not argue that if the wind appears warm to you
and cold to me, then at least one of us must be wrong; instead we
should conclude that the wind is both warm and cold, and that there
is no objective, mind-independent world. He rejects a presupposition
of Democritus's naturalist argument - the existence of some objec-
tive "nature" that can be contrasted with "convention."
In considering Protagoras we have passed beyond the succession of
naturalists. For Protagoras raises the questions about skepticism not
because he is interested in cosmological speculation, but because he
is concerned with the epistemological issues that affect his views of
Cambridge Companions Online © Cambridge University Press, 2006
58 THE CAMBRIDGE COMPANION TO PLATO
morality and moral education. To understand these views and their
effect on Plato we must turn to the influence of Socrates.
IV. POLITICAL DEVELOPMENTS AND MORAL
QUESTIONS
In Aristotle's view, Socrates turned philosophy away from the study
of nature as a whole to the study of moral and political questions
(Parts of Animals 642 a2 5-31). Greek philosophy began with the appli-
cation of rational, critical, argumentative, nonmythical thinking to
cosmology and to nature as a whole. In the lifetime of Socrates reflec-
tion on morality and human society ceased to be the monopoly of
Homer and the poets; it became another area for critical thinking.33
Critical thinkers began to recognize a potential conflict that had long
been present in traditional Greek thinking about ethics. To see the
sources of the conflict we must review some earlier developments.
The moral outlook of the Homeric poems permanently influenced
Greek thought. 34 Homer expresses the highest admiration for a hero
such as Achilles - well-born, rich, powerful, fiercely jealous of his
own honor, concerned to display his power and status, and compara-
tively indifferent to the interests of other members of his commu-
nity.35 Throughout Greek history the self-absorbed, jealous hero re-
mains an object of fear, resentment, suspicion, and admiration all at
the same time. He appears in Plato's dialogues in the figures of
Callicles and Alcibiades;36 and he appears again later on a larger
scale in Alexander the Great, who modeled himself on Achilles.
Admiration for this sort of Homeric character fits badly with the
attitudes that tended to favor Greek democracy - or indeed any sort
of constitutional system concerned with the interests and rights of
the governed. The first moves of Athens toward democracy involved
equal treatment for rich and poor under written laws; the constitu-
tion of Solon removed politics and law from the whims of aristo-
cratic families. The strengthening of democratic institutions de-
prived aristocrats of a traditional field for the expression of their
competitive impulses. Moreover, the growth of attitudes favorable
to democracy illustrates the remark of the historian Herodotus that
different societies encourage different outlooks and different pat-
terns of education and upbringing (e.g. Herodotus, II 35.2, VII
102.1).37 Herodotus has Athens especially in mind, for he remarks on
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 5 9
how the introduction of democracy in Athens increased the Athe-
nians' enthusiasm for their city (V 78). If it had this effect, it must
correspondingly have tended to create conflicts with some aspects of
the Homeric outlook.
On the other hand, democracy did not require completely uncom-
pensated sacrifice from the upper classes. Between the end of the
Persian Wars in 478 and the outbreak of the Peloponnesian War in
431, Athens reached the peak of its power and prosperity in the
Greek world. It kept a stable democratic constitution (without inter-
ruption from 506 to 411) and established an empire among Greek
states (in the Aegean islands and in Asia Minor) that had been its
allies against the Persians. The stability of the democracy and the
extension of the Athenian Empire were probably not unconnected.
Athens was governed by an assembly in which all citizens (adult,
male, free) were eligible to attend, speak, and vote; but the richer
and nobler families still tended to take a leading role as speakers, as
generals (elected officials with both military and political roles), and
as govenors of the dependent states in the Empire. Contributions
from the dependent states were used not only for their original pur-
pose of defense against the Persians, but also to build temples in
Athens and to pay Athenians for sitting on juries (large courts,
chosen by lot from the citizens). In this way the Empire both paid for
some aspects of Athenian democracy and offered the upper classes a
constructive way to display their Homeric ambitions and competi-
tive spirit. Aristocrats like Cimon, Aristeides, and Pericles could
compete for leadership in a great city with extensive military and
political responsibilities overseas,- and less preeminent members of
the upper classes could hope to govern a subject city in the Empire.
In 431 the Peloponnesian War broke out between Athens and
Sparta. It lasted for twenty-seven years (with interruptions), includ-
ing the first twenty-four years of Plato's life.*8 Athens was eventu-
ally defeated, partly as a result of treachery by an oligarchic fifth
column. After a long war, Athenian resources of money and man-
power were severely strained. The strains encouraged opponents of
democracy to plot, with Spartan help, to set up an oligarchic regime
and to abolish the democratic assembly and jury-courts. The first
result of these plots was the short-lived regime of the Four Hundred
in 411-10; the second result was the regime of the Thirty, which
came to power (with Spartan help) after the end of the war in 404 and
Cambridge Companions Online © Cambridge University Press, 2006
6O THE CAMBRIDGE COMPANION TO PLATO
was defeated by the supporters of democracy (with Spartan help,
after a change of kings in Sparta) in 403. This was the oligarchical
junta that included two of Plato's relatives, Critias and Charmides
(who appear in the Charmides).^ These two were also associates of
Socrates; and Socrates7 connections with such dubious characters
probably help to explain why he was tried and convicted, under the
restored democracy, of not recognizing the gods of the city and of
corrupting the young men.4°
Thucydides' history suggests how the Peloponnesian War may
have affected moral and political attitudes. Thucydides interprets
the war as a manifestation of the conflicts resulting from antago-
nisms between the perceived interests of different classes and groups
in a state. He suggests that a relatively stable state is the product of
some force strong enough to keep the peace and to assure some
protection for the different groups; but when one or another group
sees a chance to take the dominant place, it takes the chance (III
82.2; V 89, 105.2). Since war involves an external power willing to
support a revolution, it tends to increase political instability within
a state. Thucydides describes the civil conflict in Corcyra (esp. Ill
82-5) that resulted from Athenian support for the democrats and
Spartan support for the oligarchs. He intends this to indicate the
pattern followed by civil wars all over the Greek world - and eventu-
ally Athens itself. In these circumstances, according to Thucydides,
the basic tendencies of human nature - the desire for security for
oneself and domination over others - inevitably come to the surface.
The Peloponnesian War created the sorts of tensions in Athens
that would appear to support Thucydides' analysis. Obligations to
the community required greater sacrifice and presented a clearer
conflict with the self-seeking "Homeric" pursuit of one's status,
power, and pleasure. In political terms people had to decide whether
or not to plot against the democracy to bring off an oligarchic coup.
In moral terms they had to decide whether or not to ignore the
demands of the community, summed up in the requirements of "jus-
tice," in favor of their own honor, status, power, and in general their
perceived interest. Plato was familiar with people who preferred self-
interest over other-regarding obligation; his own relatives, Critias
and Charmides, made these choices when they joined the Thirty
Tyrants.
Arguments from natural philosophy did not restrain people like
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 61
Critias and Charmides. Democritus argues unconvincingly that the
requirements of justice and the demands of nature, as understood by
Atomism, can be expected to coincide.*1 Protagoras rejects the view
that moral beliefs are true and well grounded only if they correspond
to some reality independent of believers; admittedly they are mat-
ters of convention, but so are all other beliefs about the world. This
line of argument removes any ground for preferring nature over con-
vention, but at the same time seems to remove any rational ground
for preferring one convention over another. Matters of convention
are the product of human agreements and cannot be taken to rest on
any basis independent of these agreements. Now it seems obvious
that some provisions of law and other moral and social norms are
matters of convention, for they are established by human enact-
ment, differ from one society to another, and can be changed by new
legislation. It is easy to infer that they have no standing in any
independent reality, and that if the requirements of other-regarding
morality conflict with the demands of self-interest, there is no rea-
son to pay attention to other-regarding morality. Self-interest might
appear to be nonconventional, determined by human nature, and
therefore entitled to override the purely conventional demands of
morality.
This challenge to other-regarding morality and justice is most
easily seen in some passages in Thucydides, in Antiphon the soph-
ist, and in Plato's characters Callicles (in the Gorgias) and Thrasy-
machus (in Republic Book I); and it is presented in comically exagger-
ated form by the "Unjust Argument" in Aristophanes' Clouds
(1075-82).^ Plato might reasonably conclude that none of the philo-
sophical outlooks of his naturalist contemporaries and predecessors
promised any convincing defense of other-regarding morality.
V. POLITICAL ISSUES
The second half of the fifth century provides the dramatic setting of
the dialogues, and it is the appropriate background for understanding
many of Plato's moral and political reflections. But he wrote the
dialogues during the rather different circumstances of the fourth
century. The democratic regime that was restored in 403 lasted
through the rest of Plato's lifetime, past his death in 347, and even
beyond the conquest of Greece by Alexander (who died in 323). Plato
Cambridge Companions Online © Cambridge University Press, 2006
62 THE CAMBRIDGE COMPANION TO PLATO
did not know, as we do, that he was living in the last years of
Athenian power and independence. On the contrary, he is likely to
have been impressed - and quite rightly - by the stability of democ-
racy in Athens. In the fourth century the essential institutions of
democracy remained, even without an empire to pay for them, and
the burden of paying for them must have fallen more heavily on the
richer citizens; still, there were no further attempts at an oligarchic
revolution.^
These features of fourth-century Athens perhaps help to explain
Plato's attitude to existing political systems. He does not offer pro-
posals for the reform of democracy; ** nor does he advocate a violent
antidemocratic revolution of the sort that his relatives had at-
tempted at the end of the Peloponnesian War. However much he
objects to democracy, he assumes that, practically speaking, the
Athenian democracy is stable, and that no feasible alternative is
likely to be superior.
In dismissing practical alternatives to democracy, Plato refuses to
advocate oligarchy on the Spartan model; and on this point he stands
apart from the oligarchs of the fifth century. The main rival to demo-
cratic Athens, with its relatively free, unregulated, tolerant, and
open-minded social and cultural atmosphere, 45 was the rigidly con-
trolled, militaristic, and oligarchic society that developed in Sparta.*6
Plato certainly admires some aspects of Sparta - in particular its sys-
tematic and rigorous policy of state-supervised upbringing, educa-
tion, and indoctrination that regulated every aspect of life for the
ruling class; but this admiration does not lead him to admire the
moral and political outlook that underlay the Spartan way of life, or to
suppose that it would be better to replace the Athenian democracy
with a constitution modeled on Sparta. While he argues that the
"timocratic" type of constitution found in Sparta (Rep. 547b-548d)
is, as such, superior to democracy as such, this does not lead him to
advocate an attempt to imitate Sparta. His admiration for Sparta is
excessive and misguided, but still highly selective and critical.*? The
disastrous experiments of two pro-Spartan regimes - first the Four
Hundred and then the Thirty-had shown that oligarchy did not
arouse broad enough support in Athens to be maintained without
force, intimidation, and foreign military aid. Plato might fairly con-
clude that critics of democracy needed to reject the crude, brutal, and
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 63
ultimately self-defeating tactics that had given oligarchy a bad name
in Athens.*8
Plato's doubts about the prospects of gradual reform or a purely
political revolution rest partly on his views about the sources of
political conflict. For existing cities he agrees with part of Thucydi-
des' analysis. In Plato's view, every city contains the sources of the
instability that sometimes breaks out in open struggle; for each of
them is really not just one city, but two - the city of the rich and the
city of the poor (Rep. 42id-422a, 422e-423b). Here Plato recognizes
the conflict of perceived interests that results in class conflict and,
in the appropriate circumstances, in civil war and revolution. He
even agrees with Thucydides' view that people's perceived interests
will conflict as long as the dominant class consists of either the rich
or the poor. He argues, however, that class conflict is not inevitable;
to avoid it, the ruling class must be removed from the conflicts that
result from private property, and must educate the other citizens in a
true conception of their interests Though Plato's proposals for re-
solving class conflict are certainly open to objection, his diagnosis of
the conditions that need to be removed is defensible in the light of
Greek, and especially Athenian, historical experience.
These political problems lead Plato back to ethical problems. The
conflict between justice and self-interest for the individual, and the
conflicts between the interests of different groups and classes within
a state, both result from a particular conception of the interests of
individuals and groups. Plato seeks to show that a correct under-
standing of human interests and welfare will show why neither con-
flict needs to arise.
VI. THE SOPHISTS
Socrates and Plato could not take it for granted that moral and politi-
cal questions were appropriately treated by philosophers - by those
who recognized some allegiance to the forms of inquiry begun by the
naturalists. They had to define the subject matter and methods of
philosophy in contrast to the claims of other outlooks and ap-
proaches. The two main rival approaches that Plato confronts are
those of the sophist and the rhetorician. We must see why he regards
them as serious rivals whose claims need to be disputed.
Cambridge Companions Online © Cambridge University Press, 2006
64 THE CAMBRIDGE COMPANION TO PLATO
Once democracy was firmly established in Athens, the popular
assembly became used to exercising its power to make the vital
decisions and to scrutinize the conduct of political leaders. A suc-
cessful politician had to be a good speaker who could present a
convincing case to a critical audience.*0 Since the Athenians had no
newspapers, radio, or television, they relied on political speakers for
information as well as for advice about what to do. A speaker who
was ill-informed or incapable of using his information to reason
cogently could expect to be beaten by someone who could show
himself to be a better-informed and more reliable adviser. Gradually
the role of "politician" (politeuomenos) became more professional,
and the trained and well-informed speakers tended to dominate de-
bate in the assembly.*1 Systematic instruction that would help in
these areas would be useful to the aspiring political leader. This
instruction was provided by sophists.
"Sophist" [sophistes, derived from sophos, "wise") occurs in fifth-
century Greek as a nonpejorative term applied to experts in different
areas.52 In the second half of the century the term is applied espe-
cially to teachers offering higher education for fees. The education
varied in content from sophist to sophist, but its main goal was to
equip someone to take an active part in public life [Pit. 3i8d-3i9a).
The leading sophists traveled from city to city and gained an interna-
tional reputation as "stars" in Greek cultural life. The Protagoras
(3O9a-3i4e) describes the excitement aroused in upper-class Athe-
nian circles by the visit of the eminent sophist Protagoras, and the
sense of anticipation among his potential students.
Sophistic also aroused suspicion, especially among people who
thought that someone's birth, family, and gentlemanly upbringing
gave him a right to be listened to. From this conservative point of
view, sophistic training might seem to make people too clever by half,
and the sophists might be accused of teaching unscrupulous people
the skills they needed to achieve undeserved success. This is the
attitude of Anytus in the Meno (9 ia~92e); and it helps to explain why
Socrates at his trial suggess that his accusers want to arouse prejudice
against him by accusing him of being a sophist (Ap. i9d-e).
Many modern readers have supposed that Plato blames the soph-
ists as a group for defending a specific theoretical position that he
takes to be responsible for some decay in moral standards that he
sets out to correct.53 In particular, it is sometimes supposed that the
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 65
rejection of conventional morality by such speakers as Callicles in
the Gorgias and Thrasymachus in Republic Book I is a typical result
of sophistic teaching.54
There is no foundation, however, for this view of the sophists, or
for this view of Plato's objections to them. Certainly some sophists
hold the position of Callicles and Thrasymachus; Antiphon seems
to have held such a position.55 Other sophists, however, held quite a
different position. Protagoras was a firm defender of conventional
justice and morality; Plato never denies that this is Protagoras's
view, and never suggests that Protagoras's teaching tends to turn
people against conventional morality. In the Meno Socrates point-
edly dissociates himself from Anytus's indiscriminate hostility to
the sophists. In the Protagoras Plato's portrait of leading sophists is
sometimes humorous (Prt. 3i5c-d; cf. H. Ma. 28ia-283b), but never
hostile. Both in the Protagoras and in the Theaetetus Protagoras is
taken seriously; indeed Plato defends his views against premature
dismissal and points out that Protagoras can be defended against
Socratic objections that might satisfy us too easily [Prt. 35oc-35ib,
Tht. i65e-i68c).
Plato criticizes the sophists not primarily for their conclusions,
but for the arguments they rely on. He denies that the sophists are
the main influence on moral and political education; on the con-
trary, he argues, the prejudices of the masses determine the range of
acceptable views, and the sophists simply repeat these prejudcies
(Rep. 493a). This description most obviously fits Protagoras, whose
whole epistemological position is designed to show that the views
that appear true to the many are true (cf. Tht. 167c). But Plato thinks
it also fits the sophists in general; they do not attempt to found their
views on any rational basis that goes beyond the unexamined beliefs
and prejudices of the majority. This is why he connects sophistic
with "appearances" and "images" (Rep. 5i5a5-6; Sph. 232a-236d).
By criticizing sophists Plato sets standards for his own philosophi-
cal inquiries. At first sight it is not easy to distinguish Socrates and
Plato from sophists. For Socrates' characteristic method of inquiry is
a systematic cross-examination of an interlocutor, seeking to expose
conflicts in the interlocutor's views and to reconstruct his beliefs as
a result of reflection on the conflicts and on possible resolutions of
them. This sort of inquiry clearly begins from commonsense beliefs
and seems to rely on them at each stage,- for Socrates often insists
Cambridge Companions Online © Cambridge University Press, 2006
66 THE CAMBRIDGE COMPANION TO PLATO
that the interlocutor must state his own view (instead of maintain-
ing something for the sake of argument to avoid refutation; Pit.
331c). Why is this not the uncritical appeal to popular views that
Plato criticizes in the sophists?
The use of cross-examination and refutation does not distinguish
Socrates and Plato from the sophists. Though Protagoras is repre-
sented as being unfamiliar with Socratic conversation [Pit. 334c-
335c), he was familiar with techniques of destructive argument.*6
The Euthydemus is an exhibition of the techniques of eristic
(eristikos, "contentious") argument, a technique practiced by some
sophists.57 Techniques of cross-examination and refutation are obvi-
ously useful in debate and argument; and the young men who
learned them from Socrates enjoyed practicing them on others.58 But
they do not distinguish Socrates from an eristic. In the dialogues his
interlocutors sometimes accuse him of using eristic techniques;59
and Plato's opponents describe him as an eristic.60 If Plato tries to go
beyond eristic to constructive argument, he needs to show how he
has something more to offer than the conventional beliefs that are
the sophist's stock-in-trade. In trying to distinguish Socratic and
Platonic method from sophistic, we raise some far-reaching ques-
tions about Plato's epistemology.
VII. RHETORIC
Sophistic was closely connected with another development in higher
education: the growth of rhetorical theory and teaching.61 Many soph-
ists probably included some rhetorical training in their courses of
study; but there still seems to have been a distinction between a
rhetorician and a sophist. Rhetoricians concerned themselves primar-
ily with techniques of persuasion, and not with the general moral and
political education promised by the sophists.62 This does not mean
that they were wholly concerned with rhetorical "form" rather than
"content"; they advised their students that this opinion rather than
that one was likely to be well received. But they did not engage profes-
sionally in the concerns of the sophist and the philosopher.
One of the most influential rhetoricians among Plato's contempo-
raries was Isocrates, whose speeches contain many implicit and
explicit attacks on Plato. Isocrates regards training in rhetoric as
"philosophy" in the truest sense (Antid. 50, 270, 285). It differs,
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 67
however, from the studies practiced by naturalists, mathemati-
cians, and eristics, because it is practically relevant and applicable.
Isocrates accuses Plato of formulating absurd paradoxes about mo-
rality (Helen 1). In general, "plausible belief about useful things is
far superior to exact knowledge about useless things" (Helen 5). It
is pointless, in Isocrates' view, to examine the foundations of moral
and political theory,- one should stick to the recognized virtues
(Antid. 84-5) and present conventional views about these in a per-
suasive and attractive form.
Plato's remarks about Isocrates are not completely hostile. He
regards him as something between a philosopher and a politician,
intellectually promising but unable to distinguish genuine dialectic
from eristic (Euthd. 3O5b-3o6c).63 He suggests that Isocrates has
some philosophical ability (Phdr. 279a); but Isocrates' account of his
version of "philosophy" exposes clearly the ways in which Plato
thinks Isocrates misses the point.
The political role of rhetoric arouses Plato's suspicion and criti-
cism. He asks why the Athenian democracy should regard rhetorical
ability as a sufficient qualification for giving political advice, and he
presents two objections against rhetoric: (1) If, as Isocrates admits,
the orator does not try to reach independent rational convictions of
his own on moral and political questions, he will simply repeat
popular prejudices. If he simply follows the ignorant and prejudiced
moral and political assumptions of the majority, his advice will not
promote the common good. (2) If the orator persuades people, not
because he convinces them that the course of action he advises will
really benefit them, but because he arouses their feelings and preju-
dices, even against their better judgment, what he persuades people
to do will not even be what they want to do.
In his first charge against rhetoric Plato argues that the method of
democratic government undermines its stated goal-to govern in
the interest of all the citizens. In his second charge he argues that
government by rhetorical persuasion does not even execute the will
of the majority. The rhetorician would be bad enough if he simply
expressed the views that people, after reflection, actually hold; but
he is even worse if in fact he does not simply express public opinion,
but molds and manipulates it for his own purposes.
In attacking rhetoric, Plato also attacks a much older Athenian
institution, tragic drama (Gig. 502b, Rep. 6o2c-6o6d).64 His objec-
Cambridge Companions Online © Cambridge University Press, 2006
68 THE CAMBRIDGE COMPANION TO PLATO
tions are easier to understand if we remember that the Athenian
dramatic festivals took the place of some of the mass media familiar
to us. Plato recognizes the cultural influence of tragedy, and assumes
that people are influenced by the moral views expressed in and
through the plays. He criticizes tragedy as a form of rhetoric; it
makes particular moral views appear attractive to the ignorant and
irrational audience, and it is written by writers who do not under-
stand the moral questions any better than the audiences do. Euripi-
des arouses our sympathy for Medea not because he can rationally
convince us to see on reflection that Medea deserves our sympathy,
but because he presents some features of her situation in ways that
appeal to our prejudices. The tragedians do not know what sort of
person in what sort of situation ought to arouse admiration, sympa-
thy, or revulsion.6*
Plato's criticisms do not show that there could be no legitimate
use for rhetoric, or that everyone who practices it is morally mis-
guided.66 But they raise some legitimate questions about the particu-
lar social, educational, and political role of rhetoric in contemporary
Athens. Some of Plato's opponents, and most notably Isocrates, pre-
sented rhetoric as a sufficient moral education for a good citizen
aspiring to a leading role in public life. Plato points out that the
student of rhetoric learns the moral and political assumptions that
will seem plausible and attractive, but learns to scorn any system-
atic thinking about whether these are the right assumptions or not.
Ancient Athens is not the only society that has allowed skill in
nonrational manipulation to be a dominant influence in democratic
debate; and for that reason the force of Plato's criticisms is not
confined to his own historical situation.
VIII. SOCRATIC INQUIRY
Plato's earlier dialogues present Socrates' attempts to answer the
moral questions that arose from the conflicting views of his contem-
poraries. Socrates relies on philosophical argument, and in particular
on the sort of systematic questioning and refutation that had been
begun by Zeno.6? But in contrast to Zeno and the eristics, his aim is
not purely negative. He argues constructively, in support of his own
paradoxically uncompromising defense of the moral virtues. His ar-
gument is philosophical but he distinguishes himself sharply from
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 69
naturalists. He does not rely on premises derived from natural phi-
losophy, which (as the example of Democritus might easily suggest)
might appear both dubious in themselves and unlikely to answer the
most important questions about morality.
Socrates argues that we can be rationally convinced that there is
really no conflict between justice and self-interest; the whole "Ho-
meric" conception of self-interest, and therefore the assumption
that self-interest must conflict with morality, can be shown (in his
view) to rest on assumptions that we find, on reflection, we reject.
Socrates seems to Plato to promise a method and a line of argu-
ment that might explain and justify morality. But Plato sees that
Socrates' promises are not completely fulfilled. Socrates wants to
distinguish Socratic method from the methods of sophistic, eristic,
and rhetoric; and in some transitional dialogues Plato examines
these claims and tries to develop a theory of Socratic argument that
will justify its claims to arrive at objective truth.
More generally, Plato's early association with Cratylus suggests to
him that he must go beyond Socrates and take up the wider philo-
sophical questions that Socrates had set aside in order to concentrate
on ethics. Socrates' efforts to define the virtues assume that objec-
tively correct answers can be found, and that they must correspond
to some objective realities independent of our beliefs and inquiries.
But what sorts of objective realities could correspond to our moral
beliefs? And even if we can conceive what the relevant realities
might be like, how can we reasonably suppose that we know any-
thing about them?
For these reasons Plato finds himself returning to some of the
metaphysical and epistemological preoccupations of the Pre-Socra-
tics; but whereas the Pre-Socratics are forced into them by questions
arising in the study of nature, he is forced into them by questions
about morality.
IX. BEYOND SOCRATES
A full account of Plato's treatment of the metaphysical and episte-
mological questions that he derives from naturalism would be a full
account of most of his middle and later dialogues. I will confine
myself here to a few remarks on his treatment of the questions in
Heraclitus, Parmenides, and Protagoras that I discussed earlier. In
Cambridge Companions Online © Cambridge University Press, 2006
7O THE CAMBRIDGE COMPANION TO PLATO
each case Plato's treatment shows his developing recognition of the
interest and depth of the questions raised by his predecessors.
Plato's reflections on Socrates' search for definitions make him
aware of the second type of Heraclitean flux, involving the corn-
presence of opposites (the writing that is straight and crooked, the
water that is good and bad, and so on). Plato points out that ordi-
nary observable examples of just or good actions or people turn out
to suffer compresence of opposites; bright color, for instance, is
both beautiful (in some contexts) and ugly (in others), and giving
back what you have borrowed is both just (in normal circum-
stances) and unjust (if a suicidal person asks you to return his
sword).68
Plato, however, rejects the Heraclitean conclusion that justice it-
self suffers from the same sort of flux and instability. The examples
just given are the sort that induce Heraclitus to infer that justice is
both paying your debts and not paying them. Plato, on the other
hand, infers that a different sort of account of justice is needed.
Though observable types of just actions are in flux from just to
unjust, it does not follow, he argues, that justice itself is in flux.
Reflection on Heraclitus's problems and examples leads Plato to the
sharply non-Heraclitean conclusion that justice itself must be ex-
empt from the compresence of opposites. An adequate definition of
the Forms (of Justice, Bravery, etc.) that Socrates was trying to define
must, in Plato's view, show that the Forms display Parmenidean
stability rather than Heraclitean flux.
In a later dialogue, the Theaetetus, Plato examines the first kind of
Heraclitean flux, involving change over time in (what we naively
take to be) one and the same object. He argues that if the doctrine of
total flux is accepted without qualification, so that we deny the
existence of persisting subjects altogether, it actually refutes itself;
for we cannot say what is in flux, and since we cannot speak of flux
without saying that something is in flux, the extreme doctrine of
flux cannot be true unless it is falser Once again Plato finds
Heraclitus's doctrine of flux a stimulating point of departure toward
a strongly non-Heraclitean conclusion.
In the Republic Book V Plato presents his first reaction to Parme-
nides. He agrees (in a sense) with Parmenides' claim that we cannot
know what is not, but he disagrees (in a sense) with his claim that
we cannot speak or think of what is not. His disagreement with
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 71
Parmenides rests on implicit distinction between different ways in
which Parmenides speaks of "what is not." It is plausible to claim:
1. We cannot hit or kick what is not (= the nonexistent).
2. We cannot think or say what is not (= something that has no
content at all); all saying or thinking must be saying or think-
ing something.
3. We cannot know what is not (= what is not true).
Now Parmenides' argument rules out change only if "what is not" is
taken in the existential sense, as in (1); but since the crucial prem-
ises (2) and (3) seem to involve only the predicative sense, as in (2), or
the veridical?0 sense, as in (3), his argument seems to be invalid. In
the Republic Book V Plato claims that while we cannot know what
is not (in sense [3]; knowledge that p implies that p is true), we can
believe what is or what is not (since beliefs include true and false
beliefs); and believing what is not does not constitute believing noth-
ing (having no belief).?1 But though Plato implicitly rejects Parme-
nides' account of being, and hence his views about the objects of
knowledge and the possibility of change, the Republic does not ex-
plain where Parmenides has gone wrong.
Some explanation is needed, however; for Parmenides is not re-
futed by a simple appeal to different senses of the verb "to be.;/ He
can remove any damaging appearance of equivocation on senses of
"to be" if he can argue for a particular account of what is involved in
thinking or speaking or knowing. If he can show that they are suffi-
ciently similar to grasping and kicking and other interactions with
external objects, then he can argue that speaking or thinking of what
does not exist is really speaking or thinking of nothing, and therefore
is not really speaking or thinking at all. And some arguments for
such a conclusion seem quite plausible. We might suppose that if
thinking gives us genuine knowledege of external reality, it must
make some real contact with this external reality, in the way the
senses make contact with it; and for the senses something like
Parmenides' conclusion is true, since we cannot see or hear or touch
the nonexistent. Something similar seems to be true if we try to
understand speaking of something. We can use words as we can use
gestures to point out and identify something; and just as I cannot
point out something that is not true, I apparently cannot name what
is not there. And so if we look at certain aspects of thinking, saying,
Cambridge Companions Online © Cambridge University Press, 2006
72 THE CAMBRIDGE COMPANION TO PLATO
and knowing, it seems natural to agree with Parmenides' assump-
tion that they consist in some direct causal interaction with some
external object, and therefore require an existing object.
If a particular conception of thinking, saying, and knowing tempts
us to accept the premises that imply Parmenides' startling conclu-
sions, then a proper reply to Parmenides should present a conception
of thinking, saying, and knowing that removes any temptation to
make the fatal concessions to Parmenides. Once we have the right
conception, then we should be able to see that "You cannot think
(say, know) what is not" is true only in a sense that does not lead to
Parmenides' conclusions.
Plato's claims about knowledge and belief in the Republic Book V
show that he thinks the appropriate alternative to Parmenides' posi-
tion can be worked out. But he does not work it out in the Republic.
In the Theaetetus he reexamines the Parmenidean view that since
belief is like seeing or grasping, we cannot have beliefs about the
nonexistent (Tht. i88a-i89b). Here he explains why false belief will
be impossible if we accept a Parmenidean view of belief; and so he
asserts that we must be able, in one sense, to speak of what is not. In
the Sophist the "Eleatic stranger" remarks that the question of how
things can appear, but not be, and how one can speak of something
but speak falsely "has always been full of puzzlement, in previous
times and up to the present"; and then he introduces Parmenides'
rejection of not-being (Sph. 236e-237a). The rest of the dialogue
seeks to explain how, and in what sense, it is possible to speak and
think of what is not. The main speaker is a visitor from Elea, the
home of Parmenides and Zeno; his presence indicates the impor-
tance that Plato attaches to Parmenides' views. On the other hand,
the visitor sharply emphasizes the basic disagreement with Parme-
nides that must result from a full inquiry into not-being.
The challenge of skepticism is important for Plato, since, follow-
ing Socrates, he believes in the possibility of knowledge, which the
skeptic denies. It is equally important for him to reject Protagoras's
solution to the question raised by skepticism. For Plato believes in
the existence of a knowable mind-independent reality, whereas
Protagoras believes that we can refute skepticism only if we agree
with the skeptic that there is no knowable mind-independent real-
ity. In developing his theory of Forms, Plato makes it clear that he
rejects both skepticism and Protagoras's solution; but in the early
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 73
and middle dialogues Plato simply assumes the falsity of Protago-
ras's position and discusses it only briefly.?2
In the Theaetetus, however, Plato discusses both Protagoras and
Heraclitus at some length. He argues that Protagorean epistemol-
ogy rests on an indefensible metaphysics; for it leads to the self-
refuting extreme Heraclitean doctrine about change. To this extent
his attitude to Protagoras is similar to his attitude to Heraclitus
and Parmenides; Plato realizes fairly late in his career that the
questions raised by his predecessors require a direct and fundamen-
tal examination.
The dialogues bear out Aristotle's claim that reflection on So-
cratic inquiry about ethics leads Plato back to the study of questions
derived from the naturalists; throughout his career Plato remains a
careful and appreciative critic of his predecessors. But while he faces
some of the same questions, Plato thinks he can avoid the skeptical
conclusions that seem to threaten the foundations of Presocratic
naturalism. For he thinks we have enough firm and reliable convic-
tions about some moral questions to justify us in arguing from these
convictions to whatever conditions are required for their truth.
X. THE PLATONIC DIALOGUE
In his reflections on his predecessors and his contemporaries Plato
not only had to decide what to say in defense and explanation of
Socrates,- he also had to decide how to say it. In distinguishing him-
self from naturalists, sophists, and rhetoricians, he also chose a liter-
ary form that sets his work apart from theirs.
The modern reader, used to the surviving Aristotelian corpus^
and to the literary form of later philosophical works, naturally won-
ders why Plato chose to write dialogues rather than treatises. This
question may rest on the false presupposition that it would have
been more natural for him to choose to write continuous treatises.
Since Plato was a pioneer in writing moral philosophy (as opposed to
including some remarks on morality within a treatise on natural
philosophy), there was no established literary form for the sort of
thing he was trying to write.™ Even natural philosophy had no fixed
literary form.75 Among the earlier Pre-Socratics, Parmenides and Em-
pedocles wrote in epic verse, while Heraclitus apparently expressed
his views (at least sometimes) in aphorisms, maxims, riddles, and
Cambridge Companions Online © Cambridge University Press, 2006
74 THE CAMBRIDGE COMPANION TO PLATO
paradoxes. The Pre-Socratics offered a wide choice of literary forms;
and in any case Plato had no special reason to believe that any of
these literary forms used for natural philosophy was the right way to
develop arguments in moral philosophy.
Herodotus and Thucydides (and no doubt other historians whose
works have not survived) may have suggested to Plato some aspects
of the literary form of the dialogues. Both historians insert speeches
ostensibly delivered on particular occasions to advocate a particular
decision by a king, or generals, or an assembly, or to encourage
troops before a battle. They also use fictitious speeches and debates
to explore the moral and political issues involved in a particular
incident or situation. Herodotus inserts a debate on different Greek
political systems during an episode in Persian history (Herodotus, III
80-2); and Thucydides inserts an elaborate dialogue between Athe-
nians and Melians about whether the Athenians ought to massacre
all the Melians for their support of Sparta (Thucydides, V 84-113).
The historian probably does not intend to tell us what was said on
that occasion by that particular speaker; he wants to draw our atten-
tion to the moral and political issues raised by situations of that
kind.
The same use of debates to present the different sides of moral and
political questions is an obvious feature of Athenian tragedy. ?6
Tragic characters often face difficult decisions and debate their
rights and wrongs; and especially in the plays of Euripides and the
later plays of Sophocles,?? the debates are quite elaborate, theoreti-
cal, and sometimes rhetorical. Plato's audience was used to dialogue
as a medium for exploring moral questions.
Athenian comedy is equally relevant. The interlocutors in Plato's
dialogues are not the heroic figures from the distant and legendary
past who appear in tragedies. They are Socrates' upper-class contem-
poraries, in conversation with Socrates, who is himself in many
ways an ordinary (even exaggeratedly ordinary) plain-spoken person.
For the dramatic presentation of such characters Plato might reason-
ably turn to comedy.
A comic hero in Aristophanes - Dikaiopolis in the Acharnians
provides a good example - is usually unheroic, outside the circle of
Athenians who distinguished themselves by their wealth, breeding,
and military and political careers. Though he appears to be an igno-
rant and vulgar peasant, Dikaiopolis in fact understands (according
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 75
to the play) why the Peloponnesian War broke out, while his "bet-
ters" do not; and he stands alone against them in concluding his
own peace treaty with the Peloponnesian enemy. The "establish-
ment" figures mentioned in the play - including the "Olympian"
Pericles who started a war for the sake of a few prostitutes and the
boastful, empty-headed general Lamachus - are shown to be preten-
tious and foolish. Dikaiopolis outwits his "betters" by his clever
tongue and his understanding of the situation.
After he makes his private peace with the Peloponnesians, Dikaio-
polis begins his own idyllic peacetime existence, fulfilling the
dreams of Athenians who were suffering from the rigors of war. This
fantastic element is developed further in the Utopian Cloudcuckoo-
land of the Birds. This use of Utopia and fantasy to make a serious
moral and political point may well have helped to suggest to Plato
that the description of a Utopia (not without some humorous ele-
ments; cf. Rep. 372c-d) would be an effective way to present some of
his own moral and political views.78
While details cannot be pressed, these features of comic heroes
suggest that Athenians who had laughed at the foolishness of Aris-
tophanes' Athenian generals and cheered for Dikaiopolis or Lysi-
strata would be able to appreciate the comic aspects of Socrates and
the other characters in the dialogues. Many of the interlocutors are
quite strongly characterized in ways that make them suitable for
deflation. Often they are Socrates' social superiors - the aristocrats
in the Laches and Charmides, the leading intellectuals and experts
in the Protagoras, Gorgias, Ion, and Hippias Minor and Major. Often
they begin with a rather complacent, even patronizing, attitude to
Socrates; but it eventually turns out that he understands more than
they do. Socrates says that while he cannot claim to know that his
views are true, he has found that anyone who rejects them has
turned out to be "laughable" [Grg. 509a; cf. Prt. 355a6); Plato under-
lines the comic side of his dialogues, the unlikely hero deflating the
pretensions of people whose reputation exceeds their understanding.
Aristophanes enables us to understand the comic aspects of the dia-
logues better, not merely because the dialogues are sometimes
funny, but because a particular type of comic situation that occurs in
Aristophanes provides one of the most important elements - comic
and serious at the same time - of the dialogues.™
These precedents from historical and dramatic works may help to
Cambridge Companions Online © Cambridge University Press, 2006
j6 THE CAMBRIDGE COMPANION TO PLATO
explain why Plato decided that fictional debates and conversations
were the best way to explore the moral and political issues that
concerned him. But they do not explain the distinctive character of
the Platonic dialogues. As we have seen, Plato regards dramatic pre-
sentation, in itself, as nothing better than a form of rhetoric, tending
to move and persuade hearers or readers, regardless of the merits of
the case. Mere conversational interrogations may seem to be dis-
plays of bullying eristic. The sort of argument that Socrates thought
he had discovered required some different medium.
Plato chose the dialogue because he thought it stuck most closely
to the essential features of Socratic argument. Socrates claims that
the systematic, rule-governed form of interrogation that he practices
allows him to secure his interlocutors' agreement to moral positions
that they would have firmly, often indignantly, rejected before they
faced Socrates' questions. The interrogation is not simply the way
Socrates happens to reach his conclusion on this occasion; the fact
that the conclusion is reached through this sort of interrogation of
this sort of interlocutor is part of the reason Socrates offers us for
believing his conclusion. He claims that the arguments are not sim-
ply those that strike him, but arguments that actually convince a
normal interlocutor who approaches the questions in the right
way.80 The interlocutor (Socrates claims) is not dazzled by rhetoric,
or bullied by eristic, or overawed by elaborate disquisitions on natu-
ral philosophy,- he is rationally convinced. A Platonic dialogue is
meant to show how such conviction is possible.
Now if Socrates makes these claims about the epistemological
role of the dialogue, and if Plato agrees with Socrates about them, he
might reasonably find it difficult to present the essential elements of
Socratic philosophy in any other form than the one he chooses. He
might have been able to explain in his own voice what Socrates was
trying to do and why Socrates thought he could do it; and such an
explanation would have been useful to us. But it would scarcely
have been an effective or economical method for capturing our atten-
tion and forcing us to take Socrates seriously. Since Plato takes
Socratic philosophy seriously, he writes Socratic dialogues. We need
not suppose that the dialogues are, or were taken to be, transcripts of
actual conversations; but they are intended to communicate a cen-
tral element in Socrates' defense of his moral position.
The epistemological claims implied in Socrates' use of the dia-
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 77
logue are themselves controversial; and once Plato reflects on these
claims, he decides that they need to be modified. Modifications in
the epistemological claims also require modification in the sort of
dialogue that Plato regards as the best medium for his philosophical
views,- and so he writes different sorts of dialogues, following differ-
ent rules, to achieve different ends. In some of the late dialogues, the
conversational and adversarial character of some Socratic dialogues
is drastically modified; the Timaeus and Laws, for instance, contain
long stretches of continuous exposition. But Plato does not simply
drop the Socratic dialogue characteristic of his early period. The
Theaetetus and Philebus are late dialogues sharing many important
characteristics with the dramatic and conversational early dia-
logues. Plato's choice of the dialogue form, and of a particular vari-
ety of dialogue, is determined by his philosophical aims.
It is legitimate to point out that Plato never speaks in his own
person in the dialogues, and legitimate to wonder whether this is a
device for dissociating or detaching himself from the arguments or
conclusions attributed to the main speaker (usually Socrates).81 The
ancient evidence, however, offers no sound basis for doubting that
Plato is presenting his own philosophical views.82
In deciding how to take the dialogues, our most important exter-
nal witness is Aristotle. He has sources of information about Plato's
life and philosophy that are independent of the dialogues. He men-
tions Plato's early association with Cratylus, and he refers to Plato's
unwritten teachings. And yet he shows no hesitation in attributing
the views of the Platonic Socrates8* to Plato; his procedure would be
totally unfair if Plato were known to be putting forward these views
without endorsing them, and if this had been Plato's clear intention,
many in the Academy would know that, and would immediately
denounce Aristotle's unfairness. But while a number of ancient
Platonists defend Plato against Aristotle, none of them argues that
Plato does not accept the views he attributes to Socrates. Since Aris-
totle was in a position to know much more than we can ever know
about Plato's life, we ought to accept his estimate of Plato's inten-
tions unless we find strong reasons in the dialogues themselves for
believing that Aristotle must be wrong. Until we find such reasons
we should follow Aristotle in believing that the arguments and con-
clusions of the Platonic Socrates (and other main speakers) generally
represent the views of Plato.8*
Cambridge Companions Online © Cambridge University Press, 2006
78 THE CAMBRIDGE COMPANION TO PLATO
We would have good reasons for disagreeing with Aristotle, and
detaching Plato from the views expressed by the leading speakers in
the dialogues, if we found that the views did not display enough
unity, consistency, and coherence to be the views of one philosopher.
A reasonable case can be made for the view that there are inconsis-
tencies between different dialogues and that Plato probably recog-
nized the inconsistencies,- but we can still reasonably claim that he
endorses the inconsistent views if we can attach them to different
stages of his philosophical development. Aristotle again helps us
here, since he gives us reason to believe that some dialogues main-
tain the views of the historical Socrates and others maintain non-
Socratic views.85 When we try to assess different attempts to see a
plausible line of development in the Platonic dialogues, we pass
from questions about Plato's intellectual background to questions
about the interpretation of the dialogues themselves and about the
philosophical questions they raise.
NOTES
1 A helpful book on the general topic of this chapter is G. C. Field, Plato
and his Contemporaries (London, 1930). See the review by Harold
Cherniss in Selected Papers (Leiden, 1977), chap, n , from American
Journal of Philology 54 (1933):79-83. I have given a short account of
Greek thought before Plato in Classical Thought (Oxford, 1989), chaps.
2-4.
2 Plato's only references to himself are at Ap. 38b6 and Phd. 59bio.
3 The best of the ancient lives is Diogenes Laertius, III 1-47. We cannot
trust Diogenes even when he cites an early and well-informed source,-
for he cites Plato's own nephew Speusippus, who succeeded Plato as
head of the Academy, as a source for the story (which Speusippus is not
said to endorse) that Plato was the son of Apollo (III 2). For accounts of
the life of Plato see I. M. Crombie, An Examination of Plato's Doctrines,
vol. 1 (London, 1962), chap, i; W. K. C. Guthrie, A History of Greek
Philosophy, vol. 4 (Cambridge, 1975), chap. 2.
4 The long controversy about the authenticity of the Platonic Letters is still
not settled. I am inclined to agree with those who reject all of them. See
Ludwig Edelstein, Plato's Seventh Letter (Leiden, 1966); Norman Gulley,
"The Authenticity of Plato's Epistles," in Pseudepigrapha I (Geneva,
1972), chap. 5. For a defense of the authenticity of some of the Letters,
including the seventh, see, e.g., G. R. Morrow, Plato's Epistles, id ed.
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 79
(Indianapolis, 1962); K. von Fritz, "The Philosophical Passage in the Sev-
enth Platonic Letter," in Essays on Ancient Greek Philosophy, ed. J.P.
Anton and G.L. Kustas (Albany, 1971), 408-47. The incongruity of sev-
eral claims in the Seventh Letter (about philosophy, politics, and history)
with the dialogues constitutes a strong (though not conclusive) case for
rejecting the letter. But even if it is spurious, it was probably written by
someone who knew Plato well and who wanted his forgery to be unde-
tected; hence many of the more straightforward and (for contemporaries)
easily verifiable historical claims may be accurate. But we should not
assume that the author must be telling the truth about Plato's motives,
attitudes, or aims on political or philosophical questions.
5 See W. D. Ross, "The Problem of Socrates," Proceedings of the Classical
Association 30 (1933)17-24, esp. 16-19 (reprinted in Der historische
Sokrates, ed. A. Patzer [Darmstadt, 1987]). Cratylus is the eponymous
interlocutor in a Platonic dialogue, but Plato's unflattering treatment of
him could hardly by itself have led Aristotle to believe that he must
have been an early influence on Plato. As Crombie remarks: "It is inci-
dentally interesting, in view of the fact that Aristotle tells us that it was
Cratylus who persuaded Plato of Heracliteanism, that he is treated in
this dialogue as a noodle" (An Examination of Plato's Doctrines, vol. 2.
[London, 1963], 476).
6 In using this conventional label one should not forget that the later "pre-
Socratics" were in fact contemporaries of Socrates and Plato. Democri-
tus, indeed, is alleged to have lived to the age of 109 (and so to have died
around 350, only a few years before Plato); Diogenes Laertius, IX 43.
7 On Pythagorean mathematics and metaphysics, see D. J. Furley, The
Greek Cosmologists, vol. 1 (Cambridge, 1987), 57-60; C. H. Kahn, "Py-
thagorean Philosophy Before Plato," in The Presocratics, ed. Alexander
P. D. Mourelatos (Garden City, N.Y., 1974), chap. 6. On the importance
of mathematics see Grg. 5O7e6-5o8a8; Rep. 522C-525C. On astronomy
and cosmology see esp. G. Vlastos, Plato's Universe (Seattle, 1975).
8 See H. Ma. 289a; Cra. 4O2d, 440c; Phd. 78d-e; Smp. i87a-b, 2O7d; Rep.
485b; Tht. I52e, I79d-e ; Sph. 242c
9 See Ap. 26d6-e4; Cra. 4ooa9, 4O9a7; Phd. 72c, 97b~98c.
10 See Smp. 178D-C; Prm. I27a-i28e; Tht. 183c; Sph. 217c.
11 See Meno 76c; Tht. 152c; Sph. 242d-e; Ti. 73d7, 77C6, 78c On the
Timaeus see F. M. Cornford, Plato's Cosmology (London, 1937), 334;
A.E. Taylor, A Commentary on Plato's Timaeus (Oxford, 1928), 650-4.
12 Democritus called his atoms ideai ("shapes"), using the word that Plato
uses for his "Forms" or "Ideas"; see D.K. 68 A 57, B 141. Different views
of the relevance of Democritus to the Timaeus and Laws are taken by
Taylor, Commentary on Timaeus, 83-5, 355; Cornford, Cosmology,
Cambridge Companions Online © Cambridge University Press, 2006
8O THE CAMBRIDGE COMPANION TO PLATO
210; Vlastos, Plato's Universe, 6y. Plato's failure to name Democritus is
not by itself surprising if Democritus was still alive at the time these
dialogues were written (see note 6). Plato characteristically refrains from
discussing the views of his contemporaries by name. He normally uses
descriptions; see, e.g., Sph. 25ib6; Phil. 44D-C.
13 See Chrm. is6d—e; Snip. i86b-e; Phdr. ijoc-c-, Laws jige—yiot. On
the Timaeus, see note 11 on Empedocles.
14 On natural law, see Irwin, Classical Thought, chap. 3; Vlastos, Plato's
Universe, chap. 1.
15 This is the subject of the Elgin Marbles, the sculptures taken from the
Parthenon to the British Museum. The Panathenaic festival is men-
tioned as evidence of traditional beliefs at Euphr. 6b7-C4.
16 On popular religion, see E. R. Dodds, "The Religion of the Ordinary Man
in Classical Greece/' in The Ancient Concept of Progress, ed. E. R.
Dodds (Oxford, 1973), chap. 9, esp. p. 148.
17 On Democritus's view of the gods, see Guthrie, History of Greek Philoso-
phy, 2:478-83.
18 This passage may accurately describe both Socrates' and Plato's views,-
they may have agreed with the limitations of naturalist speculation. See
R. Hackforth, Plato's Phaedo (Indianapolis, 1955), 127-31.
19 On the importance of the contrast between teleological reason and
nonteleological necessity in Plato, see Gregory Vlastos, "Slavery in
Plato's Thought," in Platonic Studies, 2d ed. (Princeton, 1981), chap. 7;
G. R. Morrow, "The Demiurge in Politics," Proceedings of the American
Philosophical Association 27 (19541:5-23, at 7-9.
20 In the Apology (26a4~5) Socrates mentions the charge against him of not
believing in the gods of the city, but in "other newfangled supernatural
beings" (hetera de daimonia kaina). We have no way of telling whether
Socrates' accusers really knew anything about his religious views or (as
Socrates suggests at 19c) they were simply trying to exploit the religious
charges made in Aristophanes' Clouds.
21 For a protest against the view that gods are to be blamed for causing
harm to human beings, see Zeus's remarks in Odyssey I 32-43. Cf.
Aeschylus, Agamemnon 1481-8. Protests against stories of immorality
among the gods appear in Xenophanes (D.K. 21 B 11) and Euripides
(Hercules Fur ens 1340-6). On Xenophanes' influence on Euripides, see
G. W. Bond, ed., Euripides: Heracles (Oxford, 1981) adloc.
22 Contrast Aeschylus, Agamemnon 67-71 with the commonsense atti-
tude expressed by Cephalus at Republic 33od-e. Plato comments on
Cephalus's attitude at Laws 9O5d-9O7b.
23 See Aeschylus, Choephori 924-5. On the religious and legal issues
raised by Euthyphro's action, see R. Parker, Miasma (Oxford, 1983), 366-
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 81
8. For related issues about pollution, see chap. 4 and pp. 196-8. See also
W. A. Heidel, ed., Plato's Euthyphro (New York, 1902), on 4b; I. G. Kidd,
"The Case of Homicide in Plato's Euthyphro/' in Owls to Athens, ed. E.
M. Craik (Oxford, 1990), chap. 25.
24 Plato agrees with some of the spirit of Heraclitus's remark that "the only
wise thing is willing and unwilling to be called by the name of Zeus"
(D.K., 22 B 32). In Plato's view, the god Euthyphro believes in is the Zeus
of traditional religion (and so he "is willing" to be called Zeus), purified
of the elements of traditional cult and myth (and so he "is unwilling" to
be called Zeus) that conflict with his role as cosmic designer and up-
holder of morality.
25 The Greeks' sense of their identity as a people depended partly on their
shared cults and shrines of the Greek gods, according to Herodotus, VIII
144. On the role of religion and kinship (themselves closely connected)
in Greek "national" consciousness, see F. W. Walbank, "The problem of
Greek nationality," in Selected Papers (Cambridge, 1985), chap. 1 (from
Phoenix 5 [1951]: 41-60), esp. pp. 10-13.
26 See Clouds 367-381, 423-4. A naturalistic explanation of traditionally
recognized divine signs is provided at 368-411.
27 On the translation of nomos ("convention," "custom," "law," "rule,"
"norm" are appropriate on different occasions), see Irwin, Plato: Gorgias
(Oxford, 1979), 171/.
28 The accuracy of Plato's report is denied, on insufficient grounds, by
Kirk, Raven, and Schofield, The Presocratic Philosophers, id ed. (Cam-
bridge, 1983), 194-7. It is defended by G. Vlastos, "On Heracleitus,"
American Journal of Philology 76 (1955): 337-68, at 338-44. See also
Guthrie, History of Greek Philosophy, 1:488-92.
29 The claim that Heraclitus's doctrine of flux covers both types of instabil-
ity is supported by Plutarch, De Exilio 392b-c (= D.K., 22 B 91); Plu-
tarch introduces compresence in his explanation of the river fragment.
This account of Plato's interpretation of Heraclitus and Aristotle's inter-
pretation of Plato is defended in Irwin, Plato's Moral Theory (Oxford,
1977), 148-53.
30 Heraclitus takes this to be assumed by common sense; it is not clear (as
Plato and Aristotle point out) that he is right either about common sense
or about the truth of the assumption.
31 Parmenides states his main thesis in D.K., 28 B 2.7-8; 3.1; 6.1-2, and he
develops its consequences for time and change in B 8. On the interpreta-
tion of his main thesis, see G. E. L. Owen, "Eleatic Questions," in Logic,
Science and Dialectic (Ithaca, N.Y., 1986), chap.i; M. Furth, "Elements
of Eleatic Ontology," in The Presocratics, ed. A. P. D. Mourelatos (Gar-
den City, N.Y., 1974), chap. 11.
Cambridge Companions Online © Cambridge University Press, 2006
82 THE CAMBRIDGE COMPANION TO PLATO
32 Replies to Parmenides are discussed briefly by David J. Furley, The Greek
Cosmologists (Cambridge, 1987), 1:42-8, and more fully by J. Barnes,
The Presocratic Philosophers (London, 1979), vol. 2, chap. 6.
33 Hume comments on this interval between the development of natural
philosophy and the development of moral philosophy (in terms favor-
able to himself), in the Introduction to the Treatise.
34 Irwin, Classical Thought, chap. 2, is a brief introduction to the Homeric
outlook. It is discussed further by A. W. H. Adkins, Merit and Responsi-
bility, (Oxford, i960); Adkins, "Homeric Values and Homeric Society/'
Journal of Hellenic Studies 91 (1971): 1-14; A. A. Long, "Morals and
Values in Homer/' Journal of Hellenic Studies 90 (1970): 121-39; H.
Lloyd-Jones, The Justice of Zeus, id ed. (Berkeley, 1983). Some relevant
issues about the fifth century are discussed by J. L. Creed, "Moral Values
in the Age of Thucydides," Classical Quarterly 23 (1973): 213-31.
Adkins replies in "Merit, Responsibility, and Thucydides/' Classical
Quarterly 25 (1975): 209-20. A useful general book is K. J. Dover, Greek
Popular Morality in the Time of Plato and Aristotle (Oxford, 1974), well
discussed and criticized by Adkins, "Problems in Greek Popular Moral-
ity," Classical Philology 73 (1978)1143-58; and C. C. W. Taylor, "Popular
Morality and Unpopular Philosophy," in Craik, Owls to Athens, 233-43.
35 The hero's indifference to other people is only comparative; he is ex-
pected to fulfill certain obligations to others and is criticized for failing
in them, as Achilles is criticized. Still, when there is a sharp conflict
between these obligations to others and the hero's own power and sta-
tus, he is expected to choose for himself and against others, as both
Achilles and Hector do.
36 A good summary of attitudes to someone like Alcibiades is given in the
speeches of Euripides and Aeschylus in Aristophanes, Frogs 1422-32.
Significantly, it is the old-fashioned Aeschylus who turns out to be more
sympathetic to Alcibiades.
37 For possible references in Plato to Herodotus, see Rep. 566c; Ti. 25c;
Laws 6o9a-d, 692c (cf. Herodotus, VII 139), 805a (cf. IV 116-7), 947a6
(cf. II 37). All except the first passage are discussed by G. R. Morrow,
Plato's Cretan City (Princeton, i960), 91, 330, 417. P. Shorey, What
Plato Said (Chicago, 1933), asserts confidently that Plato "had certainly
read" Herodotus and Thucydides (p. 8), and that "the influence of He-
rodotus would be the theme for a dissertation, but is too obvious to need
illustration here" (p. 447); and so he does not trouble to mention any of
the evidence underlying his conviction. Nor have I been able to find any
convincing evidence of clear allusions anywhere else. For a brief intro-
duction to Herodotus and Thucydides see Irwin, Classical Thought,
chaps. 4 - 5 .
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 83
38 The major source for the history of the Peloponnesian War is the history
of Thucydides. A good introductory modern account is V. Ehrenberg,
From Solon to Socrates, id ed. (London, 1973). A useful introductory
book on Athens is J. W. Roberts, City of Sokrates (London, 1984).
39 Both Socrates and Plato, however, also had friends and connections on
the democratic side (Chaerephon the disciple of Socrates; Pyrilampes
the great-uncle and stepfather of Plato). It should not be assumed that
Plato's aristocratic background must have turned him against democ-
racy. Many upper-class Athenians must have supported the democracy.
40 I have discussed some questions about Socrates' trial and its background
in "Socrates and Athenian democracy,'7 Philosophy and Public Affairs
18 (1989)1184-205. The Seventh Letter purports to describe Plato's atti-
tude to the democracy, the Thirty, and the trial of Socrates (324C-325C).
But if the letter is spurious, the author's political aims may well have
colored the views he attributes to Plato, and it is unwise to treat them
(as most accounts of Plato's life do) as historical.
41 On Democritus's ethics, see G. Vlastos, "Ethics and Physics in Democri-
tus," in Studies in Presocratic Philosophy, vol. 2, ed. D. J. Furley and R. E
Allen (London, 1975), 381-408; C. C. W. Taylor, "Pleasure, Knowledge,
and Sensation in Democritus," Phronesis 12 (1967)16-27. The writer
known as Anonymus Iamblichi (in D.K. 89 #6) presents a defense of
conventional morality. On this writer and Democritus, see E. L. Hussey,
"Thucydidean History and Democritean Theory," in Crux [Essays Pre-
sented to G. E. M. de Ste Croix), ed. P. Cartledge and F. D. Harvey
(London, 1985), 118-38. An intriguing point about Democritus's famili-
arity with some features of political life in his native Abdera is made by
D. Lewis, "The Political Background of Democritus," in Craik, Owls to
Athens, chap. 18.
42 The Unjust and Just Arguments are characters in the school ("reflec-
tory") of Socrates, who does not take responsibility for what either of
them says. The stupidity of the Just Argument, ostensibly representing
conventional morality, suggests that Aristophanes' view of conventional
morajity may not have been wholly uncritical; see K. J. Dover, ed.,
Aristophanes' Clouds (Oxford, 1968), lvii-lxvi.
43 On fourth-century history see C. Mosse, Athens in Decline (London,
1973), chaps. 1-2; S. Hornblower, The Greek World 479-322 BC (Lon-
don, 1983), chaps. 13-15.
44 This claim needs to be qualified, though not abandoned, in the light of
the Laws.
45 See Thucydides, VII 69.2; Plato, Grg. 46iei-3, Rep. 557D-C. On Athe-
nian democracy see A. H. M. Jones, Athenian Democracy (Oxford,
1957), chaps. 3, 5.
Cambridge Companions Online © Cambridge University Press, 2006
84 THE CAMBRIDGE COMPANION TO PLATO
46 It is oversimplified, but not too misleading, to describe the Spartan con-
stitution as oligarchical, in comparison with Athens.
47 On Sparta see, e.g., Cri. 52e5; Smp. 2O9d; Rep. 544c, 545a, 547d-e; Laws
631a; Morrow, Plato's Cretan City, chap. 2.
48 If the Seventh Letter is genuine, it provides important evidence of
Plato's attitude to practical politics. In fact, it probably provides nothing
more than evidence of how the author wanted Plato to be regarded.
49 This is the solution offered in the Republic. In the Laws Plato does not
advocate the abolition of private property as a practical proposal; but he
advocates other measures for the distribution and restriction of property
with the same aim of preventing the sorts of inequalities between rulers
and ruled that provoke civil war.
50 On the sophistication of Athenian audiences, see Thucydides, III 38.7.
Cleon denounces the use of sophisticated techniques in debate; his de-
nunciation of them is itself a standard rhetorical ploy.
51 On hoi politeuomenoi, see Demosthenes, 3.30-I; cf. Plato, Grg. 473e6.
On political speakers see J. Ober, Mass and Elite in Democratic Athens
(Princeton, 1989), chap. 3.
52 See G. B. Kerferd, The Sophistic Movement (Cambridge, 1981), 24. A
striking fourth-century example is Isocrates, Antid. 235 (which, how-
ever, may involve some straining in order to suit Isocrates' argument).
At 268 Isocrates refers to the Presocratic naturalists as "the old soph-
ists." The sense and force of the term in Aeschylus (?), Prometheus
Vinctus 62, 944, raises special problems. The passage might indicate
that the term could be used in an unfavorable sense, or that it normally
had a favorable sense and it is being used ironically here. The issue is
complicated by questions about the date and authorship of the play. On
line 62, see M. Griffith, ed., Aeschylus: Prometheus Bound (Cambridge,
1983).
53 Some of the main contributions to discussion about the sophists and
Plato's attitude to them are: G. Grote, A History of Greece, 6th ed., 10
vols. (London, 1888), chap. 67; E. M. Cope, "The Sophists," Journal of
Philology 1 (18541:145-88; H. Sidgwick, "The Sophists," in Lectures on
the Philosophy of Kant and Other Philosophical Lectures and Essays
(London, 1905); Guthrie, History of Greek Philosophy, vol. 3, chaps. 1, 3;
Kerferd, Sophistic Movement, chap. 2. Sidgwick's discussion remains
the best one.
54 The attack on atheism and immorality in Laws X is directed against
naturalists rather than sophists.
5 5 The main evidence on Antiphon is found in D.K. 87 B 44. The interpreta-
tion of the fragment is disputed. I see no sufficient reason for denying
the identity of this Antiphon with the oligarchic politician mentioned
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 85
in Thucydides, VIII 68.1. See further J. S. Morrison, "Antiphon," in The
Older Sophists, ed. R. K. Sprague (Columbia, S.C., 1972).
56 On Protagoras, see D.K., 80 A 1 (= Diogenes Laertius, IX 55), B 1 (where
"Destructive Arguments" is an alternative title for his work "On
Truth"), 6.
57 The rules of the eristic game require the interlocutor to answer yes or
no; he is not allowed to qualify his reply or to point out that he did not
mean it in this sense (e.g., Euthd. iSyc-d, 295D-C, 296a-b). See also
Meno 75c—d; E. S. Thompson, The Meno of Plato (London, 1901), 272-
85; A. E. Taylor, Varia Socratica (Oxford, 1911), 91-8; Kerferd, Sophistic
Movement, 62-6. I discuss Plato's attempt to distinguish himself from
eristic in "Coercion and Objectivity in Plato's Dialectic," Revue Interna-
tionale de Philosophie 40 (1986), pp. 49-74.
58 See Ap. 23c-d, Rep. 537d-539a. An amusing example of this kind of
thing is Xenophon, Memorabilia I 2.39-46.
59 See Grg. 482d, 489b-C; Rep. 338d. In each case Plato goes to some
lengths to make it clear that this charge against Socratic method is false.
60 See Isocrates, Panath. 26; Antid. 265-6 (cf. Plato, Grg. 484c).
61 Pericles is supposed to have spoken in public from a written text instead
of improvising ("Pericles," in the medieval lexicon Suda; see R. C. Jebb,
The Attic Orators from Antiphon to Isaeos, 2 vols. [London, 1893],
ixxxviii); and the popular leader Cleon is supposed to have begun the
use of ostentatious techniques of intonation and gesture (Aristotle, Ath.
Pol. 28.3).
62 This distinction, suggested by Grg. 464b-46sc, seems to me (though not
to everyone) to be broadly supported by the other evidence.
63 It is widely and reasonably assumed that Plato alludes to Isocrates with-
out naming him. For a defense of this view, see The Phaedrus of Plato,
ed. W. H. Thompson (London, 1868), 170-83; The Euthydemus of Plato,
ed. E. H. Gifford (Oxford, 1905), 17-20. On Isocrates and the Gorgias, see
Guthrie, History of Greek Philosophy, 4:308-11; T. Irwin, Plato: Gor-
gias (Oxford, 1979), index s.v. Isocrates. On the Phaedrus see R. L.
Howland, "The attack on Isocrates in the Phaedrus/' Classical Quar-
terly 31 (1937)1151-9 (speculative but interesting).
64 For further discussion, see Irwin, Gorgias, 211-13.
65 Indeed, tragedies mistakenly induce us to suppose that people's external
fortunes and situation are the most important elements of their welfare;
for they do not see that, as Socrates argues, their moral character is far
more important. When Socrates claims that nothing can harm a good
person (Ap. 4ic-d), he upsets the scale of values that causes his audience
to be moved in the way they are moved by tragedies. Socrates thereby
rejects for himself any of the pity that would normally be excited by a
Cambridge Companions Online © Cambridge University Press, 2006
86 THE CAMBRIDGE COMPANION TO PLATO
person in his situation. For further discussion, see Irwin, "Socrates and
the Tragic Hero/' in Language and the Tragic Hero, ed. P. Pucci (Atlanta,
1988).
66 Plato himself takes a different view of rhetoric in the Phaedrus from the
one he takes in the Gorgias (partly because of developments in his moral
psychology).
67 Aristotle's remarks about the early history of dialectical reasoning seem
to suggest some role for Zeno, but insist that Socrates marks a new
development; cf. Met. iO78b23-3o with Sophist fr. 1 (see W. D. Ross,
Aristotelis Fragmenta Selecta [Oxford, 1955], 15).
68 See H. Ma. 293a-b ; Phd. 74a-c, 78c-e; Smp. 2 i ia-b; Rep. 33ic-332a,
479a-c, 485b, 495^-b.
69 Plato introduces Heracliteanism (embracing both types of flux, 152d2-3)
to explain the consequences that he takes to follow from the acceptance
of a Protagorean position. There is no basis for supposing that he himself
agrees with Heraclitus about the extent of flux (of the first type) in the
sensible world. See T. Irwin, Plato's Moral Theory (Oxford, 1977), 318 n.
27; G. Fine, "Plato on Perception," Oxford Studies in Ancient Philoso-
phy, supp. vol. (1988): 15-28; M. F. Burnyeat, ed., The Theaetetus of
Plato (Indianapolis, 1990), 7-10.
70 In Greek "what is" and "what is not" are idiomatically used interchange-
ably with "true" and "false."
71 On the argument of Rep. 475-9, see G. Fine, "Knowledge and belief in
Republic V-VII," in Companions to Ancient Thought: Epistemology,
ed. S. Everson (Cambridge, 1990), chap. 5.
72 Prt. 356d4 might be an allusion to Protagoras's characteristic doctrine,
which, however, is not the focus of discussion in this dialogue. Cra.
384C-391C contains a discussion of Protagoras's position, and 439b-
44oe contains a discussion of Heraclitean flux. In these ways the
Cratylus (which I take to be a middle dialogue, earlier than the Phaedo)
anticipates the Theaetetus and shows that the Theaetetus indicates
Plato's return to questions that had occupied him earlier (just as the
Sophist returns to some questions raised in the Euthydemus, which I
take to be an early dialogue).
73 In fact, Aristotle also wrote dialogues, though they survive only in
fragments.
74 There are some references to "Socratic discourses"; see Aristotle, Poet.
I447a28-b2o, De Poetis fr. 4 (in Ross, Aristotelis Fragmenta Selecta);
Field, Plato and His Contemporaries, chap. 11. But we do not know their
historical relation to Plato's dialogues.
75 The fragmentary character of our surviving evidence on Anaxagoras and
Democritus, for instance, makes the exact literary form of their works
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 87
obscure, though some of their works must presumably have been con-
tinuous treatises. Kirk, Raven, and Schofield, Presocratic Philosophers,
356, suggest that in fact Anaxagoras wrote only one book. Democritus
wrote a large number of works, perhaps including a body of brief ethical
aphorisms (though many of the extant ethical aphorisms are post-
Democritean). It is interesting, though we cannot tell how significant it
is, that one fragment of Democritus presents a conversation between the
intellect and the senses (D.K., 68 B 125, quoted in section 3).
76 The story that Plato wrote tragedies but burned his compositions after
hearing Socrates (Diogenes Laertius, III 5) is as worthless as most stories
about Plato's life are; it might well be the invention of someone who
was struck by the dramatic qualities of the dialogues and their criticism
of tragedy.
77 These features are especially clear in (e.g.) the Hecuba and Troades of
Euripides and in the Philoctetes and Oedipus Coloneus of Sophocles.
See further F. Solmsen, Intellectual Experiments of the Greek Enlighten-
ment (Princeton, 1975), chap. 2.
78 It has sometimes been suggested that there is some connection between
the views on marriage and property in the Republic and the views paro-
died in Aristophanes' Ecclesiazusae; but there is insufficient reason to
believe that the similarities between the two works indicate any knowl-
edge by either writer of the work of the other. The vexed questions
arising here are discussed by J. Adam, The Republic of Plato, 2 vols.
(Cambridge, 1902), 1:345-55.
79 Aristophanes' speech is a comic suggestion of some aspects of Diotima's
speech; see Smp. 191c—d, 2O5d-2o6a. On Socrates see also 2i5a-2i7a.
On the Smp. see D. Clay, "The Tragic and Comic Poet of the Sympo-
sium," in Essays in Ancient Greek Philosophy, vol. 2, ed. J. P. Anton and
A. Preus (Albany, 1983), 186-202, at 198; K. f. Dover, ed., Plato's Sympo-
sium (Cambridge, 1980), 104, 113 (who misunderstands the relation of
Aristophanes' speech to Diotima's). For Plato and comedy in general
some useful material is collected by R. Brock, "Plato and Comedy," in
Craik, Owls to Athens, chap. 5. W. C. Greene, "The Spirit of Comedy in
Plato," Harvard Studies in Classical Philology 31 (1920), 63-123, is
suggestive though diffuse.
80 The importance of the interlocutor and the contrast with eristic and
rhetoric are well emphasized by L. Coventry, "The Role of the Interlocu-
tor in Plato's Dialogues," in Characterization and Individuality in
Greek Literature, ed. C. B. R. Pelling (Oxford, 1990), chap. 8, at 174-84.
81 In this essay I am not discussing the Socratic problem. I believe that
Plato's early dialogues in fact give a substantially accurate account of
the views of the historical Socrates, and I believe that Aristotle's testi-
Cambridge Companions Online © Cambridge University Press, 2006
88 THE CAMBRIDGE COMPANION TO PLATO
mony supports this view. See Irwin, Plato's Moral Theory, 291/ (a brief
statement), and for a fuller discussion see G. Vlastos, "Socrates/7 Pro-
ceedings of the British Academy 74 (1988), 89-111.
82 I will comment very briefly on some alleged evidence that has wrongly
led some readers to discount the dialogues as sources of Plato's philo-
sophical views. (1) The comment on written philosophical treatises at
Phdr. 274b-278b insists that simply reading a treatise is neither the
same as nor a substitute for actual philosophical reasoning. It does not
say that the content of such a treatise should not be taken seriously.
(And obviously it should be noted that the Platonic dialogues come
closest, because of their conversational form, to overcoming the limita-
tions of written works.) (2) The Seventh Letter (34ib-342a) claims that
philosophical truths are inexpressible (not only incapable of being put
into writing). This claim goes far beyond the Phaedrus, and we should
note that (i) the exaggerated claim suits the writer's immediate apolo-
getic and polemical purpose (since he wants to cast doubt on the accu-
racy of all written accounts of Plato's philosophy); and (ii) the letter is
probably spurious. (3) The Second Letter (314c) claims that there is no
treatise by Plato and that the so-called writings of Plato really belong to
"Socrates become handsome and young" (or "Socrates become hand-
some and up-to-date"). The same exaggeration is detectable here as in
the Seventh Letter and the same apologetic motive is evident. (4) Some
modern writers have claimed that Plato's real views are contained in his
oral teaching (the "unwritten doctrines" referred to by Aristotle at Phys-
ics 2O9bn-i7). For an introduction to the many, and mostly unre-
warding, discussions of this question see Guthrie, History of Greek Phi-
losophy vol. 5, ch. 8. Some of the issues are set out clearly and with
devastating effect by G. Vlastos, "On Plato's Oral Doctrine," in Platonic
Studies, id ed. (Princeton, 1981), 379-403. It should be noticed: (i) There
is no evidence to suggest that there was a large body of oral doctrine
constituting a complete system that would underlie, explain, or under-
mine the dialogues, (ii) There is no evidence to suggest that Plato's oral
teachings were esoteric,- one of the firmest pieces of evidence refers to a
lecture on the Good that was delivered to a general audience (Aristotle,
On the Good, fr. 1 [in Ross, Aristotelis Fragmenta Selecta]). (iii) There is
no evidence to suggest that Plato or anyone else took his oral teaching
any more seriously than he took the dialogues. Some points may have
been "unwritten" simply because Plato regarded them as tentative.
In summary, no internal or external evidence gives us any good reason
whatever for denying that the dialogues express Plato's own philosophi-
cal views.
83 The treatment of the Republic and Laws in Politics II provides a striking
Cambridge Companions Online © Cambridge University Press, 2006
The intellectual background 89
example. After speaking of the Socrates in the Republic, Aristotle goes
on to speak of the Laws as another "Socratic discourse" (1265a]:0-13),
even though (in the surviving version of the Laws) "Socrates" is not a
speaker in the Laws. He then goes on to treat the Laws and the Republic
as evidence of Plato's views (Politics 126605, 127 ibi, 1274.159).
84 It does not follow that these main speakers represent the whole of
Plato's philosophical position at the time he wrote a given dialogue.
85 On Aristotle's devices for referring to the historical and the Platonic
Socrates see W. D. Ross, Aristotle's Metaphysics, vol. 1 (Oxford, 1924),
xxxiii-xli.
Cambridge Companions Online © Cambridge University Press, 2006
LEONARD BRANDWOOD
3 Stylometry and chronology
For a correct understanding of Plato, account needs to be taken of the
fact that his philosophical activity spanned some fifty years, during
which time certain doctrines underwent considerable changes. To
trace this development and so be able to identify the final expression
of his thought, it is essential to know in what order the dialogues were
written, but there is little help in this quest either from external
sources or from the dialogues themselves.1 Regarding the former, the
only information likely to be reliable is Aristotle's statement that the
Laws was written after the Republic.2 This is repeated by Diogenes
Laertius (III 37) and Olympiodorus [Prol. VI 24), who add that the
Laws was still in an unrevised state on wax tablets when Plato died
and was published posthumously by one of his students, Philip of
Opus. As for internal evidence, cross references in the Sophist (217a)
and Politicus (257a, 258b) indicate the prior composition of the
former, while the Timaeus (27a) mentions the Critias as its sequel.
Rather less definite is the apparent reference in the Timaeus (17b-
19b) to the Republic, in the Sophist to the Parmenides (217^3 and
Theaetetus (216a), and in the Theaetetus to the Parmenides (183c).
There is one other important piece of evidence: In the introduction to
the Theaetetus (143c), Plato renounces the reported dialogue form
with a clear indication that the use of introductory formulae, such as
xai eycb efotov ("I myself said"), and of interlocutors' replies, was be-
coming a nuisance. It seems unlikely, therefore, that any of his works
written in this form are later than the Theaetetus.
In the eighteenth century and the first half of the nineteenth at-
tempts to establish the chronological sequence were based on an
assessment of each dialogue's argument, followed by the formulation
of a line of development for the philosopher's ideas. Not surprisingly
90
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 91
the subjective nature of this approach led to a considerable discrep-
ancy among the conclusions of the various scholars/ so that by the
1860s hope was beginning to fade, and G. Grote, for example, was to
be found declaring that the problem was incapable of solution.*
Two years later, however, hope was revived with the introduction
of the stylistic method by L. Campbell.6 Observing an increased use
of technical terminology in what were then taken to be Plato's latest
works, the Timaeus, Critias, and Laws, he calculated from Ast's
Lexicon the number of words that each of twenty-four dialogues?
had in common exclusively with these three. Then dividing this
figure by the number of pages in each dialogue, he arrived at the
average occurrence per page and arranged the dialogues in a series
according to their relative degree of affinity to the three latest works
in respect of vocabulary. The series was headed by the Politicus,
Phaedrus, and Sophist, each showing an average occurrence of more
than one word per page. Mindful of the influence of subject matter
on the choice of words, Campbell did not follow his figures slavishly
in drawing conclusions about chronological order, but remarked
that, combined with some further observations on rhythm and word
order,8 they did at least support his view of the close temporal affin-
ity of the Sophist and Politicus to the Timaeus, Critias, and Laws.
The usefulness of this method for determining the chronological
order of works was discovered independently by W. Dittenberger,?
who investigated two aspects of Plato's vocabulary, the first being
the use of \ir\v ("indeed") with certain other particles. The distribu-
tion of three of these (see Table 1) enabled him to divide the dia-
logues into two groups according to their occurrence or nonoccur-
rence,10 the later group being indicated by the presence in it of the
Laws. The fact that all three expressions were found together in each
work except for the Symposium and Lysis led him to conclude that
these two works were the earliest of the second group. Since the date
of the Symposium's composition was fixed as shortly after 385 B.C.
by what is clearly a topical allusion in it to the dispersal of
Mantinea, which took place in that year,11 he believed that tt \ir\v
was a conversational idiom of the Dorians in Sicily, which Plato had
visited a few years before. In support of this view he noted that the
expression was not to be found either in earlier Attic prose or in
Aristophanes, though significantly the equivalent Doric act JI&V did
occur in the latter.12
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Table i. Frequency of occurrence of certain expressions in the dialogues, used by Dittenberger as a means
of ordering them chronologically
Pages
(Didot) xi \ir\v, aXka. . . \ir\v-, ye \ir\v &OKBQ xaOouieQ ecog(jieQ) nexQutee x&xa loutc,
I
Ap.
Cri.
Euphr.
Pit.
Chrm.
La.
H. Mi.
Euthd.
Meno
Gig.
Cm.
Phd.
19-7
95
II.7
395
18.1
17.8
IO.I
27.9
23.3
61.6
42.3
49.2
3 i
7
68
9
12
Q
O
3O
2 1
69
8O
8O
I
I
I
I
2
6
3
2
2
4
3
8
16
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Ha
Snip.
Lys.
Phdr.
Rep.
Tht.
lib
Pirn.
Phil.
Sph.
Pol.
Ti.
Criti.
Laws
39-3
14.9
39-O
194.0
53-O
3 1 2
43-2
39-6
43-2
53-O
11.2
236.8
1
1 1
34
13
6
26
1 2
2 0
48
2
4
1
1 1
1
5
7
5
8
6
1
24
55
17
27
212
47
9
9
9
16
10
2
24
27
14
34
18
5
148
5
23
10
(5)
3
3
5
3
1
16
1
1
3
4
1
16
Notes: Different texts can produce different occurrences of a given word. The figures in the tables are those provided by the scholar concerned except for
large errors, when the original figure is replaced by one in parenthesis; this always refers to the Oxford Classical Text. Further details are supplied in the
tables of L. Brandwood, The Chronology of Plato's Dialogues.
The ea)g(jteo) figures for groups I and Ha, apart from those in parenthesis, were provided later by Ritter.
tocog xdx" av at Ap. 3133 was ignored by Dittenberger, though he included the similarly inverted form at Ti. 38e2.
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Cambridge Companions Online © Cambridge University Press, 2006
94 THE CAMBRIDGE COMPANION TO PLATO
Observing the difference in frequency of ye \ir\v between the Re-
public with four instances and the Laws with twenty-five, and recall-
ing Aristotle's testimony about their relative dates, Dittenberger
concluded that the works of group Ila in the table, where the occur-
rence of this expression is sparse, were earlier than those of lib.
As the subject of his second investigation he took two pairs of
synonyms, a)OJteQ-xa6djtEQ ("like") and ea)g(jteQ)-[iixQi:rc£Q ("un-
til"), together with the pleonastic combination t&xa icoog ("maybe
perhaps"). Although these criteria did not distinguish works of the
first group from those of the second, they did reinforce the evidence
of ye \ir\v for a division of the second group, since all the lib dia-
logues with the exception of the Parmenides show a preference for
xa8&jr£Q over cnoneg and the exclusive (apart from a solitary in-
stance in the Apology) use of \xexQM£Q a n d x&xa iocog. As for the
Parmenides, he was so puzzled by its inconsistencies that he was
inclined to doubt its authenticity.
Dittenberger, like Campbell, considered his main achievement to
be the demonstration of the lateness of the Sophist and Politicus,
works that previously had been thought to be much earlier. He had
also provided evidence for a later dating of the Philebus, Phaedrus,
and Theaetetus, the former after the Republic, the latter two close to
it, the position for the Phaedrus being especially significant in that
it had frequently been regarded in the past as one of Plato's earliest
compositions.
The next worthwhile contribution came from M. Schanz/3 whose
research resembled Dittenberger's second investigation in that it
compared pairs of synonyms, in this case three in number, all denot-
ing "in reality" or "in truth" (Table 2). On the basis of his figures he
too divided the dialogues into three chronological groups. The last,
comprising the Philebus, Politicus, Timaeus, and Laws, was charac-
terized both by the complete absence of two of the synonyms, xw
OVTL and cog aA,r]0cog, and by the occurrence in it alone of a third,
ah^Qeia,1* while the middle group, consisting of the Cratylus,
Euthydemus, Phaedrus, Theaetetus, Republic, and Sophist, was dis-
tinguished from the earlier by the presence in it of ovxcog.
Comparison of his results with those of Dittenberger reveal only
slight differences,- indeed his allocation of the Sophist to the middle
group on the basis of a single occurrence of xco ovxi and three of cog
&Xr]0cog is hardly justified, considering that their frequency in rela-
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 95
Table 2. Frequency of occurrence of three pairs of synonyms in the
dialogues, used by Schanz as a means of ordering them
chronologically
Pages
(Didot) xcp ovxi ovxcog cog CXXT]6CO g a^rjGcog xfj a.hf\Qeia akr\Q£iqL
1 3
1
6
Ap.
Euphr.
Gig.
La.
Lys.
Pit.
Smp.
Phd.
Cm.
Euthd.
Tht.
Phdr.
Rep. I-IV
V-VIII
VIII-X
Sph.
Phil.
Pol.
Ti.
Laws
Chrm.
Cri.
H.Ma.
Menex.
Meno
Prm.
Epin.
19.7
11.7
61.6
17.8
14.9
39-5
3 9 3
49.2
42.3
2.7.9
53-O
39.O
80.5
60.5
53-o
39-6
43.2
43-2
53-O
236.8
18.1
9 5
19.0
11.6
2 3 3
3 1 2
14.1
5
1
9
2
6
2
5
14
1
4
6
8
13
18
1 0
1
5
6
1
1
1
1
6
5
4
2 1
15
1 1
8
5O
16
2
1
7
7
2
2
3
1 1
3
2
8
7
19
6
3
3
5
2
1
4
1 2
1 2
1 2
2 3
3 3
3 9
6
7
4 1
3 1
6 3
2
3
1 1
2
1 1
Note: The last seven works were not included in his table by Schanz, some because,
like H. Mi. and Criti., they contained examples neither of xcp ovxi nor of ovxcog, others
because they were considered unauthentic. The figures for these relate to the O.C.T.
flSee note 14.
tion to the respective synonyms is the reverse of that found in the
other works of this group. As for the Cratylus and Euthydemus, the
argument for placing them in the middle group is weak, consisting
of a solitary instance of ovtwg; in each case furthermore this is the
reading of inferior manuscripts. In this connection it is also worth
Cambridge Companions Online © Cambridge University Press, 2006
96 THE CAMBRIDGE COMPANION TO PLATO
noting that ovxcog does not appear in the first four books of the
Republic, casting further doubt on the reading of this form in the
text of the Cratylus and Euthydemus. Unlike Dittenberger he found
no evidence to suggest that the Symposium and Lysis belonged in
the middle rather than the early group.
The next feature of Plato's style to be recognized^ as useful for
chronological purposes, his varying use of reply formulae, was to
form the subject of several investigations during the following three,
decades.16 Since these cannot all be treated here, that of Ritter,
which was by far the most extensive, may be taken as representa-
tive. Convinced by the work of Dittenberger and Schanz that the
Sophist, Politicus, Philebus, Timaeus, Critias, and Laws formed a
discrete chronological group, he compiled a list of forty-three linguis-
tic features, both reply formulae and others, which supported the
view that these six dialogues marked the culmination of Plato's
literary activity. A sample of these is reproduced in Table 3. By
counting how many were present in each diaglogue he was able to
determine its degree of linguistic resemblance to the Laws, which he
regarded as the last to be written. The result was as follows: Laws
40, Phil. 37, Pol. 37, Sph. 35, Rep. 28, Tht. 25, Phdr. 21, Prm. 17,
Epin. 12, Cra. 8, Lys. 8, Phd. 7, La. 5, Euthd. 4, Pit. 4, Menex, 4, Smp.
3, Chrm. 3, Gg. 3, H.Ma. 3, Ion 3, Ap. 2, Meno 2, Cri. 2, Euphr. 1.
This seemed to confirm the view of his predecessors, that between
the late works and the mass of early dialogues there was a middle
group consisting of the Republic, Theaetetus, and Phaedrus.1'?
To determine whether the Timaeus and Critias belonged to the
last or middle group, it was necessary to exclude from consideration
those of the forty-three linguistic features, such as reply formulae,
which were connected with dialogue, these two works being almost
wholly narrative in form. Of the nineteen features remaining, the
middle and late works possessed the following share: Laws 18, Ti.
17, Phil. 16, Pol. 16, Sph. 14, Criti. 11, Rep. 9, Phdr. 8, Tht. 6, Prm.
1 - indicating a position for the Timaeus and by implication its se-
quel the Critias in the last rather than the middle group.
Turning next to the question of the order of composition within
each of these groups, Ritter first had to consider the possibility that
an extensive work such as the Republic did not all appear at one
time, but that the other dialogues were composed either contempora-
neously with or in between parts of it. When to this end he produced
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 97
statistics of his forty-three criteria for each of the ten books, it be-
came apparent to him that, while Books II-X had a fairly uniform
style, Book I exhibited a number of features that were typical of the
early rather than the middle chronological group.18 This seemed to
justify the assumption that Book I was written separately some time
before the rest, and the sporadic occurrence in it of expressions char-
acteristic of the middle group1? could be explained by the assump-
tion that it underwent some revision before being incorporated into
the larger work.
Of the dialogues in the final group Ritter considered the Sophist to
be the earliest, since certain expressions characteristic of Plato's
earlier style were still to be found in it before disappearing entirely,
for example, xco ovn and cog aXrjGcbg (cf. Table 2). On the evidence of
other features that he considered particularly significant, he ante-
dated the Politicus, to the whole of the Laws, which he took to be
the last work of all, and assigned to the Philebus a position contem-
porary with the first half of the Laws, and to the Timaeus and
Critias one contemporary with the second half.20
Regarding the middle group, he refused to draw any conclusion
from the fact that out of the forty-three expressions used as criteria for
lateness of composition the Republic contained twenty-eight, the
Theaetetus twenty-five, and the Phaedrus twenty-one, since the dif-
ference in the size of the figures corresponded to that of the works. A
direct comparison of the Theaetetus and Phaedrus revealed that
seven of the expressions favored a later date for the Phaedrus, only
four the reverse order. Where the Republic stood in relation to the
other two dialogues he could not say,- on account of the time that the
composition of such a lengthy work must have taken he was inclined
to believe that the Theaetetus, and perhaps even the Phaedrus too,
might have been written contemporaneously with it.21
About the chronological relationship of the rest of the works,
which he assigned to the earliest group, he was not prepared to
hazard any conjecture, since their style was generally uniform and
such differences as were apparent were insignificant.
Toward the end of the last century the separate threads of research
were pulled together by W. Lutoslawski.22 His method was similar to
that of Ritter, being an enumeration of the "late" linguistic features
in each work, but whereas Ritter used only about 40 criteria, he
amassed 500. This was made possible by the fact that for him a
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Table 3. Frequency of occurrence of selected linguistic features in the dialogues, used by Ritter as a means of ordering
them chronologically
.3
1
!
5 8
o
I &
as
•a §
S 5
Total of reply formu-
lae
i . akr\Qr\ Xeyeig
2 . ClA.T|0T|
3. &A.Tj0Tf (Xiyeiq),
6o0d)g Xeyeig
4. akiptmaxa
6o0OTaTa Xeyeig
5. navv ye
6. navv \ikv ovv
7. jiavrdjiaai \ikv ovv
8. xai |xaXa
9. Jtwg;
IO. Jiff;
n . (TO) mxov (6r|);
77
S
1
6
1
1 0
5
1
1
n o
2
3
6
16
5O
6
1
6
3
3
107
3
3
2 0
3
1
1
203
9
1 0
1
38
13
1
2
IO 22
2
2
3
I
64
4
5
1 2
3
1
336
5
5
48
7
176
6
1
8
4
23
17
1
1
1
182
4
5
27
5
36
5
5
1
6
285
2
9
14
8
5
16
9
4
4
13
69
1
2
3
1
2
3
3
2
4
1260
9
29
48
4 0
4 0
64
38
47
32
4
48
315
3
7
1 0
8
1 0
14
IO
4
2 0
7
32
2 5 1
1
5
8
15
7
18
4
2
17
6
36
3 H
2
2
6
2 2
9
2 1
4
7
18
3
33
13
2
1
1
568
7
4
2 2
36
4
49
13
6
14
3
47
43
3
(3)
4
1
95
5
(1)
(5)
1 2
6
71
1
(1)
3
5 120
3
6
1 0
1 18
1
1
486 9
4
18
24
7
28
15
7
2
1 0
3
3
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Pages (ed. Didot)
12. 6fj"Xov oxi
13. 6fj7.ova)g
14. oxe66vxi
15. oxeoov
16. evexa
17. x&oiv
18. xo/xavDv
19. xoecbv (ecru)
20. Ionic dative form
21. Jieoi(%)fl
18
7
7
8
13
18
2
3
1
6
1
0
4 0
15
3
14
1
1
2
28
1 1
2
5
1 0
42
17
2
9
(1)
3
2 0
7
2
5
1 1
1 0
3
2
1
2
0
1 2
5
1
4
8
62
15
2
3
3 i
3
1 0
49
6
6
2
13
1
4
23
1 2
1
2
1 0
39
6
3
16
1
8
53
1
1
1 2
4
1
1 2
39
8
3
4
9
8
3
23
194
47
2
1 2
7
69
1 2
1
6
23
4 0
1 0
8
26
6
1
5
1
2 2
43
1 0
2
13
2 2
3
5
1
4
26
43
8
5
14
19
3
9
34
53
1
4
1
9
13
7
7
3
2
16
1 1
1
4
2
2
3
2
1 0
237
16
14
2
1 2 2
i n
33
79
57
85
3 i
9
5
2
1
1
3
19
7
5
4
(1)
0
1 0
3
(1)
4
1 2
1
1
1
2 0
15
5
1
25
6
3 i
(3)
1
(1)
7
14
1
2 0
3
4
4
2
19
Note: 1. "You speak truly." 2. "Truly." 3. "(You speak) truly/rightly." 4. Superlatives of no. 3. 5 and 6. "Certainly." 7. "Most certainly." 8. "Very much so." 9. "How?" 10. "In
what way?" 11. "What sort (indeed)?" 12 and 13. "It is obvious that." 14 and 15. "Just about." 16 and 17. "For the sake of." 18. "At the present." 19. "It is necessary." 20. Attic
form with iota suffix. 21. "About."
The dialogues in the right-hand columns (Ion, etc.) were not included by Ritter in his main investigation on the grounds of unauthenticity.
"Numbers in this line indicate the anastrophic use of Jieoi, expressed as a percentage of the total occurrence of the preposition.
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Cambridge Companions Online © Cambridge University Press, 2006
IOO THE CAMBRIDGE COMPANION TO PLATO
characteristic of Plato's later style did not necessarily mean an ex-
pression occurring in the Laws, as it did for Ritter still trying to
prove the lateness of the Sophist, Politicus, Philebus, Timaeus, and
Critias, but one occurring in any of these six dialogues. To produce
this larger total he selected what he considered to be the most impor-
tant of the statistics published by earlier investigators of Plato's
style, both chronologists and philologists, in the case of the latter
determining himself which features were meaningful. To each he
allocated a value of one, two, three, or four units, according to the
degree of importance that it seemed to him to have, then after count-
ing how many of the 500 features occurred in each dialogue, evalu-
ated in terms of units of affinity its approximation to the final group
or, in the case of the works in this group, to the Laws. Like Ritter he
concluded that the last chronological group was preceded by one
consisting of the Republic II-X, Phaedrus, Theaetetus, and also the
Parmenides. His results, however, were vitiated by several flaws in
his method, the most serious being the arbitrary nature of his evalua-
tion of the importance of the various features and the use of some
that were unsuitable, such as those which instead of being character-
istic primarily of the last group were characteristic of another or of
none.2*
Contemporaneously with Lutoslawski's work, C. Baron published
the results of research into Plato's use of the anastrophe of JteQi
("about"), that is, its occurrence after instead of before the substan-
tive.2* He discovered that its incidence was noticeably higher in
works generally thought to be late (cf. Table 3, where it is expressed
as a percentage of the total occurrence of the preposition),2* though
the behavior of the Parmenides was conspicuously deviant from this
trend, as it had been in Ritter's investigation.26 There appears to
have been no one reason for the increasing use of anastrophe by
Plato, but at least in works of the final group the avoidance of hiatus
may plausibly be surmised.2?
By the turn of the century, then, research into Plato's style had
succeeded in separating the dialogues into three chronological
groups, but had failed to determine the sequence within any of
these, apart from some evidence to suggest that the Sophist was the
earliest work of the last group. Shortly afterward new ground was
broken, when G. Janell,28 following comments made much earlier
by F. Blass,29 investigated the frequency of hiatus.3° He began by
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology I O I
Table 4. Frequency of "objectionable" hiatus in the dialogues, as
calculated by Janell
Lys.
Euthd.
Prm.
Chrm.
Rep. I
H.Ma.
Phd.
Pit.
Rep. IX
Rep. IV
Ap.
Meno
H. Mi.
Cri.
Smp.
Gig.
Euphr.
Rep. I-X
Rep. VII
Rep. X
Ion
La.
Rep. VIII
Tht.
Rep. VI
Instances
of hiatus
685
1258
1376
797
901
779
2017
1591
601
757
764
892
378
342
1414
2182
413
6833
661
664
312
598
615
1733
626
Pages
(Didot)
14.9
27.9
31.2
18.1
20.5
19.0
49.2
39-5
15.1
19.1
19-7
23.3
IO.I
9 5
39-3
61.6
11.7
1937
18.8
19-3
9.1
17.8
18.6
53.0
19-3
Average
per page
46.0
45.1
44.1
44.0
44.0
41.0
41.0
40.3
39-8
39-6
38.8
38.3
37-4
36.0
36.0
35-4
35.3
35-3
3 5 2
34-4
34-3
33.6
33-i
32.7
32.4
Rep. II
Rep. Ill
Rep.V
Cm.
Men ex.
Phdr.
Laws V
Laws III
Laws XII
Laws X
Laws II
Laws XI
Laws I
Laws IX
Laws IV
Laws I-XII
Phil.
Laws VIII
Epin.
Laws VII
Laws VI
Ti.
Criti.
Sph.
Pol.
Instances
of hiatus
607
706
695
1319
327
932
126
1 2 1
152
108
89
172
95
189
77
1389
160
94
40
7 i
95
62
9
24
19
Pages
(Didot)
18.8
22.0
22.2
42.3
11.6
39.o
15.9
19.4
21.1
1 9 5
16.3
19.4
18.6
22.2
14.7
236.8
43.2
16.9
14.1
27.6
25.2
53-O
11.2
39-6
43-2
Average
per page
32.3
32.1
3i-3
31.2
28.2
2 3 9
6-7 (7.9)
6.2
57(7.2)
5.6
5-5
54(8.9)
5-i
5.1(8.5)
4 .8(5 .2)
4.7(5 .9)
3-7
3.7(5.6)
2.8
2.5
2.4(3.8)
1.2
0.8
0.6
0.4
Notes: For the figures in parenthesis, see CPD, 155 n.3.
The figures for H. Ma., Ion, and Epin., which fanell omitted from his investigation, are
the author's.
distinguishing two types, permissible and objectionable. In the
former he included two broad classes of hiatus: those which Plato
could have avoided, if he had wished, by some simple means such
as elision, crasis, or the choice of an alternative form, and those
involving words of common occurrence, such as the definite article
and xai ("and"), where avoidance seemed to be scarcely practica-
ble.*1 The incidence of the rest, which he classed as objectionable,
is shown in Table 4.
The conclusion that Janell drew from these figures was that
Cambridge Companions Online © Cambridge University Press, 2006
IO2 THE CAMBRIDGE COMPANION TO PLATO
Plato's treatment of hiatus differed considerably in two separate
periods. In the first, comprising all the works down to the Phaedrus,
Plato was not troubled by hiatus of any kind, but in the second, he
carefully avoided those types of hiatus identified as objectionable.
He made no attempt to deduce anything about chronology from the
figures for the dialogues of the first period, and clearly there would
be little point in it, since, if Plato's avoidance of hiatus was an
abrupt, not a gradual development in his style, as the statistics
would seem to indicate, then any variation in its frequency here
would probably be accidental. With the last group of dialogues, how-
ever, one might well expect to find some development, either an
increasing or a decreasing avoidance, or a third possibility, an avoid-
ance that increased up to a certain point, then decreased. Janell
himself, however, was not greatly concerned with chronology, his
sole contribution in respect of the order of the last group of works
being that " obviously Plato was prevented by death from applying
the last touches to the Phil, and Laws. This seems to be the explana-
tion of the difference in frequency of hiatus between these two and
Sph., Pol, Ti., Criti."
Lastly he remarked on the frequency of hiatus in the Phaedrus
being somewhat lower than in the rest of the earlier dialogues. To
explain this he accepted Blass's view, that Plato later revised it, and
in support of this thesis cited several passages where hiatus appears
to be avoided more carefully than usual, such as (i) 250c xairca \xkv
oirv-25ia xolg Jiai6ixolg, (2) 259b oi> jiev 6f)-259e \isKkr\, (3) 265a
&\M\v oe—26$d a%agi, and (4) 265c TO jtaAiv-267d xOevxai ovofxa,
with two, two, one, and ten instances of objectionable hiatus respec-
tively. 3* It could be argued that similar passages may occur in other
dialogues outside the last group. Without examining the whole Pla-
tonic corpus in detail it is impossible to refute this hypothesis abso-
lutely, but one may at least cast some doubt on it. On the basis of the
figures in Table 4, Menexenus and Cratylus are the works in which
there would be an expectation of finding passages with a scarcity of
hiatus similar to that in the Phaedrus; yet in reading them one is not
consciously aware, as with the Phaedrus, that some parts of the text
contain fewer instances of hiatus than the rest.
This subjective impression can be given a numerical expression;
if, for instance, with the first passage above one counts the number
of words between the hiatus immediately preceding touta \iev oirv
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 103
and that immediately succeeding xolg jtai5ixoTg, the result is 210
(O.C.T.). For purposes of comparison, therefore, it may be described
as 2 instances of hiatus in 210 words. Similarly the second passage
will be 2 in 211 words, the third 1 in 196 words, and the fourth 10 in
600 words. By contrast, in the Cratylus (a work of roughly equal
length), there are only two passages of note: 4ooc7~4Oib7 with 5 in
217 words and 4O4e7~4O5C5 with 3 in 183 words. Iti the Menexenus
there are three passages: 234c6-235b8 with 2 in 124 words, 242d8-
243b4 with 3 in 151 words, and 24oe3-242a7 with 10 in 285 words.
Since neither can be said to match the Phaedrus, there is some
evidence to support JanelPs view of an apparent tendency on Plato's
part to avoid hiatus more carefully in some parts of the Phaedrus
than in others.
Janell's inquiry confirmed the unity of the final chronological
group established by earlier research. Regarding a possible develop-
ment of hiatus avoidance within this group, if one assumes that it
was not haphazard and that the Laws, at least in part, was probably
the last to be written, it would appear that toward the end of his life
Plato was less strict in his approach. This would be psychologically
plausible in that, having demonstrated his ability to match his rival
Isocrates in this aspect of prose style, he could afford to adopt a more
relaxed attitude. As the incidence of hiatus in the Philebus is similar
to that in the Laws, one might argue that it is the closest to it of the
other five or that it represents the first serious attempt to put Isocra-
tes' principles into practice before achieving greater success in the
Timaeus, Critias, Sophist, and Politicus.
Yet another aspect of Plato's style was revealed by W. Kaluscha,
who examined the rhythm of his prose.33 Since prior to the advent of
computers it was impracticable to analyze the whole text in this
respect, he confined his investigation to the part of the sentence
considered in antiquity to be the most important rhythmically,
namely the clausula, which he interpreted as the end of a period or
colon. This, for the same reason of economy, was regarded as consist-
ing of only five syllables, either long (-) or short (̂ ) metrically,
yielding thirty-two different combinations.
He first looked at the clausulae of the Laws in order either to
corroborate or to contradict Blass's belief that in the latter part of his
life Plato began, under Isocrates' influence, to prefer certain rhythms
to others. He then compared with them the clausulae of the Sophist,
Cambridge Companions Online © Cambridge University Press, 2006
IO4 THE CAMBRIDGE COMPANION TO PLATO
Politicus, Philebus, Timaeus, and Critias, supposedly late works on
the one hand, and those of the Protagoras, Crito, and Apology, sup-
posedly early works on the other, to see which showed the greater
similarity. Since his initial survey of the Laws indicated that the
four most frequent combinations in every book were types ending in
a long vowel and only in a few books did the fifth most frequent end
in a short vowel, he decided that it was possible to regard the ambigu-
ous quantity of the final syllable^ as long in every case and reduced
his statistics accordingly to cover sixteen combinations (Table 5).35
He observed a preference for five clausulae in the Laws: II 4, III 9,
and IV 4, each of which represented one of the five highest figures in
all twelve books; II 10 likewise with the exception of Books III and
VIII; V, which had one of the five highest figures in half of the books.
These five formed the following percentage of the total number of
clausulae in each book:
I
II
III
46
55
51
9
•3
•7
IV
V
VI
54
53
55
•4
. 1
•9
VII
VIII
IX
54
5 i
56
•4
. 2
.8
X
XI
XII
60.
52.
54.
I
4
6
That is to say, in all except Books I they were more frequent than the
other eleven taken together. Conversely, there appeared to be a par-
ticular aversion to four clausulae: II 7, III 3, III 6, and III 8.
On comparing the Protagoras, Crito, and Apology with the Laws,
Kaluscha noted the following differences: First, only two of the
clausulae favored in the latter work (III 9 and IV 4) occurred fre-
quently, but then they did so in every period of Plato's literary activ-
ity^6 secondly, the clausulae to which there was an aversion in the
Laws were not avoided - on the contrary, they were common, thus
lending support to the view that their later avoidance was deliberate.
By contrast he found that the prose rhythm of the Sophist, Poli-
ticus, Philebus, Timaeus, and Critias was similar to that of the Laws.
In the case of the Philebus it was practically identical, since its five
most frequent clausulae were the same, and it avoided the four un-
popular clausulae to an equal degree. In both the Sophist and Poli-
ticus, three of the five most common Laws clausulae also showed the
highest frequencies, and the unpopular ones were avoided, though II7
and III 6 not so carefully in the Sophist as in the Politicus and
Philebus. In the Timaeus also, three of the five highest frequencies
coincided with those of the Laws, but with a difference; whereas in
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Table 5. Prose rhythm in the dialogues: relative frequency by percentage of different types of clausula, as calculated
by Kaluscha
I
II
m
IV
V
5
4
7
9
IO
3
5
6
7
8
9
i
2
3
4
wwww-
- w w w -
w w - w -
w w w —
ww-
-_ww—
ww
w-w
w——w —
w
_ w
w _
Prt.
4.2
3.8
5.2
4.O
4.2
6.3
8.0
7-3
6.9
6.3
6.9
7-1
8.1
7.6
7-7
6.3
Cri.
2.6
1.9
3 1
3.8
6.3
5-i
7.6
7.0
6.3
9 5
9-5
7.0
7.0
6.3
IO.I
6.9
Ap.
3-4
3.6
6.1
4.2
6.6
6.1
7-5
8.6
6.4
6.6
5-5
7-7
8.5
7-4
5.6
6.6
I
5.6
8.8
4.1
4.4
10.0
5.6
3.8
1.8
6.2
2.3
11.1
5 9
7.6
5 9
11.7
5 3
II
6.4
9.0
3.0
3.8
1 1 3
2.6
4-5
1.1
3-O
i-5
10.2
7-5
5.3
6.0
13.9
10.9
III
8.5
16.6
4.1
6.0
6.6
2 . 2
2.8
1.6
4-7
1.9
11.0
6.6
6.9
3-1
11.6
6.0
IV
6.4
11.7
2 . 1
6.8
10.3
4.6
5-3
1.8
2.8
2.8
10.3
4.6
5-7
2.5
11.4
10.7
V
9.1
12.3
3-7
7.8
9.9
4.1
5.8
0.8
2.5
0.8
9-5
3-3
5-3
3-7
11.5
9.9
Laws
VI
7-9
14.2
1.4
5.2
12.0
5.2
2.5
0.8
6.3
1.1
12.0
5-4
6.3
2 . 2
9 3
8.4
VII
6.7
12.4
2 . 2
4-7
8.2
2.7
5.6
1-3
6.0
2.9
12.9
5.8
4.2
3.6
11.8
9.1
VIII
5.6
12.3
1.6
6.3
7-9
2.4
3.6
0.8
5.6
2 . 0
9-5
9.1
6.7
4.8
15-5
6.3
IX
5 1
14.4
2.7
5-4
7-5
2.7
4.8
1 .2
3.0
2.7
15-3
7.2
5-4
3-O
10.8
8.7
X
3.6
14.8
3 9
6.4
8.4
3-1
4.2
1.4
3 4
1.4
10.9
3-4
5.0
4.2
18.2
7.8
XI
5.2
IO.I
2.6
5.2
6.0
6.0
3-4
0.4
3-7
2 . 2
11.2
9 7
6.7
2 . 2
17.6
7-5
XII
4-9
13.2
4-3
6.6
6.6
3.6
3.6
1 . 0
4.6
3-3
13-5
5-3
6.9
1-3
13-5
7-9
Phil.
7.3
9-3
2.8
4-5
9.4
4-7
4-3
1-5
5-7
1-9
12.0
6-7
4-5
3-1
14.5
7.8
Pol
4-5
6.5
4.2
5.6
7-5
3-9
4.2
1 .2
7-7
1-9
9 7
6.2
8.3
4.8
13.6
IO.I
Sph.
5-3
5-9
6.4
5-7
7-5
5-5
6.1
5-1
6.4
2.9
7.8
5-4
7.2
6.7
9.9
6.0
Criti.
7-3
8.0
4.0
4-7
13-3
6.0
3-3
6.0
4.0
5-3
8.0
5-3
4-7
8.0
4-7
7-3
Ti.
59
7.2
8.6
4.1
7-3
6-7
6.4
4-5
6.0
5-5
6.4
4.6
6.3
9-7
6.8
4.0
Note: The figures of the five most frequent types of clausula in each work are in bold type.
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Cambridge Companions Online © Cambridge University Press, 2006
IO6 THE CAMBRIDGE COMPANION TO PLATO
the Sophist, Politicus, and Philebus the three highest-frequency posi-
tions were occupied by members of this favored group, here the top
two positions were claimed by other clausulae, leaving only the
third, fourth, and fifth places for the Laws forms. Moreover, the
unpopular clausulae of the Laws were avoided less scrupulously
than in the other three works.^
Kaluscha concluded that these works belonged together chrono-
logically and that in accordance with the degree of similarity of their
prose rhythm to that of the Laws the probable order of composition
was Timaeus, Critias, Sophist, Politicus, Philebus, and Laws. In
support of this sequence he referred to the following: The Politicus,
Philebus, and Laws were connected by their stricter avoidance of the
four clausulae mentioned previously (II 7, III 3, III 6, and III 8), their
combined occurrence as a percentage of the total number of claus-
ulae being Ti. 25.3, Criti. 21.3, Sph. 19.9, Pol. 11.2, Phil. 10.9, Laws
10.0. On the other hand the Timaeus, Critias, and Sophist were
connected by the small variation in frequency of their various
clausulae, neither a strong preference for nor prejudice against par-
ticular forms being observable. This can be shown numerically by a
calculation based on Table 5; taking 6.3 as the mean percentage
frequency for the sixteen clausulae and observing the difference
from this of the actual percentage of each, the average deviation is
Ti. 1.3, Criti. 1.8, Sph. 1.0, Pol. 2.5, Phil. 2.9, Laws 3.1.
Turning to the remaining dialogues, Kaluscha discovered that
they resembled the Protagoras, Crito, and Apology in lacking any
consistent tendency to prefer certain clausulae and therefore con-
cluded that together they belonged to an earlier period, in which
Plato showed little or no conscious interest in prose rhythm.
This same subject was investigated again some years later by L.
Billig,*8 who condemned Kaluscha's treatment of it as "unsatisfac-
tory in many ways." Although he did not say what these were, they
may have included the failure to mention the edition used, to spec-
ify the minimum length of sentence, and to define certain principles
of scansion, all of which made it difficult to verify the accuracy of
Kaluscha's statistics. He attempted to forestall similar criticism of
his own inquiry by providing information about his procedure in
these respects.
His primary reason for a reinvestigation, however, came from ob-
serving in the Laws the frequent occurrence of the fourth paeon
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 107
(wvw|asa clausula, the rhythm recommended by Aristotle (Rheto-
ric III 8) for this position. Unlike Kaluscha he did not restrict the
clausula to the last five syllables of the sentence, but permitted such
variation between four and six syllables as seemed appropriate, the
basic fourth paeon, for instance, being extendable by one or two
extra syllables (Table 6).
In the Laws Billig noted the high incidence not only of the fourth
paeon and its variants but also of two other clausulae (—^* and
*). If there were an even distribution of the fifteen clausulae,
these six forms would constitute 40 percent. Perceiving that their
actual occurrence in the Timaeus barely exceeded this, but was al-
most double in the Laws, and assuming an increasing expertise on
Plato's part in achieving these preferred forms, he reached the con-
clusion that the chronological sequence corresponded roughly with
their increasing occurrence: Ti. 45.6 percent, Chti. ^2.2 percent,
Sph. 55.8 percent, Pol. 70.7 percent, Phil. 78.2 percent, Laws 77.9
percent.
More recently, the use of statistical techniques to evaluate the
significance first of Kaluscha's figures, then of data from a fresh
examination of the clausulae has confirmed the results obtained by
the two earlier investigations.39
The by this time somewhat hackneyed subject of reply formulae
enjoyed a late revival through a second article by H. von Arnim,
which reached book proportions.*0 The aim he set himself was to
make the results of his new inquiry conclusive. Previous investiga-
tions had failed to achieve this because, while they showed that
certain dialogues belonged together by reason of a common posses-
sion of particular stylistic features, they did not prove that an alterna-
tive arrangement according to others was impossible. Although it
had been discovered, for instance, that a large number of such fea-
tures connected the Sophist and Politicus to the Philebus, Timaeus,
and Laws, no one had thought of finding out how many connected
these same two dialogues to, say, the Symposium, Phaedo, and
Critias. Yet it was theoretically possible that such an investigation
would reveal a greater number than in the former case, necessitating
a complete revision of the "established" chronology.
In order to eliminate any doubt in this respect, each work needed
to be compared with every other, a task of several lifetimes if the
material were to be all the possible features of style. However, an
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Table 6. Prose rhythm in the dialogues: relative frequency by percentage of different types of clausula (broadly
defined), as calculated by Billig
(I) wwww
www —
www —
w
— w
Total of (I)
(II) _ ww
(III) — ^
Total of (I), (II) &
(III)
— w w
w —w
— \J w — W W
— w w w
w—w—w ^
-
—w—ww 1
_ w w —w
Total of remainder
Ti.
I2.O
6.4
2.9
3-i
24.4
12.6
8.6
45.6
5-5
10.1
16.1
3-i
1.8
12.4
5-2
54.2
Sph.
11.5
7-3
3 0
4.8
27.5
17.8
10.5
55-8
7.0
7-3
11.5
2.3
3.2
8.4
4-7
44.4
Ciiti.
16.2
14.4
1.8
32.4
13.5
6.3
52.2
5 4
9 9
9.9
2.7
2.7
11.7
5-4
47-7
Pol.
14.2
8-7
T 7
JW
6.6
332
21.2
16.3
70.7
8.1
4.1
6.8
2.4
1-3
5.8
1.1
29.6
Phil.
17.1
11.0
A A
6.2
38.7
23-5
16.0
78.2
4 4
2.9
5-9
1 . 0
1.4
5-3
0.8
21.7
I
16.2
9-3
1 a
j-°
4.1
33-4
21.3
14.2
68.9
7-9
5.2
7-9
1.1
1.9
6.0
1.1
3i.1
II
14.0
11.5
2.0
3-O
30.5
20.1
23.2
73-8
8.5
4-5
6.5
2 . 0
1 . 0
2 . 0
i-5
26.0
III
22.8
8.8
8.0
4.4
44.0
21.7
11.6
77-3
5.2
2.8
6.4
2 . 0
1.6
3-2
1 .2
22.4
IV
18.8
9-1
6.6
4.6
39-1
27.4
12.2
78.7
5.6
2 . 0
5.6
2 . 0
-
5.6
0.5
21.3
V
23-5
10.8
G 1
°'j
O.6
43-2
22.3
10.8
76.3
5-1
0.6
5-1
0.6
1-3
8-3
2.6
23.6
Laws
VI
21.8
12.4
A Q
*t-y
5-3
44.4
22.3
13.6
80.3
5-3
1-5
4-5
1-5
1-5
4.1
1.1
19-5
VII
19-7
8-4
r r
>• J
5-5
39-1
2 3 4
139
76.4
6-5
2.5
4.4
1.9
i-5
6.1
0.9
23.8
VIII
17.4
10.7
6.2
5.0
39-3
23.6
18.6
81.5
5.6
4-5
3-4
-
1.1
2.8
1.1
18.5
IX
22.4
9.6
6.9
4.1
43.O
25.0
16.8
84.8
5.0
2.7
2-3
0.9
0.5
3-2
0.9
15-5
X
17.5
10.4
6.3
4-9
39-1
30.5
1 1 . 1
80.7
3-7
3-O
4-9
0.8
0.8
4.2
2 . 2
19.6
XI
15-5
6-3
6.3
1.2
2 9 3
37-4
n - 5
78.2
7-5
2.9
5.2
i-7
1 .2
3-5
-
22.0
xn
18.6
7.7
7'7
5.0
39-0
24.9
13.6
77-5
6.8
2.7
5-9
1.4
0.9
5.0
-
22.7
Notes: The figures for the CritL, which Billig did not include in his investigation, were calculated by the author in accordance with his principles; likewise those for the
Sph., since a check failed to substantiate Billig's own figures.
The total occurrence for each work is not always exactly ioo % owing to rounding up or down of decimals.
C
am
b
rid
g
e C
o
m
p
an
io
n
s O
n
lin
e ©
C
am
b
rid
g
e U
n
iversity Press, 2006
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 109
alternative was available in the smaller, yet self-contained material
of affirmative reply formulae, which Arnim reexamined in order to
acquire as accurate data as possible, even though the ground had
been partially covered before by Ritter and himself. He then com-
pared each pair of works in respect of both types of reply formulae
employed and their relative frequency.*1 In this way he identified the
following groups, arranged in chronological sequence:
1. Ion, Pit.
2. La., Rep. I, Lys., Chrm., Euphr.
3. Euthd., Gig., Meno, H. Mi., Cia.
4. Cii., H. Ma., Smp., Phd.
5. Rep. II-X, Tht, Vim., Phdi.
6. Sph., Pol, Phil, Laws.
Despite the extensive nature of Arnim's material, which compre-
hended a much larger number of reply formulae than any previous
investigation, the reliability of his results was diminished by meth-
odological faults.*2 Nevertheless, it may be noted that the chrono-
logical sequence that he arrived at corresponded broadly with that
obtained by his predecessors.
In his later years C. Ritter returned to the subject of research into
Plato's style, which he had done so much to promote, for the specific
purpose of determining the order of composition of the early dia-
logues.43 By looking at the occurrence in them of five features** he
was able to subdivide this group into an earlier and later set, the
former comprising the Hippias Minoi, Chaimides, Laches, Piotago-
ms, Euthyphio, Apology, Ciito, with the Goigias and Meno at the end,
the latter the Hippias Majoi, Euthydemus, Menexenus, Ciatylus,
Lysis, Symposium, Phaedo, and Republic I. It must be said, however,
that this conclusion, based as it was on low frequency figures in every
case, should be regarded as no more than a probability.
Nearly a century after its inception through Campbell the stylistic
method came full circle with an examination of Plato's vocabulary
by A. Diaz Tejera.** His approach, however, was different; whereas
Campbell's standard of reference was internal, in that he measured
the degree of affinity of the other dialogues to the Laws, Diaz
Tejera's was external. Assuming that the development of the various
Greek dialects into the Koine should be traceable, he collected to-
gether what he called "the non-Attic vocabulary,*6 which is well-
Cambridge Companions Online © Cambridge University Press, 2006
IIO THE CAMBRIDGE COMPANION TO PLATO
documented in the Koine/' then examined its occurrence in Plato's
works.
Leaving aside the Laws, which he took for granted as the final
work, he found the highest incidence in the Timaeus and Critias,
followed by the Politicus, then the Sophist and finally the Philebus.
In a "later middle group/' which showed a considerably lower inci-
dence, he placed the Theaetetus, Phaedrus, and Parmenides,^ pre-
ceded in turn by an "earlier middle group" consisting of Republic II-
X, Phaedo, Symposium, and Cratylus. Observing that the incidence
in the latter three was roughly comparable with that in the early
books of the Republic (as far as VI 5O2e), he inferred that the first
part of the Republic was written between 388/7 B.C. (Plato's return
from Sicily) and 384 B.C. (terminus post quern of the Symposium)*8
and was followed by the Cratylus, Symposium, and Phaedo before
the Republic was resumed.
The main differences between Diaz Tejera's chronological order
and that arrived at by his predecessors were the separation of the
Republic by other works and the reversal of the positions in the final
group of Timaeus/Critias and Philebus. If the investigation that pro-
duced these results had been sound, they would have required seri-
ous consideration, but it was seriously flawed in both concept and
procedure. Starting, as it did, from the chronological divisions estab-
lished by earlier research and accepting as evidence of late composi-
tion words common to the Koine and the final group of dialogues, its
argument tended to be circular, and this basic fault was aggravated
by various procedural errors, such as incorrect or inconsistent classi-
fication of words, incomplete statistics, and faulty calculations.*?
The rhythm of Plato's prose was once more examined for chrono-
logical purposes by D. Wishart and S. V. Leach,*0 who analyzed in
this respect not merely the clausula, as Kaluscha and Billig had, but
the whole sentence. Owing to the exhaustive nature of the investi-
gation, samples rather than whole works were looked at, and both
the initial categorization of the text into long or short syllables and
the subsequent assessment of the statistics were carried out by
computer.
The authors took as their unit of measurement a group of five
syllables, yielding thirty-two permutations. Every sentence was ana-
lyzed into such groups sequentially, that is, first syllables 1-5, then
2-6, then 3-7, and so on, after which the occurrence of each of the
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 111
thirty-two types was expressed as a percentage of the total number
of syllable groups in the sample. The works that they considered
were as follows: Sym. (4), Phdr. (5), Rep. (3), Sph. (1), Pol. (1), Phil. (1),
Epist. VII (1), Ti. (9), Criti. (3}, Laws (5). The figures in parenthesis
indicate the number of samples taken from the work in question,
each sample comprising between two and three thousand groups of
five syllables.
To determine the interrelationships of the various samples and
works, five different statistical techniques were used: three of clus-
ter analysis, one of principal components analysis, and one of multi-
dimensional scaling. The purpose of cluster analysis was to identify
groups of works or samples exhibiting a uniform use of prose
rhythm. If it resulted in the clustering of separate samples from the
same work, it would confirm that the work in question displayed
consistent rhythms and so could be regarded as homogeneous; if, on
the other hand, any sample could not be clustered with the rest, it
would suggest either a difference of genre or a chronological separa-
tion or unauthenticity. The same would be true of whole works.*1
It turned out that the thirty-three samples were grouped together
according to their origin with the exception of those from the Re-
public and Phaedrus. In the former the sample from Book II ap-
peared to be widely separated from those from Book X, which the
authors were at a loss to explain, though they suggested as possible
causes the shortness of the Book II sample and the fact that it came
from a speech, whereas the other two were from a narrative. An-
other reason might be that the Book II sample contained several
quotations in verse which clearly should not have been included in
an analysis of prose rhythm (cf. CPD, 240). In the Phaedrus, while
the four samples from Socrates' two speeches were grouped to-
gether, that from Lysias's speech was quite different. They consid-
ered that in view of the uniformity of rhythm in the four Sympo-
sium samples, despite their being parodies, imitation of Lysias's
style would not account for its deviation and so concluded that it
was probably Lysias's own compositions
Regarding the ten works the authors decided that the chronologi-
cal sequence was Phdr., (Smp. and Rep.), Ti., Sph., Criti., [Epist. VII
and Pol.), Phil., Laws,^ thus confirming the order arrived at by ear-
lier investigators of Plato's prose rhythm, at least from the Timaeus
onward. 54
Cambridge Companions Online © Cambridge University Press, 2006
112 THE CAMBRIDGE COMPANION TO PLATO
The most recent attempt to solve the chronological problems 5 was
based on a computer analysis of the occurrence in words of certain
letters, the occurrence being classified according to whether it was
(a) anywhere in the word, (b) at the end,*6 or (c) in the penultimate
position. The incidence of the significant letters, or variables, which
were found to be thirty-seven in number,57 was determined for se-
quential samples of 1,000 words from both Plato and other contem-
porary prose authors. The statistical profiles of the samples formed
by these thirty-seven variables were then compared with one an-
other by various techniques, such as cluster analysis and discrimi-
nant analysis, in the expectation of finding insignificant differences
indicative of homogeneity between samples from the same author,
but significant ones between those from different authors. While
this expectation was for the most part fulfilled, there was a disturb-
ing number of instances where statistical analysis failed to distin-
guish the works of two authors.*8 Nevertheless Ledger concluded
that the results of comparisons with genuine works suggested the
authenticity of Alcibiades I, Theages, Epistles VII, Hippias Major,
Epinomis, and possibly also of Alcibiades II and Hippias Minor.
A comparison of the Platonic works using canonical correlation
analysis to establish the chronological order indicated the existence
of a sharply defined final group consisting of Phil, Cleit., Epist. Ill,
VII, and VIII, Sph., Pol, Laws, Epin., Ti., and Criti., written in that
order between 355 B.C. and 347 B.C.59 Prior to these came the Phae-
drus and Menexenus. While accepting this fairly late position for the
Phaedrus in compliance with his statistics, Ledger rejected it for the
Menexenus in deference to the traditional view of an earlier date,
based in part on a supposedly topical reference (245c) to the Peace of
Antalcidas of 386 B.C. Immediately before these was a "middle
group" of works written probably between 380 B.C. and 366 B.C. in
the order Euthd., Smp., Cra., Rep., Prm., Tht., Epist. XIII,6o preceded
in turn by an "early middle group" consisting of Grg., Menex., Meno,
Chrm., Ap., Phd., La., Prt., written in this order probably between
387 B.C. and 380 B.C. Finally the earliest group comprised Lys.,
Euphr., Minos, H. Mi., Ion, H. Ma., Ale. I, Theag., and Cri., the Lysis
being dated to 400 B.C. before the death of Socrates on the basis of
the anecdote in Diogenes Laertius (II 35).
The study of Plato's literary style has revealed two broad develop-
ments, an earlier one which was slow and gradual, and a later one,
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 113
starting when he was about sixty, which was sudden and rapid.
Regarding the former, where the changes concerned his vocabulary
and were for the most part probably unconscious, one would expect
the trend to be uneven and at times haphazard;61 in the latter, which
concerned the euphony of his prose and involved a deliberate choice
in respect of hiatus avoidance and rhythm, a more rational and sys-
tematic evolution might be anticipated, with any aberrations in it
explicable by known or deducible factors.
The early research on Plato's vocabulary by Campbell, Ditten-
berger, and Schanz, culminating in Ritter's book on the subject,
identified in the Sophist, Politicus, Philebus, Timaeus, Critias, and
Laws a group of dialogues distinguished from the rest by an exclu-
sive or increased occurrence in them of certain words and phrases.
Subsequent investigations into this aspect of style arrived at the
same conclusion, and the dichotomy was confirmed by two further
criteria with the discovery that only in these works, together with
the Epinomis and Epistle VII, did Plato make a consistent attempt to
avoid certain types of hiatus and achieve a different kind of prose
rhythm.
It has been argued that Plato avoided hiatus changeably rather
than consistently after a certain date.62 This is to attribute to an
elderly philosopher a fickle attitude, which is hardly compatible
with the character of one who in his works emphasizes the impor-
tance of rational, consistent behavior. Of course Plato could change
his style within a single dialogue, as in the Symposium and Phae-
drus, but these changes were made for a specific purpose that is
immediately apparent. No reason has so far been adduced why he
should have employed the principle of hiatus avoidance intermit-
tently, and in the absence of such a reason it is unsatisfactory to
resort to the use of analogy, especially of Isocrates' forensic speeches,
where the greater or lesser avoidance of hiatus is explicable on vari-
ous grounds, not least temporal and commercial, considerations that
hardly applied to Plato.
Regarding the question of sequence within the final chronological
group, in comparing the various kinds of evidence particular weight
should perhaps be attached to prose rhythm and the avoidance of
hiatus, because unlike vocabulary they appear to be independent
both of a work's form and of its content. Although the testimony of
the data for hiatus avoidance was ambiguous, three independent
Cambridge Companions Online © Cambridge University Press, 2006
114 T H E CAMBRIDGE COMPANION TO PLATO
investigations of clausula rhythm and one of sentence rhythm
agreed in concluding that the order of composition was TL, Criti.,
Sph., Pol, Phil., Laws. In the light of this the ambiguity of the hiatus
evidence regarding the place of the Philebus may be resolved in favor
of its proximity to the Laws, a position supported by particular as-
pects of hiatus63 and by other features, such as the reversion to
longer forms of reply formulae after a predominance of abbreviated
versions in the preceding works,6* the culmination of a trend toward
the more frequent use of superlative expressions,6* the high propor-
tion of jteQi,66 and an increased preference compared with the Ti-
maeus, Critias, and Sophist for a long final syllable in clausulae
(CPD, 188-90).
The final position allocated by Ledger to the Timaeus and Critias
would indicate a fluctuating level of hiatus avoidance in the final
group: less strict in the Philebus, strict in the Sophist and Politicus,
less strict in the Laws, strict in the Timaeus and Critias. Likewise
with clausula rhythm: The forms preferred by Plato in the Laws
would appear with similarly high frequency right at the start in the
Philebus, with much lower frequency in the Sophist, increasing in
the Politicus toward that of the Laws, but falling away again in the
Timaeus and Critias, to the level of the Sophist. As both these lin-
guistic features were adopted consciously by Plato, such indecisive-
ness would be remarkable. Since the Timaeus and Critias are for the
most part continuous narrative compared with the dialogue form of
the Sophist, Politicus, and Philebus, the difference in Ledger's statis-
tics might be attributable to the same cause that he adduced6? for the
odd results obtained for the Apology and Menexenus, works of a
rhetorical character, namely a difference of genre.
By comparison with the differences that distinguish the final
group, those which separate the dialogues of Plato's middle period
from all preceding it are not as sharp, connected as they are with the
earlier, gradual development of his style. Ritter, incorporating the
results of his predecessors' research with his own, found that many
of the criteria used to identify the final group also served to separate
the Parmenides, Phaedrus, and Theaetetus from the remaining dia-
logues,68 and the same division was made by later investigators.6*
On the question of the unity of the Republic, Siebeck,?° Ritter, and
Arnim arrived independently at the view that Book I, which con-
tains several features characteristic of the early dialogues, was origi-
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 115
nally a separate work written some time before the rest, but possibly
revised at the time of its incorporation.
Despite the fact that there cannot be the same certainty about the
sequence within this group as about that in the last, examination of
Ritter's criteria (cf. CPD, 7gff) suggests that the order of composition
was Republic, Theaetetus, Phaedrus, which agrees with his own
conclusion (cf. CPD, 77).71 While the Parmenides unquestionably
belongs in the same group (cf. CPD, 66), its peculiar character makes
it difficult to determine its relationship to the above three works (cf.
CPD, 84). On the other hand, if one also takes into account the
apparent reference in the Theaetetus to the Parmenides and the fact
that in the Theaetetus Plato renounces the use of the reported dia-
logue form, which seems to be merely an explicit declaration of a
practice already implicity adopted early in the Parmenides (137c),
presumably induced by recollection of the wearisome repetition of
E<fyr\ ("he said"), and so on, in the Republic and the prospect of its
still greater occurrence in a dialogue with such frequent changes of
speaker, then the correct place for the Parmenides would appear to
be between the Republic and Theaetetus.
On the sequence of dialogues in the early group little can be said.
Division into subgroups also seems out of the question. The diffi-
culty is that the statistics produced by past research usually relate to
linguistic features that are primarily characteristic of works belong-
ing to Plato's middle and late periods; consequently, their occur-
rence in the early period tends to be slight and spasmodic.
The problem is compounded by two other factors: first, many
investigations concerned the use of reply formulae, which was preju-
dicial to works containing little dialogue (e.g., Menex., Ap., Cri.)}
second, most scholars of the last century omitted certain works
altogether from their inquiries, especially those suspected at that
time of being unauthentic, making a general comparison impractica-
ble. Nevertheless, if the frequency with which features characteris-
tic of the middle and late works occur in the early dialogues is
accepted as an indication of their chronological proximity, then the
Phaedo, Cratylus, Symposium, Republic I, Lysis, Menexenus, Euthy-
demus, and Hippias Major would certainly have to be regarded as
among the last of this group. Their relative order, however, cannot
be determined on the basis of the stylistic evidence that has so far
come to light.
Cambridge Companions Online © Cambridge University Press, 2006
I l 6 THE CAMBRIDGE COMPANION TO PLATO
NOTES
1 For a full discussion of the evidence, see H. Thesleff, Studies in Platonic
Chronology (Helsinki, 1982), 7-66.
2 Pol. II 6, 1264.1)24-27.
3 Cf. Prm. I27b2, C4-5.
4 See the tables in C. Ritter, Platon (Munich, 1910), 230-1, and Thesleff,
Studies in Platonic Chronology, Sff.
5 G. Grote, Plato and the Other Companions of Socrates, id ed., 3 vols.
(London, 1867), 1:185-6, 278-9.
6 L. Campbell, The Sophistes and Politicus of Plato (Oxford, 1967),
introduction.
7 In the last century many dialogues now accepted as genuine were consid-
ered unauthentic. As scholars' views in this respect varied, so too did the
number of works forming the subject of any investigation.
8 See L. Brandwood, The Chronology of Plato's Dialogues (Cambridge,
1990) 5—7 (hereafter referred to as CPD).
9 W. Dittenberger, "Sprachliche Kriterien fur die Chronologie der platoni-
schen Dialoge," Hermes 16 (1881): 321-45.
10 it \ir\v; ("What else indeed?") and oXka. . . \ir\v} ("But what else in-
deed?"), the intermediate word normally being TI, are strong affirmative
replies, while ye \ir\v ("but indeed") is usually adversative. The absence
of the first two from Ti. and Criti. results from their lack of dialogue.
11 K. J. Dover, "The Date of Plato's Symposium," Phronesis 10 (1965): 2 -
20.
12 Acharn. 757, 784.
13 M. Schanz, "Zur Entwicklung des platonischen Stils," Hermes 21
(1886)1439-59.
14 The three examples in Prt. occur in the analysis of a poem by Simonides
and, being quotations, ought to be discounted.
15 Simultaneously and independently, it seems, by H. Siebeck, Untersuch-
ungen zur Philosophie der Griechen (Halle, 1888), 253//, and C. Ritter,
Untersuchungen uber Platon (Stuttgart, 1888).
16 CPD, chaps. 10, 11, 13, and 19.
17 At this time the Prm. was held by Ritter to be unauthentic. Otherwise
his statistics would have required it to be placed in this group.
18 E.g., (1) the complete absence of xai \iaka ("very much so"), xi \ir\v-, and
696(05 ("rightly") together with its superlative, though they occur in all
the other books; (2) six of the eleven instances in the Rep. of 6fjxa ("in-
deed") with a reply occur in this book, as do eight of the twenty instances
of ((xxtvexai ("apparently"); (3) the preponderance of n&vv ye over navv \iev
ovv (16 : 5) is the reverse of that in every other book (total 24 : 59).
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 117
19 E.g., one instance each of aoiora elorjxag ("Well said!"), xal JUOC; dv;
("How might that be?"), Ionic dative (the Attic form with an iota suffix),
ye \ir\v, Jiavr&Jtaoi [lev ovv ("Most certainly"), together with a preponder-
ance of ahf\Qr\ over akr\Qr\ Xeyeig (5:2).
20 His argument for these dispositions, however, is not convincing [CPD,
74-6).
21 A reassessment of Ritter's data (CPD, 77, 82) shows that there are reason-
able grounds for concluding that the Phdr. was written after both the
Tht. and Rep.
22 W. Lutoslawski, The Origin and Growth of Plato's Logic (London, 1897),
chap. 3, "The Style of Plato."
23 For a more detailed criticism see CPD, 130-5.
24 C. Baron, "Contributions a la chronologie des dialogues de Platon,"
Revue des Etudes grecques 10 (1897): 264-78.
25 Instances of Jtegi in the various forms of the phrase KEQI nolXov
jroieTa9ai ("to rate highly"), which never appears to admit of anastrophe,
were excluded from the calculation. The figures are the author's and
relate to the O.C.T. Baron's, referring to another text, sometimes differ,
but only slightly.
26 The high percentage for the Menex. and La. may be explained by the low
overall occurrence of the preposition in the former, and by special fac-
tors such as repetition in the latter (cf. CPD, 119).
27 See below and CPD, 120.
28 G. Janell, "Quaestiones Platonicae," fahrbucher fur classische Philo-
logie, Supp. 26 (1901): 263-336.
29 F. Blass, Die attische Beredsamkeit (Leipzig 1874), 2: 426.
30 I.e., a word beginning with a vowel following one ending in a vowel. In
the fourth century with the advance of rhetorical technique such a clash
of vowels came to be regarded as detracting from the euphony of prose.
31 Reinvestigation showed that in his later works Plato did try to avoid
hiatus with these too (CPD, 162).
32 Janell used Schanz's text. In Burnet's there are even fewer instances:
only one in the second passage, none in the third.
33 W. Kaluscha, "Zur Chronologie der platonischen Dialoge," Weiner
Studien 26 (1904): 190-204.
34 In the sense that even a short syllable would be lengthened by the pause
in speech between the end of one sentence and the beginning of the next.
35 The true situation is a little more complex. Clausulae in works of the
middle period show a preponderance of long final syllables; e.g., in Rep.
VIII-X, 478 end in a long, 251 in a short syllable, an excess of long over
short of 90%. Although in the Ti. the position is reversed, with a 2%
excess of short over long, the preponderance of the long syllable re-
Cambridge Companions Online © Cambridge University Press, 2006
I l 8 THE CAMBRIDGE COMPANION TO PLATO
appears in the Criti. with a 20% excess (though in this case the low
number of clausulae makes the calculation less reliable), and gradually
increases through the Sph. (7%), Pol. (20%), and Phil. (34%) to the Laws
(66%).
36 Cf. CPD, table 18.4.
37 About the Criti. he drew no conclusions owing to the limitations of its
size and the resulting statistics, but contented himself with attaching it
to the Ti.
38 L. Billig, "Clausulae and Platonic Chronology," Journal of Philology 35
(1920): 225-56.
39 Cf. CPD, 198/7.
40 H. von Arnim, "Sprachliche Forschungen zur Chronologie der platoni-
schen Dialogue/' Sitzungsberichte der Kaiserlichen Akademie der Wis-
senschaften in Wien: Philos. Hist. Klasse 169.1 (1912): 1-210.
41 Counting the books of the Rep. and Laws separately there were forty-
two works to be compared, the Ap., Menex., Ti., Criti., and Laws V and
XI being excluded owing to their lack of dialogue.
42 Cf. CPD, 215//.
43 C. Ritter, "Unterabteilungen innerhalb der zeitlich ersten Gruppe pla-
tonischer Schriften," Hermes 70 (1935): 1-30.
44 E.g., (a) the particle \ir\v ("indeed"), (b) cog with a superlative adjective or
adverb in the sense "as (e.g., great) as possible," (c) the change in use of
exegog ("other of two") to that of oXKoq ("other"), (d) the interchange in
function of coorceo ("like") and oiov ("like for example").
45 A. Diaz Tejera, "Ensayo de un metodo linguistico para cronologia de
Platon," Emerita 29 (1961): 241-86.
46 By this he meant neologisms, Ionicisms, and poeticisms.
47 The incidence in the Prm. was less than half that in the other two, but
he attributed this to the monotonous nature of its subject matter, espe-
cially in the ontological section.
48 See note 11.
49 Cf. CPD, esp. 233-4.
50 D. Wishart and S. V. Leach, "A Multivariate Analysis of Platonic Prose
Rhythm," Computer Studies in the Humanities and Verbal Behavior 3
(1970): 90-9.
51 For an explanation of these techniques and their respective results see
CPD, 238-46.
52 The absurdity of their further conclusion, that it was later than the other
samples from the Phdr., appears not to have struck them (cf. CPD, 247).
5 3 The relative order of works in parenthesis could not be determined.
54 Concerning the position allocated to the Phdr., serious doubts arise from
the choice of samples; all were taken from speeches rather than the
Cambridge Companions Online © Cambridge University Press, 2006
Stylometry and chronology 119
dialogue section, and Socrates' two speeches, which provided four sam-
ples, are specifically denoted by Plato himself as poetical in character
(241 ei and 25J2L/\), a fact borne out by observation (cf. CPD, 57-8).
55 G. R. Ledger, Re-counting Plato (Oxford, 1989).
56 Iota subscript was ignored, perhaps putting these statistics in some
doubt.
57 Listed at ibid, 9.
58 Cf. ibid., 66-8 and 93//. Discriminant analysis, for instance, attributed
three of the eight samples from Rep. I to Xenophon (p. 103), while in a
comparison of Xenophon's Oeconomicus and Memorabilia with several
dialogues the Phdr. proved to be closer than the Oeconomicus in style to
the Memorabilia, and the Prt. closer than the Memorabilia to the
Oeconomicus (p. 160).
59 In this scheme the unfinished state of the Criti. was ascribed to Plato's
death. However, one may also conjecture that the Tht. with its apparent
reference to the Isthmian war of 369/8 B.C., was the last work to be
written before Plato's departure for Syracuse in 367; that the exuberant
expression of delight in the attractions of the countryside near Athens,
which has induced some to regard the Phdr. as a youthful work, may
instead be attributed to Plato's relief on returning home a year later after
being subjected to a period of virtual imprisonment; and that the reason
for the interruption to the composition of the Criti. was his sudden
departure for Syracuse again in 361.
60 Like several earlier investigators Ledger found the Prm. awkward to
place, differing so much in style from the other dialogues that "most
tests of authorship would lead us to conclude that it was not written by
Plato" [Re-counting Plato, 213).
61 Nevertheless, in the case of individual linguistic features it is necessary
to assume initially that the trend is, if not even, at least unilinear.
Comparison of several such criteria provides the necessary correction.
62 E.g., G. Ryle, Plato's Progress (Cambridge, 1966), 297; R. A. H. Wa-
terfield, "The Place of the Philebus in Plato's Dialogues," Phronesis 25
(1980): 274-6.
63 E.g., the frequency in both works of addresses like d> exaiQe ("my
friend") and d> doiore ("my good man"), six times in the Phil, and seven-
teen in the Laws, whereas in the other works of this group they are not
found at all. If the Phil, had represented Plato's first serious attempt at
reducing the occurrence of hiatus, he would hardly have failed to elimi-
nate such eminently avoidable instances. Moreover, in the Ti. and Criti.
there is a temporary increase in "permissible" hiatus to a level exceed-
ing even that in works where hiatus was not avoided (cf. CPD, 162-3),
indicative perhaps of Plato's first serious attempt to avoid the "objection-
Cambridge Companions Online © Cambridge University Press, 2006
I2O THE CAMBRIDGE COMPANION TO PLATO
able" kind, which at this transitional stage succeeded only at the ex-
pense of a rise in the former.
64 E.g., &XT)6TJ Xeyeig and dXr]QEcn:axa/6QQ6xaTa Xeyeig instead of aXr]6fj and
akr\QEOxaxa/dQQ6xaxa (cf. CPD, 88 and 99//.).
65 Only in the Phil, and Laws do the superlative reply formulae equal or
surpass the positive forms (cf. CPD, 87-9).
66 See Table 3. Note also that its occurrence in Ti. and Criti. is much lower
than in the other works of the late group.
67 E.g., Re-counting Plato, 127, 145, 163.
68 CPD, 57-66.
69 E.g., Arnim [CPD, 97//.) and Baron [CPD, 116//.).
70 Untersuchungen zur Philosophie der Ghechen.
71 The late position of the Phdr. is further supported by its higher propor-
tion of ovxoog to TCO ovxi (cf. CPD, 81), a percentage of rhetorical ques-
tions as reply formulae equaling that found in works of the last group (cf.
CPD, 103), the frequency of JTEQI (cf. CPD, 121) and passages in which
there appears to be a conscious effort to avoid hiatus (cf. CPD, 155),
leading to its lowest incidence outside works of the final group (cf. CPD,
156). In addition, the evidence of an interest in prose rhythm (cf. CPD,
158) together with mention of Isocrates (cf. CPD, 160) perhaps presages
the development of this in subsequent works.
Cambridge Companions Online © Cambridge University Press, 2006
TERRY PENNER
4 Socrates and the early dialogues
Can the philosophical views of the historical Socrates be distin-
guished from those of his pupil Plato? And if so, how do the master's
views differ from the pupil's? And do these Socratic views add up to
a coherent philosophical position?
In Section I of this chapter, I explain the basis on which, following
most modern interpreters, I feel able to divide Plato's dialogues into
a group of (earlier) "Socratic" dialogues, where the character Socra-
tes speaks more or less for the historical Socrates; and a group of
(middle and later) dialogues in which the main character (now not
always Socrates) speaks rather for Plato. I argue that the Plato of the
middle and later dialogues, though some of his views remain the
same, and though he attacks some of the same enemies and for some
of the same reasons, has nevertheless in some ways gone well be-
yond the master. On some points, I suggest, he even contradicts him.
In Section II, I contrast these Socratic dialogues with the other
dialogues - first, in their form, method, tone, and subject matter;
second, in their attitude to the sciences (arts, crafts, expertises),
education, rhetoric, and mathematics; and third, in their theories of
virtue, desire, and "weakness of will." In Section III, I address myself
to the question with what right I attribute any views at all to a
philosopher who claimed that he knew only that he knew nothing -
I would like to thank Antonio Chu, Paula Gottlieb, and Ruth Saunders for reading an
earlier draft, saving me from many errors, inaccuracies, and infelicities. My greatest
debt, in this article as in several other recent articles of mine on Socrates, is to
Richard Kraut, who gave me a superb and testing set of comments on, and objections
to, my penultimate draft. To persuade him on some of these matters would be to
achieve something in Socratic studies. I fear I may still be some distance from that
goal.
1 2 1
Cambridge Companions Online © Cambridge University Press, 2006
122 THE CAMBRIDGE COMPANION TO PLATO
especially when the Socratic dialogues virtually all end negatively. I
then make a suggestion as to the sort of thing I think Socrates is
doing in these dialogues, with an illustration from the Hippias Mi-
nor. And then I draw a moral from the illustration: Never consider
any one expression of Socrates' views in isolation from other expres-
sions of Socrates' views. In Section IV, I explore the central Socratic
concern with care for the soul, showing its connection with (what I
shall, for convenience, call) Socrates' ethical egoism. In Section V, I
consider Socrates' response to those he thinks of as his principal
adversaries in his drive to get people to care for their souls: the
sophists and rhetoricians, as well as the politicians and the poets.
And I conclude, in Section VI, by addressing myself to various ques-
tions about Socratic method and Socratic ignorance. Sections III-VI,
taken together, are also meant to exemplify, if somewhat sketchily
(given the scope of the essay), ways in which one may come to see
the overall coherence of Socrates' doctrines.
I have tried to keep the exposition of my picture of Socrates in the
main text fairly uncluttered with objections and responses, so that it
can be read on its own without the footnotes. At the same time, I
should alert the reader to the fact that many of my interpretations of
Socratic passages in Sections II-VI will be controversial. I have there-
fore tried to provide some objections and replies in the footnotes,
along with some indications of where at least some opposing view-
points can be found.1
I. CAN WE MAKE A DISTINCTION WITHIN PLATO'S
DIALOGUES BETWEEN THOSE IN WHICH THE
CHARACTER SOCRATES EXPRESSES VIEWS AND
CONCERNS OF THE HISTORICAL SOCRATES AND
THOSE IN WHICH HE EXPRESSES INSTEAD
DISTINCTIVE VIEWS OF PLATO HIMSELF?
There is a tradition within recent philosophy and scholarship that
says we can. This tradition has two principal sources: testimony
from Aristotle that contrasts Socrates and Plato,- and stylometric
evidence on the dating of Plato's dialogues.
Aristotle tells us that (i) Socrates asked only, and did not reply; for
he confessed that he knew nothing. He also tells us that (ii) Socrates
concerned himself with ethical matters only, being not at all con-
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 123
cerned with nature as a whole; that (iii) he was the first to argue
"inductively"; that (iv) he was the first to search (systematically) for
the universal, and for definitions - that is, to ask What-is-it? ques-
tions: questions of the sort, What is justice?, What is courage?,
What is piety?, and so forth; but that, on the other hand, (v) he did
not "separate" these universals, as Plato did: for Plato supposed that,
since perceptibles are constantly changing, there could be no knowl-
edge of the perceptibles, but only of some other things, the Ideas (or
Forms).2
Now it is far from clear what "separation" [chohsmos] is.* But at
least it is clear from the above testimony that where in the Platonic
dialogues we see the character Socrates characterizing entities like
justice itself and the good itself as changeless, and contrasting them
with the ever-changing perceptibles, we should suppose that the
character Socrates speaks not for the historical Socrates but for
Plato. This becomes even clearer when we notice Aristotle's testi-
mony that Plato, while young, studied with Cratylus, "becoming
familiar first with Cratylus and Heraclitean opinions" (to the effect
that all perceptibles are always in flux, and that there is no knowl-
edge of them); and that Plato held these Heraclitean views even
later. This Heracliteanism of Plato's is plainly a second source of
Plato's views - a source that is relatively independent of Socrates'
views.*
There is a third possible, relatively independent, source of Plato's
views that Aristotle tells us of: the views of the Pythagoreans whom
Plato "follows in many things" [Met. I.6.987a3o). Aristotle tells us
that the Pythagoreans, noting such phenomena as the mathematical
ratios involved in musical intervals, thought the elements of all
things were the elements of numbers. He also tells us that in at least
a few cases, for example, justice, soul and reason, opportunity, and
marriage, they tried, in a rather rudimentary way, to give answers to
the Socratic What-is-it? question in terms of numbers.* The sugges-
tion seems to be that as the Pythagoreans thought perceptibles had
to be understood by reference to abstract structures (numbers), so
Plato thought perceptibles had to be understood by reference to
other abstract entities - the Forms. The Pythagoreans say percepti-
bles "imitate" the numbers, Plato says that the perceptibles "par-
take in" the Forms; only the name is different, Aristotle says.6
Thus Aristotle's testimony gives us a leg up on distinguishing
Cambridge Companions Online © Cambridge University Press, 2006
124 T H E CAMBRIDGE COMPANION TO PLATO
dialogues where the character Socrates speaks for the historical Soc-
rates and dialogues where the character Socrates speaks rather for
Plato. Wherever there is talk of flux, or of perceptibles partaking or
participating in Forms, and wherever there is substantive argument
about cosmological matters (such as teleology in nature or the im-
mortality of the soul), we are hearing Plato rather than the historical
Socrates.
The distinction between Socratic and more distinctively Platonic
dialogues has been made more secure by the stylometric investiga-
tions of the past hundred and more years.? This study has enabled
scholars to reach some considerable degree of consensus about the
placing of Plato's dialogues into three broad chronological groups:
early, middle, and late. Scholars then note that a satisfying historical
picture of the relations between Socrates and Plato can be built up if
(i) we identify (most of) the stylometrically early dialogues as
"Socratic" - that is, as dialogues in which the main character, Socra-
tes, can be taken to express views of the historical Socrates,- and (ii)
we treat the remaining dialogues as ones in which the main char-
acter (often, but now not always, Socrates) speaks for the views of
Plato - views that are sometimes, though very far from always, dis-
tinct from, and even conflicting with, the views of Socrates. This
division - with a few dialogues considered transitional - pretty well
preserves, it can be argued, the distinction between Socrates and
Plato demanded by Aristotle's testimony.
By these criteria, then, I take the following as Socratic: Hippias
Minor, Charmides, Laches, Protagoras, TLuthyph.ro, Apology, Crito,
Ion-, Gorgias, Meno; Lysis, Euthydemus, Menexenus, Hippias Major,
Republic Book I, with only the barest probability on stylistic
grounds that there are breaks at the semicolons,8 but with no further
orderings within these subgroups. This list includes all of the
stylometrically early group, except for the Cratylus, Symposium,
and Phaedo, since, by the Aristotelian criteria above, they are clearly
more Platonic than Socratic.* Following Dodds, I single out the Gor-
gias and Meno, with their Pythagorean elements, as being transi-
tional to Platonic doctrine. Indeed, parts of those dialogues may be
taken to be Platonic rather than Socratic, showing as they arguably
do the effects of Plato's encounters with Pythagoreanism during his
first visit to Sicily in 387 B.C. (I am thinking here principally of the
myth in the Gorgias and a few other oblique references in that dia-
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 125
logue to Sicily and Pythagoreanism; and of the theory of recollec-
tion, and perhaps also the introduction of the method of hypothesis,
in the Meno. As we have already seen, Aristotle seems to believe in a
connection between the [non-Socratic] Platonic theory of Forms and
Pythagorean attitudes to mathematics.)10
II. SOME CONTRASTS BETWEEN THE SOCRATIC
DIALOGUES AND THE OTHERS
On the above account, we can now advance the following generaliza-
tions, some of them rough in themselves, but cumulatively worth
attending to. First, on form, method, tone, and subject matter:
1. The Socratic dialogues tend to be short (notable exceptions:
Gig. and Pit.)-, the other dialogues tend to be much longer.
2. The Socratic dialogues tend to be aporetic and without positive
results, as befits a principal investigator who confesses his own igno-
rance; the other dialogues often have positive results, the main char-
acter often laying out much positive doctrine (notable exceptions:
Tht. and the Prm.).
3. The Socratic dialogues are amusing, bantering, extroverted, opti-
mistic, and mischievous in tone; the other dialogues are most often
more inspirational or scientific in tone, but also more introverted,
pessimistic, and brooding.11
4. The Socratic dialogues are almost exclusively ethical in con-
tent, concerned with individual ethics and individual education-
"care for the soul" both for oneself and for the young; other dia-
logues are interested in many other topics besides ethics.12
5. On the question of the immortality of the soul, there is little
interest in the Socratic dialogues. The Apology seems almost agnos-
tic about it (40-41), though it is affirmed in the horrific myth at the
end of the Gorgias (which I have suggested is Pythagorean) and (in
the mouth of the Laws) at the end of the Crito (53b-c). But there are
certainly no arguments for it. Elsewhere, beginning with the Phaedo,
the immortality of the soul is passionately embraced, and argued for
with intensity.13
Now for some considerations relating to Socrates' attitude to the
sciences:
6. The Socratic dialogues treat of virtue as an expertise (science,
art, craft) like any other expertise (such as medicine, navigation,
Cambridge Companions Online © Cambridge University Press, 2006
126 THE CAMBRIDGE COMPANION TO PLATO
cobblery, boxing, horse training, arithmetic, geometry).1* This exper-
tise is evidently thought of by Socrates as intellectual, and as involv-
ing the ability to "give an account," to explain to others, and to
teach them; it is not just a matter of "knowing how."1* It does not
involve any (propaedeutic) training of the emotions independent of
the understanding to be reached by discussion. Socrates is unlikely
to have granted that there could be any education that is not routed
via the intellect. (Even of little children, there is no reason to believe
that Socrates would think any kind of training worthwhile that
doesn't proceed by way of getting the child to understand things for
himself or herself.) By contrast, in middle and later dialogues, virtue
is no longer merely an intellectual expertise, but involves, at least as
a precondition, a degree of training of emotions and attitudes in
ways that do not involve very substantially the intellect of the
trainee. The right stories must be told, the right music listened to -
never mind any discussion - before Socratic dialogue can do the
slightest good (Rep. II 376c//).16
7. There is a similar contrast in attitudes to rhetoric. While the
denial that rhetoric is a science remains common to Socratic and
Platonic dialogues,x? it is evidently Socrates' view that the only sci-
ence there could be of persuading someone that p, is the the science
of teaching them that p. Extrapolating from the whole of the So-
cratic dialogues, we may reasonably identify this "teaching" with
bringing people to understand - for themselves, and, indispensably,
by way of Socratic cross-questioning - how it can be the case that p.
By contrast, though Plato agreed that there can only be a science of
persuading someone that p in people who themselves have knowl-
edge of p, he did not doubt that this science could proceed by way of
appeals to emotion and without what Socrates called teaching. (Cf.
the entire scheme for nursery and elementary education at Rep. II
366C-III 412b, including the "noble falsehood" of 414^-415d with
4i6e-4i7a ; cf. 382c-d, 378a, 389b-c.)18
8. Returning to expertises and sciences proper, the Socratic dia-
logues treat arithmetic and geometry as just ordinary expertises like
any other, such as cobblery and boxing; they are not marked off as
bringing us special knowledge, or as bringing us objects of some
special epistemological status. By contrast, in some of the other
dialogues, especially the Republic (Books V-VII), it is clear that the
sciences of arithmetic and geometry are being taken to be a cut
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 127
above such sciences as medicine, navigation, cobblery, and boxing. It
is also clear that the objects of the mathematical sciences are being
singled out as akin to such objects as the good, justice itself, the
beautiful itself, and so on, which emerge, in the Platonic dialogues,
as the objects of the inquiries into the "What is it?" of things that
first Socrates and then Plato undertook. ̂
The remaining differences between Socratic and other dialogues
all turn on the difference already noted in (6) and (7) above, between
Socrates' intellectual attitude to virtue, where virtuous activity will
result when and only when the agent is possessed of intellectual
understanding of what is good and bad for us as humans; and Plato's
attitude to virtue, where the training of the emotional and irrational
more or less independently of the intellectual is a necessary precon-
dition to virtue.20
9. In the Socratic dialogues, we have a doctrine that, depending
upon how one interprets it, may be called either the Unity of Virtue
(UV) or the Unity of the Virtues (UVV). On either interpretation, the
doctrine requires at least this: that a person will be brave if and only
if temperate, wise, just, and pious; temperate if and only if brave,
wise, just, and pious,- and so forth. By contrast, in the Republic
(especially Book IV) it is possible for the military class to have cour-
age and the lower class to have temperance, without either of them
having wisdom - that being a virtue confined to the intellectual
class. What the military class has instead of wisdom is true belief.
Furthermore, in the Statesman (3o6a-3iic), temperance and cour-
age are actually treated as opposite virtues: if one has the one, one is
unlikely to have the other. (Once one gives up the idea that virtue is
an intellectual expertise, it becomes a lot easier for the emotional
sides of different virtues to assert themselves in opposite ways.)21
There may be some temptation to soften the contrast between the
Socratic dialogues and the Republic by invoking the (strongly puri-
tanical) distinction between "philosophic" and "demotic" virtues at
Phaedo 68c-6gd, 82a-84b, and suggesting that the virtues as we see
them in the Republic Book IV are merely demotic virtues. It could
then be argued (as I did in my "The Unity of Virtue") that Plato
might still hold to something like a unity of the philosophic virtues.
This temptation may be reinforced by Republic 435c-d and 5O4a-b,
which suggest that the account of the virtues in the Republic Book
IV is lacking because it fails to show how the Form of the Good
Cambridge Companions Online © Cambridge University Press, 2006
128 THE CAMBRIDGE COMPANION TO PLATO
enters into the account of the virtues. But even if we succumb to
this temptation - and I am not sure we should - it cannot be denied
that any interest at all in demotic virtue represents a slide away
from the Socratic position. And in the Statesman, the abandoning of
any unity between temperance and courage represents a complete
repudiation of the Socratic position.
10. In the Socratic dialogues we find an intellectualist theory of
desire, according to which all desires, and not only the desires of the
virtuous, are desires for the good22 - that is, for whatever is best for
me2* in the circumstances I am in. The result is that the virtuous
differ from the vicious not in their motivations but in their intel-
lects. (All persons desire the same end - whatever is really best for
them in the circumstances they are in. Where they differ is in their
beliefs about what courses of action and styles of life will be the best
means to that end.)2*
The point that everyone desires the same end is a little tricky. If a
good person equates his or her good with virtue, and a bad person
equates his or her good with pleasure, it might seem Socrates must
think that only the good person desires his or her real good. But I am
denying this. What Socrates is saying, I claim, is that in these circum-
stances, the good person correctly chooses virtuous activity as a
means to his or her real good, and the bad person mistakenly chooses
pleasurable activity as a means to his or her real good. Each person
chooses his or her real good as an end. (More on this point in note 42.)
I return now to the intellectualist theory of desire. According to
this theory, all desires to do something are rational desires, in that
they always automatically adjust to the agent's beliefs about what is
the best means to their ultimate end. If in particular circumstances I
come to believe that eating this pastry is the best means to my
happiness in the circumstances, then in plugging this belief into the
desire for whatever is best in these circumstances, my (rational)
desire for whatever is best becomes the desire to eat this pastry.2* On
the other hand, if I come to believe that it would be better to abstain,
then once again my desire for whatever is best will become the
desire to abstain.26 Rational desires adjust to the agent's beliefs. In
fact, on this view the only way to influence my conduct is to change
my opinion as to what is best. Hence the intellectualism of Socrates'
theories of education and persuasion noted under (6) and (7), and the
effective identifying of any virtue whatever with wisdom under (9).
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 129
The basis of this adjustability of desire to belief is desire's being
for the good - for whatever is best. The "what is best" provides a
kind of substitution device for plugging in one's beliefs about particu-
lar actions that are in fact best in the circumstances. Different be-
liefs, different actions.
Plato, on the other hand, in the parts of the soul doctrine of Repub-
lic IV 436-440, explicitly attacks the doctrine that all desire is for
the good.2? In Plato we find desire for good only in the rational part of
the soul. (What Plato speaks of as the rational part of the soul is the
entire soul by Socrates' lights.) Desires of the appetitive part of the
soul, on the other hand, are brutely irrational, or, as we might say,
blind. That is, they are blind to such changes of belief as the example
above envisages. If my desire to eat the pastry is an irrational desire,
then even if you convince me that it would be better for me not to
eat it, the desire does not go away and I may in fact act on that
desire. Furthermore, my behavior can be changed without any
change in my beliefs - simply by acting on or awakening my irratio-
nal desires.
The (alleged) phenomenon of acting contrary to what I think is
best - called, at least since Aristotle, akrasia (weakness of will) -
gives us a way of restating the difference between Socrates and Plato.
Socrates denies there is such a phenomenon, while Plato affirms
there is. Indeed, the pessimism of the Platonic dialogues about hu-
man nature in politics28 is largely the product of Plato's coming to
believe in brutely irrational desires.2*
11. Corresponding to the preceding distinction is a distinction
between two different ways of interpreting the famous dictum "No
one errs willingly." In the Socratic dialogues, the point is precisely
that if anyone errs, it is due to ignorance. By contrast, when this
dictum is echoed in later dialogues, error is not restricted to igno-
rance (which in Plato would be a defect in the operations of the
rational part of the soul), but can be due also to the action of the two
lower parts of the soul, or to a madness that is due either to a bad
state of the body or to a bad upbringing.*0
12. Finally, though we find both in Socratic and more properly
Platonic dialogues the view that knowledge is something strong that
cannot be overcome by pleasure or passion (Prt. 352b-d, 357c), what
lies behind the doctrine is quite different in the two cases. In Socra-
tes, knowledge alone - sheer understanding - enables one to avoid
Cambridge Companions Online © Cambridge University Press, 2006
I3O THE CAMBRIDGE COMPANION TO PLATO
the kind of wavering (shifting up and down) that a person without
knowledge*1 is subject to (Prt. 3656.5—7, Euph. nb-e, Meno 976-
98a).32 It is true that in later dialogues, Plato also implies that those
who know won't go wrong, and won't be overcome by pleasure or
passion. But a great deal more is involved in the explanation of the
Platonic claim than just understanding. In the Republic, the reason
the guardians won't be overcome once they have knowledge is that
they can only get to acquire the knowledge of a guardian if they have
been so trained and conditioned in their early life by the songs they
have had sung to them, the stories they have been told, the gymastic
and military exercises they have gone through, and fifty years of
education, that one might almost be tempted to say understanding is
hardly necessary any longer to their avoidance of dangerous plea-
sures or passions.33 It is an accidental fact, I suggest, that in Plato
knowledge is strong. One who knows is strong against temptation
because one cannot get around to acquiring knowledge till true be-
lief has been so pounded into one, and emotions and desires so
checked and trained out of one, that one would have been strong
against temptation even if one were failing solid geometry. (And it is
arguable that Aristotle's account of the relation between practical
wisdom and moral virtue is in this respect closer to Plato's than
Socrates' view.)
* * * * *
Here then are some of the contrasts that can be made between the
Socratic and the Platonic dialogues. Some of these contrasts are
quite extraordinary, and their very extraordinariness helps make
them more believable. Why? Consider Plato's Seventh Letter, dat-
ing from late in his life, the sheer power and brooding, pessimistic
tone of which** almost by itself recommends the letter as genuine,
as written by the same Plato we have got to know in the Republic
and the Laws. We ask: How could a man like this have produced
the sunny, mischievous intellectual adventures in the early, So-
cratic dialogues? Only one answer readily suggests itself: that there
lies behind the character Socrates in those early dialogues an extra-
ordinary personality, whose sheer intellect and character virtually
swamped the personality of the young Plato, literary and philo-
sophical genius though he was. Not till he was around forty was
Plato's own almost entirely opposite personality, with some intel-
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 131
lectual help from the Pythagorean mathematician-philosophers,
able to begin asserting itself in his writings.
III. A PRIOR QUESTION, A TYPICAL SCENARIO, AN
ILLUSTRATION, AND A USEFUL GENERAL STRATEGY
So far I have been saying how some of Socrates' views differ from
those of Plato. It may have occurred to some to wonder whether any
views can be attributed to Socrates if he is described as someone
who always questions and never asserts anything because he con-
fesses he knows nothing, and if the dialogues from which these
views are to be drawn are aporetic and without positive results.
(Those who are, in addition, appalled by what they see as distress-
ingly many silly views and fallacious arguments in the Socratic dia-
logues may even suggest that Socrates is not trying to convey any
views to us. He is simply educating us by the deliberate production
of bad arguments and fallacies; and that is all he is doing.)
My reply is that the claim to know nothing in no way implies that
Socrates does not hold strong convictions about a great many ethical
matters. And the fact that Socrates' ethical method is to refute peo-
ple rather than tell them what he thinks is no argument against
what I now wish to suggest. This is that the whole point of the
Socratic dialectic is to get people to see things for themselves, as a
result of his refutations - and without their understanding being
short-circuited by the device of giving them a formula.
Now there are quite a few passages where Socrates lets us know
fairly directly what he thinks, though it is true that none of them
are at the conclusions of dialogues.** What I want to maintain here,
however, is that there are also many negative passages in which it
is nevertheless clear what Socrates is trying to get us to see. One
typical scenario is (a) to present us with a view we have reason to
think Socrates holds - perhaps even having his interlocutor anx-
ious to defend this Socratic view; then (b) to put it together with
another view that Socrates and his interlocutor will both have
some reason to want to defend, but which, when put together with
yet other obvious considerations, conflicts with the original view;
then, (c) when the interlocutor cannot see how to resolve the con-
flict, leave the interlocutor to stew over the difficulty; but (d) leave
enough hints in the course of the dialectic that, even though the
Cambridge Companions Online © Cambridge University Press, 2006
132 THE CAMBRIDGE COMPANION TO PLATO
interlocutor can't figure it out, a perceptive and persistent reader or
hearer can.
Let me give just one, rather powerful, example of this scenario. It
is in the Hippias Minor, where the claim that justice is a science or
power (cf.: virtue is knowledge) is subjected to the difficulty of the
"Ambivalence of the Arts (or Sciences)" - the claim that
given two people who err at (t>-ing, one who errs willingly and one who errs
unwillingly, it is the one who errs willingly who is the better <fr-er.*6
These two claims together get us the outrageous conclusion that
the more just person (the better person^ is the one who does
unjust deeds willingly rather than the one who does unjust deeds
unwillingly - contrary to what the law and our general ethical con-
sciousness seem to suggest.
Now Socrates really sticks it to Hippias over this difficulty, and
suggests that he himself is in difficulty over this conclusion (372c-
373a). But, as most interpreters agree, even though Hippias cannot
see his way out of it, Socrates' final setting up of the difficulty
dangles the solution before Hippias's eyes (376b4-6):
Therefore the person who errs willingly in doing disgraceful and unjust
things, Hippias, if there is such a person, will be no other than the good
person.
That is, if there were anyone who did err willingly at doing just (or
virtuous)*8 things, they would be more virtuous than those who err
unwillingly. But - the reader or hearer is invited to conclude - there
isn't any such person, since "no one errs willingly" at being just (or
virtuous) - that is, at getting what is best for him- or herself. (For
"no one errs willingly," see [11] in section II; and for the connection
of being just or virtuous with being good at getting what is best for
one, see the discussion of Socrates' ethical egoism in section IV.) So
the teeth of the outrageous conclusion are drawn,- and though
Hippias does not see this, the perceptive and persistent reader or
hearer can.
Socrates drops hints of this solution along the way. He goes out of
his way to establish that with other sciences, such as arithmetic,
there are such people as err willingly (367a8-b3). That is, with these
other sciences there can always be a motive for the relevant expert
to err willingly at that science, namely, whenever erring willingly
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 133
will enable one to achieve whatever is best for one overall. (Theo-
dorus the arithmetician has a motive for giving a wrong answer
because doing so will help encourage his child, and Theodorus
thinks encouraging his child best for himself overall. Or perhaps
Theodorus has a motive for giving a wrong answer because giving
the wrong answer will give Theodorus a way of testing how well his
student has understood what he was saying earlier.) But with the one
science of achieving what is best for one overall, one can't have that
motive: to err willingly at that science in order to help one achieve
what is best over all(!).39
We see then what Socrates is doing in the Hippias Minor: He is
forcing those who are both willing and able to work things out for
themselves to see how we can perfectly easily reconcile justice's
being a science (or virtue's being knowledge) with the Ambivalence
Principle for the sciences - by pointing out that with this science
alone one cannot have a motive for erring in order to achieve what is
best for one overall. Socrates is showing us how, in spite of ambiva-
lence, it can still be the case that virtue is knowledge.*0 At least one
difficulty in the way of a full answer to the question, What is virtue?
has been gotten rid of.
It is true that in the Hippias Minor we have a particularly clear
case of this scenario. There is hardly anywhere else where this sort
of argument can be made with such confidence as it can in this
humanly delicious, highly philosophical, and really superbly con-
structed little masterpiece. All the same, if we work with due cau-
tion, we can, I think, find the same or similar scenarios elsewhere.*1
The Hippias Minor argument also illustrates what I think is a
useful general strategy for responding to charges that particular So-
cratic views are silly and indefensible - that they are material for the
diagnosis of philosophical error more than anything else. The strat-
egy is this: Refuse to consider suggested expressions of Socrates'
views in isolation from other expressions of Socratic views. Yes, it
sounds strange to say "Virtue is a science," since it seems to ignore
the ethical neutrality of the sciences. It seems to make it possible for
the virtuous person to show his or her virtue by doing unjust deeds
more successfully than anyone else. It seems to show Socrates ignor-
ing the Kantian point that science is compatible with a bad or evil
will. Reply: What for Kantians and many moderns would be an
unacceptable ethical slack in the idea of virtue being a science is
Cambridge Companions Online © Cambridge University Press, 2006
134 T H E CAMBRIDGE COMPANION TO PLATO
taken care of by something Socrates says in another expression of his
views - about motivation: that we all seek what is best for us. So
while for Kant and most moderns, good people differ from bad in
their wills, for Socrates we are all the same with respect to our wills
(cf. Meno 78b4-8). We all seek what is best for us. Where good
people differ from bad is in their knowledge - their knowledge of
what is best for themselves.*2
It is true that this last defense prompts a further objection: Why
wouldn't people who know what is best for themselves use that
knowledge to seek their own benefit at the price of harming, or
doing injustice to, others? But that difficulty in turn can be met by
looking to yet other expressions of Socrates' views. For, as we shall
see in the second half of the next section, Socrates also holds that it
is never in one's interest to harm, or do injustice, others.
IV. CARE FOR THE SOUL AND ETHICAL EGOISM
So far, I have said next to nothing about what was certainly the
central concern of Socrates' life - as it was the central theme of the
Apology - namely, "care for the soul." About this care for the soul,
Socrates thought it was a science in just the way horse training is a
science.^ For Socrates this concern was both for his own soul and for
the souls of others - especially in the education of the young.** In-
deed in expressing the centrality of Socrates' concern for the educa-
tion of the young in virtue, Plato the author even receives some
unconscious help from Socrates' accusers in the Apology. For their
charge of corrupting the young brings up for Socrates the very point
on which he would most like to be examined: Who cares for the
young, and who makes them better human beings? (If the charge had
not existed, Plato would have had to invent it.)
But before looking at the question why Socrates was especially
concerned with the education of the young, let us ask first a more
general question: Why should we care for our souls at all? Socrates'
answer is this: because the soul is the human being's instrument,**
that by which the person lives - and, if it be so, lives well. Like the
horse, the pruning knife, the eye, the bow, the archer, and the doctor,
the soul has an ergon (function); and the virtue (or goodness) of each
of these is the fulfilling of that function. The goodness of a knife is
cutting; of an eye, seeing; and of a doctor, healing. What then of the
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 135
soul? The soul's function is to care for, rule, and deliberate, and so
lead to the person's doing well and being happy (Rep. I 353d-354a).
We must care for the soul, then, because what will enable us to be
happiest is the best possible soul. And we all wish to be happy
[Euthd. 278e; cf. Meno 77b-78b).*6
But what does Socrates suppose doing well, or being happy, is? Is it
to have good things, such as health, wealth, power, honor, or even
temperance, justice, and bravery? Only if we use them, and use them
well.47 To have these so-called good things and use them badly
would actually be worse for us than if we didn't have them. But that
means that the only thing good in itself is wisdom, that is, the
knowledge of how to use, for one's own happiness, the so-called
good things (Euthd. 278d-282e). This wisdom is the knowledge or
science that Socrates elsewhere (La. i95c-d, I96a2~3; cf. 199C6-7;
also Chrm. 172c-i74d, esp. i74b-d, e3) calls the science of goods
and bads.*8 This is the science of caring for the soul that, we have
seen, Socrates holds to be analogous to horse training.
Now it is amazing for a modern to be told that there is one thing
that is unconditionally good, namely wisdom, when Kant told us so
clearly - in unwitting parody of the Euthydemus - that there is only
one thing in the world, or out of the world, that is unconditionally
good, namely, a good will. It is also amazing for a modern to be told
that the goodness of a good human being is goodness at something,
namely, getting happiness. A philosopher like Kant might agree that
some good is goodness at performing a function - the goodness of
knives to cut well, the goodness of architects to build well, and even
the goodness of parents to nurture well and the goodness of friends
to care for their friend and look out for their friend's interest. But
Kant and others are likely to draw the line with good person. Here,
they will say, we are speaking of moral good, not functional good.49
How can you call people good, or just, they will ask, merely on the
grounds that they are good at doing what they need to do in order to
be happy? Surely that would make it possible to call people good or
just who get their happiness by harming others! And that surely
cannot be deemed the part of an ethical person.
This damaging implication of virtue as knowledge, of course, bids
fair to render Socrates' ethical egoism a nonstarter as an ethical
theory. And it certainly does make some versions of ethical egoism
nonstarters. But not Socrates' ethical egoism. To see why, we must
Cambridge Companions Online © Cambridge University Press, 2006
136 THE CAMBRIDGE COMPANION TO PLATO
look at a doctrine of Socrates' that I have only just mentioned so far:
the doctrine that it is never in one's interest to harm, or do injustice
to, others.
Socrates' position here is not, I maintain, that it is immoral ever
to harm others. Rather, like Gandhi and at least some other propo-
nents of nonviolence - who argue that violence is, as a sheer matter
of fact, not in your political interest - Socrates grabs the bull by the
horns and argues that harming others is, as a sheer matter of fact, not
in your interest.*0 The argument in its simplest form shows up in
Socrates' response, in the Apology, to the charge of corrupting the
young. Do I do so willingly or unwillingly? he asks Meletus. If will-
ingly, do I not know that bad people have a bad effect on those
around them, so that by willingly corrupting the young I am doing
myself harm (Ap. 2$c-26a, esp. 25d-e)? But then what I deserve is
not punishment but instruction.51 Elsewhere, at Gorgias 472c-48ib
with 482c//, Socrates argues, at greater length - though unfortu-
nately at a fairly high level of generality, and with some complexity
of structure - the comparable thesis that doing injustice to others is
always worse than having others do injustice to oneself.
Now it must be admitted that many moderns will find these argu-
ments unconvincing.*2 Certainly making out that harming others
can never be in one's interest will involve as risky and chancy an
argument for Socrates in the ethical sphere as the corresponding
argument is for Gandhi and others in the political sphere.
Some modern philosophers will in any case be uncomfortable
with having their ethical theory be a hostage to such a controversial
hypothesis as that it is never in one's interest to harm others. They
will prefer to think, like Kant, that ethical theory needs to be inde-
pendent of the actual details of human psychology.53 ("What if some
Hitler says he is happier harming others?" we will hear. To which
Socrates may reply: Saying it doesn't make it so, and neither does
sincerely believing it. We should distinguish between a bully's hav-
ing rhetorically effective words to hit us with and the bully's saying
something that is true.(54 On the other hand, if ethics is to be for us
humans, then it is arguable we must take the kind of risk involved
in having our ethical theory raise such questions as the bearing on
our happiness of doing harm to others. Few think that in educating
their children they can insulate themselves from such questions.
Why should ethical theory be any better off?55
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 137
At any rate, whatever the risks Socrates incurs by having his eth-
ics depend upon considerations of human psychology, 56 what cannot
be denied is that he does argue that harming others is never in one's
best interest. The risks that philosophers like Kant will point out to
Socrates' notion of care for the soul, and to the ethical egoism that
goes along with it, are to be met by appealing to yet another expres-
sion of Socrates' views: that it is never in one's interest to harm
others.*?
V. SOCRATES' MAIN ENEMY
If virtue is the knowledge of how best to care for one's soul in order
to be happy; and Socrates, the wisest human in Greece, knows only
that he knows nothing; then, of course, no one has the knowledge of
how best to care for one's soul - and indeed no one has virtue.*8
Why, then, is Socrates so passionate about this knowledge, forever
asking what it is (under its various different names: courage, wis-
dom, justice, piety, temperance**); whether it is teachable; whether
it is strong; whether it can be used for bad purposes; whether one is
happier with this knowledge than if one is unjust; whether politi-
cians, poets, artisans, rhetoricians, or sophists have it; and so
forth?60
To answer this question, we must look at the alternatives open to
one who does not have knowledge. They must be: either (a) continu-
ally strive to get closer and closer to this knowledge, even if one is
virtually guaranteed never to gain it totally; or (b) try some other
method for being happy and getting on in life-in particular the
methods of rhetoricians and sophists. The Socratic dialogues are
clear testimony to the fact that Socrates cleaves to (a). To see why,
let us look briefly at Socrates' rejection of (b).
For Socrates, as for Plato, there is no doubt that the main philo-
sophical enemy is the sophists and rhetoricians whose form of educa-
tion is one that offers neutral techniques for "getting on" in private
and political life - neutral in that they are indifferent to any good in
human life other than what the individual chooses to think is good -
the individual's "values" (as we say).61 The sophists and rhetoricians
purport to put persuasive means in their students' hands to achieve
whatever goals "seem best" to them [Gig. 466b//, esp. bn-e2,
467a8-468e5; and cf. 464d-46sc), without raising any questions
Cambridge Companions Online © Cambridge University Press, 2006
I38 THE CAMBRIDGE COMPANION TO PLATO
about what people take to be their ends. For Protagoras in the
Theaetetus, it is all but explicit that there are no further questions
about ends: What the individual thinks good is what is good "for"
that individual.62
When the sophists say they can impart to their students a science
of the best means to whatever the students' ends may be (the end
being whatever the students value most, a la note 61 -whatever
"seems best" to them a la Grg. 466-469), Socrates attacks the sup-
posed science, and contrasts with it sciences like medicine, carpen-
try, cobblery, and navigation. The latter sciences are no more neutral
about the end than they are about the means. As medicine looks to
what is objectively best for health (and not just to what seems best
to the patients for their health), so the science of virtue looks to
what is objectively best for humans (not just to what humans think
best for themselves). Thus Socrates urges against relativism the ob-
jectivity of the sciences, and suggests that the knowledge that is
virtue is just one more objective science.
Another form of Socrates' attack on alternatives to the Socratic
"examined life" is the characterizing of the minds of rhetoricians,
sophists, politicians, and poets as working more by divine inspira-
tion than by any understanding of what they are doing. When poets
say true things, that is purely a product of divine inspiration (phusei
tini kai enthousiazontes), and not at all a product of any kind of
wisdom or understanding of what they have said (Ap. iicff, esp.
22C3). So too, interpreters of Homer like Ion say what they say not
out of knowledge but by "divine dispensation" (see Ion 534c, 535a,
536c-d). For this "divine dispensation" as the source of politicians'
uncomprehending activities, see Meno 99e.6* Similarly, in the Euthy-
demus 2890—2909., the art of speech-making is said to be "lofty" and
"divine" (thespesia) - like the art that charms snakes, tarantulas,
and scorpions! And at Apology 2od-e, Socrates speaks of sophists
like Gorgias, Prodicus, Hippias, and Euenos as having a wisdom
"beyond that of humans." Consider now all of those pursuits of the
good for humans - the pursuits of poets, exegetes, politicians, rheto-
ricians, and sophists - that Socrates sees as rivals to the pursuit he
thinks humans should take up. To each of these, Socrates will say
what he says of rhetoric: that it is not a science (Grg. 46$aff with
454d-457c). What the references to divinity betoken is the sugges-
tion that the activities of poetry, politics (as practiced by Pericles et
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 139
al.), and the interpretation of Homer (as practiced by Ion) are, like
rhetoric, not sciences. Unlike the expertises of artisans (when they
stay within the boundaries of their own expertise), these activities
do not involve knowledge. How do rhetoricians, politicians, poets,
and so forth achieve as much as they evidently do if they have no
science, no expertise? It's by divine intervention - a gosh-darned
miracle!
This, then, is Socrates' response to relativism: to deny that the
states of mind of such alleged experts as rhetoricians, sophists, politi-
cians, poets, and the like are those of legitimate scientists.
Of course I have only hinted here at why Socrates thinks none of
these disciplines is a science. The answer has to do with the lack of
coherence involved in the ends of these would-be sciences. Since
rhetoric promises to get for you whatever seems best to you whether
or not it is in fact best, the possibility of a serious incoherence
results. (I want to be happy, and think I'll be happiest if I have a
political rival exiled; so I hire an orator to find and employ the best
persuasive means to my end. In fact I'm wrong that I'll be happiest
that way. The result is that there is an incoherence in my desire to be
happy. What I want is that exile of my political rival which will in
fact make me happiest; but given that any exile available to me in
these circumstances will not make me happy, there can be no sci-
ence of the sort rhetoric claims to be, that can secure me that end.)6*
In the next section, I offer another explanation of why Socrates
will say that poets, rhetoricians, and the like have no knowledge of
"anything they say."
VI. SOCRATIC METHOD AND SOCRATIC IGNORANCE
I conclude with some remarks on the (often doubted) sincerity of
Socrates' claim to know only that he knows nothing. We can defend
this sincerity, I suggest, if we attend to the strategy announced in
Section III - if we notice how, for Socrates, discussing any one ethi-
cal question seems inevitably to involve one in discussing (what one
might at first have thought to be) quite different ethical questions.
Consider first how Socrates argues that he knows nothing [Ap.
i9bff, esp. 20c—d). Those whom one might think wiser than Socra-
tes, because they know something substantive about how a person
should live - poets like Aeschylus, say-turn out upon Socratic ex-
Cambridge Companions Online © Cambridge University Press, 2006
I4O THE CAMBRIDGE COMPANION TO PLATO
amination to "know nothing of what they are saying" (Ap. 22C3).
The difference with Socrates is that Socrates knows that he "knows
nothing of what he is saying," while Aeschylus and others do not
realize this fact. (In spite of his concern for care of souls, Socrates
actually knows nothing about improving humans in the way horse
trainers know about improving horses; Ap. i9d-2oc.)
Now there are various problems with the claim to know only that
one knows nothing which I haven't space to deal with here.6* But the
crucial question here is: Why does Socrates maintain that no one at
all, himself included, knows anything substantive about the matters
he is inquiring into? Let us engage in a thought experiment here.
Imagine the poet Aeschylus believing that it is good to know oneself
(cf. Prometheus Bound, 309) and Socrates affirming in his own per-
son that it is good to know oneself (cf. Ale. I I27e-i35e), thus appar-
ently granting that Aeschylus's belief is true. Finally, imagine that
Aeschylus believes this because he has been told so by the Oracle;
that the Oracle is reliable in never saying anything untrue,- and that
neither Socrates nor Aeschylus has any reason to doubt what the
Oracle says on this occasion. On such grounds, should we not infer
that Aeschylus knows that it is good to know oneself? Certainly we
should by the rather nonintellectual standards that modern philoso-
phers tend to be contented with for knowledge.
Socrates, we know, will not allow that Aeschylus knows any such
thing. Faced with such a claim, Socrates will ask Aeschylus a whole
bunch of questions - not just narrowly homing in on the truth of the
sentence "It is good to know oneself," but questioning the reasons
why Aeschylus thinks this true, and indeed all sorts of other matters
connected with knowledge generally: whether such knowledge is
the whole of virtue or only a part, whether this knowledge is teach-
able, whether this knowledge by itself is enough to make one happy,
whether passions and pleasures can overcome this knowledge, and
so forth - hardly stopping short of an examination of Aeschylus's
whole life [La. i87e-i88c). If Aeschylus trips up over any such ques-
tions in traversing this Socratic minefield, Socrates will say Aeschy-
lus doesn't know even that it is good to know oneself (Ap. 22C3).
A modern would be likely to defend Aeschylus by saying, "Look,
it's true enough that Aeschylus doesn't know about these other
things,- but he does know that self-knowledge is good." (Just so, a
modern might have defended Oedipus when he said, "I don't know
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 141
what else may be true about Jocasta; but I do know I am married to
her and that she is Laius's widow." Surely Socrates couldn't deny
that?)
But Socrates' view was that unless Aeschylus can sustain a typical
Socratic cross-examination in the vicinity - near or more remote - of
the claim that it is good to know oneself, he doesn't know that it is
good to know oneself.
Now why should this be? Why should knowing that self-knowledge
is good require one also to know that virtue is not teachable? The
hypothesis I wish to suggest here is one about the identity of the thing
known. For Socrates, it is not enough for purposes of knowing that it
is good to know oneself, that one know (as we put it nowadays) the
proposition that it is good to know oneself.66 Or to make a closely
related point,6? it is not enough to know that a sentence with the same
meaning and grammar as "It is good to know oneself" is true. We do
not know that little, I suggest, if, by Socrates' lights, we know that it
is good to know oneself. For Socrates, the thing known when one
knows that it is good to know oneself includes not just what the
words mean, or what words are used for expressing the opinion, but
also what the words refer to. Is the knowledge that one has of oneself
knowledge of what is good for oneself? And is that a knowledge that
cannot be overcome by pleasure? If it is, then Socrates will suppose
that to know that it is good to know oneself must also involve know-
ing that knowledge cannot be overcome by pleasure. Now, that
knowledge cannot be overcome by pleasure is hardly included in the
proposition that it is good to know oneself; and it could hardly be
learned from the dictionary meaning of the words "it is good to know
oneself." So, for Socrates, knowing that it is good to know oneself
involves a good deal more than just knowing the proposition that it is
good to know oneself.
Another example may make this clearer. Take Nicias's offering of
"courage is the knowledge of the fearful and the hopeful" as an
account of what courage is. Nicias claims that this is a Socratic
account (La. 194C8—d2), and Socrates acknowledges that it is [La.
I94d3; cf. Prt. 358d5~7, 36oc6-d5). Yet at I97e-i99e, Socrates re-
duces this account to absurdity as follows:
The fearful = future bads, and the hopeful = future goods;
So, courage = the knowledge of future bads and goods.
Cambridge Companions Online © Cambridge University Press, 2006
142 THE CAMBRIDGE COMPANION TO PLATO
But the science of future K's = the science of all K's, past,
present, and future;
Therefore, courage is the knowledge of all bads and goods,
past, present, and future.
But the knowledge of all goods and bads = virtue;
And, at that rate, courage, which was supposed to be only a
part of virtue, would have to be the whole of virtue;
So, courage can't after all be the knowledge of the fearful and
the hopeful.
What has happened here? Socrates has apparently refuted his own
account of what courage is. Or has he? In my "Unity of Virtue," I
suggest that Nicias gets refuted because he thinks courage is only a
part of virtue. Once we see, what Nicias does not see, that Socrates
is here arguing for (UV), the difficulty disappears.
Now, evidently Socrates is not here in the business of refuting the
proposition that courage is the knowledge of the fearful and the
hopeful. If he were, he would be refuting a proposition he himself
accepts. No, it is not propositions that he is concerned with. It is the
nature of courage that he is concerned with. The courage that Nicias
believes to be the knowledge of the fearful and the hopeful is a
courage that is only a part of virtue. The courage that Socrates be-
lieves to be the knowledge of the fearful and the hopeful is the whole
of virtue - as in the Protagoras (see [e] in note 21). For Socrates, to
know that courage is the knowledge of the fearful and the hopeful is
(or at least requires) the knowledge that courage is not merely a part
of virtue.68 Similar arguments could be developed to show that know-
ing that courage is the knowledge of the fearful and the hopeful also
requires one to know that knowledge is strong, that no one errs
willingly, that virtue is not teachable, and so forth. To know any-
thing at all about human goodness, one will have to know every-
thing about it.6*
It is a mistake, then, to look at the Socratic elenchus as a process
of trying to establish certain propositions on the basis of certain
deductive arguments.?° The issue is one of the identity conditions
for things known. Where many moderns think that its being good to
know oneself is one fact to be known, while knowledge's being
strong is another fact to be known, and virtue's being knowledge
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 143
another, virtue's being unteachable another, Socrates does not see
the truths about good and bad as breaking up into these and so many
other atoms clearly marked off from each other. For Socrates, a lack
of understanding about how it is that virtue can be unteachable and
yet be knowledge will show a defect in one's knowledge that it is
good to know oneself.
Now, it is true that something like the point I am making here can
be translated into a point using propositions. The fact that Socrates
will not allow something to be knowledge that involves affirmation
of one expression of Socratic belief but denial of another can be ac-
counted for by saying what I am saying - that the thing to be known is
broader than a proposition. But it can also be accounted for by saying
that knowledge of one proposition is not possible in isolation from
knowledge of all sorts of other propositions.~?T It is because interpret-
ers often seem implicitly to grasp this interconnection of the proposi-
tions they find in Socrates, that they for the most part stay so well in
touch with Socrates' views - in spite of their being wedded to proposi-
tions.^2 The danger for such interpreters comes when, for example,
they start thinking of Socratic "definitions" as giving necessary and
sufficient conditions for the application of a virtue-word to an action.
For then they are tending to think of such "propositions" as intelligi-
ble in isolation from other propositions. (Cf. note 60.)
So much, then, in explanation of Socrates' claim that poets don't
know anything of what they are saying. Their supposed knowledge
of some one proposition, in isolation from others, won't count as
knowledge of anything. This explanation of why Socrates says that
poets (and indeed politicians, interpreters of Homer, rhetoricians,
and - outside of their specific expertise - artisans) know nothing of
anything they say, thus supplements the explanation sketched
briefly at the end of Section V>3
Besides helping us to understand why Socrates appears to attack
views he himself accepts, these remarks may also help us to under-
stand the role of the so-called Socratic paradoxes. A paradoxical
remark - Socrates sometimes speaks of "riddles" - is just the sort of
thing to force one to see the surprising interweavings of (what mod-
erns tend to think of as) quite different Socratic claims - an inter-
weaving that is central to Socrates' whole approach to ethics. (So-
cratic paradoxes are in this respect just like contradictions or absurd
conclusions reached after some Socratic dialectic.)
Cambridge Companions Online © Cambridge University Press, 2006
144 T H E CAMBRIDGE COMPANION TO PLATO
Again, my explanation helps us to see that when Socrates reduces
his interlocutors to absurdity by his elenchus, he is not turning his
back on them in a failure of love.™ For Socrates does not think giving
someone a formula, such as "Courage is the knowledge of the fearful
and the hopeful," will be any help at all to them if they don't
understand - and understand for themselves - what that broad thing
to be known is that that sentence makes reference to. (Compare "Oe-
dipus is married to Jocasta" not being of any help to the Department of
Public Health in Thebes, when, in tracking down the sources of the
plague, it asks, "Whom is Oedipus married to?" The verbal formula,
without the whole truth about the person asked about, will be no help
at all.) Only someone in possession of the whole truth about courage
and knowledge will for Socrates know that courage is the knowledge
of the fearful and the hopeful. Only such a person will have the knowl-
edge that is virtue. Love for one's interlocutors involves precisely
trying to bring them to such knowledge in probably the only way it
can be done - by Socratic elenchus - and without the corrupting in-
fluence of Socratic (or any other) authority.
My explanation also handles two other explananda. First, why
does Socrates insist on examining only people's actual beliefs?75
Because he is not interested in having them hand him a mere for-
mula for examination - any more than he thinks they will be helped
by being handed a mere formula.?6 Particular expressions of belief on
a person's part are also to cover parts of the interlocutor's belief not
given merely by the words in that expression or by their (dictionary)
meanings. As we have seen, Nicias's belief that courage is the knowl-
edge of the fearful and the hopeful is a belief about a (supposed)
courage which is only part of virtue. That is the belief of Nicias's
which is to be examined - not a belief that the proposition
Courage is the knowledge of the fearful and the hopeful
(whether courage is a part of virtue or the whole of it)
is true.
Second, what is the connection between the Socratic questioning
of beliefs and the "existential dimension" of the elenchus - the ex-
amining of people's whole lives [La. i87e-i88a)? It is not a case of a
"double objective" for the elenchus: examining the truth of a propo-
sition and examining the whole life of a person.?? No, since (a) to
examine a person's belief about courage will be to examine that
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 145
person's total body of belief about virtue, knowledge, and the good
life,- and (b) desires always automatically adjust to one's beliefs (see
[10] in section II); it will follow that to examine a person's belief
about courage will be to examine that person's whole life.
As to Socrates' own knowledge, Socrates is not saying he has no
grip whatever on any part of the whole, or indeed on the whole. He is
just claiming that what is known when one knows any part of the
whole adequately will in fact include knowledge of the whole. It
follows, as a consequence, that the only way to set people on the
path of virtue will be to begin that process of self-examination that
alone will make sure that, little by little, they come to see (all
together) that virtue is knowledge, that virtue is one, that knowl-
edge is strong against pleasure and passion, that virtue is not teach-
able (at least in the ordinary way), and that desire is for good. Or, lest
the last sentence seem to speak of five different propositions/8 that
the virtue which is knowledge, strong against pleasure and passion
(cooperating always with desire which is always for the good), and
unteachable in the ordinary way, is one. We may suspect, though
Socrates never tells us so, that Socrates thinks himself rather farther
along than anyone else in this attempt to grasp the whole. But unless
he thinks there is nothing left for him to figure out and fit together,
he may still fairly claim to know only that he knows nothing.
One final objection. Does Socrates really think there are still
things for him to figure out? Richard Kraut asks, appropriately
enough, what evidence I have, on my interpretation of Socrates, that
Socrates thought there were still problems left for him to solved He
points out that I should feel this difficulty particularly acutely be-
cause I hold (in Section III and note 41) that Socrates often seems to
know just what he is trying to get us to understand, and even dangles
before our noses solutions to puzzles he has propounded. Where
then are there examples of unsolved problems, if the solutions are
always being dangled before us? Why would Plato represent Socrates
as a searcher who lacks answers and not give us any examples of
Socrates lacking answers? (On Kraut's own approach to Socrates, we
are supposed to find examples of unsolved problems in Socrates'
really not knowing what to suggest to us at the end of the H. Mi. or
at the end of the refutation of Nicias's [Socratic] account of courage
as the knowledge of the fearful and the hopeful. See notes 40, 41.)
My response is this: First, wherever a dialogue leaves us in aporia,
Cambridge Companions Online © Cambridge University Press, 2006
I46 THE CAMBRIDGE COMPANION TO PLATO
experience has suggested to me that the strategy of Section III —
assuming that there is something Socrates wants us to see and trying
to track it down - almost always pays dividends. I am willing for
this principle to stand or fall by the accounts it yields of passages
like those just cited from the Hippias Minor and the Laches, as well
as others cited in note 41. Does this suggest that Socrates has all the
answers? I don't see why it should. The whole point of the present
section of the paper - as of Section III and of note 41 - is that to get
anything right about the human good, you have to get everything
right about it. Socrates still had lots of things to say more about, and
to get right.
But what specific things did Socrates still have to figure out? Some
examples that spring to mind right away are (a) the nature of happi-
ness (about which Socrates, unlike Aristotle, says very little that is
useful); (b) the nature of pleasure (again the contrast with Aristotle is
striking); (c) the defects of radical Protagoreanism (here it is the
contrast with Plato's Theaetetus that is striking) - though Socrates
is less likely to have reflected on the possibility of any anti-
Protagorean strategy other than the appeal to the sciences; (d) a more
articulated and more convincing set of arguments that it always
harms you to harm others; (e) more convincing accounts of the func-
tion of humans and of the art of ruling; and so forth. I feel all of these
as unsolved problems in Socrates. It is true that none of them gets
mentioned as such by Socrates. But it is hard to believe he didn't
actually feel at least the problems in (a), (b), (d), and (e). And the fact
that they don't get mentioned as unsolved problems is perhaps well
enough explained merely by the fact that Socrates is busy enough
trying to get his interlocutors to think their way through some of the
problems Socrates has more or less seen his way through.
The crux of the issue is this. Kraut sees knowing what courage is
as knowing the definition of courage; and this he sees as knowing
some one particular proposition about courage (see the references in
note 72). This proposition, Kraut maintains, Socrates doesn't know
to be true. So he can't mean us, in the Laches, to see that courage,
the knowledge of the fearful and the hopeful, is identical with the
knowledge of goods and bads which is virtue.
As against this, I say this propositional view of what courage is is
too dangerously like the view that "What is courage?" asks for the
meaning of "courage," rather than the reference. (Propositions are
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 147
meanings of sentences.) There is no reason why Socrates should not
have wanted us to see - as a partial account of what courage is, and
as providing the solution to a problem Nicias can't see his way
through - that courage is the whole of virtue. But to have such a
partial account is not yet to know what the reference of "courage"
is. To know what courage is, it is not enough to offer some one
proposition (the meaning, or sense, of some sentence) as a definition.
One needs to know what the reference of courage is. Frege says
about reference: "Comprehensive knowledge of the reference would
require us to be able to say immediately whether any given sense
belongs to it. To such knowledge we never attain."80 That is the kind
of knowledge of virtue, knowledge, power, desire, good, and so forth
that I see Socrates striving for.81
NOTES
1 The most considerable alternative interpretations of Socrates overall
that I know of, and the ones I most admire, are those of Terence Irwin,
Plato's Moral Theory (Oxford, 1977); Richard Kraut, Socrates and the
State (Princeton, 1984); G. X. Santas, Socrates: Philosophy in Plato's
Early Dialogues, (London, 1979); and Gregory Vlastos, Platonic Studies,
id ed. (Princeton, 1981), and a plethora of earlier articles by Vlastos.
Though on a number of points Irwin's views are close to mine (in, for
example, my "The Unity of Virtue/' Philosophical Review 82 [1973]:
35-68), he also seems to me rather to straitjacket Socrates for purposes
of getting a neat exegetical opposition with Plato. (See, as an indication,
note 14 below.) In Kraut I find both more to agree with (notes 72, 14, 16
below) and more to disagree with (notes 12, 16, 40, 51, 54, 60, 63) than in
any of the others. Vlastos was of course the greatest Socrates scholar of
the century, and his Socrates: Ironist and Moral Philosopher (Cam-
bridge, 1991) adds to the many important Socratic studies he had already
published. Like most people's in the field, my views about Socrates have
been formed in great part by thinking about what Vlastos says about
Socrates. My own differences with Vlastos, which are many, flow
mainly, I think, from philosophical differences between us.
2 For (i), see Soph. El. i83b6-8; for (ii)-(v), see esp. Met. I.6.987a29-bi4
(with 6.987022-24, 27-33, and 5.987a2O-25) and XIII.4.io78bi2-34, as
well as the discussion in W. D. Ross, Aristotle's Metaphysics (Oxford
1924), 1 :xxxiii-xlv, xlvii, 158-161; 2: 420-3. It has been plausibly sug-
gested that the term "induction" [epagoge) refers to the wide use of
analogy in Plato's early dialogues, especially the Analogy of the Arts: If,
Cambridge Companions Online © Cambridge University Press, 2006
I48 THE CAMBRIDGE COMPANION TO PLATO
in medicine, the doctor looks to the health of his or her patients,- and, in
navigation, the pilot looks to the safe passage of his or her passengers;
then surely in ruling, the ruler will look to the interests of those over
whom he or she rules. As the horse trainer stands to horses, so those
who would teach virtue stand to humans. And so forth. See Vlastos, ed.,
Plato: Protagoras (Indianapolis, 1956), xxix n. 18, with n. 49; Richard
Robinson, Plato's Earlier Dialectic (Oxford, 1953), 41//.
3 The question has become controversial in recent years. Though I dis-
agree strongly with recent accounts, adequate treatment of the notion is
beyond the scope of the present paper.
4 There is reason to believe Aristotle's testimony is based on more than
just Plato's dialogues. For example, Aristotle could not have learned
from the dialogues that Cratylus was an early teacher of Plato. (Vlastos,
"Socrates," Proceedings of the British Academy 74 [1988]: 104-5, citing
a 1934 remark of Ross's quoted in Andreas Patzer, ed., Der Historische
Sokrates [Darmstadt, 1987].)
5 For example, the what-is-it of justice is the square, or the number 4: As
the factors of 4 (2 x 2) behave to each other in exactly the same way, so it
is with the citizens of a just state. See Ross's commentary in Aristotle's
Metaphysics on 985029, 987a22, 30, 1078023; see also Grg. soye6ff, esp.
5o8a6.
6 The point of Aristotle's remark is presumably that both notions are prob-
lematic in the same way. (Aristotle seems content to leave the relation
between particulars and his own universals unexplained and primitive.)
7 For an explanation of stylometry and its coordination with our meager
knowledge of the chronology of Plato's dialogues, along with a sketch of
the history of stylometric scholarship, see Leonard Brandwood's authori-
tative essay that is Chapter 3 of this volume and his The Chronology of
Plato's Dialogues (Cambridge, 1990). The latter volume describes in
some detail, and for the first time in print, the results of Brandwood's
important doctoral dissertation (1958). All Plato scholars will applaud
its appearance at long last. Unfortunately, this volume appeared too late
to be considered in my article. I was able, however, thanks to Brand-
wood's kindness, to see a draft of his contribution to this volume while
preparing my own.
8 See Brandwood's guarded remarks in Chapter 3 ("no more than a proba-
bility") on Ritter's 1935 examination of the early dialogues, as well as
his remarks at the end of the essay. In Brandwood's A Word Index to
Plato (Leeds, 1976), xvii, the Grg. and Meno are not distinguished from
the dialogues of the third subgroup. I see no difficulty in the possibility
that the Lys., Menex., and Euthd. may be later than the Grg. and Meno.
Rep. I may be a second edition, stylistically revised for inclusion in the
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 149
Rep. as a whole. As Brandwood stresses, the great triumphs of stylom-
etry reside in the treatment of later dialogues. A general grouping of
early dialogues separate from middle and late is assured, but not much is
certain about the ordering within the early dialogues.
9 This is true of the Phd. more or less throughout, of the upward path of
the Smp. (210-212), and of the discussion of flux and the Forms in the
Cra., especially at the end of the dialogue. The placing of these three
dialogues as closer to middle dialogues like the Rep. (Books II-X), Prm.,
Tht, and Phdr. coheres with Brandwood's cautious stylometric judg-
ment. Cf. also his Word Index, xviii.
10 See E. R. Dodds, Plato: Gorgias (Oxford, 1959), 18-30; also W. K. C.
Guthrie, A History of Greek Philosophy (Cambridge, 1969, 1975),
3:29-39; 4: 39-56. Very little can be said about exact dates. See, for
example, K. J. Dover, "The Date of Plato's Symposium/' Phronesis 10
(1965): 2-20.
11 On optimism vs. pessimism, I have this sort of thing in mind: Socrates
thought he could make a difference (at any rate some small difference) in
improving himself and those around him - especially the young-by
engaging in what he enjoyed most: rational discussion of questions as to
how to live. The attitudes to the masses and to the military in Plato's
totalitarian political theory reflects Plato's despair about reaching any
but a few via rational discussion - and then only after long training.
12 For the small interest even in political theory in the Socratic dialogues -
in sharp contrast to the later dialogues - see Vlastos, "Socrates," 97-8.
For a remarkable and instructive discussion of the Crito, which some-
times (esp. in chap. 8) differs sharply from my own views, see Kraut,
Socrates.
13 See Vlastos, "Socrates," 94-5.
14 Socrates uses techne ("art" or "craft") and episteme ("knowledge" or
"science") interchangeably. "Expertise" seems to me the best word for
the single conception Socrates has in mind here, and "science" next best
(for the continuity it keeps with Plato's episteme).
One needs to be careful with "craft" if that word is taken as suggesting
a discontinuity with "science" - and even more if it is taken as suggest-
ing that Socrates has a "craft-knowledge" conception of ethics (Irwin,
Plato's Moral Theory, 71-101, thence structuring the discussions in
chaps. 4, 5). On this supposed "craft-knowledge" conception, ethics is a
science studying only objectionably narrow instrumental means to
some further (independently identifiable) end. Here Irwin contrasts Soc-
rates unfavorably with Aristotle and Plato, in whom the virtuous activ-
ity that ethics studies is itself an ingredient means to happiness - not a
mere instrument for gaining some further goal of happiness, but itself
Cambridge Companions Online © Cambridge University Press, 2006
I5O THE CAMBRIDGE COMPANION TO PLATO
part of the activity in which happiness consists. (To use f. L. Ackrill's
examples from lectures in the 1960s, a bathing suit, or taking the train
to Eastbourne, are instrumental means to having a good holiday, while
going swimming is an ingredient of having a good holiday: cf. Ackrill,
"Aristotle on Eudaimonia," Proceedings of the British Academy [1974]:
339-59-)
Now I am not at all convinced that Socrates thinks virtue a mere
instrumental means to happiness. For Socrates surely thinks that, other
circumstances being favorable, the life of wisdom is the happy life
(Euthd. 28id-e: cf. Kraut, Socrates, 211 n. 41); and he surely thinks that
care for the soul and care for the souls of the young - the unremitting
and lifelong inquiry into how to live (Rep. I 352d and Grg. 500c, 487c-
488a, 472c-d), the constant and thorough testing and examining of oth-
ers and oneself - is a principal ingredient of happiness. Cf. also Ap. 41c,
where Socrates tells us that if, after his execution, there happens to be an
afterlife, what an incredible happiness it will be to be with, discuss with,
and examine the heroes of the Trojan War. Why need Socrates have
thought happiness to be recognizable independently of the sort of hard
ethical thinking and discrimination that he so evidently spent his whole
life engaging in and getting others to engage in?
15 See Vlastos, "Socratic Knowledge and Platonic 'Pessimism/ " Philo-
sophical Review 66 (1957): 226-38.
16 Cf. the modern either-or question, Do we let our younger children watch
such and such programs on TV or not? For Plato, as for Aristotle, that
would be exactly the right question. For Socrates, this would be sidestep-
ping the central issue, which is: Do we, or do we not, discuss with our
children the content of what they are watching?
If I am right, Socrates would have thought of Platonic or Freudian atti-
tudes to childhood training, or Aristotelian or Skinnerian conditioning -
all of them emphasizing an early more or less reason-independent
training - as exceedingly blunt instruments for childhood nurture. He
would have resisted the (Protagorean!) suggestion in Kraut, Socrates,
219-25, 296, that the laws of Athens by themselves provide (something of
a) training in virtue - leading to the additions of some true propositions to
one's beliefs. (On propositions, cf. note 72 below, with notes 68, 78.) On
the other hand, I have no problem with the quite different suggestion in
Socrates, 226-8, that Socrates would have approved of the laws of Athens
for their facilitating of free inquiry.
17 Grg. 463a (cf. 454C-456C); Rep. VI 493a-C; Phdr. 259c//, esp. 26ie-262c,
272d-274a. For parallel attitudes to poetry, see Ion 536c, 54ie-542b;
Rep. X 598b-d. Cf. also my "Socrates on the Impossibility of Belief-
Relative Sciences/' in Proceedings of the Boston Area Colloquium in
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 151
Ancient Philosophy, ed. John J. Cleary, 3:263-325, as well as section V of
this chapter.
18 Some may wonder, How exactly are Socrates and Plato supposed to
differ here? (a) It is true that Socrates will say that when we persuade via
the emotions we are not teaching. But couldn't Plato grant that point?
On the other hand, (b) surely Socrates will have to grant that when we
persuade via the emotions - and indeed even when Gorgias persuades -
we are just as much getting people to hold the beliefs we want them to
hold as we do when we teach them? (I am grateful to Richard Kraut for
pointing out that what I have said so far leaves me open to objections of
this general sort.)
As to (a), Plato will still need a distinction between teaching and
correct indoctrination (knowledge and correct, or true, belief). But the
distinction between knowledge and true belief is virtually absent from
the Socratic dialogues. It is true that to say this is not to say much, since
this absence is itself an urgent explanandum (see further note 31). But
the difference between Socrates and Plato remains.
On (b), see my discussion of why rhetoric is not a science in my 'Tower
and Desire in Socrates: The Argument of Gorgias 466a-468e that Orators
and Tyrants Have no Power in the City/' Apeiron 24 (1991): 147-202; also
my "Belief-Relative Sciences." I argue in the former paper that Socrates
held that without the science of goods and bads, one can never do the very
thing one wants to do. (For a little on his reasoning here, see note 25
below, as well as section V, next-to-last paragraph.) It will be a corollary of
this claim that without getting the persons to be persuaded to understand
for themselves just what is good or bad about what is to be done, one will
not be able to persuade them of the very thing one wants to persuade them
of. This has Socrates placing limitations on persuasion via appeals to the
emotions that Plato simply ignores.
19 I have suggested in my The Ascent from Nominalism: Some Existence
Arguments in Plato's Middle Dialogues (Dordrecht, 1987), xi-xii, 12-
16, 26-33, 40—3/ t nat: Plato, for as long as he is obsessed with the theory
of recollection, tends to suppose that our knowledge of the Forms, once
attained, can have the kind of self-evidence that knowledge of the axi-
oms of geometry seemed to many Greeks to have. This tempts Plato to
give mathematical sciences a special status, higher than that belonging
to, say, carpentry; and to think of the square, the odd, equality, and the
number one (one itself) as special objects to which he may liken the just
itself, the good itself, and so forth.
Against this more Pythagorean attitude to mathematics, Plato is else-
where more evenhanded about the sciences, showing himself willing to
grant to the shuttle itself and the bed itself, along with Forms of the
Cambridge Companions Online © Cambridge University Press, 2006
152 THE CAMBRIDGE COMPANION TO PLATO
elements Earth, Air, Fire, and Water, quite as much right to exist as
Forms as do the Good Itself or the Square Itself. If the tendency to give
greater weight to mathematics is more Pythagorean, the evenhanded-
ness about Forms for the shuttle, the bed, and so forth, is more heir to a
Socratic strain about the sciences.
20 Why might not persons act virtuously by acting as virtuous persons
would - though by good fortune, since they do not understand why they
should act that way? I have tried to respond to this question in a number
of places elsewhere. See "Belief-Relative Sciences/' Appendix II, 316-
20, where I ask what the chances are that someone entirely ignorant of
the science of engineering should by good fortune hit on just the right
specifications for building the Golden Gate Bridge. (Living a good life, I
suggest, is not fundamentally easier than building complicated bridges.)
21 A word on the differences between (UV) and (UW). According to (UV) -
for which, see my 1973 paper "The Unity of Virtue" - there is really just
one virtue, with five different names. (Cf. Prt. 329c6-di, 349b2-3, and
compare "The Morning Star" and "The Evening Star": though these two
expressions have different meanings, they refer to the same object, the
planet Venus. They have the same reference.) According to (UW) - for
which see Vlastos, "The Unity of the Virtues in the Protagoras," in
Platonic Studies, 221//, with 410//, 418//-Socrates believes there are
five different virtues, and whoever has any one of them has all the
others.
(UV) is naturally put together with the well-known Socratic doctrine
that virtue is knowledge, and the specification of the knowledge in
question as the knowledge of goods and bads, to yield
that one and the same thing that is named by these various
virtue-names is the knowledge of goods and bads.
Thus (UV) + "Virtue is knowledge" says that the one thing whose pres-
ence explains not only courageous deeds (in situations of danger), but
also temperate deeds (in situations of temptations of desire and plea-
sure), and wise, just, and pious deeds in other sorts of situations, is - and
is no more than — the knowledge of goods and bads.
Those interpreters who claim that Socrates holds only (UW) must
obviously give a somewhat different account of the relation the knowl-
edge of goods and bads has to the five supposedly different virtues. They
must also offer a somewhat more complex explanation of Socrates' sug-
gesting in some places that (a) wisdom is identical with virtue (Euthd.
28 ie, Meno 88b-89a), and in others that (b) justice is the whole of virtue
(asserted in Rep. I 335c, cf. 35ob-d, 35ia-c, 353e-354b; implied by H.
Mi. 375d-376b); perhaps also implying in others that (c) courage is the
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 153
whole of virtue [La. 197c—199c) and in others that (d) temperance, as the
knowledge of good and bad (Chrm. i74b-d), is the whole of virtue [La.
i99d-e). Cf. also the suggestion (implication?) that (e) the temperance at
Pit. 356c8-e4, 357a5-b6, is identical with the courage at 36oc6-d5,
both being sciences for measuring goods and bads, the pleasant and the
painful (36oa8; so also C. C. W. Taylor, Plato's Protagoras [Oxford, 1976],
162-3, 209, 213-14, and J. C. B. Gosling and C. C. W. Taylor, The Greeks
on Pleasure [Oxford, 1982], 55).
One other important difference between (UV) and (UVV) is in their
reading of answers to the "What is it?" question. For (UW), "Courage is
the knowledge of the fearful and the hopeful" gives the meaning
(Fregean sense) of "courage," while for (UV) it gives the reference - i n
exact parallel with "The knowledge of future K's is the knowledge of all
K's past, present, and future" (an obvious identity of reference rather
than meaning; cf. La. I94e-i95ai with I98di-i99a4, as well as the brief
discussion of this account of courage in section VI).
The real issue between (UV) and (UW) lies in the question whether
Socrates sought the meaning of "courage" or the reference. (Or, put in
another way, whether the unity Socrates speaks of is a matter of equiva-
lence of meanings or of identity of reference: see my "Unity of Virtue.")
That the issue is one of meaning vs. reference may be obscured if too
much attention is paid to my speaking of the virtues as causal entities in
my "Unity of Virtue." In fact I only introduced this expression under
pressure from the editors of the Philosophical Review (whose concern
for the welfare of the paper I, as a young philosopher, much appreciated,
in spite of my feeling that they were wrong on this point). I had wished
to keep the discussion on the level of virtues being references rather
than meaning (or things with the identity conditions of meanings). On
the other hand, the talk of causal entities will do no harm, provided one
realizes that virtues are causal entities not because all abstract nouns
stand for causal entities rather than meanings, but because all abstract
nouns stand for their references rather than their meanings. In the case
of virtue-words, the references just happen to be causal entities. Vlastos,
in "What Did Socrates Understand by His 'What Is F Question?" Pla-
tonic Studies, 410-17, misses this last point. He takes it to be a refuta-
tion of (UV) that "shape" in the Meno does not refer to a causal entity.
But it is no refutation at all, but rather an ignoratio elenchi, since the
reference of "shape" is not a causal entity. The Meno passages are in fact
inconclusive as between the meaning and the reference of "shape." (One
might note in this connection, however, that the two possible accounts
of shape that are offered are very far from synonymous. But shouldn't
they look something like synonymous if both are supposed to be good
Cambridge Companions Online © Cambridge University Press, 2006
154 T H E CAMBRIDGE COMPANION TO PLATO
candidates for the meaning?) Vlastos also argues that "quickness" at La.
I92ai-b4 can't be a state of soul, so must be a meaning. But (a) that
doesn't show that "quickness" isn't the reference rather than the mean-
ing; and, in any case, (b) on Vlastos's reading of "quickness" here, he
should say that the "quietness" or "slowness" that is contrasted with
"quickness" at Chrm. 15^3// , esp. C4, 6-9, ds, 10-11, e3-io, i6oai -
d3, also can't be a state of soul. But plainly it is: The contrast being
debated in the Chrm. passage is precisely that between people in whose
souls there is quietness and those in whose souls there is energy. Other
arguments of Vlastos on this point will have to be postponed to another
time.
Again, (UV) has the advantage of giving a far simpler and cleaner
reading of La. 197c—199c As the identity of
the science of future goods and bads
with
the science of all goods and bads, past, present, and future
is an identity of references and not of meanings, so for the identity of the
science of the fearful and the hopeful with the science of all goods and
bads which is identical with virtue. (Again, see the brief discussion in
Section VI.) Contrast the complicated reading Vlastos must give to this
argument in Platonic Studies, 266-9 - with Socrates making gratuitous
meaning-reference confusions.
Finally, on the other side, I should note that (UV) lies best open to
attack in passages about justice, courage, and so forth being "parts" of
virtue in the way that odd is a part of number (cf. Gig. 462c//): as at
Euphr. n e - i 2 e (though piety's being a part of justice is not followed up
here; and cf. Gig. 5O7a7-b4) and Meno 73d-75a.
(Something should be said about the last passage, since Vlastos makes
so much of it in "Socrates on 'The Parts of Virtue,' " Platonic Studies,
418-23. The force of the crucial passage 73d7-e2-about justice just
being a virtue - is considerably lessened when we notice the same theme
reappearing later at 78d3~79cio. The latter passage in turn leads us on
toward the discussion of whether virtue as a whole is knowledge at 87b//.
As already noted in [a] above, Socrates in 87b// in effect argues that wis-
dom is identical with virtue as a whole - which can only happen if cour-
age, etc., are identical with wisdom. See esp. 88bi-d3. More precisely,
what Socrates is saying here is that virtue is wisdom, or a part of wisdom.
[Cf, "a kind of wisdom" at 88d3, cf. La. i94dio, and compare 89ai with
88d2.] Note that, like other recent commentators, including Bluck, I here
depart from Guthrie, who translates that wisdom is virtue or a part of
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 155
virtue, rather than virtue being wisdom or a part of wisdom. Aside from
the parallels I have just cited against this, there is the point that if Socrates
were saying here that wisdom might be only a part of virtue, then 89C2-3
would be a nonsequitur: if wisdom might be only a part of virtue, we
could not infer that virtue is teachable.)
I should record here that it was Vlastos who, some twenty-five years
ago, brought home to me the importance of the point that the military
class in Rep. IV can have (at least demotic) courage without having
wisdom, just by having true belief - thus contradicting anything that
could have been said in the Socratic dialogues, whether we interpret
them in terms of (UV) or in terms of (UW). Also that the credit for using
Meno 87b// against the earlier part of the Meno belongs to Irwin, Plato's
Moral Theory, 301 n. 57 and 305-6 n. 3.
22 Cf. Meno 77b//, esp. 78b, with passages cited in note 24.
23 "best for me": I here assume what I shall argue briefly in Section IV, that
Socrates is an ethical egoist. For the move from "good" to "good for me,"
see Meno 77C8, as noted by R. S. Bluck, Plato's Meno (Cambridge, 1961),
71, as well as Grg. 468b6. For the move from "good" to "best," see Grg.
466a-468e.
24 On the importance of the means-end distinction to the Socratic theory
of desire, see esp. Grg. 466a-468e, Lys. 2i9b-22ob, La. i8sc-e, as well
as Euthd. 28id-282a, and possibly Meno 77e5~78a8. Usually, the So-
cratic theory that all desire is for the good is interpreted without refer-
ence to the means—end distinction - as saying that all desire is for the
apparent good. [Apparent good because of the following sort of case: I
desire to do this action, thinking it good, though in fact it is a bad
action - a mistake. So how is my desire for the good? It's for what I think
good - the apparent good.)
But once one notices the centrality of the means-end distinction to
Socrates' account of desire, it becomes clear that the usual interpreta-
tion cannot be right - especially when we are speaking of ends. (On
means, see also notes 42, 64.) On the usual interpretation, to desire
something as a means to an end can only be to desire what one thinks
the best means to what one thinks is the best end. But then on that
interpretation, one will desire the apparently best end, not the really
best end.
As against this, careful study of the passage cited above will make it
clear that Socrates held that one desires the really best end. (For Plato's
grasp on this point, and Aristotle's apparent missing of it, see Rep. VI
5O5e-5o6a, Top. VI I46b36-i47an. See also my "Power and Desire,"
esp. Section 12. The latter paper is the first in a series of discussions in
which I shall be opposing the view that Socrates thinks we desire the
Cambridge Companions Online © Cambridge University Press, 2006
I56 THE CAMBRIDGE COMPANION TO PLATO
apparently best end, not the really best end, and examining the Socratic
theory of desire that results from paying due attention to the means-end
distinction.)
25 "Desire to eat this pastry": Strictly, this should be: desire to eat this
pastry which will lead to the end which is in fact best for me in the
circumstances. (The latter is in fact the "very thing one wants to do/; of
note 18; cf. also note 24.) When the pastry is in fact bad for me, there
results a certain incoherence in the desire for the means, in the way in
which when we apply beliefs we suppose to be about the real world to
the real world, we get an incoherence in our beliefs. (Compare: "I believe
of Jocasta — whoever she is [you plug it in - that's whom my belief is
about] - that she is not my mother/7) I discuss this incoherence briefly
in Section 12 of "Power and Desire"; and in a little more detail in two
hitherto unpublished manuscripts: "Plato and Frege" and "Plato and
Protagoras." Cf. also the next-to-last paragraph of Section V of this chap-
ter, with note 64.
26 Strictly: the desire to do that abstaining which will lead to . . . (as in note
25).
27 So N. R. Murphy, The Interpretation of Plato's Republic (Oxford, 1951),
28-9, and Penner, "Thought and Desire in Plato," in Plato, vol. 2., ed.
Gregory Vlastos, 96—118. Cf. also Phdr. 237d-238c.
28 See note 11.
29 The simpler picture Plato gives us of irrational and weak human behav-
ior has been widely believed to be correct, more or less from the time of
Plato's parts of the soul doctrine; and Socrates' account has been
thought to be incorrect. See, for example, the impressive attacks on
Socratic intellectualism in Gregory Vlastos, Plato: Protagoras, xxxix-xl,
xlii-xliii, and Vlastos, "Introduction: The Paradox of Socrates," in The
Philosophy of Socrates, ed. G. Vlastos (Garden City, N.Y., 1971), 15-16,
and the endorsement of the parts of the soul doctrine in Vlastos, "Socra-
tes," 99, with n. 63. See also my exposition of that doctrine in "Thought
and Desire" and my attack on it in "Plato and Davidson: Parts of the
Soul and Weakness of Will," in Canadian Journal of Philosophy, supp.
vol. 16(1990): 35-74.
30 Cf. Laws 86od-863e, Sph. nje-i^oe-, Ti. 86c-e, Laws 731c and 734b.
31 Even a person with true opinion, be it noted (though true opinion is not
mentioned in the Prt. - or indeed anywhere in the Socratic dialogues,
save only the transitional Meno and Grg.). This absence of true opinion
is a key explanandum for interpreters of the Socratic dialogues.
32 The question of the instability of opinion is omitted in almost all ac-
counts of the actual argument for the strength of knowledge in the Prt.
These accounts of the strength of knowledge simply follow from "No
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 157
one errs willingly" together with "Whoever knows that p also believes
(or truly believes) that p." So, for example, the fine pieces by G. X.
Santas, Socrates: Philosophy in Plato's Early Dialogues (London, 1979),
chap. 7, and by James J. Walsh, Aristotle's Conception of Moral Weak-
ness (New York, 1963), chap. 1. Such accounts have the consequence
that Socrates should hold that knowledge is strong because belief is
strong (or-if erring is a form of weakness - because true belief is
strong). For a sketch of the view presented in the text (and to be elabo-
rated elsewhere) see my "Plato and Davidson."
33 I am not here denying that knowledge is necessary to the rational part.
What I am questioning is whether that knowledge is the real source of
the strength that the rational part has in holding out against appetite.
And I am suggesting that the source of the strength resides in the train-
ing of emotions and desires to follow whatever reason says. (If one's
reason perchance came to a false conclusion, reason would still be
strong. This is at the furthest remove from Socrates' conception of the
strength of knowledge.) Richard Kraut has suggested to me that Plato
would have said that understanding of the Forms would intensify one's
desire to be virtuous. This can be granted, I think, without entirely
removing the suspicions about the source of the strength of reason that I
have been raising here.
34 Again, see note 11.
35 Just a few examples: La. i94di-9; Euthd. 278d-e; Rep. I 339b with
354a; Euphr. 14b—c. (Euthyphro turns away [ i4an] just when he is on
the verge of an answer [ i 3 e i o - n ] . If justice is the service [therapeia] of
humans and piety the service of gods [ne//, esp. i2e6-8], what is the
work [ergon] the gods undertake when our service helps them? Euphr.
14c// simply shows the bad effects of this turning aside.)
In addition, Meno 77b-78c; Lys. 2O7d-2iod with 2i ia4-5; Ap. 24C-
26a, 26b-27a; Chrm. I72b-i74b7 (Socrates'dream); Grg. 453b7~455a6;
are just a few examples of actual bits of dialectic where Socrates lets us
know what positive result he seeks, and indeed that he has gained it. Cf.
also La. I92b-i93d where Nicias at i95C5-dio sees what Socrates
wanted Laches to see.
36 H. Mi. 375<1//; cf. Rep. I 335c, cf. 35ob-d, 35ia-c, 3536-354^
37 Socrates evidently uses justice and virtue in the passage interchange-
ably: cf. also Rep. I 3sod, 35ia-c, 353c, and (b) in note 21.
38 See preceding note.
39 In my "Socrates on Virtue and Motivation," in Exegesis and Argument,
ed. E. N. Lee, A. P. D Mourelatos, and R. M. Rorty (Assem, 1973), 133-
51, I use the following analogy: Economic man never errs willingly at
maximizing his own profit. For he can never have an economic motive
Cambridge Companions Online © Cambridge University Press, 2006
I58 THE CAMBRIDGE COMPANION TO PLATO
to fail to maximize his own profit. But any other expert can have an
economic motive for erring willingly at their particular science. For
example, it may be more economically profitable for a doctor to subtly
kill off a patient in order to gain earlier and more profitable access to the
patient's legacy. Cf. also Gig. 5O2C2-d7, as well as the similar argument
I make at note 49. Cf. also Rep. sosd-e for a close relative of this point in
Plato.
40 As against the interpretation of the H. Mi. offered here, Kraut, Socrates,
311-16, argues that Socrates is honestly perplexed by the conflict be-
tween virtue's being a science and the sciences being ambivalent, appeal-
ing in particular to the sincerity of 372d7-ei. I would argue that the
immediate context (b4—d7, e3-d6) suggests irony rather than sincerity.
Mainly, however, I just don't see the dictum "No one errs willingly" not
occurring to Socrates as relevant to resolving the announced conflict as
he wrote the last lines of this dialogue. Kraut and I are agreed, however,
on the seriousness of the issue raised by ambivalence for Socrates' ethi-
cal egoism. See also notes 51 and 54.
41 Here are some (sketchily indicated) examples:
a. Lys. 2i6d—217a plus 2 i7a -2 i8c - in spite of 218c//, esp. 22oci-e6
with escape clause at 22oe6//, esp. 22ias-ci (with more puzzles for
the interlocutors to follow).
b. La. 1976-1996, esp. 198C6-7 (cf. i94di-9 with I94en-i95ai)plus
I98a4—5 - and cf. the dropped hint at I95a4~5 of what Nicias is
missing at the end. (This passage is discussed briefly in Section 6.)
c. Meno ySc-ygc, esp. 78C4-5 (Socrates' own answer to the question
what virtue is!) plus C5~d3 (Meno's disastrous view of what things
are good) plus 79a3~5.
d. Chrm. 172c-174d. Here Socrates reduces to absurdity Critias's view
that temperance is the supposed knowledge (science) of the things
one knows and doesn't know, by showing that the knowledge that
makes medicine, house building, and all the other sciences benefi-
cial is the knowledge (science) of good and bad (i74bio-c4, d.5-6,
ti). It is the knowledge of good and bad, and not temperance - i.e.,
not the supposed knowledge (science) of the things one knows and
doesn't know (i74bi2-c2 with i74d3-6). And though the hint is
plainly dropped that Critias should identify temperance with the
knowledge (science) of good and bad, Critias is too anxious of victory
(cf. earlier 162c-e, 169c-d) and hangs on tenaciously to the supposed
knowledge (science) of what one knows and doesn't know (i74d-
I75a8). It is also noteworthy that Critias's main account of temper-
ance (to which Socrates steers him at i64a-c), as knowing what one
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 159
knows and doesn't know, is plainly intended by Critias to be (what
Critias thinks of as) the Socratic virtue, the human wisdom of Ap.
22e-23b: being able to examine oneself and others to see whether
they think they know (when they don't) or whether they know they
don't know (16 7a 1-7, ijod$). (Critias, like Nicias on courage, seems
in general to put forward accounts of temperance that aim at being
Socratic.) But Socrates doesn't let Critias get away with this appar-
ently Socratic position, since Critias doesn't see the whole picture -
and in particular doesn't see that true virtue is the substantive
knowledge of good and bad that Socrates acknowledges he doesn't
have.
e. Meno 89d-96d (virtue is teachable if and only if there are teachers -
leading briefly to the doubt that there may not even be any good
men) plus 96d-iooc (supposedly virtuous politicians, not being
able to impart their virtue, have virtue by "divine dispensation"
and without intelligence [nous: 99e6], but no politician has the
kind of virtue that can be taught to others). See further notes 58
and 63.
For opposing readings of the above passages, and of other similar pas-
sages, see the powerful and all-but-persuasive Kraut, Socrates, chap. 8.
42 I first suggested this general strategy, as well as the claim that the good
differ from the bad not in their wills but in their intellects, in my
"Socrates on Virtue and Motivation." On the latter claim, we can make
this much accommodation to the usual view, in the light of note 24:
People don't differ in the ultimate end they desire, merely in the means
they choose to that end. So if we distinguish the will's desire for means
from the will's desire for the ultimate end, we can say good people
differ from bad in their wills. But it has to be clear that it is the will's
desire for means that is in question. And the fact is that the will's
desire for means derives immediately from (a) desire for the ultimate
end (the same in everyone), together with (b) the agent's beliefs as to
the best means for achieving that end in the circumstances the agent is
in. So, once more, the good differ from the bad solely in their beliefs (as
to what is a means to what). (Compare and contrast Aristotle on "igno-
rance in the choice" vs. ignorance of circumstances at N.E. III.i.) See
also notes 24-26 and 64.
Notice that in saying people all desire the same ultimate end, I am
implicitly rejecting the view that good people differ from bad people by
good people desiring virtue for its own sake and bad people desiring such
things as pleasure for its own sake. This peculiar Aristotelian use of "for
its own sake" - where one can both desire sight for its own sake and for
Cambridge Companions Online © Cambridge University Press, 2006
I6O THE CAMBRIDGE COMPANION TO PLATO
the sake of happiness - only invites confusion, I think. On the view I
attribute to Socrates, bad people only seem to desire pleasure above
everything else because they think falsely that getting pleasure is the
best means to their real good. That is, they don't desire pleasure for its
own sake. What they desire for its own sake is their own real happiness -
even if what that consists in is different from what they think it is. (See
note 24, 25, 66, esp. the remarks about incoherence in the desire in note
25; also my 'Tower and Desire," Section 12.)
43 See Ap. 2oa-b, 24C-25C, with Cri. 46c-47d; Euphr. i3b-c, Grg. 52oa4
with C4-5; and on care for the soul generally, cf. also Ap. 29d-3oa, 31b,
32d, 36c, 39d.
44 La. 180c, cf. i8oe-i8ia, i8sai; Lys. 2O3a-2O4b; Chrm. 153d- 154c
45 For "instrument," see H. Mi. 37463. Cf. also the souls of horses, dogs,
and other animals; archers, doctors, and lute players; slaves; and our
own souls at 375a4, 6-7, 8, bs, 7-8, C4, and 6-7. Notice also the compari-
son of the soul with such instruments as eyes, ears, and pruning knives
at Rep. I 3526-3543, esp. 353a9-n with d3-en.
46 Thus Socrates takes it here that the good person is the person good at
getting happiness - an outrageous conclusion by modern lights. See the
discussion in the paragraph after next, as well as in note 49.
47 Socrates here implies — inconsistently with either (UV) or (UW| - that
it is possible to have courage and justice without having wisdom - when
one uses one's courage or justice not well. Cf. note 21.
48 In the first protreptic of the Euthd. (278c-282e), as at Chim. I73a-i74d,
Socrates all but speaks of a science of happiness. There is of course much
more that needs to be said about the first protreptic. For example, there
is the paradox Socrates propounds here that wisdom is happiness, on
which see my "Belief-Relative Sciences." There is also the second
protreptic in the Euthd. at 288c-292e, where Socrates raises difficulties
about identifying this science he calls wisdom, and of which we must
give an account that will cohere with our account of the first protreptic.
49 To think that the mere making of the distinction between moral and
functional good is, by itself, enough to defeat the Socratic argument is to
underestimate the resources of the Socratic position. Socrates can ask:
Why isn't moral good (or, better, ethical good, given that "moral" sug-
gests a rule-based ethics as opposed to a happiness-based ethics) func-
tional? Must the difference between a good parent or a good friend
(which are functional) and a good person be all that great? Must there be
a difference of kind here? (This line of thought is familiar in modern
times from P. T. Geach, "Good and Evil," Analysis 17(1956): 33-42, and
Stuart Hampshire, Thought and Action (London, 1959), 227//, esp. 229,
231-2, 236.
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 161
It might be thought that the following sort of argument could be made
against Socrates' use of "function":
The function of a hammer is to fulfill the standard expectations
for a hammer — certain purposes of other beings (humans). But
what is the function of the human being? Not now the purposes
of others, but of the self: doing well, happiness! So the function
argument breaks down for good human being.
(For an argument of this sort, see Irwin, Plato's Moral Theory, 14-15.)
But the function argument no more breaks down, I think, than the
following argument:
The function of expertises like medicine and cobblery is not the
maximizing of the experts' own economic good; but it is the
function of economic experts to maximize their own economic
good. Therefore, the function argument breaks down for
economic experts.
The objection that follows in the main text - about the permissibility of
harming others on the ethical egoist view - is of course quite another
matter.
50 My analogy between Socrates and harm on the one hand, and Gandhi
and violence on the other, may tempt some - impressed by Socrates'
apparently willing military service - first, to object that there is no evi-
dence that Socrates was nonviolent, and, second, to suggest that by
"harm" Socrates simply meant psychic harm in the form of "making
people more unjust." These reactions seem to me to go too far. I would
grant only that he thought psychic harm a greater harm than physical
harm. I myself find it hard to dissociate physical harm from psychic
harm.
What of self-defense and punishment? Would Socrates not have
thought violence in the pursuit of these goals acceptable? It might be
thought that Socrates could have no (political) objections to pre-
announced postures of self-defense designed to avoid violence on the
part of bullies. But then Socrates was not a very political person. Still,
it is not clear that he has ruled out individual, unpremeditated acts of
self-defense. In any case, I suggest in the next note that Socrates might
well have had doubts about punishment. (I am grateful to Richard
Kraut for pressing me on the question of harm.)
51 There is an interesting corollary to this argument: Since no one errs
willingly, the only "punishment" that is ever appropriate is instruction
(Socratic questioning)! Merely another instance of Socrates' political
innocence (or insouciance)? Or is he serious about the point? I believe he
Cambridge Companions Online © Cambridge University Press, 2006
162 THE CAMBRIDGE C O M P A N I O N TO PLATO
is. Certainly he would not have supposed that legal punishment has, by
itself, much to do with care for anyone's soul. See the interesting discus-
sion of this issue at Kraut, Socrates, 313 - which cites passages both for
and against the possibility that we should take Socrates as expressing
serious doubts about the institution of legal punishment. But of the
passages Kraut cites against, only Cri. 49a is at all troubling. And an
explanation of it too is possible. (For example, along the lines of Ap. 25c
on willingly harming those around one: Try to avoid willing injustice =
try to be clear that nothing you willingly do is in fact unjust. Cf. Kraut's
own remarks at Socrates, 213 n. 46.)
52 Socrates also maintains that it is bad to harm others at Cri. 49a-e and
Rep. I 335b-e, though we do not have in these passages a clear intima-
tion that the argument is made on the basis of self-interest. Socrates
simply assumes that what is best for us will be best for others at, for
example, Grg. 520a//, esp. 520c, 52ib-c.
53 In my "Belief-Relative Sciences/' I try to explain why Socrates would
reject the possibility of such sciences as the psychology-free moral sci-
ence that Kant envisages. For Kant, most of the brilliant argumentation
that Aristotle gives us on pleasure and happiness in Books I, VII, and X of
the Nicomachean Ethics is entirely irrelevant to moral philosophy.
54 We confront here the claim that ethics must deal with moral "evil ge-
niuses," which I deny in my "Virtue and Motivation." Kraut, Socrates,
314-15, thinks Socrates would have worried about this question.
55 Of course many modern philosophers will agree with Socrates (as with
Plato and Aristotle) that ethics should not be insulated, in the Kantian
manner, from facts about human nature. Their quarrel with ethical ego-
ism will not be for bringing psychology into ethics; it will reside in the
charge that egoism cannot provide a complex enough psychology on
which to base ethics. (I am indebted here to Richard Kraut.) Can ethical
egoists love their children, for example? This is not the place to defend
egoism on such points.
5 6 Another controversial psychological hypothesis to which Socratic ethi-
cal theory is committed if its ethical egoism is to be defensible has to do
with the fear of death. Ethical egoists, it is alleged - at least those who
don't believe in immortality - must always in the end go to any lengths
(including immorality) to avoid death. So, once again, it might be
thought, ethical egoism is a nonstarter. But it is clear, especially in the
Ap., that Socrates rejects the claim that one must go to any lengths to
avoid death. What Socrates will refuse to give up, even under the threat
of death, is caring for wisdom and truth and that one's soul be the best it
can be (29a-3oe, 32a-e, 35a, 36c, 38e~39b; cf. Grg. 522d-e). In such
care, indeed, lies one's happiness. Socrates is well aware that the medi-
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 163
cal question, Can I keep this patient alive? is a quite different question
from Is it better for this patient to live or die? (La. i95c-d). And he
would have agreed with John Stuart Mill that "the conscious ability to
do without happiness gives the best prospect for realizing such happi-
ness as is attainable" (Utilitarianism, chap. 2). It is not in our interest to
do just anything to escape death. That is no recipe for happiness.
57 Thus I meet the objection in the last paragraph of Section III by yet
another use of the general strategy in the next-to-last paragraph of Sec-
tion III.
58 I have suggested in Appendix II of my "Belief-Relative Sciences/7 as well
as in "Virtue and Motivation/' n. 32, that the fact that no one has virtue
is the resolution of the apparent conflict - central to both the Protagoras
and Meno — between Socrates' claims that (a) virtue is knowledge, yet (b)
virtue is not teachable. If we add to (a) and (b) that something is knowl-
edge if and only if it is teachable, we get a contradiction. The Meno
makes it clear that Socrates' claim that virtue isn't teachable is based
upon the claim that there are no teachers, since the politicians who are
supposed to be the teachers of virtue evidently don't have the knowledge
in question. And, sure enough, there can't be any teachers of virtue if
even the wisest of mortals has no substantive knowledge of goods and
bads. (See further [e] in note 41; also note 63.)
59 Prt. 329c6-di, 349D2-3.
60 This is a representative list of central Socratic questions. Many interpret-
ers claim that a subclass of these - the What-is-it? questions, asked of
the virtues - are the central philosophical questions for Socrates. Prop-
erly understood, this is not far wrong. But if it is supposed that the
correct answers to these questions will give us in each case one true
proposition as the "definition"; and that the point of the definition is to
give us "standards" - necessary and sufficient conditions - for judging
whether particular actions qualify as just, brave, temperate, and so forth;
then I have some difficulties with the claim. Since I speak of my difficul-
ties with propositions below (Section VI, with note 66, 72, and 78), I will
say nothing more on that point here. I note only that several important
What-is-it? questions have no very strong connections with standards.
Take, for example, What is friendship? (or, perhaps, What is desire for
good?, since the Lys. is really a dialogue about desire for good rather than
friendship), What is the experience of being overcome by pleasure? (Prt.
352e7-353a6, 353CI-2, 35466-7, 357C7-di, e2, cf. 35562-3), and What
is rhetoric? (Grg. 447c with 448e-449a; 45ia-b; 453a4~5 with 6-7;
462b, 463c). (The latter two are hardly ever listed as amongst the What-
is-it? questions.)
For two of the best of these authors who see the What-is-it? questions
Cambridge Companions Online © Cambridge University Press, 2006
164 THE CAMBRIDGE COMPANION TO PLATO
as directed at standards, see Kraut, Socrates, 209, 213-14, 233-4, 247;
251-2, 300, 309, and esp. 254-8; Irwin, Plato's Moral Theory, 42-4, 61-
2, 65, 68, 72. As against this view, I don't myself think Socrates was
particularly interested in casuistry, except where, as at Euphr. 4e-9e (the
principal evidence to which these interpretations appeal), it gives him a
way into the examination of those characteristics of people (as opposed
to actions) that will lead to their good. Euphr. 9e -nb is profoundly
irrelevant to the "standards" Socrates is supposed to be especially inter-
ested in; and the end of the dialogue (13a//) seems to concern itself
entirely with sciences — which are characteristics of people: cf. i3a-b.
Indeed, it seems likely Socrates thinks the answer to the question What
is piety? lies here - in some sort of science. (See esp. I4b8-ci, which
locates Euthyphro's turning away at i4an from deciding [a] what the
science the gods have is to accomplish, in order to see [b] what the
science of serving the gods [ i3e io-n] is to accomplish.)
Kraut thinks Socrates is interested in standards for actions, in part
because of (political) passages that seem to him to imply that an expert
in virtue should command others and others should obey. But most of
the passages Kraut cites in support of this claim seem to me have to do
less with politics, ruling, and casuistry than with one-on-one education
of the individual (Socrates, 196-9, 257). (That is, once more, the concern
is with characteristics of people, not of actions.) That Socrates should,
even in these passages, be more concerned with individual acquisition of
virtue rather than moral judgments on actions is what one would expect,
I think, given how little a practical proposition political life seemed to
Socrates for a person bent upon acquiring human goodness. (Cf. Socra-
tes, 208-15.)
61 An individual's "values" are what the individual "thinks-good/' where
the hyphen indicates we have one inseparable word. Though what some-
body thinks red can be contrasted with what is red, "thinks-good" does
not contrast with "is good" because of the hyphen. (Cf. the modern
notion of a "value judgment" as opposed to judgments on scientific
matters or matters of fact.) That is the notion of values that Socrates and
Plato were fighting. It is still around today.
62 The phrase "for that individual" of course takes some unraveling. All I
am saying here is that the upshot of "What appears good to A is good for
A" is that on that view, one's final end is whatever it appears to one to
be. For Protagoras in the Tht. see i52a-b, i66d4-8 ("is and appears")
i72ai-5, b2-6.
The argument of the present paragraph in the main text of course
presupposes that there is a distinction between getting what is best and
getting what seems best to you. And, as I remarked three paragraphs ago,
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 165
Protagoras would deny this. Though Socrates' argument here will work
against positions like those of Gorgias and Polus, it will not work
against any position so radical as that of Protagoras. The radical Protagor-
ean position is not argued against head-on until the Tht. Even Cra.
385e-386d is simply a replay of Socratic appeals to the sciences.
63 Obviously I am taking the Meno to be ironic in its suggestion that there
could be virtue based on true belief rather than knowledge. In other
words, I have this part of the Meno looking backward to the Ion on
"divine dispensation." (How is it that politicians attain virtue, given
that they don't have knowledge? It's a gosh-darned miracle! Socrates
says.)
Other interpreters take the Meno rather to be looking forward to the
virtues of the military class in the Republic. That is, they take the
Meno's broaching of true belief without knowledge as a form of virtue to
be making a serious suggestion about virtue - opening the way to the
Rep.'s virtue of courage in a military class that does not itself possess
knowledge, but merely true belief (cf. point 9 in Section II). I was myself
a little soft on this interpretation in my "Unity of Virtue," n. 32.
Against such interpreters, it may be urged that Socrates' rough treat-
ment of Anytus in this context in the Meno counts strongly against the
view that Socrates is seriously suggesting politicians have a pretty ade-
quate sort of virtue. For it is the dangerously angry Anytus - most formi-
dable of the eventual accusers of Socrates - whom Socrates chooses as
the representative and defender of politicians,- see esp. 95c with 100b,
and notice Burnet's reasonable conjecture that it is Anytus who is re-
ferred to at Ap. 2ic. Cf. also the rough treatment of Pericles and
Themistocles (Meno 93b—94c, taken very ill by Anytus at 95 a; and Grg.
515C-5 i9d). I treat of this point in a little more detail in "Belief-Relative
Sciences," Appendix II. For an opposing view, see Kraut, Socrates, 278,
285//, esp. n. 81; and see further notes 41 and 58 above.
64 For my insistence on "wanting that exile of my rival which will in fact
make me happiest" as opposed to simply ("simply": Grg. 468C3) "want-
ing the exile of my rival," see notes 25 and 26, with note 24. There is a
rather fuller discussion of the entire issue here in my "Power and Desire."
65 For example, the alleged contradiction in knowing only that one doesn't
know; the apparent contradiction of this last claim by claims at Ap. 29a6-
b9 and 37b5~9; and Socrates' alleged claims of moral superiority at Ap.
34e-35a which have suggested to some a claim to moral knowledge.
66 It is axiomatic within modern philosophy that (a) things known, (b)
things truly believed, and even (c) things falsely believed are all the
same kind of things: propositions. (So that if any one of these kinds of
things exists independently of our thought and language [as one might
Cambridge Companions Online © Cambridge University Press, 2006
166 THE CAMBRIDGE COMPANION TO PLATO
hope facts and things known would], so do the others,- and if any one of
these is a mere convenient artifact of theory [as one might suspect false
beliefs and even some true beliefs would be], they all are.) Though I
understand the convenience to philosophers of regimenting the objects
of knowledge and belief in this way, I have some doubts about the under-
lying assumptions — reinforced by other doubts about propositions made
familiar by Quine, Davidson, and others. (Cf. also my "Belief-Relative
Sciences/7) But this is not the place to discuss these doubts.
67 Due to Kent Anderson.
68 This example of course depends upon the correctness of my interpreta-
tion of La. I97e-i99e. Both Vlastos, Platonic Studies, 422-3, and Kraut,
Socrates, 260, argue that Socrates cannot be attacking the claim that
courage is only a part of virtue since he himself introduces it at 190C8-
d8. The whole first pass with Laches at igocSff needs discussion. I shall
be offering such a discussion, with a response to the Vlastos-Kraut objec-
tion, elsewhere.
69 Arguments parallel to those just given about Nicias in the La. can be
made about Critias's obvious attempts to give Socratic accounts of tem-
perance in the Chrm., and about Meno's giving of a Socratic account of
what virtue is in the Meno. See examples (c) and (d) in note 41.
70 As in accounts of the elenchus, like Vlastos, "The Socratic Elenchus,"
Oxford Studies in Ancient Philosophy 1 (1983): 39-42, 47-9, which
work with a sharp deductive-inductive distinction and the narrowest
possible criteria of propositional identity.
71 I owe this point to Ruth Saunders. Those who believe in propositions
have merely to attribute the view to Socrates that you can't know the
proposition that self-knowledge is good if you don't know the (quite
different) proposition that knowledge is strong. (Some defeasibility theo-
rists might find a way for knowledge of the one proposition to be de-
feated by defects in one's knowledge of the other.)
72 The position of Kraut, Socrates, 280-5, is interesting in this connection.
For it comes very close to the position I am maintaining here, as earlier
in my "Virtue and Motivation" - in spite of a very strong commitment
on Kraut's part to propositions of neo-Fregean ilk. (For Kraut's commit-
ment to propositions, see Socrates, 197 n. 8, 211, 220-2, 225, 231, 241,
246, 260, 264, 269, 272, 278-9, 283-4; m many of these passages Kraut
is actually counting propositions ["some true beliefs, some false be-
liefs"].) Kraut shows this same closeness to my position in his review of
Irwin's Plato's Moral Theory, Philosophical Review 88 (1979): 633-9.
Curiously, we can also find in Irwin [Plato's Moral Theory, 63-4, 69-70)
passages reminiscent of the similar positions in Kraut and myself. Thus
it is both the case that Kraut rightly criticizes Irwin in places for too
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 167
little attention to the broader picture (too narrow an interest in individ-
ual propositions, as I would put it) and that Irwin too sometimes does
quite adequately have his eye on the broader picture.
73 On the connection between the two explanations: In the preceding sec-
tion, one fairly trivial false belief ('Tor want of a nail . . .") can throw off
an entire means-end structure, just as in the present section, one's
knowledge that virtue is knowledge may be thrown off by not knowing
why there are no teachers of it. In both cases, one is failing to see the
whole picture.
74 See Vlastos, "Introduction: The Paradox of Socrates/' 16-17. Here
Vlastos claims that behind (what Vlastos sees as) Socrates7 failure to see
that knowledge by itself cannot be the way to the saving of human souls,
lies "a failure of love./;
In saying this, I am not taking over-seriously the prickly exterior and the
pugilist's postures. I have already argued that he does care for the souls of
his fellows. But the care is limited and conditional. If men's souls are to be
saved, they must be saved this way. And when he sees they cannot, he
watches them go down the road to perdition with regret, but without an-
guish. Jesus wept for Jerusalem. Socrates warns Athens, scolds, exhorts it,
condemns it. But he has no tears for it. One wonders if Plato, who raged
against Athens, did not love it more in his rage and hate than ever did
Socrates in his sad and good-tempered rebukes. One feels there is a last zone
of frigidity in the soul of the great erotic; had he loved his fellows more, he
could hardly have laid on them the burdens of his (as Nietzsche has it)
"despotic logic," impossible to be borne.
75 Vlastos, "The Socratic Elenchus," 35.
76 See the preceding paragraph.
77 Ibid., 37.
78 No doubt most of Socrates' interlocutors would have considered them to
be five different beliefs, since they will have thought that any one of
them could be true while any other one is false. Such is the standard
Fregean and neo-Fregean criterion of identity for propositions: if Oedi-
pus thinks that (a) he is married to Jocasta, but that it is false that (b) he
is married to his mother, then even for Tiresias, who knows the truth, (a)
and (b) express different beliefs, and (a) and (b) are therefore different
facts. (Cf. note 66.) But is it clear this is right? If we want Oedipus's
beliefs to be beliefs about Jocasta, is it clear we can leave out of the fact
(a) just everything about her except perhaps her being named "Jocasta"?
In any case, my suggestion about Socrates here is this: To know that
virtue is one, we shall have to know of this virtue that it is the virtue
which is knowledge-as in knowing that 11 + 1 = 12, we had better
know that 12 is the successor of 11, and maybe even that 11 is the
Cambridge Companions Online © Cambridge University Press, 2006
168 THE CAMBRIDGE COMPANION TO PLATO
successor of 10. The idea is that knowing that virtue is such and such
will, for Socrates, be knowledge about virtue; so the question will arise:
Which virtue? A virtue which is the mere genus of five different virtues,
wisdom, temperance, and so forth? Or a virtue which is identical with
wisdom, which is identical with temperance, and so forth? We must
know the reference of "virtue," not just the meaning. (On orthodox
Fregean propositional analyses of belief and knowledge contexts, the
belief that virtue is knowledge is not about the reference of "virtue" so
much as the meaning of virtue.) I discuss these topics further in the
unpublished manuscripts cited in note 25.
It will be obvious that the remarks of the previous paragraph also
speak directly to the alleged "Socratic fallacy" which has long puzzled
and annoyed modern readers. (See P. T. Geach, "Plato's Euthyphro: An
Analysis and Commentary," Monist 50 (1966): 369-82,- also Taylor,
Plato's Protagoras, 212-13, on Prt. 36oe6-36ia3.) The "Socratic fallacy"
is found in Socrates' claim that we can't know whether x is F until we
know what x is. Moderns deny this claim. They say that we can know
that x is F without (exhaustive or essential) knowledge about what x is.
What is more, we only get to know what the reference of "x" is (= what
x is) by knowing first lots of "facts" like that x if F. (Facts being true
propositions, and propositions being meanings, all we need access to, on
this view, in order to know that x is F, is the meaning of "x." We don't
need to know the reference of "x.") It is this picture I am denying when I
say that for Socrates knowledge that x is F is knowledge of the reference
of "x."
79 Cf. also Kraut, Socrates, i^sff
80 Frege, "On Sense and Reference," in Translations from the Philosophi-
cal Writings of Gottlob Frege, ed. Peter Geach and Max Black, 2d ed.
(Oxford, i960), 58.
81 By the same token, what is important in an interpretation of Socrates is
that (as one might say) the different pieces - about virtue being knowl-
edge; about desire being for the good; about ambivalence, advantage,
rhetoric, power, and poetry,- about harming others; about being over-
come by pleasure; and so forth - all hang together as a single whole.
Though I have not been able to show this in the space available, but
merely to hint at it, I hope this is true of my interpretation, as it is (in
fact) of the quite different interpretations of Vlastos, Santas, Irwin,
Kraut, and others. There is no substitute for working out a whole view of
all these matters simultaneously. This means, of course, that the risk of
error in interpretation is correspondingly greater. As will become clear, I
believe this has something to do with explaining the fact that though
Socrates appears to hold a great many beliefs very seriously, he never
Cambridge Companions Online © Cambridge University Press, 2006
Socrates and the early dialogues 169
claims to know anything substantive. Knowledge of anything substan-
tive could only come if one could be sure of every part of the picture. By
the same token, though I believe my interpretation is better than its
rivals, I would not claim to know it is correct.
Cambridge Companions Online © Cambridge University Press, 2006
IAN MUELLER
5 Mathematical method and
philosophical truth
I. PLATO'S ACADEMY AND THE SCIENCES
At some time between the early 380s and the middle 360s Plato
founded what came to be known as the Academy.1 Our information
about the early Academy is very scant. We know that Plato was the
leader (scholarch) of the Academy until his death and that his
nephew Speusippus succeeded him in this position. We know that
young people came from around the Greek world to be at the Acad-
emy and that the most famous of such people, Aristotle, stayed there
for approximately twenty years. However, it appears that, at least in
Plato's time, there were no fees attached to being at the Academy.2
Thus it does not seem likely that it had any official "professorial
staff" or that "students" took a set of courses to qualify them to fill
certain positions in life. The Academy was more likely a commu-
nity of self-supporting intellectuals gathered around Plato and pursu-
ing a variety of interests ranging from the abstractions of metaphys-
ics to more concrete issues of politics and ethics.
In Book VII of the Republic Socrates describes a plan of higher
education designed to turn the most promising young people of a
Utopian city-state into ideal rulers. It is frequently assumed (and quite
naturally) that this curriculum bears a significant relation to Plato's
plans for the Academy; sometimes it has even been described as essen-
tially the plans themselves.3 It is important to see that this assump-
tion is subject to major qualifications. For, first of all, fourth-century
Athens is not even an approximation to Plato's Utopia; Plato could
I would like to thank Richard Kraut for his comments on an earlier version of this
paper.
1 7 0
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 171
not expect entrants in the Academy to have been honed in the way the
Utopian citizens are supposed to be. Secondly, the educational timeta-
ble of the Republic seems totally impracticable for a privately orga-
nized institution in a free city: ten years of mathematics - that is, of
numbers, geometry, stereometry, mathematical astronomy, and har-
monics;* five years of dialectic; fifteen years of practical experience;
and then, for a few select fifty-year-olds, ascent to the Good, followed
by alternating periods of ruling and philosophizing. We do not know
whether the Academy had any curricular requirements at all, but it
seems to me highly likely that the Academy would have been still-
born if Plato had announced to new enrollees that they would begin
their most important studies thirty years later.
We must, I think, assume that Academic "education" was more
compressed than this, that mathematics, dialectic, and discussion of
goodness were carried on simultaneously. But how were they carried
on? Again, I think one should stress informality. Groups of people
gathered together to discuss subjects of common interest. In these
discussions there would obviously be leaders, teachers. We know
that Plato gave at least one public lecture on the Good, and several
references in Aristotle give us grounds for thinking that Plato put
forward some ideas in discussion that he did not express in the
dialogues.* Presumably, mathematics, too, would involve some lec-
turing, but there is reason to think that forms of Socratic discussion
were also common.
As for subjects of scientific discussion, it is important to be aware
that our evidence suggests that more disciplines than those men-
tioned in the Republic were treated in the Academy. The most gen-
eral kind of evidence is just the interests of various people closely
associated with the Academy.6 But we have valuable more specific
indicators as well. One comes from a conversation between un-
named speakers in a fragment (Theodorus Kock, ed., Comicorum
Atticorum Fragmenta, 3 vols. [Leipzig, 1880-8], 2: 287-8) of a com-
edy by Plato's contemporary Epicrates:
What about Plato, Speusippus, and Menedemus?7 What subjects are they
dealing with now? What thought, what argument are they investigating? If
you've come knowing anything please tell these things to me with discretion.
I can talk about these things clearly. At the Panathenaic festival I saw a
band of gay youths in the gymnasium of the Academy8 and heard them say
Cambridge Companions Online © Cambridge University Press, 2006
172 THE CAMBRIDGE COMPANION TO PLATO
unutterably weird things. They were making distinctions concerning na-
ture, the life of animals, the nature of trees, and the genera of vegetables.
Among other things they were studying the genus of the pumpkin.
How did they define it? What is the genus of the plant? Reveal this to me
if you know.
Well, first they all stood silently, bent over, and they thought for a consid-
erable time. Suddenly, while the young men were still bending over and
reflecting, one of them pronounced it a round vegetable, another a grass, a
third a tree. A Sicilian doctor who heard these things blew a fart at the fools.
That must have made the students very angry. I suppose they shouted out
against the man's derision. For it is out of place to do such things during a
discussion.
It didn't bother them. Plato was there, and he enjoined them, very gently
and without agitation, to try again from the beginning to distinguish the
genus of the pumpkin. They proceeded to do so.
The reliability of a comic representation is always subject to the
skepticism of scholars who hold theories incompatible with the repre-
sentation. This representation of Plato overseeing a biological classifi-
cation in the Academy does not fit well with the educational scheme
of the Republic. But, as I have already indicated, that scheme is an
ideal one for an ideal state. And it is also pitched to the specific
philosophical purpose of showing how certain studies turn the soul
from the sensible world to the intelligible one. (Note especially 521c-
d.) This philosophical purpose very much colors Socrates' description
of the course of higher education,- although it would be wrong to
downplay things he says to the extent of denying that Plato thinks
them true, one should not suppose that what he says exhausts Plato's
position on science or is free of rhetorical exaggeration.
The other piece of evidence I wish to consider brings us directly
into the domain of mathematics. It is an account of Plato's activities
found in Philodemus's history of the Platonic school, written in the
first century B.C.9 Unfortunately, it is preserved in a papyrus roll in
tenuous condition, and requires supplementation of varying degrees
of certainty. In my translation I indicate some of the major problem-
atic places.10
At that time great progress was seen in mathematics, with Plato serving as
general director (architektonountos) and setting out problems, and the
mathematicians investigating them earnestly. In this way the subject of
metrology (metrologia) and the problems concerning (. . . ) " then reached
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 173
their high point for the first time, as E(udo)[x]us12 and his followers trans-
formed the old-fashioned work (a[rch]aismon) o(f Hip)po(cra)tes.I3 Geome-
try, too, made great progress,- for analysis and the [lemma] concerning
diohsmoi were created, and in general the subject of geometry was advanced
greatly. And (op)t(ic)s and mechanics were not at all ignored.
There is much to say about this passage, but for the moment I wish
only to consider the subjects that it associates with Plato. The term
metrologia occurs nowhere else in surviving Greek literature. Its
best translation is "theory of measurement," but it is not clear what
such a theory might be.1* Eudoxus's best-known work in pure mathe-
matics concerns the theory of ratios and the measurement of areas
and volumes by indirect procedures (Euclid, Elements, books V and
XII); it is particularly characterized by the logical scrupulousness of
its methods. If readers of the Republic are not surprised to be told
that geometry advanced under Plato's direction, they may be by the
reference to optics (conjectural) and mechanics (unquestionable).
Some may wish to resort to conjectures of their own to explain away
this reference, but, as I have indicated, it seems more reasonable to
accept as a fact that Plato's Academy was not nearly as "Platonic" as
the institution of higher education of the Republic.
The Philodemus passage speaks of Plato's directorship of mathe-
matics, of his setting out problems that the mathematicians eagerly
investigated with great success. There are two well-known anec-
dotes relating to this aspect of Plato's activity. The first concerns
the so-called duplication of the cube, the construction of a cube
twice the volume of a given one.1* According to ancient stories
interest in this problem was stimulated by a Delian appeal to Plato
to help them to appease the god Apollo who had commanded them
to double the size of an altar. According to another story Plato
reproached Eudoxus, Ajxhytas, and Menaechmus for reducing the
duplication problem to mechanical constructions, thereby destroy-
ing the goodness of geometry, "turning it back to sensible things
instead of rising upwards to grasp eternal and incorporeal images"
(Plutarch, Quaestiones Convivales ["Table-Talk"], 7i8e-f). This is
good "Platonism," quite in keeping with the Republic. Unfortu-
nately the solution of the duplication problem ascribed to Plato is
more mechanical than the ones he is said to have censured, in the
sense that it involves the construction of an instrument. Of course,
Cambridge Companions Online © Cambridge University Press, 2006
174 T H E CAMBRIDGE COMPANION TO PLATO
we have the option of rejecting the ascription to Plato, but that
ascription is harder to explain than the story of his reproach of the
other solutions.
The second example of Plato's setting out problems relates to the
anomalous motions of the planets by comparison with the sun or
moon.16 The sun and moon appear to make a uniform daily trip
through the heavens from east to west and a uniform yearly or
monthly trip from west to east. The planets make the same uniform
daily east-west trip, but their west-east trips involve striking
anomalies including periods of apparent motion from east to west.
In his commentary on Aristotle's On the Heavens Simplicius (sixth
century A.D.) remarks on the problem of "saving" these anomalous
motions, that is, producing an explanation of them:
In order to save these many motions in each case, some assume eccentrics
[circular orbits with centers other than the earth] and epicycles [circles with
centers on the circumference of revolving circles], while others hypothesize
so-called counteractive homocentries.17 In the true account the planets do
not stop or retrogress nor is there any increase or decrease in their speeds,
even if they appear to move in such ways; nor are hypotheses that they are
this way introduced, but the heavenly motions are shown to be simple and
circular and uniform and ordered from the evidence of their own substance.
For since it is not possible for a faculty restricted to appearances (phantasia)
to grasp accurately how the planets are disposed and since the consequences
drawn by such a faculty are not the truth, it was asked that one try to
discover how the apparent motions of the planets could be saved by uni-
form, ordered, and circular motions. And, as Eudemus [an associate of Aris-
totle] reports in the second book of his history of astronomy - and so does
Sosigenes [second century A.D.], who is relying on Eudemus - , Eudoxus of
Cnidus is said to be the first Greek to have concerned himself with such
hypotheses; according to Sosigenes, this problem was made up by Plato for
those who concerned themselves with these subjects: by hypothesizing
what uniform and ordered motions is it possible to save the appearances
relating to planetary motions.
(Simplicius, Commentary on Aristotle's "On the Heavens," 488.7-24)
Astronomy is included in the Republic curriculum, but, as we will
see, Socrates' description of it is not at first sight reconcilable with
Plato's reported interest in "saving the phenomena." Again there is
an apparent contrast between the practice of a science and Plato's
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 175
attempt in the Republic to incorporate the science into the educa-
tion of rulers in an ideal state.
However, the point I wish to emphasize now is the evidence that
Plato did play some kind of role as general mathematical director,
posing problems to the mathematicians of his time, sometimes with
striking results. We need not suppose that the work associated with
Plato's inspiration was all done at the Academy, and, in the case of
Eudoxus, there is good reason to suppose it was not. Nor need we
think that Plato's role as director precluded his applying his own
talents to the solution of scientific problems. However, there is no
compelling evidence that Plato showed any great success in this
arena, and many of the mathematical and scientific passages in his
writing are fraught with impenetrable obscurity. It is best, then, to
think of Plato as a source of challenge and inspiration to mathemati-
cians and not as a mathematician of real significance.18
II. MATHEMATICAL METHOD! ANALYSIS,
SYNTHESIS, DIORISMOI, AND LEMMAS
In addition to referring to branches of mathematics, the Philodemus
passage mentions "analysis and the lemma concerning diorismoi."
The notions of analysis and of a diorismos are treated in a somewhat
confusing way in Greek discussions, ̂ although the fundamental
ideas are not difficult. My treatment will be somewhat simplified.
Analysis can be thought of as the process of looking for the proof of an
assertion P by searching for propositions that imply P, propositions
that imply those, and so on until one reaches propositions already
established; in synthesis one simply writes down the proof discov-
ered by analysis, that is, one goes through the steps of analysis in
reverse order. In the most common case one focuses on a single estab-
lished proposition Q which (conjoined with propositions Qlf . . . Qn
taken as given) implies P, that is, is a sufficient condition for the truth
of P-, it might happen that P also implies Q in which case Q will also
be a necessary condition for the truth of P.
A diorismos is usually explained as the determination of the neces-
sary and sufficient conditions for the solution of a problem or the
truth of a proposition. The standard example is provided by proposi-
tion 22 of Book I of Euclid's Elements:
Cambridge Companions Online © Cambridge University Press, 2006
I76 THE CAMBRIDGE COMPANION TO PLATO
I.22 Out of three straight lines which are equal to three
given ones to construct a triangle; thus it is necessary that
two of the straight lines taken together in any manner
should be greater than the third.
Here the second sentence, the diorismos, states the necessary and
sufficient condition that a triangle be constructive out of three
given lines. Euclid, however, formulates it as a necessary condition
and shows (by carrying out the construction) that it is sufficient.20
He has already proved that the condition is necessary in proposition
I.20:
1.20 In any triangle two sides taken together in any
manner are greater than the remaining one.
In his Commentary on the First Book of Euclid's Elements,
Proclus explains what a lemma is:
The term "lemma" is frequently predicated of any premiss assumed in
establishing something else, as when people say they have made a proof
from so and so many lemmas. But in geometry a lemma is specifically a
premiss which needs verification (pistis). Whenever in a construction or
proof we assume something which has not been shown but needs an ac-
count {logos), we call the assumption a lemma because we consider it worth
investigating although doubtful in itself; we distinguish it from a postulate
or axiom because it is provable whereas we assume them directly without
proof to verify other things. The best thing for finding lemmas is mental
address. . . . Nevertheless, methods have been transmitted. The best is reduc-
tion of what is sought to an agreed-upon principle by analysis, a method
they say Plato transmitted to Leodamas; from it Leodamas is said to have
become the discoverer of much in geometry.
(Commentary on Euclid, 211.1 -23)
Proclus mentions Leodamas21 in an account of the history of mathe-
matics before Euclid, particularly the history of geometry:
Plato made geometry and the rest of mathematics undergo great progress
because of his earnestness concerning them, which is evident from the
density of mathematical considerations (logoi) in his writings22 and from his
everywhere awakening in adherents of philosophy an admiration for mathe-
matics. Also alive at this time were Leodamas of Thasos, Archytas of
Tarentum, and Theaetetus of Athens. . . . Neoclides and his pupil Leon were
younger than Leodamas, and they added discoveries to those of their prede-
cessors, so that Leon both composed an Elements which was superior in the
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 177
number and the usefulness of its results and discovered diohsmoi, [which
indicate] when a problem under consideration can be solved and when it
cannot. (Commentary on Euclid, 66.8-67.1)23
Although Philodemus's phrase "lemma concerning diorismoi" is
hardly transparent, it seems to me quite likely that it has no more
specific meaning than the term "analysis/' and that the Philodemus
passage ascribes to the time of Plato a concern with the search for
lemmas and diorismoi, that is, for propositions sufficient (and per-
haps necessary) for the proof of other theorems and for conditions
under which a problem can be solved (or a theorem proved). Clearly,
despite the variety of terms, we are dealing with one central piece of
methodology. The search for premises (analysis) needed to establish
a proposition or solve a construction problem may lead back to
established propositions or constructions (successful analysis), or to
a lemma in need of proof, or to a restriction on the proposition or
construction to conditions under which it can be proved or carried
out (diorismos). Plato himself is credited with transmitting this
methodology to others.2* I shall not be further concerned with this
aspect of Plato's activity, but rather with certain crucial passages
that show the influence of these mathematical methods and con-
cepts on Plato's own methodological thinking.
III. INVESTIGATION FROM A HYPOTHESIS IN THE
MENO
Plato does not use the words "lemma," "diorismos," "analysis," or
"synthesis" in their technical sense, but in the Meno, he invokes as
a procedural precedent a mathematical practice of setting out the
conditions under which a problem can be solved. Meno asks Socra-
tes to tell him whether virtue can be taught, and Socrates asks to be
able to consider the question "from a hypothesis."
What I mean by "from a hypothesis" is like the way in which the geometers
often consider some question someone asks them, for example, whether it
is possible for this area to be inscribed in this circle as a triangle. Someone
might say, "I don't yet know whether this is such that it can be inscribed,
but I think I have a certain hypothesis, as it were, which is useful for the
question, as follows: if this area is such that, when one places it alongside
its given line, it falls short by a figure similar to the one that was placed
Cambridge Companions Online © Cambridge University Press, 2006
I78 THE CAMBRIDGE COMPANION TO PLATO
alongside, I think one result will follow, and another, on the other hand, if
this cannot happen to it. Making a hypothesis, then, I am willing to tell you
the result concerning the inscribing of it in a circle, whether it is possible or
not./; [Meno 86
Socrates here seems to be describing a situation in which a geometer
is considering the problem that Euclid would formulate as
Problem. To inscribe a triangle of a given area in a given
circle.
Socrates' geometer "solves" this problem by giving a condition that
the area must satisfy. Euclid would add this condition to his state-
ment of the problem as a diorismos:
Diorismos. Thus it is necessary that, "if one places the
area alongside its given line, it falls short by a figure
similar to the one that was placed alongside."
Clearly for this diorismos to be effective one will need to know (or
assume) a theorem to the effect that
Theorem. If the area of a triangle inscribed in a circle is
"placed alongside its given line, it falls short by a figure
similar to the one that was placed alongside."
Socrates' presentation of the geometric example does not make clear
whether he takes the diorismos or the theorem to be the hypothesis
on which the problem depends. In fact, of course, it depends on both:
To solve the problem one needs to impose the condition given by the
diorismos and rely on the theorem. When Socrates returns to the
topic of virtue, he says.
Similarly then concerning virtue, since we don't know either what it is or
what sort of thing it is, let's make a hypothesis and consider whether it is
teachable or not, as follows: what sort of thing among those connected with
the soul would virtue be to make it teachable or not teachable? First, if it is
different from or like knowledge, is it teachable or not? . . . Or is this at least
clear to everyone, that a person isn't taught anything other than knowledge?
But if virtue is some sort of knowledge, it's clear that it will be teachable.
Then we've quickly finished with this point: if virtue is of one sort it's
teachable, and, if of another, not. (Meno 8yb-87c)
In this application of the hypothetical method Socrates does not
describe a diorismos, but performs what I have called an analysis,
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 179
that is, he reduces the question of establishing that virtue is teach-
able to the claim that virtue is knowledge if and only if it is teach-
able, or at least:
Hypothesis-theorem. If virtue is knowledge, then it is
teachable.
But corresponding to the need for a diorismos in the case of the
geometric example, the hypothesis-theorem is of use only if one can
establish
Hypothesis-lemma. Virtue is knowledge.
There has been scholarly disagreement as to which of these two
hypotheses Socrates considers to be the hypothesis to which he has
reduced the question of teachability. The most explicit texts (89c-d)
suggest the hypothesis-theorem, and this is what one would expect
in terms of the model of geometric analysis. But, of course, the
hypothesis-lemma is also an assumption, and needs to be estab-
lished in order to show (using the hypothesis-theorem) that virtue is
teachable. And Socrates proceeds to establish it by using the further
hypothesis that virtue is good (87C-89a; Socrates refers to "virtue is
good" as a hypothesis at 87d). It is not clear whether this new hy-
pothesis is conceived as a "theorem" or as a "lemma" still needing
justification. Socrates speaks of it as holding (menein, 89d) and it is
maintained to the end of the Meno, as is the hypothesis-theorem. To
this extent the Meno involves an adaptation of the method of analy-
sis to reduce the teachability of virtue to two hypotheses-theorems.
However, there can be no question of a perfect fit with successful
mathematical analysis since the dialogue ends with Socrates arguing
against both the hypothesis-lemma and the teachability of virtue
(89c//).
The absence of a perfect fit is, I think, a reflection of a practical
difference between mathematics and philosophy. When one looks at
mathematics one cannot help but be impressed by its success, at the
apparently definitive way in which it solves open questions and
resolves disputes. This perspective on mathematics is reflected in a
Greek tendency to think of geometric analysis as successful analy-
sis, as a method of finding rather than a method of searching. It may
also explain why, in the Meno, no attempt is made to relate the
subsequent refutation of the claim that virtue is knowledge to the
Cambridge Companions Online © Cambridge University Press, 2006
I8O THE CAMBRIDGE COMPANION TO PLATO
mathematician's investigation from a hypothesis. However, in phi-
losophy, "analysis" and the discovery of lemmas is much less likely
to produce a definitive answer to a question,- for, as in the Meno, a
lemma will frequently be found questionable. If it is relatively clear
that in the Meno a philosophical hypothesis is a "theorem/' it will
become clear that Plato comes to apply the word "hypothesis" to
lemmas that are thought of as tentative and subject to investigation.
Indeed, one might say that Plato's development of the hypothetical
method involves an attempt to unite the generally smooth working
of mathematics with the rough-and-tumble of the Socratic examina-
tion of doctrines.
The absence of a perfect fit between mathematical method and
Plato's adaptation of it might not present serious obstacles to inter-
pretation if Plato himself were clear about the discrepancies. But the
Meno is a good example of Plato's tendency to ignore differences.
This tendency and the looseness of fit has led some interpreters to
downplay the connection between Platonic methodology and mathe-
matics. But the historical evidence of a connection is too strong to
make this approach viable. Our task should be to make as much of
the connection as we can without losing sight of the imperfect fit.
This task is not made easier by Plato's general unwillingness to use
precise vocabulary. Where Plato uses one word, "hypothesis," we
find it advisable to distinguish among theorems, lemmas, and
diohsmoi. As I proceed in this chapter I will point to further exam-
ples of problematic vocabulary and looseness of fit. I do not intend
thereby to disparage Plato's accomplishments, but simply to im-
prove our understanding of Plato's adaptation of mathematical
method.
IV. THE METHOD OF HYPOTHESIS IN THE PHAEDO
In the Phaedo, starting at 95e7,26 Socrates gives a general description
of a philosophical method that seems to be based on mathematical
analysis and synthesis, but goes well beyond them in important
ways. In the passage, Socrates describes, as a preliminary to an argu-
ment for the immortality of the soul, a method he has worked out
for determining "the explanation (aitia) of each thing, why it comes-
to-be, why it ceases-to-be, why it is" (96a9-io):
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 181
On each occasion I hypothesize the thing [logos) which I judge to be strong-
est, and I lay down as true whatever seems to me to agree (sumphonein)
with it, whether the subject is cause or anything else, and I lay down as false
whatever does not seem to me to agree with it. (Phd. iooa3~7)
Socrates' recommendation here should be understood to be relativ-
ized to a subject of inquiry. He suggests that on any subject of inquiry
one should take as hypothesis the relevant belief in which one has
most confidence, add further relevant ideas that (in some sense) agree
with the hypothesis and reject relevant ideas that do not agree.2? He
illustrates what he has in mind in the case of the question of immor-
tality (or the explanation of each thing) by hypothesizing that each
Form is something (apparently the assumption that the Forms exist)
and adding the belief that each thing is or comes to be (what it is) by
participating in the appropriate Form. The example (and also the later
example at iosb-c) indicates that the method espoused is to answer a
given question by building up a consistent theory applicable to an
issue through the addition of compatible beliefs. In later ancient
logical texts, the word "agreement" used by Socrates can mean sim-
ple logical consistency. Here it includes the notion of logical consis-
tency but is presumably stronger; many explanations of being and
coming to be are consistent with the existence of Forms, but explana-
tion by participation is, in some reasonably clear but not easily expli-
cable sense, suitable for the believer in Forms.
At IOIC Socrates says that, confronted by alternative explanations
of coming to be, the person following his method should leave them
for others and "hang on to the safety of the hypothesis." Prior to this
Socrates has only referred to the original assumption as a hypothe-
sis, but what he is referring to now must include the additional
explanation of being and coming to be. I suspect he means the whole
theory that has been built up by accreting harmonious assump-
tions.28 Socrates now turns to the status of the "hypothesis":
But if anyone were to grab on to the hypothesis itself, you would say good-
bye to him and not answer until you had investigated whether [it seemed] to
you that the things which came from it [ta hormethenta) agreed or disagreed
with one another. And when it was necessary for you to give an account of
the hypothesis itself, you would give it in the same way, hypothesizing
another hypothesis, whichever among higher hypotheses seemed best, until
you came to something sufficient. [Phd. ioid3-ei)
Cambridge Companions Online © Cambridge University Press, 2006
182 THE CAMBRIDGE COMPANION TO PLATO
It is easy enough to see how the method of analysis can be related to
this last sentence. Forced to justify an assumption one has made, one
finds an assumption that would justify it, and, if demands for justifica-
tion continue, one proceeds in the same way until one finds an as-
sumption not needing justification. Socrates does not indicate what
conditions a hypothesis would have to fulfill to be "sufficient/7 but
obviously mathematics itself would provide examples of successful
analysis in which sufficiency was attained, or was at least thought to
be. However, even at this point it is clear that Plato's philosophical
interest is widening the gap between his hypothetical method and
geometrical analysis. For, although the ideal of an ultimately satisfac-
tory justification remains in play, the earlier hypotheses, however
strong or good they may be, are not established "theorems," but provi-
sional lemmas subject to test and possibly still needing justification.
Moreover, in successful analysis the hypothesis-theorem becomes
a starting point from which the proposition under consideration is
deduced in the synthesis. However, we have already seen that in his
initial description Socrates treats the original hypothesis as a basis
for the acceptance of additional ideas judged harmonious with it and
the rejection of those judged inharmonious. Socrates' only illustra-
tion of the rejection of beliefs is the rejection of explanations of
being and coming to be other than participation in the appropriate
Form (iooc-ioid), that is to say, the rejection of beliefs blatantly
incompatible with beliefs already accepted as harmonious with the
original hypothesis. In the second passage Socrates speaks of check-
ing whether the things that come after the hypothesis are harmo-
nious with one another.^ He seems to have in mind the sort of
testing of people's views that he practices in other dialogues such as
the Euthyphro. This procedure does not seem to be important in
mathematics, and it is not easy to see how it can be fit into the
method Socrates is introducing. Perhaps what he has in mind can be
understood in terms of his example. The problem Socrates confronts
is the explanation of things in our world, why they come to be, cease
to be, and are. To attain such an explanation Socrates posits that
there are Forms and adds the apparently harmonious assumption
that things come to be and are what they are by participating in the
Forms. (We are not told how Socrates would use his theory to ex-
plain the ceasing to be of things.) To investigate fully the adequacy
of this theory one would investigate the "things which come from
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 183
it/' not only consequences but also other assumptions apparently
harmonious with it, for example, about the character and relation of
Forms, the nature of coming to be, and so on. Ultimately this investi-
gation will involve tests for consistency, but the tests will be applied
to a rich set of apparently harmonious beliefs about the world. Given
the apparent harmoniousness of this rich set, one may still find it
insufficient for some reason - for example, one may doubt that there
are Forms. The task of the defender of Forms is to search for a
"higher" hypothesis harmonious with the developed set. Socrates
says nothing about what such a higher hypothesis might be, but the
present passage and others in the Phaedo (e.g., 107b) suggest at the
very least that he does not consider the demand for a hypothesis
higher than that of the Forms inappropriate.
In analysis the attack on a problem is the search among proposi-
tions until one finds a hypothesis-theorem from which a solution to
the problem can be deduced (synthesis). In the method of the Phaedo
the attack on a problem involves making a hypothesis judged to be
the strongest available and building up through the addition of har-
monious ideas a theory adequate to solve the problem. The closest
one gets to a hypothesis-theorem is a satisfactory hypothesis. And
the closest one gets to a deduction is the harmonious expansion of a
hypothesis. From our point of view there is a considerable difference
between deduction and harmonious expansion, between a conse-
quence of and a plausible addition to a theory. I suspect that Plato
did not assign this difference fundamental importance. By this I do
not mean that Plato would have been willing to overlook the substi-
tution of plausibility considerations for proof in mathematics. I only
mean that for philosophical purposes he was willing to class to-
gether deduction and less formal methods of serious argumentation.
The model of mathematical method remains, but it has been ex-
panded in its adaptation to philosophy. This expansion of the notion
of synthesis to include harmonious elaboration will be important in
the next section.
V. MATHEMATICS AND DIALECTIC IN REPUBLIC VI
AND VII
Mathematics comes to the fore in the Republic in the famous
divided-line passage at the end of Book VI. I shall pass over many of
Cambridge Companions Online © Cambridge University Press, 2006
184 THE CAMBRIDGE COMPANION TO PLATO
the interpretive issues associated with this passage to discuss a few
of concern to me.3° The divisions of the line apparently include all
things, the sensible world consisting of objects and their images and
presided over by the sun, and the intelligible world presided over by
the idea of the Good. Socrates is not clear about the relationship
between these two worlds. The line and the comparison between the
sun and the Good suggest a strong division, but the cave allegory and
the mathematical curriculum suggest a considerable continuity
which is likely to have been an important feature of Plato's general
outlook. Socrates divides the intelligible world by reference to two
mental conditions (pathemata) or what we might call modes of cog-
nition^1 He identifies one of these [noesis)^2 with the use of dialectic
and illustrates the other (dianoia) by reference to mathematics, "ge-
ometry and its sister arts" (5iibi-2). There is some question about
whether the illustration exhausts the content of the relevant section
or whether there are nonmathematical instances of dianoia. For my
purposes it is sufficient that the section includes mathematics,
which totally dominates Socrates' discussion.
Socrates makes two contrasts between dianoia and noesis:
1. Dianoia is compelled to study its objects by proceeding from
a hypothesis toward an ending, but noesis studies its objects
by proceeding from a hypothesis to an unhypothetical begin-
ning (principle).
2. Dianoia uses sensible things as images, but noesis uses no
images and proceeds through Forms in a systematic way.
There is little doubt that Socrates has in mind here two features of
mathematics that we associate particularly with geometry: the use
of diagrams in arguments and the derivation of conclusions from
initial assumptions (synthesis). The first point to notice about Socra-
tes' explication of the mathematicians' use of diagrams is that, ac-
cording to him, although mathematicians make their arguments
(logoi) about images,
they are not thinking (dianoein) about them, but about the things which
resemble them; they make their arguments for the sake of [heneka) the
square itself and the diagonal itself [i.e., the Forms of square and diagonal]
and not for the sake of the one they draw. . . .33 The things which they mold
and draw . . . they use as images seeking to apprehend things which cannot
be apprehended except by dianoia. (5 iod6-51 iai)
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 185
A modern philosopher of mathematics might say that although ge-
ometers use drawn figures in their arguments, they are not arguing
about figures (since figures satisfy their hypotheses only approxi-
mately) but about something else (which does satisfy the hypotheses
exactly). Socrates says instead that geometers argue about visible
figures, but do so for the sake of, that is, in order to apprehend,
something else, namely (intelligible but not perceptible) mathemati-
cal Forms. It is difficult to be certain how far to press Socrates7
vocabulary here ("arguing about/7 "thinking about,77 "arguing for
the sake of77), but it certainly looks as though he thinks of mathemat-
ics as an attempt to understand the intelligible world by reasoning
about sensible things rather than (as we might suppose) as an at-
tempt to reason about the intelligible world using sensible things.
The importance of this contrast may be made clearer by reference
to the fact that Socrates twice speaks of the mathematicians being
compelled [anangkadzomai) to use hypotheses, but he never speaks
of them as being compelled to use images. Moreover, what he says at
510b may mean that the use of images compels the soul to inquire
hypothetically. Thus Socrates may mean that the mathematician is
forced to use hypotheses because he is reasoning about sensible
things in an attempt to understand intelligible ones. In another pas-
sage Socrates speaks similarly of geometers being forced to use the
language of action, although this language is misleading with re-
spect to the intelligible objects for the sake of which they are pursu-
ing geometry:
They speak in a way which is ridiculous and compulsory [anangkaios); for
they are always talking about squaring and applying and adding as if they
were doing things and were developing all their propositions for the sake of ac-
tion; but, in fact, the whole subject is pursued for the sake of understanding.
(Rep. 527a6-bi)
The general picture then is that the mathematician is in the position
of trying to apprehend an intelligible, static world of Forms, but
attempting to do so by arguing about visible things. This mode of
arguing forces the mathematician to speak about acting on the
things and to argue from hypotheses.
It is presumably clear enough why argument about diagrams neces-
sitates talk about activities or operations, but not immediately clear
why the use of diagrams necessitates the making of hypotheses.
Cambridge Companions Online © Cambridge University Press, 2006
186 THE CAMBRIDGE COMPANION TO PLATO
This is made somewhat clearer by looking at what Socrates has to
say about the hypotheses of the mathematicians, who, according to
him,
hypothesize the odd and the even and the figures and the three kinds of
angles [acute, right, obtuse] and related things in the case of each science
(methodos); they hypothesize these things as known, and do not think it
worthwhile to give any account of them to themselves or others, as being
clear to all; beginning from these, they proceed through the remainder and
finish consistently with that for which the investigation was undertaken.
(Rep. 5ioc3-d3)
Here the parallel with Socrates' hypothesizing in the Phaedo is strik-
ing. There the discussion shows that Socrates' initial hypothesis is a
rather elaborate one about Forms, but he formulates the hypothesis
simply as "that a beautiful itself by itself is something, and a good,
and a large, and the rest." Similarly whereas we think of the mathe-
matician as making rather elaborate assumptions about parallel
lines, equality, and the meaning of certain terms, Socrates in the
Republic mentions only "the odd and the even and the figures and
the three kinds of angles and related things in the case of each
science." We do not know enough about the way mathematics was
presented in the early fourth century to judge the ^curacy of Socra-
tes' characterization, but we can see that he is not worried about
details. Nothing he says implies that the mathematician deduces
consequences from assumed propositions as opposed to making argu-
ments on the basis of some assumed background knowledge about
various concepts.34 It may be exactly this latter sort of knowledge
that Socrates thinks must be presupposed if one "argues about"
figures.
The situation becomes more opaque when we turn to dialectic.
Socrates tells us that, whereas the mathematician inquires from
hypotheses and proceeds to a finishing point rather than to a starting
point, the dialectician proceeds from a hypothesis to an "unhypo-
thetical starting point," and proceeds "by means of Forms and
through Forms," not using images. The directional contrast is pre-
sumably related to the analysis-synthesis contrast, but here the
upward movement is assigned to the philosopher and distinguished
from the mathematical procedure by more than its direction. Socra-
tes goes on to assign a downward method to the dialectician as well:
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 187
[Dialectical argument] does not make hypotheses starting points but genu-
ine things laid down^ like footholds and sources of impetus [hormai) in
order to proceed to the unhypothetical, the starting point of everything:
having grasped that and again the things right after it, it moves down to a
finish, using nothing sensible at all, but only Ideas, moving through Ideas
and into Ideas and finishing in them. (Rep. 51 ib5-c2)
The differences between this passage and the methodological de-
scription of the Phaedo need not be as significant as they are some-
times suggested to be. Some of them, notably, the insistence on the
use of images, seem to stem from the fact that Socrates is concerned
specifically with mathematics in the Republic, whereas in the
Phaedo he is making a general methodological point, albeit one
based on mathematical method. In the Republic there is no clear
difference between the upward and downward movements of dialec-
tic except for their direction. It is natural for us to assume that the
downward method of dialectic is the same as the downward method
of mathematics and to assimilate the former to what we know of the
latter, namely that it is propositional deduction (synthesis). It is then
natural to suppose as well that the upward method is propositional
deduction or something reasonably like it. But in the Phaedo the
upward "method" is simply a matter of making "higher" hypothe-
ses, and the downward method seems to include the making of
additional hypotheses harmonious with a given one. I see no good
reason to think Plato is being more restrictive in the Republic. For
Socrates' examples of the mathematicians' hypotheses in the Repub-
lic are of a piece with his taking the Forms as his sample hypothesis
in the Phaedo. The most striking difference between the two pas-
sages is perhaps the contrast between Socrates' invocation of the
unhypothetical principle of all things in the Republic and his rather
bland reference to something satisfactory in the Phaedo. However,
the latter reference is sufficiently bland to accommodate what is
said in the Republic.*6
We have seen that in the Phaedo and the Meno the presentation of
hypothetical reasoning gets connected with ideas of refutation
which seem to have less of a role to play in mathematics than in
philosophy. In the divided-line passage of the Republic there is no
sense that either the mathematician or the dialectician ever finds a
hypothesis unsatisfactory. Mathematicians end up consistently with
Cambridge Companions Online © Cambridge University Press, 2006
188 THE CAMBRIDGE COMPANION TO PLATO
their hypotheses, and dialecticans move up from their hypotheses
and back down, presumably to the same "hypotheses." Later, in
Book VII, Socrates indicates that dialectic does involve refuting argu-
ments, but he makes clear that successful dialecticans will be able
to defend their position against all attempts at refutation (534b-d).
However, just before this passage Socrates describes dialectic as "do-
ing away with or destroying (anairein) hypotheses" (533C8), and he
speaks rather denigratingly about mathematics:
Geometry and the studies associated with i t . . . do apprehend something of
being, but. . . they are dreaming about it. They cannot have a waking vision
of it as long as they use hypotheses and keep them fixed, unable to give an
account of them. For when the starting point is not known and the finishing
point and what comes in between are woven together out of what is not
known, there is no way that such a consistency will ever become knowledge.
(Rep. 533b-c)
Some later Platonists used this passage to belittle mathematics, ̂
and modern scholars have debated what Socrates could have in mind
by destroying the hypotheses of mathematics. I think it is fair to say
that there is now consensus that the only destruction Socrates has in
mind is the destruction of the hypothetical character of mathemati-
cal hypotheses through subsumption under an unhypothetical start-
ing point. Equally when he denies that ordinary mathematics is
knowledge he does not mean that it is false, but only that it lacks the
requisite foundation to count as known. Insofar as mathematics
provides dialectic with its hypotheses, dialectic starts with truths
that it will test but not refute.
The divided-line passage, then, emphasizes the following features
of mathematics:
1. Reasoning about sensible objects, figures, for the sake of,
that is, in order to understand, intelligible ones.
2. The laying down of hypotheses, presented as the assumption
of certain objects (the odd and even, the figures, the kinds of
angle), but in fact involving assumptions about the nature of
these objects and the ways they can be manipulated.
3. The downward development of these hypotheses, including,
but not necessarily restricted to, deduction.
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 189
Plato sees the first of these as the cause of the second and, presum-
ably, of the third: Because mathematicians reason about sensible
things, they must make hypotheses and they must move downward
from them, as they must speak about acting on sensibles. It is reason-
able to suppose that Plato's description of mathematics as depending
on hypotheses that the mathematician never attempts to justify is
an accurate description of the mathematics of his time. But why
does Plato think of the downward direction as a necessary feature of
mathematics? After all, mathematicians do perform analyses on
propositions below the level of their ultimate hypotheses. Why
couldn't they attempt to do the same thing on those hypotheses?
The answer here may be a simple matter of definition: For Plato
such an upward movement would take one outside the domain of
mathematics. But there may be more involved. For Plato, to justify
the hypotheses of mathematics will be to answer questions like What
is a figure? and What is an angle? Answering this kind of question
satisfactorily requires one to move from arguing about sensibles to
arguing about Forms. Obviously the same person can switch from
mathematically developing hypotheses to asking Platonic/Socratic
questions about the hypotheses, but this change is a change from
arguing about sensibles to arguing about intelligibles, that is, a
change from mathematics to dialectic. Socrates makes something
like this point at 523a//when he is describing the mathematical cur-
riculum. There Socrates is no longer interested in the downward
aspects of mathematics, but in its power to turn the soul's attention
upward from sensibles to intelligibles. To argue that arithmetic, prop-
erly pursued, has this power, he distinguishes between aspects of
things that seem contradictory to the senses and those that do not.
Supposedly, seeing a finger does not lead or compel (anagkadzein)*8
an ordinary person to ask what a finger is, but seeing that one finger is
larger than a second but smaller than a third does lead or compel a
person to ask what largeness is, that is, to ask a question about Forms.
Glaucon volunteers that unity falls in the second category because we
see the same thing simultaneously as one and infinitely many. 39 Soc-
rates adds that the same will be true of all number.
Socrates' way of speaking about the abandonment of sensibles in
dialectical argument was taken by the Neoplatonists to involve refer-
ence to a mysterious "nondiscursive" thought, which, among other
Cambridge Companions Online © Cambridge University Press, 2006
I9O THE CAMBRIDGE COMPANION TO PLATO
things, violates Aristotle's dictum (On the Soul III.7.43 iai6-i7) that
"the soul never thinks (noein) without an image [phantasma)."*°
Nothing in the Republic seems to me to justify this Neo-Platonist
reading, although one cannot preclude the possibility that Plato had
something of the sort in mind. However, I am inclined to think that
when Socrates describes dialectic as restricted to Forms, he is not
talking about what goes on in the consciousness of a dialectician at
work, but is simply developing the contrast between dialecticians
and mathematicians. Mathematicians reason about sensibles for the
sake of intelligibles; they use sensibles. Dialecticians reason about
intelligibles for the sake of intelligibles; whether or not images occur
in their minds or they refer to sensible things, they do not reason
about sensible things, they do not use them.
It becomes clear from Book VII of the Republic that the unhypo-
thetical first principle of all things is the Idea of the Good (532a//). It
also becomes reasonably clear (534b-d) that its unhypothetical char-
acter depends upon the fact that people who apprehend it fully can
defend themselves when someone tries to "grab onto" their hypothe-
sis. That is to say, for a principle to be unhypothetical is for it to
require no higher hypothesis for justification, that is, to be capable of
withstanding argumentative assault on its own. The notion of being
the first principle of all things seems to me impossible to construe
sensibly in terms of a strictly deductive model of the downward
path. The Good becomes such a hypothesis only by the addition of
other hypotheses harmonious with it. From a modern logical point
of view additional hypotheses are additional hypotheses, but for
Plato the hypothesis of the Good is the condition that restricts these
further hypotheses and so is higher than they. I do not believe that
one can make ultimately satisfactory logical sense of Plato's posi-
tion here,*1 but appreciating it seems to me a precondition of under-
standing the implication of the Republic that mathematical hypothe-
ses are subsidiary to the Idea of the Good. Plato is not suggesting
that mathematical hypotheses can be deduced from assumptions
about the Good/2 but "only" that they will fit harmoniously into a
fully developed theory anchored in the Good. For Plato it is good
that a sum of even numbers is even and that the planets move
uniformly in circular orbits. We may try to explain the latter of these
beliefs by reference to a teleological conception of the world and the
former by reference to the beauty and goodness of mathematical
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 191
truth. But Plato is unlikely to have distinguished the two features
clearly; for him, as far as we can tell, it is for the best that even
numbers add up to an even number and it is true that the world
system is a beautiful thing.
I have already mentioned that when Socrates expounds his mathe-
matical curriculum, he is interested almost exclusively in the power
of mathematics to draw the soul away from the sensible to the
intelligible. After his argument that arithmetic has this power, he
adds a second consideration to show that the arithmetician really is
concerned with intelligibles:
For I imagine that you know that those who are formidable in these
matters, if someone tries to divide the one itself by argument, laugh and will
not accept it, but if you chop it up, they multiply, taking care that the one
will never appear not one but many parts.43. . . Glaucon, what do you think
would happen if someone were to ask them, "What kind of numbers are you
discussing in which the one is the sort you deem it to be: each equal to every
other, not differing even slightly, and containing no part?" What do you
think they would answer?
I suppose they would say that they are talking about things which can
only be thought and cannot be treated in any other way.
{Rep.
Here Socrates indicates that the way arithmeticians talk commits
them to an intelligible world. He does not draw the distinction made
in the divided-line passage between what the arithmetician reasons
about and what he reasons for the sake of. Indeed, he says explicitly
that the arithmetician discusses intelligible units rather than saying
(as I think he should for consistency) that the arithmetician dis-
cusses sensibles for the sake of intelligibles. The reason for this
discrepancy is that Plato wants to use the way arithmeticians talk as
an indication that their concern is with the intelligible. In his imme-
diately succeeding treatment of geometry and his subsequent treat-
ment of astronomy and harmonics, he wants to stress that mathe-
matical practice is misleading. Geometers, he says, speak as if they
were doing something, but their knowledge does not concern chang-
ing things but what always is. Socrates does not argue for this conclu-
sion; 44 it is volunteered by Glaucon, who, we may suppose, has been
carried along by Socrates' treatment of mathematics since the
divided-line passage. But even Glaucon does not seem totally pre-
Cambridge Companions Online © Cambridge University Press, 2006
192 THE CAMBRIDGE COMPANION TO PLATO
pared for Socrates' discussion of astronomy,** a discussion that must
be interpreted in the light of the specific role of drawing upward that
Socrates assigns to mathematics.
Geometers reason about sensible figures for the sake of intelligible
things, the square itself, the diagonal itself, and so forth. In astronomy
what is observed in the heavens takes the place of sensible figures.
What corresponds to the mathematical Forms is, in one of Plato's
more opaque phrases, "the true things, the movements by which are
moved real speed and real slowness in true number both by all true
figure and in relation to one another, and by which they move the
things in them" (529di-5). Socrates goes on to say that no one should
expect to find the truth about ratios in the visible heavens or think
that the periods of the various heavenly bodies will remain constant
through time; that is to say, as sensible objects the heavens cannot
perfectly embody scientific laws. Socrates concludes,
We will pursue astronomy as we also pursue geometry, using problems, but
we will leave things in heaven alone, if, sharing in genuine astronomy, we
are going to change what is naturally intelligent in the soul from uselessness
to use. (Rep. 53ob6-ci)
Socrates' treatment of astronomy has caused Plato's admirers a good
deal of discomfort. Various expedients have been proposed, but none
can eliminate the fact that "true" astronomy does not concern the
visible heavens any more than arithmetic and geometry concern sen-
sible objects. Our custom is to distinguish applied and pure sciences.
Even if some of us do not accept Plato's position on the pure sciences,
most of us will at least acknowledge the force of the idea that arithme-
tic and geometry, for example, deal with imperceptible realities. But
the suggestion that true astronomy deals with such realities seems
bizarre. Can anything be said to ameliorate the difficulty?
First of all, we can be reasonably certain that Plato did not himself
ignore the importance of giving some account of the apparent mo-
tions of the heavenly bodies. For, if we can believe Simplicius, he set
astronomers the task of accounting for these apparent motions
through a hypothesis of uniform circular motions. Plato himself
sketches the beginnings of such an account in the Timaeus (36b-d),
and he reaffirms the importance of understanding planetary motion
at 822a of the LawsA6 Simplicius refers to the task as a problem, and
in the Philodemus passage Plato is also described as setting prob-
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 193
lems. It would seem likely, then, that in the Republic Socrates has in
mind for astronomy the attempt to resolve outstanding questions by
reduction to better-understood things. But whereas in geometry
Plato can cite the (true) hypotheses that the geometer does not ques-
tion, he has no analogue in astronomy. In other words in geometry
successful reductions or analyses of problems move to hypotheses-
theorems, but in astronomy there are no such theorems; the task of
analysis is to move to hypotheses-lemmas - in the case of Plato's
astronomical problem, uniform circular motions. Plato does not ad-
dress the question of the status of these lemmas when they are
hypothesized, but I think we can reasonably say that in the down-
ward movement of dialectic they will be established as "theorems"
harmonious with the Idea of the Good.
There remains the question of how Plato understood the relation-
ship between astronomical phenomena and the true astronomer's
hypotheses. If we follow the Republic those hypotheses cannot be
about the phenomena any more than what the geometer apprehends
is truths about sensible things. And when Socrates says that in true
astronomy one should leave things in heaven alone, it is perhaps
natural to take him to be implying that astronomy might be devel-
oped without anyone ever looking at the heavens. But such a view is
so implausible that one is reluctant to ascribe it to anyone. We may
prefer to rely on Socrates' comparison between astronomy and ge-
ometry. Geometers reason about sensible things for the sake of intel-
ligibles,- this means that their truths, like the truths of arithmeti-
cians, are not truths about sensible things. However, geometers
know this last fact, so there is no reason to tell them to leave sensi-
ble things alone in the sense of attending to intelligibles. However,
we need not suppose that Plato would urge geometers to stop using
diagrams in their reasoning.*? By analogy we can say that Plato is
urging astronomers to stop thinking that their subject is sensibles,
but he is not urging them to stop using astronomical appearances as
astronomical appearances. Astronomers can attend to appearances,
argue about them, but they must do so for the sake of, that is, in
order to understand, an intelligible world containing "the true
things, the movements by which are moved real speed and real slow-
ness in true number both by all true figure and in relation to one
another, and by which they move the things in them."
We find this position hard to accept because for us astronomy is
Cambridge Companions Online © Cambridge University Press, 2006
194 T H E CAMBRIDGE COMPANION TO PLATO
about the phenomena, not about an intelligible world. But Socrates,
in the Republic, holds that scientific knowledge is of eternal un-
changing truths, and he does not think that the heavens or anything
sensible is unchanging in a way that permits such knowledge. How-
ever, this does not mean that astronomical truth makes no contribu-
tion to our understanding of the sensible world any more than the
fact that arithmetic and geometry are about the intelligible world
means that they make no contribution to our understanding of the
sensible world.*8 The crucial point is that for Plato such understand-
ing depends upon understanding another, ideal world over which the
Good reigns.
NOTES
1 There is a useful discussion of Plato's Academy in chap. 2 of John
Patrick Lynch, Aristotle's School (Berkeley, 1972). The evidence for
most assertions about Plato and the Academy is very complicated. I try
to indicate clearly when what I say is generally accepted and when it is
more controversial.
2 See Diogenes Laertius, Lives of the Philosophers, IV.2; Olympiodorus,
Commentary on theFirstAlcibiad.es, 140.16-17 Creutzer,- and Olympio-
dorus, Anonymous Prolegomena to the Philosophy of Plato, 5.24-27
Westerink.
3 For two influential examples see Paul Shorey, What Plato Said (Chicago,
I933)/ 3°; a n d F. M. Cornford, "Mathematics and Dialectic in the Repub-
lic VI-VII," Mind 41 (1932): 173-4 (reprinted in Studies in Plato's Meta-
physics, ed. R. E. Allen [London, 1965], 77-8). For criticism see Harold
Cherniss, The Riddle of the Early Academy (Berkeley, 1945), 66-82.
4 Socrates calls these subjects mathemata, a general term for things to be
learned. Because of the influence of the Republic the word came to be
applied to these subjects specifically, and so mathemata became a tech-
nical term that it is customary to translate "mathematics." I shall use
this translation, but it is important to realize that for Plato and other
ancient writers "mathematics" includes subjects we associate with phys-
ics as well as some we associate with pure mathematics.
5 For an introduction to this very complex topic see Konrad Gaiser,
"Plato's Enigmatic Lecture On the Good," Phronesis is (1980): 5-37.
6 See G. C. Field, Plato and his Contemporaries, 3d ed. (London, 1967),
40-5.
7 Menedemus, a pupil of Plato, was almost elected scholarch of the Acad-
emy after the death of Speusippus in 339. See Francois Lasserre, De
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 195
Leodamas de Thasos a Philippe d'Oponte, vol. 2: La scuola di Platone
(Naples, 1987), 93-6 with commentary.
8 Here the Academy is the public area in the outskirts of Athens from
which Plato's Academy took its name. Plato taught in the public area
and established a residence nearby. The two uses of the term "Academy"
give rise to some confusion in our sources.
9 The so-called Academicorum Philosophorum Index Herculanensis. Phil-
odemus preserves excerpts from earlier authors, but the question of
what author he is quoting in our passage is disputed. For discussion see
Konrad Gaiser, Philodemus: Academica (Supplementum Platonicum 1)
(Stuttgart-Bad Cannstatt, 1988), 76-7, 88-91, whose reconstruction
(pp. 152-3) I have largely followed.
10 Letters enclosed in angle brackets (()) correspond to gaps in the papyrus,
and letters within square brackets ([]) to letters that cannot be read with
certainty. I have introduced these niceties only in cases of significance
for my topic.
11 The gap here is about seven letters long, followed by the legible letters
2MOY2. Conjectures: definitions, numbers, ratios, diorismoi, altars,
astronomy, atoms. Gaiser mentions as other possibilities rhythms and
sections.
12 Perhaps the greatest mathematician and astronomer of the fourth cen-
tury, Eudoxus probably spent some time at the Academy, although he
also spent considerable time elsewhere and led a school in Cnidus. Mate-
rials relating to him can be found in Francois Lasserre, Die Fragmente
des Eudoxos von Knidos (Texte und Kommentare IV) (Berlin, 1966).
There is a brief summary of his accomplishments in Charles C. Gil-
lispie, ed., Dictionary of Scientific Biography (New York, 1970-80). The
Dictionary is a generally reliable source of information on the scientific
achievements of most of the Greek mathematicians mentioned in this
paper and in other works on Greek science.
13 If the restoration is correct, the reference is to Hippocrates of Chios, the
earliest person (late fifth century) to whom we can ascribe specific
mathematical accomplishments with confidence. According to Proclus
(A Commentary on the First Book of Euclid's Elements, 66.7-8),
Hippocrates was the first person known to have written a book of Ele-
ments (more than a century before Euclid).
14 Like Gaiser I am inclined to think that it has something to do with the
treatment of common measures and their absence (i.e., incommensura-
bility). However, it might also concern the determination of areas and
volumes.
15 The problem must have been considered before Plato's time since
Hippocrates of Chios is said to have been the first person to realize that
Cambridge Companions Online © Cambridge University Press, 2006
I96 THE CAMBRIDGE COMPANION TO PLATO
the problem of constructing a cube that is double the volume of one
with side of length 1 was solvable by finding x and y such that 1: x :: x : y
:: y : il. For detailed information about the Greek treatment of this
problem see Thomas Heath, A History of Greek Mathematics, vol. 1
(Oxford, 1921), 244-70. Heath describes Plato's alleged solution on pp.
255-8. Another mathematical accomplishment ascribed to Plato in an-
tiquity was a procedure for finding square integers equal to the sum of
two square integers; see ibid., 79-82.
16 The description that follows is simplified. The Greeks classed sun and
moon as planets because, unlike the fixed stars, they had an apparent
west-east motion. Among the many sources one may consult on an-
cient Greek astronomy I mention D. R. Dicks, Early Greek Astronomy
to Aristotle (Ithaca, N.Y., 1970).
17 Simplicius here refers to theories of the kind put forward by Eudoxus.
They involve the view that the sun, moon, and planets are attached to
spheres rotating about the earth as center. To explain the anomalous
motions, Eudoxus postulated additional spheres rotating in other direc-
tions and counteracting the motion of the primary sphere of a heavenly
body.
18 See on this topic the fundamental paper of Harold Cherniss, " Plato as
Mathematician," Review of Metaphysics 4 (1951): 395-426, reprinted in
his Selected Papers, ed. Leonardo Taran (Leiden, 1977).
19 The major source of perplexity (and controversy) stems from descrip-
tions of analysis that represent it as a matter of deducing conclusions
rather than of searching for presuppositions. For discussion see Norman
Gulley, "Greek Geometrical Analysis/7 Phronesis 3 (1958): 1-14.
20 Clearly there might be situations in which one had to settle for suffi-
cient but not necessary conditions or know that certain conditions were
necessary but not be able to prove them sufficient, but the Greeks do not
mention this point in discussing diorismoi.
21 Leodamas is also the addressee of Plato's Eleventh Letter, the contents of
which concern politics. Otherwise we know nothing about him except
what Proclus tells us.
22 There is a quite good list of mathematical passages in Plato with discus-
sion in Attilio Frajese, Platone e la matematica nel mondo antico
(Rome, 1963).
23 The whole of this passage (which extends to 68.6 and can be read in an
English translation by Glenn R. Morrow in Proclus: A Commentary on
the First Book of Euclid's Elements [Princeton, 1970]) is a fundamental
document for interpreting Plato's relationship to the mathematics of his
time. The clear implication of the passage (which probably derives ulti-
mately from Eudemus) is that all mathematical work done in the fourth
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 197
century was done under the influence of Plato and probably at the Acad-
emy. But Plato himself is described only in the passage quoted, where he
is treated as an enthusiast able to inspire others.
24 Clearly the task of "saving the apparent motions of the planets'7 is also a
demand for the analysis of the apparent motions and a reduction of them
to uniform circular motions.
25 My translations from the Meno are very slight revisions of those of R. W.
Sharpies, Plato, Meno (Warminster, 1985). Sharpies briefly discusses the
obscurities of the mathematical example on pp. 158-61.1 shall write as
if the meaning of the example is transparent, that is, I simply reproduce
the example without explanation. My interpretation of the whole pas-
sage is indebted to Ernst Heitsch, "Platons hypothetisches Verfahren im
Menon," Hermes 105 (1977): 257-68.
26 I do not undertake to give a full treatment of the issues raised by this
passage. For thorough discussion see the notes in Plato, Phaedo, trans-
lated with notes by David Gallop (Oxford, 1975).
27 The parallel between Socrates' methodological proposal and Plato's chal-
lenge to the astronomers is suggestive: To save the phenomena the as-
tronomer is urged to hypothesize uniform ciruclar motions and refine
their description until they characterize the phenomena.
28 The phrase used by Socrates is more literally translated "that [part?] of
the hypothesis which is safe," and hence may refer to the explanation by
participation as a "safe" addition to the original hypothesis of Forms.
See Paul Plass, "Socrates' Method of Hypothesis in the Phaedo/' Phrone-
sis 5 (i960): 111-12.
29 At ioie Socrates insists that the hypothesis should not be questioned
until one has tested the things that come from it for agreement. His
separation of the question of the possibility of justifying a hypothetical
theory by reference to a higher one and checking the internal soundness
of a theory is no doubt methodologically sound, but in practice it seems
highly unlikely that people could be restrained from asking about the
doctrine of Forms and participation until the doctrine had been fully
tested for harmoniousness. Moreover (at least from a contemporary
point of view), if the hypothesis were shown to be harmonious and to
give a reasonable account of the coming to be, ceasing to be, and being of
each thing, one might find the question of satisfactoriness relatively
insignificant.
30 The reader may wish to consult chaps. 10 and 11 of Julia Annas, An
Introduction to Plato's Republic (Oxford, 1981).
31 The term "mode of cognition" is intended to indicate rather than ex-
plain what Socrates is talking about. As examples of different modes of
cognition one might consider the difference between knowledge and
Cambridge Companions Online © Cambridge University Press, 2006
198 THE CAMBRIDGE C O M P A N I O N TO PLATO
belief or between a person who has witnessed an event and one who has
heard about it or inferred that it must have happened.
32 Socrates' terms for the two modes of cognition are of no real help in
understanding the distinction he intends. I prefer to leave them untrans-
lated to avoid importing misleading connotations.
33 Here and elsewhere I have transformed a Socratic rhetorical question
into an assertion.
34 For further discussion of the mathematical hypotheses mentioned by
Socrates and of mathematical principles in early Greek mathematics
and philosophy see my paper "On the Notion of a Mathematical Starting
Point in Plato, Aristotle, and Euclid/7 in Science and Philosophy in
Classical Greece, ed. Alan Bowen (London and New York, 1991), 59-97.
35 Socrates here relies on the etymology of the Greek hupothesis.
36 For arguments that the unhypothetical principle of the Republic is an
instance of something satisfactory in the sense of the Phaedo, see Har-
old Cherniss, "Some War-Time Publications concerning Plato. I," Ameri-
can Journal of Philology 68 (1947): 141 (reprinted in his Selected Papers).
37 See Proclus, Commentary on Euclid, 29.14-24.
38 Socrates mentions that mathematics compels one to move upward to
the intelligible world already in the Divided Line passage at 51IC7.
39 It seems that Plato is here taking for granted Zeno's argument that an
extended thing can be divided into indefinitely many parts and his own
conviction (cf. Prm. I27d-i3oa) that Zeno's arguments apply to visible
rather than intelligible things. This conviction is perhaps justified by the
belief that all and only visible things are extended, but the claim that we
actually see things as one and many (rather than argue that extended
things are one and many) would seem to need more justification than
Socrates provides.
40 For discussion of nondiscursive thought see A. C. Lloyd, "Non-Discur-
sive Thought-An Enigma of Greek Philosophy/7 Proceedings of the
Aristotelian Society 70 (1969-70), 261-74.
41 There seems to be something reasonable about the idea that the hypothe-
sis of Forms is "higher than" the hypothesis that things are what they
are by participation in the Forms. But if the second is really a further
hypothesis not implicit in the first one, it is hard to see how the first can
be thought to rule out alternatives to the second in any strictly logical
sense of "rule out."
42 Contrast, for example, F. M. Cornford's position in "Mathematics and
Dialectic," esp. 178-81, 187-90 {Studies in Plato's Metaphysics, ed.
Allen, 82-5, 91-5).
43 It is not clear what, if anything, Socrates has in mind by this attempt to
divide up the one or the actual division and multiplication of the one. For
Cambridge Companions Online © Cambridge University Press, 2006
Mathematical method, philosophical truth 199
one attempt to relate what Socrates says to Greek mathematical practice
see B. L. van der Waerden, Science Awakening (New York, 1963), 115-16.
For a different and more plausible reading see M. F. Burnyeat, "Platonism
and Mathematics: A Prelude to Discussion/' in Mathematics and Meta-
physics in Aristotle, ed. Andreas Graeser (Bern and Stuttgart, 1987), 226.
44 Later philosophers, including ancient ones, argue that sensibles do not
satisfy the conditions laid down by geometers, e.g., that widthless lines
cannot be perceived. Such arguments are grist to Plato's mill, but, if he
was aware of them, he never invokes them explicitly. For an attempt to
assign some such argument to the Republic see Burnyeat, "Platonism
and Mathematics/' 221-5.
45 I here pass over Socrates' remarks about stereometry. His claims about
the backward state of stereometry have been taken to reflect Plato's
conception of the situation in mathematics in the fourth century. It is
hard to resist the suggestion that Socrates' call for a director of ste-
reometrical studies has some connection with Plato's role in the Acad-
emy. I shall also pass over Socrates' treatment of harmonics, which
seems to me quite of a piece with his description of astronomy. For some
discussion of both astronomy and harmonics see my paper "Ascending
to Problems: Astronomy and Harmonics in Republic VII," in Science
and the Sciences in Plato, ed. John P. Anton (Albany, 1980), 103-21. The
papers by Mourelatos and Vlastos in the same volume are very useful
treatments of the same material, focusing on astronomy.
46 See Gregory Vlastos, Plato's Universe (Seattle, 1975), 49-61 with the
relevant appendices.
47 I here take a position on a very difficult issue. In the Divided Line passage
Socrates describes the mathematician as using figures and hypotheses.
The dialectician destroys the hypothetical character of those hypotheses,
but there is no reason why the mathematician might not still continue to
draw conclusions from them. What about the use of figures? Does the
dialectician somehow make possible a geometry in which figures are no
longer used in argumentation? One can see how one might infer such a
possibility from what Socrates says. But he doesn't say it, and customary
ways of trying to make sense of the possibility are anachronistic. (For an
extreme example of such anachronism see A. E. Taylor, Plato the Man
and his Works, 5th ed. [London, 1948], 289-95.) I doubt very much that
Plato envisaged the possibility, but I am uncertain what Plato took the
connection between the use of diagrams and the apprehension of truth
about the intelligible world to be. His lack of explicitness on this question
is paralleled by his lack of explicitness on the relation between astronomi-
cal phenomena and astronomical knowledge.
48 On the importance of applied mathematics see the Phil. 53 d//.
Cambridge Companions Online © Cambridge University Press, 2006
GAIL FINE
6 Inquiry in the Meno
In most of the Socratic dialogues, Socrates professes to inquire into
some virtue.1 At the same time, he professes not to know what the
virtue in question is. How, then, can he inquire into it? Doesn't he
need some knowledge to guide his inquiry? Socrates' disclaimer of
knowledge seems to preclude Socratic inquiry.2 This difficulty must
confront any reader of the Socratic dialogues; but one searches them
in vain for any explicit statement of the problem or for any explicit
solution to it. The Meno, by contrast, both raises it explicitly and
proposes a solution.
I. THE PRIORITY OF KNOWLEDGE WHAT ( )
Meno begins the dialogue by asking whether virtue is teachable
(7oai-2). Socrates replies that he doesn't know the answer to
Meno's question; nor does he at all (to parapan, 712.7) know what
virtue is. The latter failure of knowledge explains the former; for "if
I do not know what a thing is, how could I know what it is like?" [ho
de me oida ti estin, pos an hopoion ge ti eideien; 7^3-4). Nonethe-
less, he proposes to inquire with Meno into what virtue is. Here, as
in the Socratic dialogues, Socrates both disclaims knowledge and
proposes to inquire. Socrates' disclaimer rests on his belief that he
satisfies the antecedent of the following conditional, when "virtue"
is substituted for x:
The first version of this paper was written while I was on leave in Oxford in the spring
of 1987. Since then, various versions have received helpful comments. I am especially
indebted to Jyl Gentzler, both to discussions with her and to her writings, and also to
Lesley Brown, David Brink, Terry Irwin, and Richard Kraut.
2OO
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 201
(PKW) If one doesn't at all know what x is, one can't
know anything about x.
I shall call this claim the Principle of the Priority of Knowledge
What (PKW). There is considerable dispute about how to interpret
PKW. One suggestion is that it means:3
(A) If one has no idea what x is - has no beliefs at all
about x - then one can't (intend to) say anything about x.
(A) is independently plausible; but it is difficult to believe it is what
Socrates intends. For the self-confident way in which he examines
Meno about virtue suggests he satisfies neither the antecendent nor
the consequent of (A): He appears to have some ideas, some beliefs,
about virtue; and he proceeds to say various things about it. More-
over (focusing for the moment on the antecedent of PKW), Socrates
doesn't say he has no ideas or beliefs about virtue; he says he
doesn't know (oida) what virtue is.* Here there are two points: first,
he claims to lack knowledge, not all beliefs or ideas; second, he
claims to lack knowledge about what virtue is. This second claim,
taken in context, suggests that the knowledge he (believes he) lacks
is knowledge of the definition of virtue, of its nature or essence; he
doesn't know the answer to the Socratic What is F? question, where
F is virtue. We might then try altering the antecedent of (A) so as to
yield (B):s
(B) If one doesn't at all know the definition of x, one can't
(intend to) say anything about x.
In contrast to (A), (B)'s antecedent is one Socrates seems to satisfy
(believe he satisfies). However, unlike (A), (B) seems self-defeatingly
strong. For if it is a precondition of saying anything (intending to say
anything) about x that one know what x is, and one does not know
what x is, then it is difficult to see how one can inquire into x. How
can one inquire into something if one can't even (intend to) say
anything about it? Moreover, we have seen that Socrates acts as
though he doesn't satisfy the consequent of (B), for he says quite a lot
about virtue. If he is committed to (B), and (believes he) doesn't
know what virtue is, yet continues to talk about virtue, then his
theory and practice conflict.
There is an alternative to (B) that is worth considering. Just as
Cambridge Companions Online © Cambridge University Press, 2006
2O2 THE CAMBRIDGE COMPANION TO PLATO
Socrates says, not that he lacks beliefs about what virtue is, but that
he lacks knowledge of what virtue is, so he says that such knowl-
edge is necessary, not for saying (intending to say) anything about
virtue, but for knowing anything about virtue:6
(C) If one doesn't at all know the definition of x, one can't
know anything about x.
In contrast to (B), (C) uses "know'' in both clauses,- so too does
Socrates. What he says is that one needs to know what virtue is, not
in order to say anything about virtue, or in order to have any beliefs
about virtue, but in order to know anything about virtue. This claim
leaves open the possibility that one could have beliefs about virtue,
and (intend to) say various things about virtue, without knowing
what it is; and, if beliefs that fall short of knowledge are adequate to
guide inquiry, then even if Socrates lacks all knowledge about virtue,
he can still inquire if he has and relies on suitable beliefs.
Although (B) and (C) can thus be read so as to be quite different, it
does not follow that Plato is alive to, or exploits, their difference. For
him to do so, he must be clear, among other things, about the differ-
ence between knowledge and belief. Yet it has been argued that he is
unclear about their difference, at least in the Socratic dialogues J We
shall need to see, then, whether Plato is able to exploit the difference
between (B) and (C) that their phrasing leaves open.
Whether we read PKW as (B) or (C), Socrates claims not to know
what virtue is, in the sense of not knowing the definition of virtue.8
One might wonder why he does so. Does not his ability to pick out
examples of virtuous actions, to use virtue terms coherently, and the
like, show that he knows the definition of virtue? If to know the
definition of virtue were simply to know the meaning of the term
"virtue," in the sense of knowing something like a dictionary or
lexical definition of it, then it would indeed be odd for Socrates to
claim not to know the definition of virtue.9 But for Socrates, to
know the definition of virtue is not simply to know the ordinary
meaning of the term "virtue"; it is to know what the thing, virtue,
really is, its explanatory properties. Knowing what virtue is, for Soc-
rates, is more like knowing a Lockean real than nominal essence -
more like knowing, say, the inner constitution, the atomic number,
of gold, than like knowing, or having some idea of, the surface,
observable features of gold, such as that it is yellow and shiny.10 If
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 203
this is right, then Socrates' claim not to know what virtue is is
reasonable; for as the progress of science reveals, real essences are
difficult to discover.
Even if it's reasonable for Socrates to disclaim knowledge of the
real essence of virtue, one might wonder whether it's reasonable for
him to claim that such knowledge has the sort of priority he accords
it. Here (B) and (C) demand different verdicts.11 It certainly doesn't
seem reasonable to say, as (B) does, that such knowledge is necessary
for saying (intending to say) anything about virtue. Surely most of us
lack knowledge of the real essence of virtue, but have some reliable
beliefs about virtue. Socrates seems to agree; at least, in the Socratic
dialogues, certain beliefs about virtue - for example, that it is admi-
rable (kalon), good (agathon), and beneficial (ophelimon) - are regu-
larly relied on.12 But (C) seems more reasonable; it says that knowl-
edge of the real essence of virtue is necessary for knowing anything
else about virtue. If we place strong conditions on knowledge, and
clearly distinguish knowledge from belief, then it is reasonable if
controversial to claim that knowledge of the nonessential properties
of a thing must be suitably rooted in knowledge of its nature.^
PKW raises a prima facie problem. For it says that if one doesn't
know what x is, one can't know anything about x. Socrates claims
not to know what virtue is, yet he proposes to inquire into what
virtue is. How can he inquire, or be justified in inquiring, given his
disclaimer of knowledge? Doesn't inquiry demand some initial
knowledge?
On some conceptions of his disclaimer and his project, there is no
difficulty. One conception, for example, is that (a) Socrates is not
really inquiring into virtue, at least, not in the sense of seeking
knowledge of what virtue is; he seeks only to expose the ignorance
of others, and this less demanding aim does not require moral knowl-
edge. Hence, his disclaimer does not conflict with his project after
all.1* Another conception is that (b) although Socrates wants to
know what virtue is, he is hypocritical or ironical in disclaiming
moral knowledge or, more charitably, he disclaims knowledge only
in an effort to force interlocutors to think for themselves. If Socrates
does not intend the disclaimer seriously, then, again, there is no
difficulty in squaring his project with his disclaimer.1*
Yet a third conception is that (c) both the disclaimer of knowledge
and the desire to know what virtue is are genuine-but the dis-
Cambridge Companions Online © Cambridge University Press, 2006
2O4 THE CAMBRIDGE COMPANION TO PLATO
claimer is less sweeping than it is sometimes thought to be. On one
version of this view (ci), Socrates disclaims knowledge in the sense
of "certainty/' but not in the different sense of "justified true be-
lief"; and knowledge in this latter sense can guide inquiry. On an-
other version of this view (c2), Socrates disclaims knowledge of
what virtue is - of its essence or nature - but not of what virtue is
like; he does not, for example, disclaim knowledge of instances of
virtuous action, and his knowledge of them can guide inquiry.16
If any of these conceptions is correct, then the problem disappears.
Unfortunately, however, none of them works for the Meno.^ Plato
claims in what follows that inquiry can achieve knowledge (85C9-
di). It would be perverse to suggest that although he believes it can
achieve knowledge, its true goal is only to perplex interlocutors,-
rather, as we shall see, he elicits perplexity only as an interim stage
in a journey whose ultimate destination is the acquisition of moral
knowledge. Conception (a) is thus inadequate.
Conception (b) is also inadequate. As Irwin argues, Socrates' "re-
peated disclaimers of knowledge are too frequent and emphatic to be
dismissed as ironical without strong reason; Aristotle takes them
seriously {Soph. El. i83b6-8), and so should we."18
Nor is (ci) adequate. At Meno 98a, Plato offers just one definition
of knowledge - as justified true belief (true belief coupled with an
aitias logismos) - and it is presumably knowledge of this sort that
he disclaims. Yet it is just this sort of knowledge that, according to
(ci), Socrates (believes he) possesses.1?
Nor is (c2) adequate. For Socrates claims not to know what virtue
is,- given PKW, it follows that he knows nothing at all about virtue.
He can't then, as (c2) proposes, know some things about virtue.
Our problem remains, then. On the one hand, Socrates professes
not to know what virtue is; given his affirmation of PKW, it follows
that (he believes) he knows nothing at all about virtue. How, then,
can he inquire into virtue? As we shall see, this is just the question
Meno asks him.
II. MENO'S PARADOX
Although Socrates claims not to know what virtue is, and so not to
know anything about virtue, he proposes to inquire into what virtue
is. Meno valiantly offers several suggestions; but Socrates rebuts
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 205
him at every turn, using his familiar elenctic method. He asks an
interlocutor a What-is-/7? question - What is courage [Laches), or
friendship (Lysis), or piety (Euthyphro). He cross-examines the inter-
locutor, appealing to various agreed examples and principles. Eventu-
ally the interlocutor discovers that, contrary to his initial beliefs, he
does not know the answer to Socrates' question. For it emerges that
he has contradictory beliefs about the matter at hand and so lacks
knowledge about it; if I have contradictory beliefs about x, then I
lack knowledge about x.2° The Socratic dialogues typically end at
this stage, with the interlocutor at a loss (in a state of aporia). In the
Meno, however, matters are carried further. Meno turns to the offen-
sive, challenging Socrates' right to question him when Socrates him-
self lacks knowledge. He poses a paradox, generally known as the
eristic paradox or as Meno's paradox:
How will you inquire into something, Socrates, when you don't at all know
what it is? Which of the things that you don't know will you suppose it is,
when you are inquiring into it? And even if you happen upon it, how will
you know it is the thing you didn't know? (8od5~8)
Socrates reformulates the paradox as follows:
I understand what you mean, Meno. Do you see what an eristic argument
you're introducing, that it isn't possible for one to inquire either into what
one knows, or into what one doesn't know? For one wouldn't inquire into
what one knows - for one knows it, and there's no need to inquire into such
a thing; nor into what one doesn't know - for one doesn't know what one is
inquiring into. (8oe 1 -6)
Meno poses three questions:
a. How can one inquire into something if one doesn't at all
know what it is?
b. Which of the things one doesn't know is one inquiring into?
c. How will one recognize the object of one's inquiry, even if
one finds it?
Socrates recasts Meno's paradox into the form of a constructive
dilemma:21
1. For any x, one either knows, or does not know, x.
2. If one knows x, one cannot inquire into x.
3. If one does not know x, one cannot inquire into x.
Cambridge Companions Online © Cambridge University Press, 2006
2O6 THE CAMBRIDGE COMPANION TO PLATO
4. Therefore, whether or not one knows x, one cannot inquire
into x.
The argument seems valid. (1) is a harmless instantiation of the law
of the excluded middle - either one does, or does not, know p}
tertium non datur. (2) and (3) tell us that whichever of these exclu-
sive and exhaustive options obtains, inquiry is impossible. (4) then
validly concludes that inquiry is impossible.
Although the argument seems valid, one might question its sound-
ness.22 (1) is harmless; but what about (2) and (3)?
In defense of (2), Socrates says only that if one already knows,
there is no need to inquire. This is not a very good defense of (2); I
don't need another meal at Lutece but, for all that, I might still go
there. Nor is it clear why, if one knows x, one cannot inquire into it.
I might know who Meno is, but seek to know where he is,- I might
know something about physics, but seek to know more about it. Of
course, if I know everything there is to know about Meno, or phys-
ics, then there is no need - or possibility - of inquiring about them.
But surely not all knowledge of a thing is tantamount to complete or
total knowledge of it; generally, one has only partial knowledge. (2)
thus seems false.2*
(3) also seems false. To be sure, if I do not know x in the sense that
my mind is a complete blank about it, if I am totally ignorant about
it, have no ideas whatsoever about it, then I cannot inquire into it.
But being totally ignorant about x does not seem to be the only way
to lack knowledge about it. I might lack all knowledge about x, but
have some (true) beliefs about it; and perhaps they are adequate for
inquiry. Having (true) beliefs that fall short of knowledge is one way
of lacking knowledge; but it is not a way of lacking knowledge that
seems to preclude inquiry. (3) thus also seems false.2*
Notice that in arguing that (3) is false, I appealed to the sort of
distinction I mentioned above, in distinguishing between two read-
ings, (B) and (C), of PKW. I suggested earlier that PKW (if it is read as
[C], and [C] is carefully distinguished from [B]) is controversial but
not outrageous. It is not outrageous, I suggested, because, although
one might need to know what x is to know what x is like, one need
not know what x is to have beliefs about what x is like, and perhaps
beliefs can guide inquiry. If, however, one speaks of a lack of knowl-
edge without differentiating between total ignorance (being a blank)
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 207
and having beliefs that fall short of knowledge, then this way of
distinguishing between (B) and (C) will not be available. One might
expect, then, that if Plato is aware of the falsity of (3), he can also
distinguish between (B) and (C); if, however, he is not aware of the
falsity of (3), then perhaps, equally, he will not be able to distinguish
between (B) and (C).
I have argued so far that the eristic paradox is unsound, since both
(2) and (3) are false. But how does Plato view the matter?
We might expect him to be especially troubled by (3). At least, we
should expect him to be especially troubled by (3) if he takes Socrates'
disclaimer of knowledge seriously, thinks Socrates was genuinely
inquiring, through the elenchus, in an effort to find moral knowledge,
and also wants to defend Socrates. For if (3) is true, Socrates cannot, as
he claims to, inquire in the absence of knowledge. (3) thus threatens
the core of Socratic inquiry. However, the Meno is a transitional
dialogue - transitional between the thought of the Socratic dia-
logues, on the one hand, and that of the middle dialogues, on the
other - and one might think that it is transitional in part because it
finds fault with Socrates' claim to be able to inquire in the absence of
knowledge.2* Let us see, then, whether Plato attempts to dislodge (3),
and so to vindicate Socrates,- or, alternatively, whether he abandons
the Socratic procedure - a procedure that requires the falsity of (3) -
in favor of some other epistemological program.
III. THE ELENCTIC REPLY TO THE PARADOX
In reply to Meno's paradox, Socrates initially describes a priests' and
priestesses' story, according to which:
Since the soul is immortal and has been born many times, and has seen
(heorakuia) all the things both here and in Hades, there is nothing it has not
learned. Hence it is no wonder if it can recall virtue and other things which
it previously knew. For since all nature is akin, and since the soul has
learned all things, there is nothing to prevent it, when it has recollected one
thing - which men call learning - from discovering all the other things, if
he is brave and does not tire of inquiring. For inquiring and learning are just
recollection. (8ic5-d5)
This is Plato's famous theory of recollection, according to which the
soul is immortal and, in a prior life, knew "virtue and other things,"
Cambridge Companions Online © Cambridge University Press, 2006
2O8 THE CAMBRIDGE COMPANION TO PLATO
so that what is called learning is really just recollection of things
previously known.26 Meno professes not to understand what it
means to say that learning is just recollection, and so he asks Socra-
tes to teach him that it is (8ie3~5). Socrates points out that, since
learning is just recollection, there is no such thing as teaching, and
so he cannot teach him that learning is just recollection; but, he
says, he will show [epideixomai, Sibi) Meno that it is. He then
embarks on a standard Socratic-style elenchus, along with a running
commentary, with one of Meno's slaves (82b-85d). He then reverts
to the theory of recollection (85d-86c). There are, then, two ac-
counts of the theory of recollection; sandwiched in between, there is
a sample elenchus.2?
How do the elenchus and the theory of recollection fit together?
How does either reply to the paradox? I begin by considering the
elenctic reply.
Socrates draws a square with sides two feet long, and asks the
slave how long a side is needed for a square with double the area of
the original square (82c-e). The slave replies that we need a side
with double the length of the original side (82e). Like most of Socra-
tes' interlocutors in the earlier dialogues, and like Meno at the begin-
ning of this one, the slave thinks he knows the answer to Socrates'
questions, though he does not. Socrates then questions him further,
until the slave realizes that he doesn't know what he thought he did;
he is then puzzled and confused (84a-b). This aporetic result is often
reached in the Socratic dialogues too, as it was earlier in the Meno.
The Socratic dialogues typically end at this point, which is one
reason Socrates is often thought to be purely negative and destruc-
tive in his use of the elenchus. Here, however, Plato defends him
against that charge. Initially the slave thought he knew the answer
to Socrates' geometrical question, but he didn't; he then realizes
that he doesn't know the answer. But realizing this is not merely
destructive,- it makes (or ought to make) him more willing and able
to inquire (84b-c). The exposure of ignorance is thus of positive
value.
Plato also points out that although the Socratic dialogues typically
end aporetically, elenchus need not end in aporia; the elenctic
method can take one all the way to knowledge. To show this, Socra-
tes questions the slave further, until the slave eventually states the
right answer (84d-85b); this further stage of questioning involves
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 209
the elenctic method no less than does the initial stage, and so Plato
shows that the elenchus can go beyond the exposure of ignorance to
the articulation of true beliefs.28 For although the slave still lacks
knowledge, he "has in himself true beliefs about the things he does
not know" (85C2-8). Not only that, but "if someone asks him the
very same [sorts of] questions [alternative translation: "asks him
questions about the very same things"] often and in different ways,
you can see that in the end he will know these things as accurately
as anybody" (85Cio-di).
Now if the slave can inquire about geometry in the absence of
knowledge then so too, Socrates assumes, can we all. Nor is there
anything special about geometry; inquiry in the absence of knowl-
edge is likewise possible for "every other subject" (8562-3), includ-
ing virtue, and so, Socrates concludes, they ought to resume their
inquiry into virtue, even though they don't know what it is (86c).
Plato has just rejected (3) of the paradox. Contrary to (3), one can
inquire even if one lacks all knowledge of the subject, for the slave
has just done so. The slave can inquire, although he entirely lacks
knowledge, because he has both true beliefs, and also the capacity
for rational reflection and revision of his beliefs, and these are ade-
quate for inquiry.29 Similarly, Socrates was justified, in the Socratic
dialogues and earlier in the Meno, in claiming to be able to inquire
into virtue in the absence of knowledge. For although he disavows
all moral knowledge, he never claims to lack true moral beliefs; and,
indeed, he seems to believe he has them.3° Moreover, in clearly
distinguishing between knowledge and (true) belief, and in insisting
that inquiry requires only the latter, Plato shows that PKW is not
self-defeatingly strong, that he can distinguish between (B) and (C),
and accepts only (C).*1
Various objections to Plato's claims about the powers of the
elenchus might be raised, however; let us consider some of them,
along with some possible replies.
Objection 1: I am overly optimistic about the force of 85C6-7,
where Plata distinguishes between knowledge and true be-
lief. Alexander Nehamas, for example, argues that the pas-
sage is merely an intermediate step in Plato's resolution of
the paradox, not his final conclusion.*2 Even if it is true that
using the elenchus can take us all the way to knowledge, and
Cambridge Companions Online © Cambridge University Press, 2006
2IO THE CAMBRIDGE COMPANION TO PLATO
that knowledge and true belief differ, surely 8506-7 is not
the core of Plato's reply to the paradox?
Reply: It is true that there is more to come: We have not yet looked
at the theory of recollection. But there is reason to believe
that, whatever the importance of the theory of recollection,
the distinction Plato draws between knowledge and true be-
lief is of vital importance, not only to Plato's epistemology
generally, but also to his resolution of the paradox. For at the
close of the dialogue, he recurs to, and elaborates on, the
distinction between knowledge and true belief (97a-98c). He
argues first that true belief is as good a guide for right action
as knowledge is. This reemphasizes the present point that
one can inquire on the basis of true belief; knowledge is not
necessary for inquiry. Nonetheless, he insists, knowledge is
more valuable. For although both knowledge and true belief
are truth-entailing, knowledge is true belief tethered with an
aitias logismos, an explanatory account. He also insists that
the process by which one works out an explanatory account
is recollection. Hence the difference between knowledge and
true belief is not a mere aside; Plato returns to it later, and
connects it to the present contexts
Objection 2: Socrates has not shown that inquiry is possible in the
absence of knowledge. For even if the slave lacks knowledge,
Socrates, in this case if not in the moral case, has the rele-
vant knowledge, and that is what makes progress possible.
Reply: Socrates does not claim that he knows the answers to the
questions he asks, and it is not clear that (he believes) he
does; perhaps he only has a correct belief about the answers.
In just the same way, Socrates can guide elenchi in the So-
cratic dialogues, not because he knows the answers, but be-
cause he has true beliefs.^
But even if Socrates knows, or believes he knows, the an-
swers, the point of the elenctic demonstration is not under-
mined. For although Socrates asks the slave leading ques-
tions, he does not feed him the answers. On the contrary,
Socrates emphasizes that the slave should not rely on Socra-
tes' authority, but should say what he believes (83d); this
point is brought home by the fact that the slave twice offers
wrong answers by relying uncritically on what Socrates says.
The slave's progress - from initial misguided confidence, to a
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 211
realization of his ignorance, to the discovery of the right
answer - ultimately comes from his own independent reflec-
tion. At each stage he decides to resolve a conflict in his
beliefs by discarding those beliefs that seem less reasonable,
or less well entrenched - just as other interlocutors do in
moral inquiry. Socrates' geometrical knowledge (or true be-
lief) makes the elenchus proceed more quickly and smoothly;
but it is not what makes it possible. What makes it possible is
the slave's own true beliefs and his capacity for reflection and
revision. 3 5
Objection 3: The slave can make progress in the geometrical case
because geometry is a deductively closed system, consisting
of necessary, a priori truths; since morality is not like this,
progress cannot be achieved in the same way in its case.
Reply: If the objector concedes that one can inquire in the absence
of knowledge in the geometrical case, then he concedes that
the paradox has been disarmed; for that concession involves
abandoning (3). Still, the objector has a point; for we want to
be able to vindicate Socratic moral inquiry in the absence of
moral knowledge. But are the geometrical and moral cases
so different? To be sure, geometry (unlike morality) may
well be a deductively closed system, consisting of necessary,
a priori truths. But these are not the facts about it that Socra-
tes emphasizes. He describes the mathematical inquiry in
much the same way that we would describe scientific in-
quiry. We begin with a variety of beliefs about, say, gold.
Some of these are true, others false,- we gradually refine our
beliefs - discovering, for example, that fool's gold is not
gold-until we arrive at knowledge of the real essence of
gold, its atomic constitution. We have moved from belief
about gold to knowledge of its real essence, not by rigorous
deduction, but by trial and error; in just the same way, we
can make progress in the moral sphere.
Objection 4: The scientific analogy is unhelpful. After all, one
thing that enables us to make progress in the scientific case
is the availability of samples or examples of gold; but what
samples or examples are available in the moral case?
Reply: The answer is: examples of virtuous behavior. Although Soc-
rates denies that we know what virtue is, he never denies,
Cambridge Companions Online © Cambridge University Press, 2006
212 THE CAMBRIDGE COMPANION TO PLATO
but in fact assumes, that we have fairly reliable beliefs about
virtue.*6 To be sure, sometimes we make mistakes-we
falsely believe, for example, that lions are courageous [La.
i96ei-i97C4). But then, people once believed that fool's
gold was gold. We also correctly believe that returning a
sword to a madman is not just (Rep. 331).
But doesn't Socrates insist, contrary to my suggestion,
that there are irresoluble disputes in the moral cases? Well,
he does believe there is great dispute about the correct defini-
tion of virtue terms, and of course there is some dispute
about particular moral cases. But there is also considerable
agreement, enough agreement to secure the reference of the
terms and so to ground inquiry.
Objection 5; How can we know which of our beliefs are true, which
false? Beliefs don't come neatly labeled "true" and "false";
what's to stop us from relying on the false ones instead? The
mere fact that I have true beliefs is not sufficient to ground
inquiry. 37
Reply: Plato's claim is that one can inquire, even if one lacks knowl-
edge, so long as one in fact relies on one's true beliefs; he
does not claim that one can inquire, even if one lacks knowl-
edge, only if one knows that one is relying on true beliefs. Of
course, from a first-person perspective, I will be subjectively
justified in inquiring only if I believe that I am relying on
true beliefs. But I do not need to be able to identify my true
beliefs as such in order to be able to inquire. We need to
distinguish the question of what makes inquiry possible
from the question of what subjectively justifies one in think-
ing one is in a position to inquire. In neither case, however,
do I need to know (or even have true beliefs about) which of
my beliefs are true, which are false. In the first case, I need to
rely on some beliefs that are in fact true; in the second case, I
need to believe I have some true beliefs. Neither of these
ways of appealing to true beliefs requires one to know (or
have true beliefs about) which of one's beliefs are in fact
true.38
Of course, someone might rely on false, rather than on true, be-
liefs. As in science, one can follow a false track; progress requires
luck. Socrates seems to assume, however, that everyone, or at least
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 213
everyone rational, will, if they inquire systematically, progress in
the same direction. That's because he also seems to assume that
some important true beliefs are better entrenched than are various
false beliefs (or will seem more reasonable to us when we first con-
sider them) so that, in cases of conflict, we tend, upon reflection, to
reject the false beliefs.39 This is a substantial, and optimistic, claim
about human nature-one that requires and, as we shall see, re-
ceives, further explanation.
IV. THE THEORY OF RECOLLECTION
The elenctic reply disarms the paradox by arguing that, contrary to
(3), inquiry is possible in the absence of knowledge. It seems to be a
good, and complete, reply. Why, then, does Plato supplement the
elenctic reply with the theory of recollection? What role does it play
in replying to the paradox?
I suggest that the theory of recollection is introduced, not as a
direct reply to the paradox (the elenctic reply plays that role)/0 but
to explain certain facts assumed in the elenctic reply. For example,
the elenctic reply assumes that in inquiring, we tend to favor true
over false beliefs. Plato believes that this remarkable tendency can-
not be a brute fact, but requires further explanation,- the best such
explanation, in his view, is the theory of recollection. We can all
inquire, and tend toward the truth in doing so, because, although we
now lack the relevant knowledge, we once had it, in a prior life. Like
advocates of innate knowledge, Plato believes that certain remark-
able features of human beings require explanation in terms of prior
knowledge - though for Plato, in contrast to the innatists, the knowl-
edge is had not from birth, but only in a previous existence.
Even though the theory of recollection is thus not a theory of
innate knowledge/1 its motivation is similar to the motivation for
innatist theories of knowledge. As such, it is vulnerable to similar
objections. Many would prefer to say that even if a given tendency is
remarkable, still, it is just a brute fact that we have it; there is no
further explanation. Or, if there is a further explanation, it consists
not in immortal souls that had knowledge in some previous life, nor
in innate knowledge, but in, for example, evolution.
In claiming that we once had the relevant knowledge, Plato inevi-
tably invites the question of how we acquired it. If the answer is
Cambridge Companions Online © Cambridge University Press, 2006
214 THE CAMBRIDGE COMPANION TO PLATO
"through inquiry/7 then we can raise Meno's paradox again, and we
might then seem to be launched on a vicious infinite regress. But
there is no regress, vicious or otherwise; for Plato does not commit
himself to the claim that we acquired the previous knowledge at all.
He seems to think we simply had it at some previous stage, without
having gone through any process of acquiring it.*2
Notice that whatever account one favors of how the soul once
knew, the theory of recollection (in contrast to the elenctic reply)
does not by itself provide a sufficient answer to the paradox - for if
one once knew, but now lacks the ability to inquire, the prior knowl-
edge is idle. We should therefore be reluctant to put the whole
weight of Plato's reply to the paradox on the theory of recollection;
and on my account of its role, we need not do so. It is also important
to be clear that, no matter what account one favors of how the soul
once knew, Plato's introduction of the theory of recollection does
not show that he has abandoned the elenchus as the sole method of
inquiry (in this life). The demonstration with the slave is just a
standard elenchus; and in it, Socrates claims that if one follows it
long enough one will achieve knowledge. The theory of recollection
goes beyond Socrates, not by replacing the elenchus with an alterna-
tive route to knowledge, but by explaining how something he took
for granted (the possibility of inquiry in the absense of knowledge,
and the remarkable fact that in so inquiring we tend toward the
truth) is possible. To say that p is best explained by q, or is possible
because of q, is not to abandon p. The theory of recollection is
introduced to vindicate, not to vitiate, Socrates' claims about the
powers of the elenchus.
However one spells out the details of the theory of recollection,
few nowadays are likely to believe it. The elenctic reply, however,
remains convincing, and it can be accepted even by one who rejects
the theory of recollection; one can accept Plato's claim that one can
inquire in the absence of knowledge, because of one's capacity for
reflection and because of one's true beliefs, without accepting his
account of what explains the capacity and the beliefs. It is thus
pleasing to see that Plato himself seems to place less weight on the
theory of recollection than on the elenchus. He introduces the
theory as something said by priests and priestesses and by Pindar
and other poets (8ia5-6, aio-b2); later he makes it plain that he
thinks such people lack knowledge (99c). Socrates says he would not
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 215
like to take an oath on all that he has said (86b); but later he says
that if he were to claim to know anything, one of the few things he
would claim to know is that knowledge differs from true belief
(98bi-5 (.43 And it is of course the difference between knowledge and
true belief that is crucial to the elenctic reply.
At least in the Meno, then, Plato replies to the eristic paradox by
reaffirming the powers of the elenchus and by vindicating Socrates'
claim to be able to inquire, through the elenchus, in the absence of
knowledge. At least in this respect, the Meno does not depart from,
but is continuous with, the project begun in the Socratic dialogues.
NOTES
1 The Platonic dialogues are often divided into four groups: (i) early So-
cratic dialogues; (ii) transitional dialogues: (iii) middle dialogues; and (iv)
late dialogues. Group (i) includes Ap., Cri., Euphr., Chrm., La., Lys., H.
ML, Euthd., Ion, and Pit. (ii) includes Gig., Meno, H. Ma., and Cia. (iii)
includes Phd., Smp., Rep., and Phdr. (iv) includes Pirn., Tht, TL, CritL,
Sph., Pol., Phil., and Laws. Some scholars favor a tripartite division
instead, into early, middle, and late. The dates of some of these dialogues
are disputed, but it is generally agreed that the Meno belongs after the
dialogues I have included in (i), and before the dialogues I have included
in (iii). For some discussion of the dating of the dialogues, see Leonard
Brandwood, "The Dating of Plato's Works by the Stylistic Method: A
Historical and Critical Survey/7 Ph.D. thesis (University of London,
1958) (available from University Microfilms); see also Brandwood's
"Stylometry and Chronology/' Chapter 3 of this volume. The Meno is
often taken to be transitional on the grounds that (a) it is more self-
conscious about methodology and epistemology than are the dialogues
in group (i); on the other hand, (b) the theory of Forms that is prominent
in group (iii) is muted in the Meno.
1 There are actually two questions here: (a) Does Socrates need to have
some knowledge in order to inquire? (b) Does he need to believe he has
some knowledge in order to be subjectively justified in inquiring? If
Socrates believes he lacks knowledge but in fact has knowledge, then (b)
but not (a) arises. If he believes he has knowledge but does not in fact
have any knowledge, then (a) but not (b) arises. I shall generally ignore
the difference between (a) and (b), though see my reply to Objection 5 in
section III.
I take inquiry to be a directed, intentional search for knowledge one
lacks. Hence, perceiving, happening upon an object one is looking for,
Cambridge Companions Online © Cambridge University Press, 2006
2 l 6 THE CAMBRIDGE COMPANION TO PLATO
and being told are not forms of inquiry. Ordinary scientific research, on
the other hand, is an example of inquiry. Inquiry can take various forms.
In the Socratic dialogues (and, I shall suggest, in the Meno), it takes the
form of elenchus, on which see further below.
3 See Alexander Nehamas, "Meno's Paradox and Socrates as a Teacher/'
Oxford Studies in Ancient Philosophy 3 (1985): 5-6.
4 Unfortunately, many translations obscure this crucial point. For exam-
ple, although W. K. C. Guthrie (Plato: Protagoras and Meno [Harmonds-
worth: Penguin Classics, 1956]) translates 7ib3~4 accurately enough, he
mistranslates similar passages in the surrounding context. He translates
71 as-7 as: "The fact is that far from knowing whether it can be taught, I
have no idea what virtue itself is." Here "I have no idea;/ should be
rendered as "I do not know (eidos)." He translates 7ib4-6 as: "Do you
suppose that somebody entirely ignorant who Meno is could say
whether. . .," when he should have: "Do you suppose that someone
who does not at all know (gignoskei) who Meno is could know (eidenai)
whether . . . ." Nehamas ("Meno's Paradox/7 8) translates 8ods-6 as "in
what way can you search for something when you are altogether igno-
rant of what it is?"; but the last clause is better rendered by "when you
don't at all know what it is."
5 See P. T. Geach, in "Plato's Euthyphro: An Analysis and Commentary,"
Monist 50 (1966): 369-82.
6 What he literally says is that one needs to know what x is (ti) to know
what x is like (poion). But I take the present contrast between ti and
poion to be exhaustive. (In the case of things like virtue, this will be the
exhaustive contrast between the essence of x [ti] and its nonessential
properties [poion]} cf. the contrast at Euphr. na -b between ti and pa-
thos. ) Hence, if one needs to know what x is to know what x is like, then
one needs to know what x is to know anything at all about x.
Strictly speaking, I would prefer to say that (C) is an instance of, rather
than a version of, PKW. That is, PKW claims quite generally that in order
to know anything about x, one must know what x is. But I think Plato
believes that the relevant knowlege-what differs from case to case: To
know anything about virtue, one must know its definition; but to know,
e.g., who Meno is, one need not know his definition - here, the relevant
knowledge-what consists in something other than knowing a definition.
Hence, even if Meno cannot be defined, it does not follow that he cannot
be known. I focus on definitions in this chapter, since I shall not be
discussing Plato's views about knowledge of such things as Meno; but
see notes 19, 21, and 26.
7 See, e.g., John Beversluis, "Socratic Definition," American Philosophi-
cal Quarterly 11 (1974): 331—6.
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 217
8 I assume PKW is not to be read as (A), and so I shall not consider it
further here.
9 This might be disputed. David Bostock, e.g., in Plato's Phaedo (Oxford:
Clarendon Press, 1986), 69-72, argues that knowing the meanings of
terms is actually quite difficult. That is no doubt true on some views
about meanings; but if we take meanings simply to articulate ordinary
usage, then even if not everyone can readily state them, it would be odd
for Socrates to believe that knowledge is as difficult to come by as he
seems to assume it is. In other respects too, it is unlikely that Socrates is
searching for knowledge of the meanings of terms. For some brief consid-
erations against the meaning view, see my "The One over Many/' Philo-
sophical Review 89 (1980): 197-240; see also note 10.
10 For Locke on real vs. nominal essences, see John Locke, An Essay Con-
cerning Human Understanding, ed. P. Nidditch (Oxford: Clarendon
Press, 1975) (originally published in 1690), III. iii; III. vi; III. x; IV. vi.4-9;
IV. xii.9. For a defense of the claim that Socrates is more interested in
something like the real essence of virtue than in the meaning of virtue
terms, see Terry Penner, "The Unity of Virtue/' Philosophical Review
82 (1973): 35-68; and Terence Irwin, Plato's Moral Theory (Oxford: Cla-
rendon Press, 1977), esp. chap. 3.
11 At least, this is so if (B) and (C) are taken to differ in the way described
above.
12 See, e.g., Chrm. 159C1, i6oe6; La. 192C5-7; Prt. 34963-5, 35964-7. See
also Meno 8761-3. Socrates also seems to believe that his, and his
interlocutors', beliefs about examples of virtuous actions are generally
reliable.
13 In Posterior Analytics, i 1-10, Aristotle defends a version of (C), claim-
ing that one can know the nonessential properties of a thing only by
deducing certain propositions about them from its real definition.
14 For conception (a), see Gregory Vlastos, "Introduction," in Plato: Prota-
goras (New York: Bobbs-Merrill, 1956), esp. xxvi-xxxi; contrast Vlastos,
"The Socratic Elenchus," Oxford Studies in Ancient Philosophy 1
(1983): 27-58, esp. 45//. In "Elenchus and Mathematics," American Jour-
nal of Philology 109 (1988): 362-96, Vlastos rejects (a) for the Socratic
dialogues; but he argues that in the Meno Plato believes that elenchus
(the form of inquiry favored in the Socratic dialogues) can do no more
than detect contradictions.
15 For conception (b), see, e.g., Richard Robinson, Plato's Earlier Dialectic
(Oxford: Clarendon Press, 1953), chap. 2.
16 Conception (ci) distinguishes between two types or kinds of knowledge
(certainty and justified true belief); (c2) distinguishes between different
ranges of things known or not known (what virtue is, what it is like). For
Cambridge Companions Online © Cambridge University Press, 2006
2 l 8 THE CAMBRIDGE COMPANION TO PLATO
(ci), see Gregory Vlastos, "Socrates' Disavowal of Knowledge/' Philo-
sophical Quarterly 35 (1985): 1-31. For (02), see Richard Kraut, Socrates
and the State (Princeton: Princeton University Press, 1984), chap. 8.
There are traces of (ci) in Vlastos's article, though he focuses on (ci).
Alexander Nehamas, in "Socratic Intellectualism," in Proceedings of
the Boston Area Colloquium in Ancient Philosophy, vol. 2, ed. John J.
Cleary (Lanham, Md.: University Press of America, 1987), esp. 284-93,
also seems to endorse a version of (c2). One might try to defend (ci) by
appealing to two types of knowledge other than those Vlastos appeals to,-
for such an attempt, see Paul Woodruff, "Plato's Early Theory of Knowl-
edge," in Companions to Ancient Thought 1: Epistemology, ed. Stephen
Everson (Cambridge: Cambridge University Press, 1990), 60-84. The
consideration I adduce below against Vlastos;s version of (ci) applies to
other versions of it as well.
17 It is of course possible that one or more of them is adequate for one or
more of the Socratic dialogues. Kraut, e.g., explicitly defends (c2) only for
some of the Socratic dialogues; he agrees that it is inadequate for the
Meno.
18 Irwin, Plato's Moral Theory, 39-40.
19 Even if one denies that 98a defines knowledge as justified true belief, the
fact remains that Plato offers only one definition of knowledge, and that
would be odd if he intended his disclaimer to apply only to some other
sort of knowledge. Hence we should be reluctant to endorse any version
of (ci).
It is sometimes denied that 98a defines knowledge as justified true
belief. For example, it is sometimes said that (i) what Plato believes
must be added to true belief to get episteme is not justification but
explanation, so that (ii) episteme is not knowledge but understanding.
(Further arguments have also been offered in support of [ii]. For various
versions of this view, see, e.g., Nehamas, "Meno's Paradox/7 esp. 24-30;
M. F. Burnyeat, "Socrates and the Jury/' Proceedings of the Aristotelian
Society, supp. vol. 54 [1980]: 173-92, esp. 186-8; and Burnyeat, "Witt-
genstein and De Magistro," Proceedings of the Aristotelian Society,
supp. vol. 61 [1988]: 1-24, esp. 17-24.) However, (i) does not imply (ii);
for Plato might believe that knowledge requires explanation. I also
doubt that (i) is true. Plato of course believes that in many cases, ade-
quate justification consists in explaining the natures of the entities one
claims to know; but he doesn't believe that it always does - one can
adequately justify one's claim to know, e.g., who Meno is, or the road to
Larissa, without explaining their essences; here some less demanding
sort of justification will do. (Hence, contrary to Nehamas, if we take
episteme to be knowledge, Plato does not have an impossibly demand-
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 219
ing conception of knowledge; this removes one reason for wanting to
believe [ii].) Plato focuses on explanation, not because he thinks it is
necessary for episteme as such, but because he thinks it is necessary for
knowledge of things like virtue, which are his primary concern here. See
also notes 6, 21, and 26.
Gregory Vlastos, "Anamnesis in the Meno," Dialogue 4 (1965): 154-
5, believes that Plato's definition of knowledge as true belief "bound" by
an aitias logismos is meant to restrict knowledge to necessary truths.
But this is too narrow, nor are Vlastos's arguments convincing. He ap-
peals, e.g., to pre-Socratic usage of anangke, a word the Meno does not at
this stage even use. He himself notes that logismos is often used for
rational thought in general, and it is of course well known that aitia can
be used quite broadly.
20 It might be argued that having contradictory beliefs does not automati-
cally debar one from having any knowledge about the subject matter; for
this argument, see, e.g., Alvin Goldman, Epistemology and Cognition
(Cambridge, Mass: Harvard University Press, 1987); and Gilbert Har-
man, Change in View (Cambridge, Mass.: Bradford Books, 1986). How-
ever, the contradictions Socrates uncovers are so blatant that it seems
reasonable to conclude that his interlocutors lack knowledge.
21 For this point, see Nicholas P. White, "Inquiry," Review of Metaphysics
28 (1974): 290 n. 4.
Some differences between Meno's and Socrates' formulations are
worth noting: (a) (2) has no analogue in Meno's formulation, (b) Con-
versely, Socrates' formulation ignores Meno's third question, (c) Meno
asks how one can inquire into what one does not know at all (to
parapan, 8od6; cf. 7ia7, 7ib3, 5); Socrates asks only how one can inquire
into what one does not know. (For further discussion of [c], see note 29.)
Some details about the scope of the paradox are also worth noting. The
paradox does not ask whether, in general, one can acquire knowledge,- it
asks only whether one can come to know things like virtue through
inquiry. Hence Meno's paradox does not question one's ability to come
to know things quite unlike virtue (e.g., the road to Larissa); nor does it
question one's ability to come to know things in some way other than
through inquiry (e.g., through perception, or by being told). Correspond-
ingly, Plato's reply does not address the question of whether it is possible
to know things unlike virtue; nor does it say whether it is possible to
know things in some way other than through inquiry. It does not follow
that Plato restricts knowledge to things like virtue, or restricts the
method of achieving knowledge to inquiry. For what it is worth, I think
the Meno leaves open the possibility of knowing, e.g., who Meno is and
the road to Larissa; I also think it leaves open the possibility of achieving
Cambridge Companions Online © Cambridge University Press, 2006
22O THE CAMBRIDGE COMPANION TO PLATO
knowledge by means other than inquiry. I shall not defend these claims
in any detail here, but see notes 6, 19, and 26.
22 The argument can be read so as to be invalid, through equivocation on
"know." But since Socrates' reply seems to attack only its soundness, I
shall assume it is valid.
23 (2) can be read so as to be more plausible than I have made it seem. It is
more plausible, e.g., if it is read to say that if I know that p, then there is
no need to inquire whether p. One reason to favor a plausible reading of
(2) is that although Socrates (I shall argue) goes on to argue that (3) is
false, he does not explicitly reject (2). Though rejecting (3) is sufficient
for showing that the paradox is unsound, one might expect him to reject
(2) as well, if it is indeed false. Perhaps the fact that he does not do so
suggests (2) should be read so as to be true. On the other hand, there are
good reasons for him to be especially troubled about (3), and so it is not
surprising that he focuses on it. Moreover, if (2) is read so as to be
plausible in the way just suggested, it does not fit as well into the overall
argument as it does if it is read in the way suggested in the text; see note
24.
24 Although (2) and (3) seem false, they would seem true to anyone who
believed that there was an exclusive and exhaustive dichotomy between
complete knowledge and total ignorance, such that if one has any knowl-
edge at all about a thing one has complete knowledge of it, and such that
if one lacks any knowledge at all about a thing one is totally ignorant of
it. And someone who accepts a certain sort of acquaintance model of
knowledge would believe just this. (This perhaps provides some reason
to read [2] as I read it in the text, rather than in the more plausible way
suggested in the previous note.) For a lucid account of how one might be
seduced by the paradox in virtue of accepting such a model, see John
McDowell, Plato: Theaetetus (Oxford: Clarendon Press, 1973), 194-7.
He is explaining a puzzle about false belief raised in Tht. i88a-c which,
like the eristic paradox, begins from the assumption that for all x, one
either knows, or does not know, x. However, McDowell believes Plato
accepts the underlying acquaintance model. If he were right, then Plato
might well be seduced by the eristic paradox. In "False Belief in the
Theaetetus," Phronesis 24 (1979): 70-80, however, I argue that Plato
does not accept the underlying acquaintance model in the Theaetetus.
The account I go on to give here suggests he does not accept it in the
Meno either, but I cannot defend this view in detail here. The similarity
between the eristic paradox and the puzzle at Tht. i88a-c, and the
possible connection to some sort of acquaintance, are also noted by
Irwin, Plato's Moral Theory, 315 n. 12.
25 For the dating of the dialogues, see note 1.
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 221
26 At 8106-7 (cf. di), Socrates says that (a) the soul has been "all things/7
which might suggest omniscience. However, at 8ic8, he talks instead of
(b) "virtue and other things"; and at 8id4~5 he talks about (c) all zetein
kai manthanein ("inquiry and learning"). I assume that (b) and (c) re-
strict the scope of (a), which makes it reasonable to assume that recollec-
tion is restricted to general truths about such things as virtue and geome-
try. Hence there is no implication that our discarnate souls knew truths
falling outside the scope of such disciplines, or even particular (as op-
posed to general) truths within such disciplines. One might argue that if
such truths are not recollected, then they cannot be known. For at Meno
98a Plato says that this — working out a suitable aitias logismos - is
recollection, and having a suitable aitias logismos is necessary for knowl-
edge. However, Plato means only that working out a suitable aitias
logismos is, in certain cases, a case of recollection. He does not mean
that every case of working out an aitias logismos involves recollection,
or that working out an aitias logismos is all there is to recollection.
Hence he leaves open the possibility that one can have a knowledge-
constituting aitias logismos of, e.g., the way to Larissa, but one that does
not involve recollection. See Irwin, Plato's Moral Theory, 316-17 n. IJ}
contrast Nehamas, "Meno's Paradox," 10-11.
27 See Irwin, Plato's Moral Theory, 139, 315-16, for a lucid and detailed
defense of the claim that the whole of the demonstration with the slave
is a standard Socratic elenchus. For a different view, see Vlastos,
"Elenchus and Mathematics," esp. 375. See also note 28.
28 Vlastos, in "Elenchus and Mathematics," 375, agrees that the negative
stage of the inquiry involves the elenchus, but he argues that the posi-
tive stage does not: "[e]lenchus is good for this, and only this [i.e., only
for "convicting him of error"]. It does not begin to bring him to the truth
he seeks." However, his reasons for this claim are weak. They seem to
be that (a) in the positive stage of inquiry, Socrates sheds his adversarial
role; and that (b) the inquiry concerns geometry rather than morality.
But as against (a), elenchus does not require anyone to play an adversar-
ial role in the sense Vlastos seems to intend. As to (b), the initial stage of
inquiry equally involves geometry, yet Vlastos allows that it involves
elenchus. See also the replies to objections at the end of this section for a
defense of the claim that geometrical and moral inquiries are (for Socra-
tes) quite similar. For cogent criticism of Vlastos's view, see Jyl Gentzler,
"Knowledge and Method in Plato's Early through Middle Dialogues,"
Ph.D. thesis (Cornell University, 1991).
29 I pointed out above (note 21, difference [c]) that one difference between
Meno's and Socrates' formulations of the paradox is that Meno asks
whether one can inquire into what one doesn't at all know, whereas
Cambridge Companions Online © Cambridge University Press, 2006
222 THE CAMBRIDGE COMPANION TO PLATO
Socrates asks only whether one can inquire into what one doesn't know.
This difference is sometimes thought to suggest that Plato believes that
one can't inquire into what one doesn't at all know, but can inquire into
what one doesn't know in a way (so long as one knows it in some
different way); see, e.g., Julius M. E. Moravcsik, "Learning as Recollec-
tion," in Plato, vol. i: Metaphysics and Epistemology, ed. Gregory
Vlastos (Garden City, N.Y.: Anchor Books, Doubleday, 1970), 57. On my
account, however, Plato allows that one can inquire into what one
doesn't at all know, in any way, and so his omission of "at all" is not
significant, at least, not in the suggested way. Plato is cavalier in his use
of toparapan elsewhere too: Despite its occurrence in 7ia7 and b3, 5, it
is omitted in the statement of PKW at 7^3 -4 .
30 For a justification of this claim, see Irwin, Plato's Moral Theory, chap. 3.
The frequency with which Socrates relies on various claims in the dia-
logues (see note 12) also supports this view. It is sometimes wrongly
thought that Irwin's view is that Socrates identifies knowledge and true
belief; see, e.g., Woodruff, "Plato's Early Theory of Knowledge," 64.
Vlastos, in "Socrates' Disavowal of Knowledge," 6 n. 12, on the other
hand, wrongly suggests that Irwin's view involves conflating knowledge
and certainty. Irwin's view is that for Socrates, knowledge is justified
true belief (which does not involve certainty); Socrates disavows all
moral knowledge, but he thinks he has (not moral knowledge but) true
beliefs about virtue.
31 Further, in noting that knowledge and total ignorance are not exhaustive
options, that (true) belief is a tertium quid, Plato says something that is
incompatible with the acquaintance model on which the paradox argu-
ably rests; see note 24. This suggests that, at least in the Meno, he does
not accept that sort of acquaintance model of knowledge.
32 Nehamas, "Meno's Paradox," 29.
33 On Plato's account of knowledge as true belief coupled with an aitias
logismos, see note 19. For his claim that working out an account is
recollection, see note 26.
34 Socrates (believes he) has, not just true beliefs, but also true beliefs that
are better justified than are those of his interlocutors (though not well
enough justified to count as knowledge - justification comes in degrees).
Hence, Socrates belongs at the second stage of the Line; he has pistis,
whereas his interlocutors have only eikasia, about morality. See my
"Knowledge and Belief in Republic V-VII," in Companions to Ancient
Thought 1: Epistemology, ed. Stephen Everson (Cambridge: Cambridge
University Press, 1990), esp. 101-4.
35 See Vlastos, "Anamnesis in the Meno," esp. 158-9, and "Elenchus and
Mathematics," 374 n. 42.
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 223
36 See notes 12 and 30. One might argue that although we have fairly
reliable beliefs about the nature of virtue and about what sort of behav-
ior and what sort of person would count as virtuous, there are no actual
examples of virtuous actions or of virtuous people available for us to rely
on. For Socrates thinks that knowledge of what virtue is is necessary for
being a virtuous person; since no one knows what virtue is, no one is
virtuous, and so, one might also think, neither are there any virtuous
actions, although, of course, some people and actions might nonetheless
be better than others. (A can be better than B even if neither is good.) If
this is so, then we could not say, in reply to Objection 4, that examples
of virtuous behavior play the role in moral inquiry that examples of gold,
e.g., play in scientific inquiry into the nature of gold. However, reliable
moral intuitions and beliefs could still guide inquiry.
37 For something like this objection, see Nehamas, "Meno's Paradox/'
16-17.
38 Cf. note 2. One might argue that it is not necessary to have true beliefs
in order to inquire; all that is necessary is that one's use of a term be on a
suitable causal chain,- see Saul Kripke, Naming and Necessity (Cam-
bridge, Mass: Harvard University Press, 1980) (originally published in
1972); and Hilary Putnam, "The Meaning of 'Meaning/ " in Philosophi-
cal Papers (Cambridge: Cambridge University Press, 1975), 2: 215-71.
39 See Irwin, Plato's Moral Theory, 41-2, 66-70.
40 It is sometimes thought that the theory of recollection is the direct
reply, and that it replies by denying (2): The slave can inquire because
he has knowledge; more generally, everyone can inquire, because every-
one has some relevant knowledge. If Plato does claim that everyone
now knows, then he contradicts his claim at 85b-d, that the slave does
not now know. But Plato does not claim that everyone (or the slave)
now knows; all of his references to knowledge are either forward-
referring (to the time when, by further questioning, the slave will ac-
quire knowledge) or backward-referring (to our previous lives, when we
did know). The passage that seems most difficult to square with this
claim is 86a8; but it too can be accommodated, for all it says is that the
slave's soul "has for all time been in the state of having once been [in a]
learned [condition]" - that is, it is always true of him (and so it is now
true of him) that he was once in a learned condition, i.e., once had
knowledge. To claim that it is always (and so is now) true of him that
he once had knowledge is not to claim or imply that it is now true of
him that he now knows. (For a similar suggestion, see Vlastos,
"Anamnesis in the Meno/' 153 n. 14.) I count 8sdi, 3-4 [epistesetai],
d6, 9 as forward-referring; 81C9 as backward-referring. 86bi says that
the truth about the things that are is always in the soul, but that is not
Cambridge Companions Online © Cambridge University Press, 2006
224 T H E CAMBRIDGE COMPANION TO PLATO
to say that knowledge is; perhaps we always have the truth in our souls
in that we once knew, can come to know again, and indeed are predis-
posed to the truth though we do not now know it. Similarly, Leibniz
says that innate ideas are in us in something like the way in which
Hercules is in rough marble before it has been carved, because its veins
make it easier to carve a Hercules shape than various other shapes,- the
use of "in" is very weak. See G. W. Leibniz, "Meditations on Knowl-
edge, Truth, and Ideas," in Philosophical Essays, trans. R. Ariew and D.
Garber (Indianapolis: Hackett, 1989), 27.
Nicholas P. White, Plato on Knowledge and Reality (Indianapolis:
Hackett, 1976), 47//, attempts to disarm the seeming contradiction by
claiming that in those passages where Plato says that the slave has belief
but not knowledge, he is speaking with the vulgar. Given the impor-
tance the Meno attaches to the difference between knowledge and (true)
belief, this seems unlikely.
If Plato nowhere claims that anyone now knows, then he nowhere
replies to the paradox by denying (2). I do not think he believes (2) any
more than he believes (3)-at least, that is so if (2) is construed as I
construed it in the text above, rather than as it is construed in note 23.
But he focuses only on (3), and does not address (2) - because, I take it, he
wants to vindicate Socrates' claim to be able to inquire in the absence of
knowledge, and to encourage us (who, in his view, lack knowledge) to
inquire. Aristotle, by contrast, disarms the paradox by denying (2): We
can inquire into what we know in one way, so long as we do not know it
in some other way; indeed, Aristotle suggests that inquiry requires some
prior knowledge. See Posterior Analytics, i 1.
41 Contrast Leibniz, New Essays on Human Understanding, tr. and ed.
Peter Remnant and Jonathan Bennett (Cambridge: Cambridge Univer-
sity Press, 1981), book I, chap, i; and Dominic Scott, "Platonic Anamne-
sis Revisited," Classical Quarterly 37 (1987): 346-66, esp. 351-3. Nor,
contrary to Moravcsik, "Learning as Recollection," 59, 61-2, is the
theory of recollection a theory of innate beliefs or concepts. The theory
of recollection does accord us innate abilities, and it is sometimes sug-
gested that concepts or beliefs are abilities. But as Aristotle points out
(De An. ii i), there are two different ways of construing abilities. A child
is able to be a general in that she might become one under certain
circumstances when she grows up (this is what Aristotle calls a first
potentiality); and I am able to study Greek, even though I am not doing
so now, because I can do so immediately if I choose (a second potential-
ity). If one construes abilities as first potentialities, then concepts and
beliefs are not abilities; but that is the only way in which the theory of
recollection postulates innate abilities.
Cambridge Companions Online © Cambridge University Press, 2006
Inquiry in the Meno 225
Nor is the theory of recollection a theory of concept acquisition. It
explains, not how one acquires concepts, but how one can move from
concepts and beliefs to knowledge, how, given our various beliefs (how-
ever they are acquired in the first place, a question Plato does not ad-
dress), we tend to favor the true ones over the false ones. For this point,
see further my "The Object of Thought Argument," Apeiron 21 (1988):
esp. 137-42; and Scott, "Platonic Anamnesis Revisited."
For some discussion of innatist theories, see Stephen Stich, ed., Innate
Ideas (Berkeley: University of California Press, 1975).
42 As I noted (in note 40), 86a8 says that the soul is always in a state of once
having been in a learned condition; the perfect tense leaves open the
possibility that there was no process of learning (contrast Vlastos's trans-
lation of the passage, "Anamnesis in the Meno," 153 n. 14). The claim
that it is always true of the soul that it earlier knew also suggests that
there was no initial stage in which it lacked, and then acquired, knowl-
edge. Even if (contrary to my suggestion) the soul did go through an
initial process of learning, a vicious regress can be avoided - even if the
process was simply (elenctic) inquiry all over again. Plato might argue,
e.g., that when the soul was discarnate, it was not hampered by percep-
tion and bodily desires; without such distractions, it could acquire
knowledge through inquiry even if it did not know in some still earlier
life. On this account, the theory of recollection would be introduced to
explain not how inquiry is in general possible, but how inquiry is possi-
ble in this life, or when incarnate.
Plato's claim that the soul "saw" various things is sometimes thought
to suggest that the soul acquired its previous knowledge through some
sort of acquaintance,- see, e.g., R. S. Bluck, ed., Plato's Meno (Cambridge:
Cambridge University Press, 1961), 286-7; Vlastos, "Anamnesis in the
Meno/' 164-5; Harold Cherniss, "The Philosophical Economy of the
Theory of Ideas," American Journal of Philology 57 (1936): 44 5-56. How-
ever, I take it that we saw all things in that we saw their point, i.e.,
understood them; for this sort of interpretation of Plato's visual vocabu-
lary, see J. C. B. Gosling, Plato (London: Routledge & Kegan Paul, 1973),
chap. 8; Burnyeat, "Wittgenstein and De Magistro," esp. 19-21.
Other attempts to argue that in the Meno Plato accepts some sort of
acquaintance model of knowledge are similarly weak. For example, at
71D4-7, Plato illustrates PKW by saying: "Does it seem to you possible
for one who does not at all know who Meno is [Menona me gignoskei to
parapan hostis estin) to know whether he is fine or wealthy or well-born
or the opposites of these?" It is sometimes inferred that for Plato, all
knowledge is like knowledge of persons which, it is assumed, consists in
or involves acquaintance; see, e.g., Bluck, Plato's Meno, 213-14. But
Cambridge Companions Online © Cambridge University Press, 2006
226 THE CAMBRIDGE COMPANION TO PLATO
Plato speaks here, not of knowing Meno, but of knowing who Meno is;
and it is not at all clear that I need to be acquainted with Meno to know
who he is - 1 know who he is from having read Plato's dialogues. (The
syntax is "know Meno who he is"; but the natural sense of the phrase is
"know who Meno is." White seems to agree, though he also argues that
for Plato, one can know who Meno is only if one knows Meno, by being
acquainted with him,- see Plato on Knowledge and Reality, 36-7, 54 n.
8.) Or again, at 97a9~b3, Plato seems to suggest that someone can know
the road to Larissa only if she travels along it (although he does not
actually quite say so); this too is sometimes thought to suggest that all
knowledge involves some sort of acquaintance. But even if one in some
sense needs to be acquainted with a route to know it, it does not follow
that knowledge in general requires acquaintance. Plato's point is that in
order to know something, one must have some sort of first-hand under-
standing or experience. In the case of a route, this first-hand understand-
ing may require traveling along it (a not implausible claim in Plato's day,
when there were no detailed road maps), and so in some sense it may
require being acquainted with it; but in other cases, understanding will
be gained by independent thought and reflection, which does not involve
acquaintance in any interesting sense. For this point, see Burnyeat, "Soc-
rates and the Jury" and "Wittgenstein and De Magistro." Further, if, as I
suggested, Plato's reply to the eristic paradox presupposes a rejection of
an acquaintance model of knowledge (see notes 24 and 31), then we
should be reluctant to saddle him with it elsewhere in the dialogue,- and
there is no need to do so.
43 The passage can be read to say that one of the few things Socrates
actually claims to know is that knowledge and true belief differ, in
which case it provides even stronger support for my view. But if the
passage is taken that way, then Socrates would be claiming, contrary to
PKW, that he knows something about knowledge and true belief (that
they differ) without knowing what they are. Although, on my view,
Socrates does not flatly claim to know that knowledge and true belief
differ, he expresses considerable confidence in that claim, saying that he
does not issue it on the basis of eikasia. Perhaps he thinks he has pistis
about it. (On the differences between eikasia and pistis, see Rep. VI-VII
and my "Knowledge and Belief in Republic V-VIL") I am indebted to
Hannes Jarka for discussion of this passage.
Cambridge Companions Online © Cambridge University Press, 2006
MICHAEL L. MORGAN
7 Plato and Greek Religion
Religion permeated life in classical Athens and in classical Greece
generally.1 It is hardly surprising, then, that religious vocabulary -
mention of gods, festivals, beliefs, and rites - also pervades Plato's
dialogues. These dialogues reveal a man struggling to understand
human life and how it ought to be lived, a man engaged in deep
reflection about rational inquiry, the human roles in society and in
the cosmos, and man's relationship to the divine. Religion, as rite,
conception, motif, and vocabulary, is integral to his thinking. By
showing how this is so we can illuminate Plato's thinking from the
religious side, as it were, and thereby exhibit Plato's relationship to
Greek religion and piety.
It is hard to exaggerate the prominence of religion in Greek life.
Greek religion was pluralistic and heterogeneous,- there was a host
of divinities with overlapping roles and features. A dozen gods
formed the conventional core of this pantheon (Zeus, Hera, Posei-
don, Athena, Apollo, Artemis, Aphrodite, Hermes, Demeter, Diony-
sos, Hephaistos, Ares); broadly conceived as Zeus's family, these,
and lesser divinities such as the goat-god Pan, are the Olympians,
so called after Mount Olympus, site of Zeus's palace. They and
other gods, such as Hades and Persephone, were themselves varied
and multiple, each present at dozens of places in various guises,
serving a variety of purposes and roles. Zeus, for example, manifest
as a thunderbolt, was the strongest of the gods and the father of
gods and men.2 But in fact there were many Zeuses present in
many places and with many specifications - for example, "Zeus of
the city," "Zeus of the stranger," "Zeus of boundaries," and "Zeus
of the mountain tops."^ Greek polytheism, then, incorporated a
plurality of gods, each with many domains and roles. At the local
227
Cambridge Companions Online © Cambridge University Press, 2006
228 THE CAMBRIDGE COMPANION TO PLATO
level, there was centralization and continuity, for the family, the
phratry (a subdivision of a tribe, of which there were originally four
in Athens), the deme (the local community of which each Athenian
was a citizen), and the polis. And there was some weak unity at the
international level, through the celebration of certain Panhellenic
festivals like the Great Panathenaea, held every four years, and the
Olympic games, and through the international use of oracles, pri-
marily at the sanctuary of Apollo at Delphi and the sanctuaries of
Zeus at Dodona and Ammon. Moreover, Homer and Hesiod were
universally honored to some degree by all Greeks.* Nonetheless,
Greek religion was disparate and diverse. The unifying factors of
Greek religion notwithstanding, not all gods were worshiped every-
where and where they were, their character and status differed.
This tremendous plurality and variety in part enabled Greek religi-
osity to be so pervasive and so complexly interconnected. In classi-
cal Greece everything - politics, ethics, science, painting, music,
dance, drama, agriculture - had a religious character.*
Each of the twelve months was filled with festivals of varying
degrees of significance,6 from the monthly festival celebrating the
new month to deme or phratry festivals and general Athenian festi-
vals. Perhaps as many as half of the days of the Athenian year in-
volved festivals and their processions, sacrifices, dancing, hymns,
and competitions. It is clear that festivals and new moons shaped
the calendar and that the Greeks lived from festival to festival. Each
month was named after a festival, some minor and some major. The
month of Thargelion, for example, was named after the Thargelia, a
celebration of Apollo that included the creation of a scapegoat
(pharmakos) and the offering of a pot of boiled grain and vegetables
to the god. But the months contained many other festivals as well;
another one that took place in Thargelion was introduced into Ath-
ens in 429 B.C. It was a celebration of Bendis, a Thracian goddess,
akin to Artemis, the huntress, and held in the Peiraeus, the Athe-
nian port; it involved a procession and a torch race on horseback, a
novelty, so we are told.? Indeed, it is on the occasion of the inaugural
Bendidia that Plato sets the Republic, when Socrates, after worship-
ing and seeing the sights, meets Cephalus and his son Polemarchus
on the way back to Athens and is invited to their home.
Sacrifice was the central feature of Greek religious life. Oxen,
sheep, goats, and pigs were the most common victims of such acts of
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 229
ritual slaughter and communal consumption, acts that were con-
ducted constantly, some for the polis as a whole, some for the deme
or the phratry, and some for the family.8 Gifts were continually being
offered to the gods; solidarity of the community was thereby secured
and the proper relationship between gods and humans was estab-
lished.? Oracles too were consulted for advice and counsel, and acts
of divination were performed by priests, local diviners, and less digni-
fied peddlers of prophecy.10 Temples, with their sacred trees and
boundary stones, sacrificial altars, statues, and cult images, were
numerous,- herms (small stone representations of Hermes placed out-
side Athenian homes) were everywhere. In any given year the aver-
age Athenian would participate in hundreds of religious acts and
inhabit thousands of regions of religious space. In a sense, his entire
world of time and space was a religious one, a complex, variegated
symbiosis of land and architecture in which divinity was perva-
sive.11 His life and the writings of his literary tradition expressed
this sense of divine presence, of divinity that was both ubiquitously
intimate and yet awesome and separate.
Let me focus on two features of the religious life of Plato's time:
the prominence of new cults and the existence of a common theol-
ogy underlying the traditional pluralistic religion. Many new cults,
rites, and practices flourished in Athens in the late fifth century B.C.,
during and after the Peloponnesian War (431-404), that traumatic
struggle between Athens and Sparta that culminated in Athens's
defeat and the ruin of Athenian hegemony and the Athenian Empire.
To be sure, Greek religion was always changing and especially in
Athens. Throughout the fifth century, Athens was a major importer
and exporter of forms of religiosity. The spiritual movement of
Orphism, with its mythic poetry of the origins and fate of human-
kind, its emphasis on mystery rites of initiation and purification,
and its associations with Pythagoreanism, came early in the century.
The cult of Pan arrived following the battle of Marathon (490). The
Eleusinian Mysteries, with their own Orphic elements and their
associations with Dionysos and Bacchic rites of purification, became
an Athenian festival in the first half of the fifth century.12 All of this
is true. But still it is clear that the plague, isolation, and self-doubts
of the war years, the fears and anxieties of the lingering siege and
ultimately of the premonition and reality of defeat, led to a prolifera-
tion of novel religious forms.
Cambridge Companions Online © Cambridge University Press, 2006
230 THE CAMBRIDGE COMPANION TO PLATO
We have already mentioned the celebration of Bendis, imported
when the war had just begun. There was also the worship of
Asclepius, son of Apollo and god of healing, imported from Epi-
daurus during the war and temporarily housed in Sophocles' resi-
dence. There is also evidence of increased interest in various forms
of ecstatic rituals and personally oriented salvation rites, Bacchic
(associated with Dionysos) and corybantic rites among them.1* The
central feature of these practices was an aspiration to purification
and catharsis from earthly ills, to transcend the physical world, to
achieve an ecstatic kinship with the divine, to become in some
sense divine, what Plato in the Theaetetus calls homoiosis theo
(i76bi). These rites often involved the use of wine and erotic stimu-
lation to bring the celebrant to a state of raving or frenzy. Their goal
was to gain release from distress and the pressures of physical life
and to achieve a postcarnate blessedness. This aspiration was specifi-
cally associated with the immortality of the soul, metempsychosis,
and the achievement of divinity, as we find, for example, in Pindar's
second Olympian Ode. Walter Burkert compares Pindar's poetic
paean to a blessed afterlife with similar thoughts in Herodotus, Em-
pedocles, and Plato, thus showing how pervasive these ideas were in
the late fifth century.1*
This concern with the state of divinity as a real human possibility
was exemplified in other developments, such as the prominence of
charismatic, almost shamanistic religious figures like Empedocles
and Pythagoras, and the deification of heroes, an old tradition now
newly revived in important ways.1* These developments, of course,
did not all involve the human aspiration to divinity in the same way.
But they did contribute to the existence of a context in which the
gap between gods and humankind was conceived as frequently tra-
versable. One key to this set of changes, which Burkert calls a revolu-
tion, was the belief in the immortality of the human soul.16
In Athens, then, there was, at least during the fifth century, a
dominant religious attitude and a variety of alternative religious
styles, such as mystery cults, ecstatic rites, and salvific societies
organized around charismatic leaders. This dominant religiosity is
what Burkert has called the "polis tradition."1? It was a conglomer-
ate of traditional religious practice - sacrifices, festivals, oracles,
divination, and more - that hardly had a single, uniform theology.
Religious belief and mythology in classical Greece were as complex
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 231
and pluralistic as was religious practice. Even amid this pluralism,
however, with its world of separated, powerful, and immortal dei-
ties, we can distinguish a common theological posture, one suc-
cinctly framed by the slogans associated with the Pythia, the oracle
at Delphi: nothing too much, and know thyself. What these maxims
meant was that human beings should recognize their limitations
with respect to the gods: The gods are immortal, perfect in knowl-
edge, and exceedingly powerful; human beings are mortal and lim-
ited in power and knowledge. Human beings should not want more
than they as humans should; they should not overestimate their
knowledge or capabilities, nor should they confuse who they are and
who the gods are. Underlying the world of polis religion, then, was
this theological attitude of separation between the divine and the
human, of discontinuity, of human limits and hence of the tempta-
tion to illicit self-esteem and pride (hubris). I call it the Delphic
theology.18
Contrasted with this posture was the attitude of those committed
to the alternative religious styles that involved ecstatic rites and
salvation-oriented cults. Unlike traditional Athenian piety, this atti-
tude assumed that there was continuity between the human and the
divine-for example, that both were immortal, and that the gap
between them could be bridged by the divine possession of human
beings (as in shamanism) or by human attainment of the status of
divinity or by both.1* In short, the Orphic-Bacchic-Pythagorean-
Eleusinian world assumes that relief from our physical world and its
distress could be achieved by human beings becoming as completely
divine as they could possibly be. There is an element in human life,
the soul or psyche, that has a quasi-divine nature; it is immortal.
And that element, through ecstatic ritual performance or perhaps
through a life of ecstatic practice, could grow stronger and aid in the
attainment of salvation.
Let us locate Plato in this religious setting and in this historical
context. We find someone trying to conceptualize and articulate an
understanding of the good life and its relation to the philosophical
life. In so doing Plato appropriates the two religious tendencies to
which we have called attention, but in both cases his appropriation
is qualified. At certain points he accepts but also criticizes the polis
tradition of festivals, sacrifices, oracles, and so forth. At other times,
however, Plato adopts the alternative mode of piety, the mode that
Cambridge Companions Online © Cambridge University Press, 2006
232 THE CAMBRIDGE COMPANION TO PLATO
includes mystery cults and ecstatic rites of initiation, purification,
and salvation. In adopting this latter mode, moreover, Plato modifies
it significantly. The model is ecstatic insofar as it involves a kind of
transformation whereby the soul of the human initiate steps out of
its physical environment and becomes purified of worldly distress; it
thereby gains a kind of divine blessedness and becomes as divine as
it can be. The soul, in a sense, crosses the divide that separates the
human from the divine; already divine to a degree, it seeks to perfect
its divinity. In the Orphic-Bacchic rites, this transformation is
achieved through a process of emotional excitement, induced by
music, dance, and other means. Plato accepts the ecstatic model,
that human beings can, by bringing their souls to a certain state,
achieve divine or nearly divine status. But he replaces the emotional
character of the ritual process with cognitive content. For Plato, that
is, a life aimed at salvation takes the form of a life of rational inquiry,
a philosophical life.
Once he realizes this, Plato then develops an epistemological and
metaphysical view of what such inquiry requires in order for it to
occur; this is Plato's attempt to understand what philosophy is. It
involves in part showing that inquiry aims at knowledge of divine
objects, the Forms, and that the gaining of such knowledge makes
the soul more and more like these objects. Platonic learning, then, is
an ecstatic ritual process because it is precisely organized, reli-
giously motivated by the desire to become divine, and facilitated by
the assumption that the human soul, which is immortal, can be-
come divine or nearly divine. The result of this Platonic appropria-
tion of the ecstatic model, then, is a conception of philosophy as a
lifelong quest for salvation.
What I have just said, of course, is only a proposal, but it is one
that gains support when we look at the dialogues for confirmation.
Here I will only be able to make some modest steps in that direction.
To begin, what features of Socratic piety does Plato choose to
emphasize? Socrates is Plato's model of the philosopher and the
philosophical life. In his early dialogues Plato portrays him as a
tenacious elenctic interrogator, a moral philosopher, and a devoted
Athenian citizen. In later dialogues, when Socrates also appears as
a dramatic participant, Plato's portrait changes. Often Socrates is
no longer engaged in elenchus, and his interests and views become
much broader, including mathematics and metaphysics, politics
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 233
and psychology. It is possible, then, that there is a shift in the
dialogues. That shift may be either from a historically attentive
portrait to one that employs Socrates as a Platonic mouthpiece or
from an earlier to a later Platonic perception of Socrates. For now,
we do not have to decide which is the case. By the time we reach a
middle dialogue such as the Symposium, on either reading, we
clearly have Plato's perception of Socrates as he thinks Socrates
suits Plato's own views and interests. Hence, in the Symposium
Plato can be taken to have portrayed Socrates religiously in a way
that he finds congenial to his own thinking. How does Plato draw
this portrait of Socratic piety?
In the Symposium we should notice two features of Plato's Socra-
tes. First, the dialogue, like the Phaedo, is a eulogy to Socrates and
the Socratic life. But in the central passage, Socrates does not engage
in his characteristic elenctic interrogation; rather he reports a con-
versation between himself and Diotima, a female diviner from
Mantinea, in Arkadia.20 In the first part of their conversation
Diotima plays the role of a Socratic interrogator in a dialectical
exchange about the nature and effects of Love (Eros). Then, at a
crucial juncture, when Socrates expresses his bewilderment about
the implications of their results, Diotima jettisons dialectic in favor
of soliloquy and presentation, indeed revelation. Moreover, what the
prophetess reveals in her speech to Socrates is an account of the
ascent of desire and love that culminates in the grasping of the Form
of Beauty and then spills over into virtuous conduct, what she had
earlier called "procreation in beauty" (tokos en kalo, 2o6b7-8).
Diotima's description of the ascent of philosophical desire uses
vocabulary from the mystery rites - myesis, epopteia, and orgia-
zein - and it is widely agreed that Plato's model for this account
was the Eleusinian Mysteries. These famous rites were celebrated
in the month of Boedromion by large numbers of Athenian and
non-Athenian initiates (mystai) and included a spectacular proces-
sion (pompe) from Athens to Eleusis, fourteen miles along the Sa-
cred Way. The parade passed through several stages, culminating in
an initial entrance of the mystai into the temple, the Telesterion.
Later, the initiates reentered the temple and received, in the deep-
est recesses of the sanctuary, a final, secret revelation (in the words
of Symposium 2ioai, ta telea kai epoptika)-, the Hiera (sacred
things) included things enacted [dromena, possibly the sacred mar-
Cambridge Companions Online © Cambridge University Press, 2006
234 T H E CAMBRIDGE COMPANION TO PLATO
riage), objects shown (deiknymena, possibly an ear of grain), and
words spoken (legomena).
In the central passage of the Symposium, then, Plato, using termi-
nology reminiscent of the Eleusinian Mysteries, depicts Socrates
simultaneously as the interlocutor in a preliminary elenchus, as the
recipient of a religio-philosophical teaching, and as an initiate in a
mystery rite. Moreover, the content of Diotima's teaching shows
that Socrates is being initiated not into an episode of emotional
frenzy and psychological disorder, but into the practice of philoso-
phy and the philosophical ascent to a kind of knowledge that will
result in cognitive contact with the Forms.
This picture of Socrates as a philosophical initiate is comple-
mented by Alcibiades' remarkable, inebriated reminiscence of Socra-
tes (Smp. 2i5b-2i6e). Alcibiades, a pupil and associate of Socrates,
had been a brilliant but unscrupulous Athenian general and states-
man. Notoriously, he was implicated in profanations of the Eleus-
inian Mysteries that supposedly took place in his home in the sum-
mer of 415, events alleged to have enraged the gods and thereby led
to the failure of the Sicilian expedition and Athens' demise.21 Here,
in the Symposium, set dramatically at the party celebrating Aga-
thon's victory at the dramatic competition at the festival of the
Lenaia in 416, Alcibiades enters drunk and proceeds to portray Socra-
tes as a Bacchic satyr whose ugly exterior hides a divine core. Accord-
ing to Alcibiades, Socrates throws people into a trance by use of
words alone and turns them into corybantic celebrants. Once, more-
over, while on a military campaign, Socrates himself remained in a
daylong trance, engaged in silent inquiry, immune to cold and ice,
showing his true character and his intense devotion to wisdom and
truth. Plato, in short, has Alcibiades exhibit a Socrates very much
like a Bacchic celebrant. Like them, Socrates prepares for a postcarn-
ate journey, seeks a final blessed state, and experiences a trancelike
detachment. Unlike them, he eschews wine and drunkenness (the
counterpoint with Alcibiades' own drunkenness is clear), music, and
frenzied dance; instead he engages in intense rational thought. He is,
therefore, an ecstatic of a unique sort, indeed of just the sort that I
earlier described.
The ascent passage in the Symposium does not of course give a
complete account of philosophical inquiry. It is primarily about the
desire for knowledge and beauty and does not deal with the cogni-
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 235
tive stages of inquiry that begin with belief and culminate in knowl-
edge. Plato attends to these matters in the Meno, Phaedo, and Repub-
lic. I have suggested that Plato's conception of this process, and
indeed of the philosophical life, is religious and particularly ecstatic.
This means that not only does he characterize it in terms drawn
from mystery rites and ecstatic practices; he also develops his ac-
count of inquiry and education in part from these traditions and in
the end takes philosophy to be a form of such initiation rites. If this
is so, the dialogues should show that the philosopher moves through
systematically related steps in order to gain divine wisdom, thereby
becoming virtually divine himself. The soul, that is, which is immor-
tal by nature, becomes intelligent and as much like the divine,
which is eternal, stable, and pure, as human beings can be.
We have already sketched some features of the alternative, ecstatic
type of piety that Plato is adopting. From many sources, including
Plato himself in dialogues such as the Euthydemus, Ion, Phaedo,
Republic, and Phaedrus, we gain a rich understanding of the prac-
tices, cults, and theoretical foundations of the ecstatic tradition in
classical Greece. What we learn is this. The central divinity associ-
ated with the Orphic writings and mystery rites was Dionysos, the
god of wine and ecstasy. As one of the twelve central Olympian dei-
ties, Dionysos was widely worshiped and in Athens was the object of
several festivals: the Anthesteria, the Lenaia, the rustic Dionysia, and
the Great Dionysia.22 The dark side of Dionysos involved worship of a
heterodox kind; this worship first developed in Italy and became wide-
spread on the Greek peninsula as the mysteries of Dionysos. Some-
times the frenzied worshiper went mad for Dionysos alone and in
private, but more frequently there was a group or religious commu-
nity (thiasos), especially of maenads (female celebrants) and male
satyrs, who danced to wild music and used other means to achieve a
delirious state (baccheia). The goal of all this was a "change effected
in the soul (psyche)" in which the soul was purified and gained new
powers appropriate to its blessed state.2^ Plato, in the Phaedrus
(249d-256e, esp. 254b-c), portrays this state as a kind of amazement
or stunned recoil in which the celebrant becomes immobilized and
also as a kind of madness (mania) by means of which misery is cured
(244d5-245ai). At its most extreme - one recalls the fate of Pentheus
in Euripides' Bacchae — Bacchic worship resulted in omophagy, the
eating of raw flesh as an act of ingesting the god.2*
Cambridge Companions Online © Cambridge University Press, 2006
236 THE CAMBRIDGE COMPANION TO PLATO
A central feature of these mystery rites and the Orphic-
Pythagorean conglomerate associated with them was the belief in
the soul's immortality. In the Homeric poems and elsewhere, until
the fifth century, the dominant view of soul (psyche) was that of a
complex of features and functions associated with different parts of
the body, with dreams, trances, and such phenomena, and with
death. Although it is clear that Greeks of the Archaic period could
say "I," it is doubtful that they had the notion of a unitary soul
that was the locus of conscious events.2* Nonetheless, even in Ho-
mer, there is the view that some kind of soul, a shadow or vaporous
image of the body, continues to exist after death. But reincarnation
only seems to have become prominent through the teachings of
Pythagoras of Samos in the late sixth century and then in the
Orphic texts of the fifth century. It was in this Orphic-Pythagorean
context, moreover, that the soul was taken to be immortal and
divine. Whether the Homeric poems entertain the possibility of the
soul's immortality is doubtful, but by the fifth century the idea was
available, if not yet widely accepted.26
Plato clearly did accept it. In the Meno Plato shows that his under-
standing of inquiry or learning was closely tied to the belief in the
soul's immortality. At Meno 8od, frustrated by the aporiai ("perplexi-
ties") that have arisen out of Socrates' interrogation about the nature
of arete ("excellence"), Meno puts forth a puzzle about inquiry. Soc-
rates rephrases the puzzle into a dilemma: If one knows what one is
seeking, there is no point in searching for it, and if one does not
know it, search cannot be directed to it; therefore inquiry is impossi-
ble, for it cannot be initiated. Socrates then replies to this paradox by
offering a teaching that he claims to have learned from "priests and
priestesses who make it their business to be able to give an account
concerning what they do"; Pindar and other poets who are theoi
("divine") also hold the view.
The teaching is presented in two parts, a poetic quotation and
Plato's account of what the quotation implies to him about inquiry
or learning. For our purposes we shall forgo detailed discussion of
the meaning of the paradox, the doctrine of learning as recollection,
and the questioning of the slave that follows.2? Two points are impor-
tant. First, Plato is decisive about the connection between the doc-
trine that he finds in the quoted text, that learning is really recollec-
tion of truths that are always present in the soul but not in one's
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 237
mental grasp, and the soul's immortality. He believes that the
former position entails the latter; the desirability of accepting the
account of learning as recollection counts in favor of accepting the
immortality of the soul. Moreover, this text is not the last in which
Plato advocates and argues for the soul's immortality. The Phaedo of
course contains a series of such arguments, and there are further
examples in Republic X and the Phaedrus. In at least one argument
in the Phaedo and again in the Phaedrus, the soul's immortality is
explicitly tied to Plato's understanding of inquiry and learning. Here
in the Meno the tie is secured by Socrates' affirmation, as this
stretch of dialogue concludes, that "if the truth of things that are is
always in the soul, then the soul must be immortal" (86b 1-2). Learn-
ing as recollection solves the paradox of inquiry, and it does so only
if its objects are in the soul always. But this permanence requires,
Plato thinks, the soul's immortality, which means that Plato can̂
only save the possibility of inquiry if the soul is immortal. Hence,
there are good philosophical reasons for him to adhere to the Orphic-
Pythagorean doctrine.
The second important point is that Plato tells us that he is draw-
ing on Orphic-Pythagorean sources for these twin doctrines. He cites
a fragment from Pindar, the fifth-century poet influenced by Orphic
teachings and mystery rites:
[Those from whom] Persephone receives requital for ancient grief,
In the ninth year she restores again
Their suns to the sun above,
From whom arise noble kings,
And men mighty in strength and greatest in wisdom,
And for the rest of time
They are called heroes and sanctified by men.
(8ib7-C4; trans. W. K. C. Guthrie)
Here Pindar alludes to the Orphic myth recounting Dionysos's birth.
Burkert tells it this way:
Zeus raped his mother Rhea-Demeter and sired Persephone; he raped Per-
sephone in the form of a snake and sired Dionysos. To the child Dionysos he
hands over the rule of the world, places him on a throne, and has him guarded
by Korybantes. But Hera sends the Titans who distract the child with toys,
and while the child is looking into a mirror he is dragged from the throne,
killed, and torn to pieces, then boiled, roasted, and eaten. Zeus thereupon
Cambridge Companions Online © Cambridge University Press, 2006
238 THE CAMBRIDGE COMPANION TO PLATO
hurls his thunderbolt to burn the Titans, and from the rising soot there spring
men, rebels against the gods who nevertheless participate in the divine. From
the remains that were rescued and collected, Dionysos rises again.28
It is a well-known myth - Plato alludes to it again in the Cratylus
(400c) - and one that clarifies "the ancient grief" for which Per-
sephone receives requital; that is, when people die, their souls are
held by Persephone beneath the earth, as compensation for the
crime of their Titanic ancestors. Then, however, she allows these
human souls to become reincarnate, and from them come heroes,
kings, and "men of wisdom." Plato latches onto this last phrase. The
soul is immortal and undergoes reincarnation; from this experience
it has learned all that is,- hence, when it engages in inquiry and
learns, what is really occurring is a process of recollection. The men
of wisdom and indeed wisdom itself - knowledge - depend upon the
soul's immortality and the conception of learning as recollection.
But immortality is a credential of divinity, as is perfect wisdom. By
becoming heroes and men of wisdom, those who inquire - later in
the Phaedo and Republic Plato will call them "lovers of wisdom
[philosophoi]" - enhance their divinity.
At Meno 86b, near the conclusion of his response to the paradox,
Socrates says, "I shouldn't like to take an oath on the whole story,"
and it is a crux how much of the previous account Plato would
disavow. My sense is that he would have held to his understanding
of inquiry and the belief in the soul's immortality; the myth of the
Titans and its details are another matter. This is after all myth, and
it includes some horrifying acts, just the type that later, in the Re-
public, Plato will criticize. In the Meno, then, Plato's resources are
Orphic-Pythagorean,- he both appropriates and rejects features of
that inheritance.
In the Phaedo, Republic, and Phaedrus, Plato elaborates his ac-
count of rational inquiry and learning, his understanding of the soul's
nature and its immortality, the religious character of the philosophi-
cal life, and the ways in which the polis tradition of Greek religion can
be adapted to his purposes. The Phaedo contains several arguments
for the soul's immortality and employs a variety of expressions from
Orphic and Bacchic rites to characterize the philosopher. Early in the
dialogue, for example, Plato compares philosophical thinking to puri-
fication and philosophers to Bacchic initiates (bacchoi, 69C8-12). His
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 239
religious terminology in the Phaedo is Bacchic, Orphic, and Pythago-
rean all at once. In addition, Plato here first introduces the Forms as
unchanging, pure, eternal objects of knowledge; some of their fea-
tures arise from his critique of physical items and characteristics as
proper objects of knowledge and from his critique of sensory experi-
ence as a vehicle for genuine inquiry, but some derive from the need
for such objects to be of divine status.2*
The Republic does not use the vocabulary of ecstatic ritual as
explicitly as the Phaedo, but the general framework is still present
in Plato's mind. One might interpret his conception of philosophical
education in Books VI and VII, a conversion (pehagoge) of the soul as
it passes through the stages of the mathematical curriculum on its
way to a dialectical understanding of the Forms, as Plato's Pythago-
rean revision of the stages of the Eleusinian Mysteries. Moreover, in
the Republic, Plato addresses the role of religion in the polis in a
number of ways. In Books II and III, for example, he inveighs against
bogus healers and charlatans who peddle Orphic placebos to the
superstitious; he also offers his famous critique of Homeric religion
and the Olympian pantheon, charging that the disgusting and im-
moral acts in which the gods are depicted are poor models for the
young and hence inappropriate models for public education. In Book
X, finally, Plato argues for the soul's immortality and then presents a
myth of transition, in which he shows how the conduct of one's life
influences subsequent incarnations and hence how worldly virtue
has its importance, even within the context of the philosopher's
ultimate otherworldly goal.
The Phaedrus culminates this set of dialogues. Within the frame-
work of an account of interpersonal love, Plato argues for the soul's
immortality once again and then constructs an extraordinary myth
of the soul and its experiences and aspirations, of personal psychol-
ogy, cognitive resources, and philosophical method. It is a portrait of
philosophy as a special kind of mania, akin to but distinct from the
madness of poets, oracles, mantics, and cathartic initiates. Here
again, as in the Phaedo and Symposium, Plato uses the vocabulary of
ecstatic ritual to characterize the soul's vision of the Forms. More-
over, with more detail than elsewhere, Plato describes phenomeno-
logically the soul's experience of rationally grasping the Forms.
There is a further aspect of Plato's appropriation of Greek religios-
ity that takes us beyond the Phaedrus. Plato's thinking about the
Cambridge Companions Online © Cambridge University Press, 2006
24O THE CAMBRIDGE COMPANION TO PLATO
nature of divinity takes two paths. On the one hand, he identifies as
divine the highest objects of knowledge and rational aspiration, the
Forms, and he conies to associate one kind of divinity with such
generic features of the Forms as changelessness, purity, and simplic-
ity. On the other hand, Plato figures in a tradition that criticizes the
Olympian gods and seeks to rethink the notion of divinity in terms
of our understanding of the cosmos and especially of soul, life, and
motion. In Laws X, Plato pursues this second line of inquiry as he
tries to say something about impiety and religious ritual in the ideal
polis.
This is what we might call Plato's natural theology. Natural theol-
ogy involves arguing for the nature and existence of the divine from
the understanding of nature. It is an attempt to show that nature,
and especially change or order, requires the divine as an ultimate
causal explanation, and hence natural theology seeks to show the
continuity between science and religion. The early Greek natural
philosophers, the first figures in this tradition, explored the nature of
divinity, ho theos, by associating it with a variety of attributes (con-
trol, power, indestructibility) and identifying the ways in which di-
vinity occurred as natural substances, like air and fire.3° Such natu-
ral divinities function in a variety of ways in relation to the cosmos
and to human aspiration. Heraclitus, for example, says that thunder-
bolt (fire) steers all things, Xenophanes says that the one god moves
all things with the effortlessness of its mind. Plato, in many ways,
figures in the tradition that includes such thinkers, together with
Parmenides, Anaxagoras, and Empedocles. In the Republic (II 380c-
383c and Book X) and Phaedrus [24SCS-2463.2), for example, he re-
flects on the causal and hence providential dimensions of divinity,
how divinity is responsible for motion, life, and goodness in the
natural world. 3 * But these developments, which later are manifest in
Aristotle's accounts of the unmoved movers in the Metaphysics and
Physics, reach new heights in the Timaeus and the Laws as Plato
ties together divinity, motion, and soul. Here I restrict myself to
some comments on the Laws.
The argument that Plato offers in the Laws for the existence of the
gods, reminiscent of the Phaedrus (the argument for the soul's immor-
tality at 245C5-246a2) and an anticipation of Aristotle's argument in
the Physics, associates divinity with priority, self-sufficiency, mo-
tion, and life.*2 Plato's reasoning is of course focused on soul as self-
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 241
generating motion, but he is explicit that soul is a divinity (897b2; cf.
899a7-ci). Indeed, Plato goes further: Heavenly motions, like rotary
motion, reflect the order of rationality;33 hence it must be "the best
kind of soul [i.e., rational and supremely virtuous] that cares for the
entire universe and directs it along the best path" (897C7-9). Each
rational soul that guides a heavenly body in its orbit, moreover, is a
god. As Burkert puts it, "astronomy becomes the foundation of reli-
gion. "34 Plato then concludes that in order to cope with his argument,
one must either deny the motive priority of soul, reject his reasoning,
or agree to believe in the existence of gods.
Even in the face of such reasoning, there is contrary evidence. The
apparent "good fortune of scoundrels and criminals in private and
public life" (899d8-ei) compels many to doubt divine providence
and to contend that the gods are indifferent to human affairs.35 To
show that this result is false Plato draws on a feature of the previous
argument, a feature that recalls a famous caveat in Republic II, that
the gods are good, for they possess moderation, rationality, and order
and in no way their opposites.36 Being good, fully capable, and atten-
dant to details (Laws 9ooc-9O3b), then, the gods must be thought to
care for each individual through "the control of ruling powers that
have perfected the minutest constituents of the universe" (9O3b7-
9). We need not attempt a detailed discussion of Plato's psychologi-
cal description of how this process works. Suffice it to say that he
believes it does: Individuals act according to psychological laws and
their actions contribute to the cosmic order and suit the laws of
destiny (9O4a-b). Eventually appropriate reward and punishment
come to all.
Finally, since the gods do indeed care for individuals and since
destiny is fixed by laws, one should not believe that sinners can buy
off the gods with gifts (9O5d). Those who believe this demean the
gods' affinity with justice and commit the most impious of acts.
Reasoning, then, should persuade citizens of the worth of genuine
piety, that is, of as much piety as a normal citizen is capable of. In
the polis only a small number of citizens will be competent to attain
the richest level of ecstatic transcendence and philosophical piety
portrayed in the Phaedo, Republic, and Phaedrus. Others, however,
can at least achieve an understanding of what the Laws teaches
about the divine and about how to live in terms of such an account
of the gods. Many others can be made to conduct their lives in
Cambridge Companions Online © Cambridge University Press, 2006
242 THE CAMBRIDGE COMPANION TO PLATO
harmony with divine providence, even if they have no clear under-
standing of it.
In the Republic Plato had vilified Bacchic-Orphic charlatans who
proliferated in the late fifth and early fourth centuries and who sold
ritual panaceas like the phony medicines of itinerant quacks.^ He
also presented a revised and refined picture of education, music, art,
and dramatic poetry. In the Laws he levels his criticisms at those
who believe that the gods can be bought off and who become them-
selves "sub-human" who "take everybody for fools and delude many
a man during his life. . . . by promising to influence the gods through
the alleged magic powers of sacrifices and prayers arid charms, they
try to wreck completely whole homes and states for filthy lucre"
(909b; trans. Trevor Saunders). This is a group of fakes similar to
those attacked in the Republic, religious entrepreneurs who, in the
service of hubris and self-aggrandizement, capitalize on the remote-
ness of divinity.
What, however, is the political and cultural impact of this opposi-
tion between philosophical and nonphilosophical piety? Should a
polis abolish and prohibit shrines, altars, sacrifices, and so forth?
Burkert raises this question and gives Plato's response: "the religion
of the Platonic polis appears altogether familiar," with sanctuaries,
temples, altars, images, priests, seers, exegetes, sacrifices, prayers,
processions, and festivals.*8
Burkert of course is right. Plato's polis must be constructed in
strict accord with the oracular directives of Delphi, Dodona, or Am-
mon about the gods that ought to be worshiped and the temples that
ought to be founded (738b). A sacred acropolis is to be set aside as a
central precinct for worship of Hestia, Zeus, and Athena (745d), and
each of the city's twelve tribes, with its own festivals and temples,
will occupy a plot dedicated to its particular god (738d; cf. 77ic-d).
The law courts, marriage, childrearing, and much else are to be
conducted under divine auspices, and the polis has a full sacred
calendar, filled with festivals, competitions, processions, and all
their accoutrements. Much of this and more is certainly familiar and
normal for a fourth-century Greek polis.
But, we might ask, how can this be? And if it is, what is the
relation between the philosophical ecstasy that is the goal and orient-
ing experience of only a few and this traditional, widely accessible
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 243
religion of the polis? To answer these questions, Plato must first
explain why polis religion of the traditional kind is necessary at all.
Then he must show how both philosophical and nonphilosophical
piety are related to the state.
In the Laws Plato makes it clear that not all citizens are equally
equipped to become philosophers and members of the Nocturnal
Council, the supreme governmental institution in his polis. This
Council combines a variety of functions, among which is the articu-
lation of the state's moral aims based on philosophical inquiry into
human excellence (arete). Not all are capable of such inquiry and
hence of the moral knowledge to which philosophy aspires. Mem-
bers of the Council must be "particularly well qualified by natural
abilities and education" (96ia-b) and able to gain an "adequate un-
derstanding of virtue (arete)/' the goal of each person's soul. They
must also be articulate, open-minded, and committed to the truth.
Furthermore, this same distinction, between those who are and
those who are not suited to philosophy, applies to theological knowl-
edge. "The man in the street may be forgiven if he simply follows
the letter of the law, but if any intended guardian fails to work hard
to master every theological proof there is, we must certainly not
grant him the same indulgence" (966c; trans. Trevor Saunders).
Plato fixes precisely the content of what the philosopher must grasp,
that the soul is immortal and controls the material world and that
"reason is the supreme power among the heavenly bodies" (967d-e).
To orient their attention and practice to the gods in appropriate
ways, on the other hand, most people require a regimen of rituals
and celebrations. Plato is keenly aware of the needs of the masses
and of the excesses to which they are prone. This is especially clear
in his prohibition against private shrines or altars (9O9d3~9ioe4).
One must do what is necessary to prevent ordinary citizens from
responding to anxiety and distress by seeking the gods' help through
sacrifice and prayer.
Familiar Greek religious practices, then, are required for those
incapable of an accurate understanding of the divine. But how is this
mundane piety compatible with philosophy and with the state?
Let us distinguish between two views about the relation between
politics and religion. On one view, the two are continuous; politics
expresses the religious ideal and is aimed at implementing that
Cambridge Companions Online © Cambridge University Press, 2006
244 T H E CAMBRIDGE COMPANION TO PLATO
ideal. On the other, politics is independent of any particular reli-
gious conception of the good life; it seeks to facilitate individual
self-expression and to minimize conflict while avoiding advocacy.39
Both views are founded on the assumption that politics and religion
are distinct domains, and in a sense Plato would not have accepted
or even understood such a premise. But if he had, clearly he would
have been attracted to the former of these conceptions. The problem
is that he has more than one understanding of the religious life of the
citizens in the ideal polis. How can a single polis encourage and
indeed facilitate two ideals of the religious life?
The answer must be that the polis provides an institutional and
cultural environment in which both philosophical piety and non-
philosophical polis religiosity flourish and function. A regimen of
sacrifices, festivals, and celebrations for the ordinary citizens, filled
with expressions of praise and gratitude, serves to enhance political
life and to enable the good life to flourish for all. "Every man,"
Plato says, "must resolve to belong to those who follow in the
company of the divine" (7i6b8-9). But what is it to "follow in the
company of the divine"? For Plato, the polis citizen "follows in the
company of the divine" in two ways, primarily by being a good
person, moderate, wise, just, and so on, and secondarily through
praise, gratitude, and offerings of all kinds. Both together constitute
true polis piety. "If a good man sacrifices to the gods and keeps
them constant company in his prayers and offerings and every kind
of worship he can give them, this will be the best and noblest policy
he can follow" (7i6d6-ei). The philosopher also "follows in the
company of the divine," but he does so by living a life of rational
aspiration and study and by serving the polis, thereby becoming
like a god. In these ways, then, both philosophical and nonphilo-
sophical piety coexist in the state, and each, in its own way, leads to
a life with the gods.
I have tried to show how Plato's thinking is immersed in the very
complex, variegated phenomenon of Greek religion. He takes its
existence for granted, adopts features of it, adapts others, and rejects
much of it. His relation to Greek piety, however, is deeper than an
encounter with the world of Olympian deities, sacrifices, festivals,
and so on; Plato also appropriates aspects of Greek ecstatic ritual as
a framework for philosophical inquiry. It is this religious dimension
that helps to show what makes philosophy so important to him.
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 245
NOTES
1 The best recent comprehensive account of Greek religion is Walter
Burkert, Greek Religion, trans. John Raff an (Cambridge, Mass.: Harvard
University Press, 1985). One might also consult Martin Nilsson, Greek
Folk Religion (New York: Columbia University Press, 1940), Greek Pi-
ety (Oxford: Oxford University Press, 1948), and A History of Greek
Religion (Oxford: Oxford University Press, 1952); Jon D. Mikalson, Athe-
nian Popular Religion (Chapel Hill: University of North Carolina Press,
1983); and Robert Parker, "Greek Religion/7 in The Oxford History of
the Classical World, ed. John Boardman, Jasper Griffin, and Oswyn Mur-
ray (Oxford: Oxford University Press, 1986), 254-74.
2 "Most excellent and just among gods" (Euphr. se). Cf. Phdr. 246c
3 See Laws 843a, Euthd. 301b.
4 To see that not all Greeks revered Homer in the same way, one need only
recall Xenophanes (D.K., Bn), Heraclitus (D.K., B40), and Plato's critical
account of the Homeric poems in the Republic.
5 One need only remember the almost superstitious fear that spread after
the mutilation of the herms and the profanation of the Mysteries in 415
B.C. See C. Powell, "Religion and the Sicilian Expedition/7 Historia 28
(1979): 15-31; and Douglas M. MacDowell, Andokides: On the Myster-
ies (Oxford: Oxford University Press, 1962).
6 For discussion of Greek and Athenian festivals, see Burkert, Greek Reli-
gion; H. W. Parke, Festivals of the Athenians (Ithaca, N.Y.: Cornell Uni-
versity Press, 1977); L. Deubner, Attische Feste (Berlin: H. Keller, 1932).
7 See Parke, Festivals, 149—57.
8 Walter Burkert has a fascinating discussion of sacrifice and myth in
Homo Necans: The Anthropology of Ancient Greek Sacrificial Ritual
and Myth (Berkeley: University of California Press, 1983).
9 Burkert, Greek Religion, 55—75, and Homo Necans.
10 On oracles and divination, see H. W. Parke, Greek Oracles (London:
Hutchinson, 1967) and The Oracles of Zeus (Cambridge: Harvard Univer-
sity Press, 1967); and Joseph Fontenrose, The Delphic Oracle (Berkeley:
University of California Press, 1978).
11 Vincent Scully, The Earth, the Temple, and the Gods: Greek Sacred
Architecture (New Haven: Yale University Press, 1979).
12 See G. Mylonas, Eleusis and the Eleusinian Mysteries (Princeton: Prince-
ton University Press, 1961).
13 Corybantes were frenzied devotees of Kybele, the mother goddess from
Asia Minor,- they are often portrayed, in literature and vase painting, as
dancing deliriously to Phrygian music of the flute [aulos).
14 Burkert, Greek Religion, 299. Burkert emphasizes that Pythagorean, Or-
Cambridge Companions Online © Cambridge University Press, 2006
246 THE CAMBRIDGE COMPANION TO PLATO
phic, and Bacchic myths and rites, although distinct, do overlap: Greek
Religion, 300; Walter Burkert, Orphism and Bacchic Mysteries: New
Evidence and Old Problems of Interpretation (Berkeley: Center for Her-
meneutical Studies in Hellenistic and Modern Culture, 1977) and An-
cient Mystery Cults (Cambridge: Harvard University Press, 1987). See
also Parker, "Greek Religion," 263-4.
15 See Michael L. Morgan, Platonic Piety: Philosophy and Ritual in Fourth
Century Athens (New Haven: Yale University Press, 1990), 19, 199 nn.
5i-3.
16 I shall say more about this belief in the soul's immortality when I dis-
cuss the Meno. Guthrie remarked that it was a very difficult idea for
Greeks to accept, but there is evidence that the notion of an afterlife
reaches far back in Greek literature and religious thought; see W. K. C.
Guthrie, The Greeks and Their Gods (London: Methuen, 1950), 176,
180, 260—1.
17 Burkert, Orphism and Bacchic Mysteries.
18 The term is mine and is intended only as a helpful abbreviation. The
themes of this polis tradition and this theology as articulated in tragic
literature are explored by Hugh Lloyd-Jones, The Justice of Zeus, 2d ed.
(Berkeley: University of California Press, 1983).
19 On Greek ecstatic religion and the Orphic-Bacchic-Pythagorean conglom-
erate, see E. R. Dodds, The Greeks and the Irrational (Berkeley: Univer-
sity of California Press, 1951); Burkert, Ancient Mystery Cults; Susan G.
Cole, "New Evidence for the Mysteries of Dionysos/; Greek, Roman and
Byzantine Studies 21, no. 3 (1980): 223-38; Marcel Detienne, Dionysus
Slain (Baltimore: Johns Hopkins University Press, 1979).
20 I believe that Plato wants us to treat Diotima as a prophetess associated
with Dionysos and Pan. Pan was originally an Arkadian divinity, was
brought to Athens shortly after Marathon (c. 490), and flourished in
Athens as a divinity of fertility, love, beauty, and wealth. Pan is also
associated with the panic of sudden threat and surprise. In Plato's mind,
Pan, together with nymphs, sileni, and satyrs, are part of a Bacchic
conglomerate. At the end of the Phaedrus, Plato has Socrates offer a
prayer to Pan that eulogizes wisdom and the soul's well-being. Often Pan
is viewed as an agent of religious possession and madness (mania). For a
brilliant study of Pan, see Phillipe Borgeaud, The Cult of Pan in Ancient
Greece (Chicago: University of Chicago Press, 1988).
21 See Powell, "Religion and the Sicilian Expedition"; MacDowell,
Andokides.
22 Burkert, Greek Religion, 163.
23 Burkert, Ancient Mystery Cults, 97.
24 Burkert, Greek Religion, 290-5.
Cambridge Companions Online © Cambridge University Press, 2006
Plato and Greek religion 247
25 Jan Bremmer, The Early Greek Concept of the Soul (Princeton: Prince-
ton University Press, 1983).
26 N. J. Richardson, "Early Greek Views about Life after Death/ ' in Greek
Religion and Society, ed. P. E. Easterling and J. V. Muir (Cambridge:
Cambridge University Press, 1985), 65.
27 See Morgan, Platonic Piety, 47-54.
28 Burkert, Greek Religion, 297—8.
29 At Phd. J9&2 and 8od6, the Forms are called katharon ("pure"); at 8oa3
and 80b 1, they are called theion ("divine").
30 See Werner Jaeger, The Theology of the Early Greek Philosophers (Ox-
ford: Oxford University Press, 1947); and Lloyd Gerson, God and Greek
Philosophy (London: Routledge, 19 91).
31 See Morgan, Platonic Piety, 115-16.
32 Plato's argument is also an early antecedent of the cosmological argu-
ments of the Middle Ages; for example, see St. Thomas Aquinas,
Summa Theologica I, 2, 3. See Laws 89ie-899d; cf. Sph. 248.
33 Laws 898a3~6; see 897en-898c8 in general.
34 Burkert, Greek Religion, 326-9, esp. 327.
35 Laws 899d-9oob; cf. 883d.
36 Laws 9ooe; cf. Rep. 379a—c.
37 See Rep. 363a-366e and, for discussion, Morgan, Platonic Piety, 108-14.
38 Burkert, Greek Religion, 334; cf. Laws 738D-C, 759a-c, 848c-e; and
Rep. 427D-C.
39 This distinction is adapted from Charles Larmore's distinction between
the expressivist and modus vivendi approaches to the relation of politics
and morality. See Patterns of Moral Complexity (Cambridge: Cambridge
University Press, 1987), chaps. 3-5.
Cambridge Companions Online © Cambridge University Press, 2006
G. R. F. FERRARI
8 Platonic love
Plato does not have a comprehensive theory of love. Rather, he di-
verts certain received opinions about love1 to his own peculiarly
philosophic ends. He is not interested in telling us what it would be
like to live with someone as a platonic lover. Or so I shall argue,
from a reading of the Symposium2 and the Phaedrus.^ I shall ignore
the social regulations for sexuality proposed in the Republic (III
4O2d-4O3C, V 459-461) and the Laws (VIII 835C-842a) as not di-
rectly relevant to what has most fascinated Plato's readers about his
approach to love, and is the topic of this essay: namely, the bridge he
constructs between love and philosophy. I shall also ignore the poten-
tially relevant discussion of friendship (against the background of a
love affair) in that subtle and complex dialogue, the Lysis,* in order
to leave room to stretch myself to something worthwhile on the
Symposium and Phaedrus. Of these two works, I shall focus on the
former, for it alone among the dialogues is concerned exclusively
with love.
The speechmaking of the Symposium is rooted in bad faith. The
series of speeches in praise of love that makes up the bulk of the
work is set in motion by a complaint attributed to Phaedrus (whom
we shall meet again in the dialogue that bears his name). Is it not
shocking, he is reported as saying, that, when it comes to eulogies,
Eros, so ancient and so powerful a god, has been neglected in both
This essay is the better for discussion with Kate Toll and for the written comments of
Richard Kraut and Anthony Price.
248
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 249
poetry and prose by authors who do not hesitate to laud other gods,
or even, in one case, to sing the praises of salt (iyja-c)l The com-
plaint gains a certain plausibility from the piety of its terms (why
stint any god of worship?), but loses it once we consider that eros in
Greek is not only the name of a god but a word that commonly
means "love." When Anacreon writes, "With his huge hammer
again Eros knocked me like a blacksmith and doused me in a wintry
ditch" (fragment 413 in Denys L. Page, ed., Poetae Melici Graeci
[Oxford, 1962]), he is not, or is not simply, narrating an exploit of the
god, but is describing the hot and cold of love. With Aphrodite, now,
it is different. Sex is her gift, but only figuratively can her name
mean "sex" (just as the ordinary Greek for "wine" was oinos, and
only figuratively Dionysos).* She is a goddess with a life and char-
acter of her own (see, e.g., the Homeric Hymn to Aphrodite). But
there are no pre-Platonic narratives about the winged boy-god Eros;
Eros is just love, however you spell him (and the Greeks lacked
capital letters).6 To praise Eros, then, is to praise love. But seen in
this way, the task is not so obviously inviting as Phaedrus suggests.
Phaedrus would know perfectly well that the poets, whom he can
quote as well as anyone (witness 178b), had plenty to say about Eros,
and that the foregoing snippet of Anacreon was a fair sample of
much of it.? This force that falls like a blow, shivering the limbs,
piercing the bones of its helpless victim, madding the mind - is this
a candidate for praise? And yet, how not to welcome one of the
greatest of human joys? The Greeks were thoroughly ambivalent
about love; Eros, in Sappho's piquant formulation, is "bittersweet."8
And so praise of Eros is rather more surprising than Phaedrus, his
focus resolutely fixed on Love's godhead, seems prepared to admit.
As surprising, perhaps, as praising salt (and remember, love is bitter
as well as sweet).
This, then, is one element of bad faith in the instigation of the
speeches we are about to examine. There is another. The round-
robin of speeches might seem the epitome of moderation, for it ousts
the enforced circulation of half-gallon toasts and the charms of the
flute-girl (17664-10). Eryximachus, the doctor, is happy to introduce
moderation as a prescription for healthy living (i76c5-d4). He gets
his chance, however, only because the indulgence of the previous
night has been so extreme as to leave even the heavy drinkers in the
company reluctant for an encore (i76a4-b8). This is an insalubrious
Cambridge Companions Online © Cambridge University Press, 2006
25O THE CAMBRIDGE COMPANION TO PLATO
motive for self-control. And it puts a sickly complexion on the fact
that not one of the invited speakers - for the pattern will change
with the incursion of Alcibiades - praises Eros for how he makes us
feel when in love and excited.* (Instead his gifts will include moral
growth, wholeness, social peace, graceful living, philosophic enlight-
enment). These men are hung over, after all (i76d4); and that is a
state in which excitement loses its allure. Socrates, however, is ex-
empt from this aspersion, both because he was not present the night
before, and because he never shows his drink J22oa4-5). But we shall
see that he takes advantage of Phaedrus's sanctimony and of the
company's remarkable level of abstractedness. In order to consider
more closely how he does this, let us begin by examining, briefly, the
structure and development of the five encomia of love toward their
climax in Socrates' report of the teachings of Diotima. (There are as
many readings of this structure as there are readers of the Sympo-
sium.10 I offer what follows not in order to add to their number but
in the belief that the issue that I raise in its course is central to
Plato's thoughts on love. I shall take it up again in discussing the
Phaedrus.11)
The chief structural marker in this series is Aristophanes' attack
of hiccups (i85d-e). Whatever they signify,12 they serve the function
of forcing the only change in the order of speakers (Eryximachus and
Aristophanes are switched), and so compel the reader to notice that
Plato, who plans all the coincidences in this work, felt that Eryxi-
machus's speech properly belongs where he has arranged for it to be
put, alongside the earlier speeches, while Aristophanes' speech is to
be dissociated from them, and mark a fresh start.1* In short, the
hiccups seem to divide the speakers into two groups: Phaedrus,
Pausanias, and Eryximachus, on the one hand; Aristophanes, Aga-
thon, and Socrates, on the other. And the substantive issue that
distinguishes them is this: The speakers of the first group draw a
fundamental distinction (Phaedrus implicitly, the others explicitly)
between a good and a bad variety of love, while those of the second
group do not. This development comes to a head with Diotima's
teaching that love in any of its manifestations is directed toward the
good (2O5e7-2o6ai, 2o6an-i2).I4
It may seem that Phaedrus does not belong with the first group so
defined, since Pausanias takes him to task for having enjoined the
company simply to praise Eros, without first specifying which Eros
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 251
they were to praise; for there are two kinds of love, one good and
praiseworthy, the other bad (i79C-i8ia, i83d8-e2). But Pausanias is
not contradicting Phaedrus, only setting his opinions off against the
contrast that he has chosen to ignore. Both men agree to praise love
only for the fillip it gives to virtuous conduct.1* Phaedrus speaks of
lovers stimulating one another to courage in battle and to self-
sacrifice, Pausanias of lovers who fall for the good character of their
sweethearts (18365) and of sweethearts who will put themselves at
the romantic disposal only of potential mentors (184C4-7). (In this
emphasis the two men rely on the moral expectations deriving from
the conventional asymmetry of love between Athenian males - the
elder partner being expected to provide moral guidance and help
with civic initiation in return for the sexual favors of the boy, who
was to remain uninterested in the sex for its own sake.16 They appeal
to this connection with virtue even while mitigating its asymmetry.)
Pausanias draws the obvious contrast with lovers who are more
concerned for the body than the soul (i8ib3-4, i83d8-ei), and who
therefore do not take virtue into account (i8ib5-6). Phaedrus does
not; but the contrast is implicit in his claim to know of no greater
good for a young boy than a "decent" or "worthy" (khrestos) lover
(178C3-5). Whatever Pausanias's claims (i8oc4-d3), he does not
really alter the terms of Phaedrus's speech, nor does he rectify the
bad faith at the root of things. He continues to sanitize love - this
lover of the soul, who approves of those who pursue youths for their
nobility of character even if they are uglier than some (i82d7), but
who, we are reminded, is himself the longtime lover of a noted
beauty, Agathon.1?
Next, Eryximachus the doctor, conveniently repositioned to follow
Pausanias, announces that he will bring to completion Pausanias's
contrast between good and bad, "heavenly" and "vulgar" love
(i85e6-i86a3), and proceeds to inflate it into a cosmic principle of
balance and imbalance in everything from medicine to music to me-
teorology, in an unconscious parody of pre-Socratic philosophers.
Aristophanes announces a break with Pausanias's and Eryxi-
machus's scheme of things (189C2-3). He tells a tale that makes of
love a search for primordial wholeness (i92eio), each of us being as
if a half-tally cut from a two-headed, eight-limbed ancestor, and
longing to find our other half. The break, then, is to see love, not
sundered into good and bad, but as a single aspiration, common to
Cambridge Companions Online © Cambridge University Press, 2006
252 THE CAMBRIDGE COMPANION TO PLATO
all, and directed (despite differences of sexual orientation) at the
same generic object - wholeness. This emerges all the more clearly
when we consider the list of fragmented descendants whose sexual-
ity he chooses to trace - adulterers, Lesbians, eagerly passive homo-
sexuals destined to a career in politics, but not, say, married couples
or modest boys (i9ie-i92a). By choosing his examples (at least the
first and third)18 from among social deviants, but betraying no hint
of censure - indeed, going so far as to defend in this context the
passive homosexuals whom he satirized in his plays1? - he ostenta-
tiously subordinates "good" and "bad" in love to the universal de-
sire revealed by his allegorical depth psychology.
Agathon then returns to the importance of love's connection with
the good, but within the new parameters that Aristophanes has set.
Aristophanes had argued that all love is of the same generic object,
wholeness; Agathon agrees with the structural claim, but contends
that the single object of love is kallos — one word in Greek, but with
a semantic range requiring many in English: the beautiful, the fine,
the noble, the good (i97b3-9).2° The speakers of the first group had
inferred the goodness and praiseworthiness of Eros largely from the
goodness of his effects - masculine virtue, universal well-being.
This left open the possibility, exploited by Pausanias and Eryxi-
machus, that Eros could be blameworthy to the extent that his ef-
fects were bad. Agathon reverses this pattern - announcing the
change in rhetorical strategy with a flourish (i9464-195a5). He first
argues that Eros himself is entirely good, and only as a final matter
notes that his effects are like him (197c 1-3). Moreover, Eros for
Agathon is beautiful and good, not because his effects are, but be-
cause he is love of the beautiful and good (he flees old age, resides in
gentle souls, inspires artists of all sorts). Now, this is a most serious
conclusion, and a crucial shift of interest, however frothy the argu-
ments on which it rests (cf. 19767-8). It is what enables Socrates,
next to speak, to introduce Diotima's teaching that all love, ulti-
mately, is love of the good, and for that reason commendable.21
He does so, however, by first correcting a misapprehension of Aga-
thon's. That love is always of the beautiful and good does not imply
that love is beautiful and good. Love, after all, is desire, and to desire
is not yet to have (200a); so that if love is of beauty and goodness,
love is without beauty and goodness. How then can love be beautiful
and good (2oia-c)? Whatever the merits of this argument,22 it per-
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 253
mits Socrates to move on, in his report of Diotima, to a doctrine that
manages to subsume into itself important elements from the entire
series of preceding speeches, while capping them with its special
contribution.23 For if love does not have beauty and goodness, this is
not to say that love is ugly and bad (2oie). Love is no god, but an
intermediary, a "daemon" who communicates between gods and
humans; not wise, but a lover of wisdom-a philosopher (2O2d-
204c). Agathon failed to see that love's nature is to seek the good,
rather than to possess it; but it turns out that he was not wrong to
claim that love is praiseworthy in its very nature; for to seek the
good is praiseworthy. And this is to reinstate the message of Aris-
tophanes' tale: that love is above all a search for what has been lost.
Aristophanes only misidentified the loss: It is not of our other half,
but of our good (the point is made explicit at 2O5dio-2o6ai). And if
love is an ongoing search for the good, then even the distinction
between good and bad varieties of love developed by the first group
of speakers, or something like it, can be accommodated; for a search
is dynamic, and a search for the good will be better, the closer it
comes to its goal. All love is one, all love is praiseworthy, insofar as
it is love of the good; but some manifestations of it will be superior,
others inferior. Thus, while all love is praiseworthy, not all lovers
will be.2*
11
If Diotima's teaching is large enough to embrace elements from all
of the opinions heard so far, it may be thought to do so on pain of
forgetting the topic. Love, it turns out, is a search for the good - but
just as Eryximachus, crowning speaker of the first group, had bidden
farewell to love between persons and taken us for a tour of the
stratosphere, so Diotima, in the corresponding position within the
second group, threatens with a description of great breadth to lose
sight of the love that one human being conceives for another. Unlike
Eryximachus, however, she recognizes the problem; indeed, it can be
shown to govern the structure of her teaching.
Everyone wants the good, she gets Socrates to agree, because good
things bring happiness, and everyone wants happiness. In this sense,
everyone may be said to have "love" [eros) for the good (205a). In-
deed, the good is the only object of love, in that we love only what
Cambridge Companions Online © Cambridge University Press, 2006
254 T H E CAMBRIDGE COMPANION TO PLATO
we consider good; for it makes no sense to say that we could love
what we consider bad- tha t is, harmful-for us (2O5dio-2o6ai).
(Here we see Socrates fictionally acquiring the philosophic position
that he puts to extensive and notorious use in the early dialogues,
e.g., Meno 77b-78b, Gig. 467C-468C) Since, however, we differ in
our conceptions of what will bring happiness, lovers of the good
come in many varieties, and common parlance usually reserves the
words "love" and "lover" (eras, eran, erastes) for those who seek
that particular species of the good which is sexual love between
persons (2O5di-8).25
Thus Diotima distinguishes between a generic and a specific sense
of eros.26 Clearly, only the latter is what Phaedrus intended in the
complaint that sparked the evening's discourse. What is Diotima's
warrant for enlarging the topic? In fact it is rather vague at this
point. Before the discussion of generic love just summarized (205a-
206a), she seizes upon Socrates7 agreement that the lover seeks to
possess the beautiful and asks him what the consequences of suc-
cess would be for such a lover. Socrates is stumped. Diotima there-
fore proposed that they substitute "the good" for "the beautiful,"
correctly foreseeing that Socrates will find the question easier to
answer in this form. Lovers of the good who manage to gain posses-
sion of good things, he responds, will be happy (2O4d-2O5a).27 Before
we know it, they are launched on the discussion of generic love. But
what about the original question that stumps the young Socrates?
We might have thought that Diotima's purpose in substituting the
apparently easier question is to help him cope with the more diffi-
cult one. Yet she does not return to it - introducing instead a fresh
question (2o6bi-4). Socrates, whom Diotima assumes to run after
beautiful boys like all young men of his age (2iid3-8), cannot say
why he does so - cannot say what a person gains by possessing the
beautiful. It turns out that he will not have the answer to Diotima's
question until, after much preparatory teaching, she has spoken to
him of the mysteries of the Beautiful itself (2ioa-2i2a). And it is
here, we shall find, that her warrant for enlarging the topic to in-
clude generic love is finally made clear.
In order to prepare her initiate, Diotima returns to the topic of
specific love ("[what] would be called love," 2o6b3)28 and compli-
cates its traffic with beauty. Specific love is in fact not, as Socrates
(and Agathon) suppose, love of the beautiful, but rather "of begetting
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 255
and giving birth in the beautiful" (2o6e5). Why so? Because it has
been agreed that, speaking generically, love is not only love of the
good but also the desire to possess the good always (206a); but the
desire to possess the good always is really two desires, one for the
good and one for immortality;2* and "begetting" is the closest that a
mortal creature can come to immortality (2o6e7-2O7a4).3° In the
specific case, beauty takes the role of midwife to generation (2o6d2-
3), prompting those fertile in body, both animal and human, to engen-
der offspring who can renew their line and (for humans) keep their
name alive (206C1-5, 2O7a7-d3, 2o8ei-5), while those men who are
more fertile in soul than body will be inspired by a boy who com-
bines bodily beauty and beauty of character to give birth to fine
discourse about civic virtue with a view to his education (2O9a5-c7).
Such lovers successfully establish a connection between the beau-
tiful and the good. Beauty, whether of body or soul, is instrumental
to their virtuous (that is, socially esteemed) conduct - procreation of
loyal heirs, tutelage of the rising generation. Nevertheless, they can-
not be said to have examined and understood that connection.
Rather, it works its way through them,- they are conduits for it. That
is why Diotima relegates their love to the status of Lesser Mysteries
(2O9e5-2ioa2).3I What might have prompted them to a greater
awareness is made clear for us, in the case of the lover fertile in soul,
when Diotima describes his edifying discourse - his joint offspring
with the beautiful boy -a s itself "beautiful" (209C6-7). This is the
first time that what issues from the act of "begetting in the beauti-
ful" has itself been described as beautiful, and it reveals something
that will be crucially important to love's Greater Mysteries: namely,
that beauty can belong to the product as well as to the instrument of
specific love. At the lesser level, however, the lover stimulated to
"beautiful" conduct by beauty takes for granted what the beautiful
is. His reason for preferring to beget fine discourse rather than fine
children is that he appreciates the immortal name won through
discourse by such poets as Homer and Hesiod, such lawgivers as
Lycurgus and Solon. This is a pious roll call of cultural heroes. But
Diotima is about to challenge such pieties. She will have harsh
words, in the course of describing the Greater Mysteries, for those
whose horizons are bounded by custom (2iodi-2).32 Similarly, at the
level of the Lesser Mysteries she describes the ultimate good - the
goal of generic love, toward which all human actions are directed
Cambridge Companions Online © Cambridge University Press, 2006
256 THE CAMBRIDGE COMPANION TO PLATO
(2o8d8) - as "immortal virtue and the glorious fame that follows"
(2o8d7-8). The "love of honor" [philotimia, 208C3) is the highest
human aspiration here. But when introducing the topic of generic
love, Diotima had prepared us to accept the "love of wisdom"
[philosophia, 205d$) as one of its manifestations; and in the Greater
Mysteries it will be philosophy that leads us to the ultimate goal
(2iod6). The transition from Lesser to Greater bears comparison,
then, with the crucial shift of focus in the Republic from institu-
tions grounded in the honor code (Books II-IV) to those derived from
rule by philosopher-kings (Books V-VII)."
The final revelation in the Greater Mysteries of love - the vision
of the Beautiful itself - will be disclosed only to those who follow a
particular path of eroticism (2ioa2). The mark of the suitable ini-
tiate is that he does not take the nature of the beautiful for granted
as would an honor lover, but is prone to become more deeply fasci-
nated by the beauty that issues from his love than by the beauty
that first attracted it. This displacement of attention is what moti-
vates his climb to each new level of the upward path. Let us trace
his progress. His point of departure is to be smitten with the bodily
beauty of a particular person, which stimulates him to generate
"beautiful discourse" (logous kalous, 2ioa8). Since the effect of this
discourse is to prompt reflection on physical, but not also spiritual,
beauty (2ioa8-b2), and since the production of edifying advice in-
spired by beauty of soul and body combined - comparable to the
reaction of the spiritually fertile lover at the level of the Lesser
Mysteries (2O9a8-C2)-is reserved for a later stage in the ascent
(2iob8-C3), it would seem that, whatever form the discourse takes,
its content is limited to enthusiasm for the physical beauty and
prowess of the beloved.34 The philosophic initiate begins, then, at a
level lower than that attained by the honor lover in the Lesser
Mysteries (whom he will overtake in due course.^ His starting
point is higher than the level of the fertile merely in body, how-
ever,- for their love engenders human offspring (20862-4), whereas
his produces discourse. But what marks him out as having the
potential to scale the philosophic heights is his further reaction to
that discourse. His "beautiful" words have beauty as their topic36 -
not the beauty of this body alone, but also bodily beauty in general,
because to praise something is to insert it in its comparison class.
And now his thoughts turn toward the beauty to which the product
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 257
of his love has drawn his gaze, at the expense of the beautiful
individual who delivered him of that product. He reflects upon
"beauty of [outward] form" in general, and sees that, since the
beauty of any one beautiful body is akin to that of another, he
would be "very foolish" not to think of bodily beauty as one and
the same in all cases; which thought makes his feelings for the
single beauty who inspired it seem to him now an overvaluation,
and causes him to divide his enthusiasm among all exemplars of
bodily beauty (2ioa8-b6). That all bodily beauty is "akin" is a
thought likely to strike anyone who makes comparisons; that it is
"one and the same" is a claim that perhaps only the Platonist
would find it "very foolish" to deny. But that is surely the point:
This lover is marked out for philosophy, the preserve of the few (cf.
2O9e5-2ioa2), by the peculiarity in his reaction.
We next find him having come to prize beauty of soul over beauty of
body (2iob6-7). We are not told how he made the transition, but only
that these are stages along the way that he "must" visit (dei, 2ioa4) if
he is to achieve the highest goal. Nevertheless, since a guide for the
journey is only optional (21 ib7-ci), we are entitled to assume that he
is not simply following authority, and to take this as an invitation to
supply a reason for his development. (This may be Plato's way of
testing his reader's own aptitude for the journey.) It is not far to seek.
The lover who comes to be more occupied by thoughts and expres-
sions of beauty than by the beauty of the body that prompted them,
being marked thereby as a reflective and cultured type, will naturally
be open to the attractions of the soul. 37 Again it is the beauty of an
individual (a boy, as would be conventional in Athenian society for a
love of this high-minded sort) that delivers him of his progeny, which
again takes the form of discourse - this time of edifying speeches
intended to bring out the beauty of the soul (that is, decency of char-
acter, 2iob8)38 entrusted to his care (2iob8-C3). And again his focus
shifts as a result from the beautiful target of his discourse to its beauti-
ful topic. Compelled, in his role as mentor, to consider the beauty of
activites and laws, he comes to a conclusion about it that is indepen-
dent of his educative purpose (just as he spent more thought on bodily
beauty in general than was necessary for the purpose of seduction): It
too, like bodily beauty, is all of a kind (2ioc3~5).39 (Here he goes
beyond the fertile-souled pederast of the Lesser Mysteries, who
merely produced his beautiful advice, but did not contemplate its
Cambridge Companions Online © Cambridge University Press, 2006
258 THE CAMBRIDGE COMPANION TO PLATO
beauty.) The result of the earlier conclusion had been to cut the physi-
cally beautiful individual down to size in the lover's estimation; the
result now is to cut an individual category down to size:4° The lover
will think the beauty of bodies a thing of no importance (210C5-6).
This is not a state of mind that he had already achieved when we
found him prizing the beauty of soul over that of body (2iob6-7).4I
That had been a natural accompaniment to his growing thoughtful-
ness, and a matter of what he found attractive in individuals (what is
more, bodily beauty had continued to count, however marginally, as a
condition of an individual's attractiveness to him - 2 iob8-ci); this is
a considered judgment, which holds only categories in focus. The boy
of beautiful soul has disappeared from view. We are not told what
happens to him. While there is no reason to deny that the lover could
continue to share with him the results of his later philosophic quest,
the center of his emotional attention is no longer occupied by an
individual person, or indeed by persons at all.*2 As an index of this
development, it is when the beloved person falls from view that
Diotima ceases to use the term "to love" (eran, uoaj, 2ioci) and
"lover" (erastes, 2iobs) to describe the initiate's relation to what he
finds beautiful.43 From now on, the fact that he continues to find the
objects of his concern beautiful - indeed, increasingly beautiful
(2iid8-e3)-must do most of the work in convincing us that
Diotima has not changed the subject.
At the next stage of his development, accordingly, he is not at-
tached to an individual, but is attracted rather by the beauty of knowl-
edge in its various forms, which causes him to give birth once again to
beautiful discourse - now the discourse of philosophy (2iob6-d6).
While the upward move is, again, not explicitly justified, it is natural
enough.** The lover has been contemplating the beauty of activities
and laws — the principles by which we live. Any reader of the early
Platonic dialogues can bear witness to the ease with which such
contemplation gives rise to the question whether anyone has genuine
knowledge of these matters - and so to the question of expertise in
general. And now, as before, the initiate's concern is transferred from
the beauty that enticed him to the beauty that he has generated. Since
the range of knowledge is not limited in advance to a particular sub-
ject, but only to what can be known, the initiate has been led to
contemplate "the great ocean of beauty" (2iod4). He is looking back
from the height he has scaled/s and sees beauty as a whole, but a
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 259
whole of great multiplicity. Now he turns his face to the peak, and
comes to see beauty as a unity. He catches sight of a single knowledge,
which is knowledge of a single beauty - an individual, the Beautiful
itself, which is described in the classic vocabulary of the Platonic
Forms (2iod6-bs). What has happened, then, is that, in a by-now
familiar shift of focus, the initiate has turned from simply "doing"
beautiful philosophy (considering what is beautiful in the varieties of
knowledge - the beauty that attracted him, 210C7) to grasping the
beauty o/his philosophy (the beauty that he engendered, 2iod5). He
comes to understand what it is that makes possible the single knowl-
edge of the beautiful toward which, as philosopher, he has been work-
ing: namely, the existence of the Beautiful itself.
Now, Plato has deliberately left as a mystery just what this claim
amounts to (as he always does when introducing the Forms into his
dialogues). Here I shall limit myself to the following: To say that the
Beautiful itself exists is to claim that there is such a thing as beauty,
independently of what we or any creature find beautiful (cf. 2 i ib3 -
5); and that were there not, nothing at all would be beautiful (cf.
Phd. iood). But awareness of the truth of this claim, as Diotima
describes it, is meant to hit like a hammer blow. It is not an aca-
demic decision to commit oneself to a variety of philosophic real-
ism; it is more like an encounter with a beautiful individual, in
whose presence one longs to stand (2ioe4-6, 2iid8-2i2a2).46 In
short, it is more like love.
And so the initiate reaches the end of the line. The mechanism of
shift in attention can lift him no further. It had worked by fastening
on what rendered beautiful the discourse generated at each level; but
at the summit, no discourse is generated. The Beautiful itself is
beyond words (cf. 2iia7). To attempt to describe it, as at 2iia-b, is
not to give birth to discourse in its presence; rather, it is to set out
the prospect of standing in its presence as a goal, and as a spur to
philosophic discourse, which is the means of achieving that goal (cf.
Socrates' reaction at 2i2bi-4). What is generated at the summit is,
for the first time, not described as a kind of discourse, but rather as
"true virtue," which enables the initiate to become "beloved of the
gods" and "immortal, if any human being can be" (212a). This is also
the only offspring produced by the lover in his ascent that is not
itself qualified as beautiful, and so far from there being any hint that
he could transfer his concern from the Beautiful itself to the beauty
Cambridge Companions Online © Cambridge University Press, 2006
26O THE CAMBRIDGE COMPANION TO PLATO
of virtue, he is explicitly envisaged as spending his life in contempla-
tion of the former (2iie4-2i2a2). In marked contrast to the Lesser
Mysteries, what virtue amounts to here is not clearly something
other than the vision of the Beautiful that gives it birth. That is, it is
unclear whether that vision is the cause of virtue or its occasion. At
any rate, even if the two are to be distinguished, the emotional
weight of the climax falls rather on the vision of the Beautiful
(2ioe3-2iib5, 2iidi—2i2a2) than on its offspring (niai-y).^
Let me draw together the threads of this account. I said that only
after hearing of the Greater Mysteries would we understand Dio-
tima's warrant for enlarging the topic from that of specific love (the
love that people "fall in/; with one another) to generic love (the love
of the good which motivates all human action). Her warrant turns
out to be this: In describing the initiate's progress, she has connected
sexual passion to the life of true virtue (the pursuit of the good) by a
series of plausible steps; and keeps touch throughout, by her appeal
to the motive force of beauty, with how it is to be in love. The
initiate falls in love with a fresh beauty at each stage. (This is what
was missing from the Lesser Mysteries.) Moreover, he subsumes in
his ascent the inferior forms of love, being led first by sexual desire,
then by the ambition for honor, and finally by the love of learning
(cf. 21 ic); so that his development seems a natural unfolding of the
manner in which all would choose to realize the good, had they the
ability (cf. Republic IX 58od-583b). Admittedly, what he communes
with at the summit is the Beautiful itself, not the Good itself; and
the relation between the beautiful and the good, here as elsewhere in
Plato, is problematic. In view of such passages as 201c and Phaedrus
25oc-d, let us say that the beautiful is thought of as the quality by
which the good shines and shows itself to us. We can then claim that
the ascent to the Beautiful itself is indeed also an ascent to the Good
itself/8 but described so as to bring out at every turn what it is about
the good that captivates us.49
I also said that with Diotima's teaching Socrates would take advan-
tage of the company's abstractedness and the whiff of bad faith that
hangs about Phaedrus's and Eryximachus's challenge to the party.
We can now see that Diotima's teaching exploits to the highest
degree the conventional possibility of high-mindedness in the Athe-
nian homosexual love affair - the idea that erotic energy should fuel
the development of virtue. (This despite the fact that in the Greater
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 261
Mysteries Diotima, unlike the speakers of the first group, finds a
place at the foot of the ascent for sexual desire by itself as a legiti-
mate and essential species of love.) Socrates' speech is unabashedly
an account of what love can do for philosophy, rather than an ac-
count of love for its own sake. Hence at the end he singles out as his
reason for honoring Eros the fact that one could find no better "col-
laborator" [synergon) for the task of acquiring the ultimate posses-
sion (2i2b2-6) - that is, the vision of the Beautiful itself, together
with true virtue and immortality. But this is only to say that his
rhetorical strategy is all of a piece with the tone of the occasion and
its effect on the five previous speakers. Phaedrus and Pausanias had
described what love can do for manly virtue, Eryximachus what love
can do for scientific investigation (e.g., i86b2~3, i8ye6-8), Agathon
what love can do for the appreciation of beauty (e.g., I96d6-ei,
i97b8-9), and even Aristophanes, who has seemed to many modern
readers to come closest to addressing what love is all about, offers
his account (political conservative that he was) as a cautionary tale
that shows what love can do for piety (i93a3-d5).5° Not that Plato
means by this to present Socrates as failing to measure up to his
promise to tell the truth about love (i99b).5I But it is a selective
truth, reflecting only those facets of love that a philosopher would
find most beautiful (cf. I98d5~6).
In order to emphasize what the occasion has caused us to miss so
far, and why, Plato ends the Symposium with a seventh, unsched-
uled speech - the speech of Alcibiades. The new arrival restores the
party to its proper level of alcoholic indulgence (213e7-2i4ai), and
he alone speaks drunk (214C6-8), speaks from infatuation (cf.
222C1-3), and addresses love in its traditional bittersweetness (e.g.,
2i6b5-c3). But this is to say that he alone betrays that he is in
love, not that he alone understands love. Indeed, he does not even
understand Socrates - that much is clear from his attempt to se-
duce him, and especially from his refusal to learn from its failure.
Being in love, Alcibiades is the first speaker to praise his love
object - Socrates, in his case - for how Socrates makes him, person-
ally, feel (2i5ei-2i6c3, 2i9d3-e3). Moreover, of all the amazing
exploits that he attributes to his hero (2i9e-22ic), it seems that he
regards as most amazing of all (for it is the emotional and struc-
tural core of his encomium) Socrates' success in making him feel as
he does about Socrates - in captivating and shaming the beautiful
Cambridge Companions Online © Cambridge University Press, 2006
262 THE CAMBRIDGE COMPANION TO PLATO
Alcibiades (see especially 2i6a8-b2). His is the version on a heroic
scale of the danger that Apollodorus, hawking memberships to the
Socratic fan club, had illustrated in the prologue on the level of
farce: Instead of loving wisdom, he falls in love with the wisdom
lover*2 - exactly the danger that Diotima attempts to exclude from
her ladder of love by banishing individuals from the center of atten-
tion when the rung of philosophy has been reached. Socrates, good
student of Diotima that he is, had attempted to tap the energy of
Alcibiades' love and channel it away from himself and toward phi-
losophy (2i8d6-b2).53 But it is a volatile force to tap, and the saga
of Alcibiades' disastrous political ambitions waits in the wings of
this dialogue to testify to the difficulty of the task (cf. 2i6bs).
(Perhaps also Plato thought the intellectual free-for-all at Athens a
special obstacle to such a gradual doling out of wisdom as Diotima
proposed; cf. Rep. VI 498a-c). And so we are reminded at the last
that love is more awkward to praise than Phaedrus had initially
made out. 54
in
The atmosphere of the Symposium, until the incursion of Alcibiades
(and with exception made for Aristophanes' sound effects), is formal,
elegant, rule-bound, and restrained; that of the Phaedrus is intimate
and improvisatory. Socrates gets away from it all. Alone with Phae-
drus in a beautiful pastoral setting outside the city - unusual sur-
roundings for him (Phdr. 23oc6-e4)55 - he is inspired by its influence
and by Phaedrus's reading of a speech on love by the orator Lysias*6
to improvise a speech on the same topic, and then suddenly to repu-
diate it and improvise a further speech of recantation. These are
speeches of uncharacteristic length, fluency, rhetorical grandeur,
and poetic color (cf. 238c5-d3, 24iei-5, 257ai-6, ci-2). (We should
not forget, however, that in the Symposium Socrates had been por-
trayed as equally at home with the drunk and the sober; Smp.
176C3-5, 22oai~5. It is typical of him to be capable of the untypi-
cal.) What the Symposium is only after Alcibiades has come on the
scene, the Phaedrus is throughout. It should not surprise us, then,
that love in the Phaedrus is presented in its traditional bitter-
sweetness and with vivid emphasis on how the lover feels.*7
To be bittersweet is to offer material for censure as well as praise.
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 263
(Alcibiades had both loved and hated Socrates for the feelings he
aroused; Smp. 2i5e-2i6c.) Lysias's speech and the corresponding
first speech of Socrates are critical of love. They marshal against love
the traditional poetic complaint, as when Anacreon laments its ham-
mer blow (cf. Phdr. 235c). The lover is knocked out of his senses
(23idi~3, 24ia2-4); how then can the boy he loves expect him to
behave other than unreasonably? The lover is possessive (232C3-6,
239a7-b3), indiscreet (23ie3-232a4), selfish - caring more for his
own satisfaction than for the boy's development (232c-d, 239a,
239e-24oa); above all he is fickle, and upon returning to his senses
will cast the boy off without a thought (23ia-d, 232c, 24oe8-
241b).58 This is to see love-madness from the outside, from the view-
point of the boy, who is not expected to share in its intensity and
who can therefore be made to find it alien and off-putting.
After a violent change of heart, however (24ie-243e), Socrates in
his second and climactic speech finds the good in love-madness
(244a6-8). Seeing the world now through the eyes of the lover,
whose feelings he analyzes in depth (249d-254e), he derives the
soul-shaking quality of the lover's experience from its origins in the
vision that crowned Diotima's Greater Mysteries, the vision of the
Beautiful itself - considered here not as a state of enlightenment
attained only after long struggle, but as a memory of enlightenment
stirred by the boy's beauty (254b), a memory that acts as a potential
stimulant to that longer quest (256a7-b7). The fresher the memory,
the less the lover will think indulgence of his bodily desire for the
boy an appropriate response to it (2506-25 ia). Thus it is not only in
his first speech that Socrates faults the sort of lover who allows his
sexual hunger to dictate his behavior; so that in retrospect he recon-
ciles the two speeches as containing, respectively, appropriate con-
demnation of an inferior type of love-madness and appropriate praise
for a superior, "divine" type (265e-266a).
This contrast may remind us of nothing so much as Pausanias's
sundering of "heavenly" from "vulgar" love in the Symposium.
Does Socrates in the Phaedrus, then, contradict Diotima's teaching
that all love is directed, ultimately, toward the good? He does not;
but his account is psychologically more complex. Diotima discrimi-
nated among us according to the inferior and superior brands of
eternal good for which each of our souls yearns; Socrates in the
Phaedrus divides the soul itself into three parts, and assesses the
Cambridge Companions Online © Cambridge University Press, 2006
264 THE CAMBRIDGE COMPANION TO PLATO
superiority and inferiority of the individual lover according to the
outcome of the power struggle among the yearnings represented by
each part. None of these parts - allegorically portrayed as a chario-
teer with his team of two horses (246a) - is a yearning for the good as
such.59 Rather (and as in the similar tripartition developed in Repub-
lic IV and VIII-IX), that part of the soul which ought to dominate
over the others, the charioteer, seeks wisdom, or truth (hence he
alone can behold the realm of true being - the Forms - and feed on
pure knowledge, 247c); the well-behaved white horse represents the
love of honor and propriety (253d6; cf. the level of aspiration at-
tained in the Lesser Mysteries); and the violent black horse (25363),
corresponding to what in the Republic is called, among other things,
the "love of advantage" (e.g., at IX 58ia7), represents the simple
need to have one's way - limited in this context to sexual need. The
charioteer and white horse tend naturally to pull in the opposite
direction from the black horse, which will weigh them down in their
ascent to the Forms (for this is a winged chariot) unless it receives
proper training (247b2-5). Yet there is a unity to the parts also, for
the entire soul, before its fall to earth, was feathered (25 ib7), and it is
the nature of the feathers to be nourished by truth (248b5-c2).
Put without allegory: It is only at the level of the whole person,
not the parts, that all love is directed, by its nature, toward the good.
And to say that the whole person is naturally oriented toward the
good is, more accurately, to say that, first, the charioteer of his soul
is naturally oriented toward wisdom and truth; second, that the
charioteer is natural ruler of his soul (although not necessarily the
actual ruler of his soul) - as symbolized by his being in the driver's
seat; and third (a corollary of the previous point), that it is when the
charioteer is in charge that the entire soul is at its best, and the
person lives the good life, the philosophic or "wisdom-loving" life
(256b2-7). (Compare Rep. IX 586d: The other parts will attain their
"truest" satisfactions when they follow the wisdom-loving part.
This I take to be what Socrates means by describing the feathers of
the entire soul as nourished by truth.)
There is, indeed, a simpler sense in which Plato believes that we are
all naturally oriented toward the good: namely, that we all look to our
ultimate benefit. (This, after all, was Diotima's teaching at Smp.
2O5dio-2o6ai; cf. Rep. VI 5O5d-e.) But the tripartite analysis-in
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 265
particular, the fact that the wisdom-loving part is our natural ruler -
reveals the deeper sense in which those in whom the charioteer lacks
authority remain, as persons, lovers of the good, no less than those in
whom it dominates. It is not just that they continue to seek what they
(mistakenly) think will benefit them. Rather - as Plato argues most
fully in the Republic - the natural authority of the wisdom-loving
part continues to make itself felt, even in defeat, in the wretched
emptiness of their lives. They live with the twistedness (though they
could not acknowledge it is such) of those who deny their own
nature - their nature as lovers of the good.60
The tripartite analysis of the soul in the Phaedrus also permits a
more exact appreciation than in the Symposium of how it feels to
fall in love, and why this feeling should awaken aspiration to the
philosophic life. As in the Symposium, lovers are approved of who
are not simply attracted to the beauty of their beloved's body but
seek also to mold his character {Phdr. 2506-25ia, 252C-253C); un-
like in the Symposium, Socrates takes us behind the scene of such a
lover's modest and respectful behavior toward his boy (25468-
255ai), and the spectacle he reveals to us within the lover's soul is a
struggle of the utmost violence (253d-254e). The charioteer begins
by trying to persuade the black horse, which if it had its way would
lead the lover to do something "shocking and unlawful" (254^1; i.e.,
to attack the boy sexually),61 but ends by dragging it to its haunches
and drawing blood with the bit, not once but repeatedly. Let us look
more closely at this development from inner persuasion to inner
violence, for it is at the heart of what the Phaedrus has to say about
love. It is crucial that the charioteer loses the struggle when he
argues (for he agrees to do the horse's bidding and approaches the
boy, 25^3-4), and wins it when he is violent (falling back on the
reins in awe as the boy's flashing glance, seen close, recalls him to
his primeval vision of Beauty; 254b5-C3). The difference in outcome
is explained by a difference in motivation. While the parties in the
soul are arguing, no distinction is made between the charioteer and
the noble white horse, who share motivation and reaction: anger at
the black horse for pushing them toward an unlawful deed (254a7-
b3).62 Yet we know from the earlier narrative of the fall of souls that
only the charioteer saw anything of the Forms; neither horse (even
in the best souls) managed to break the surface of the heavens to
Cambridge Companions Online © Cambridge University Press, 2006
266 THE CAMBRIDGE COMPANION TO PLATO
glimpse the place beyond (24801-5). So it is that, although all three
characters in the allegory catch the flashing glance of the boy, only
in the charioteer does it spark a vision of the Forms.
What we witness at that moment is the charioteer coming into
his own. When arguing, the charioteer's emotional attention had
been focused on what we might call internal politics: establishing
correct order in the team. But when he comes to recollect Beauty
itself, his emotional attention is swept away by the object of this
private memory, and he effectively forgets about his team. That is
why his violence against them is not (as the attempt to persuade the
black horse is) a purposeful act of control. Had it been, he might
have thought to use main force only on the wayward black horse,
not also on the obedient white (254C1-2). Rather, he falls back on
the reins because compelled by awe at the sudden vision of the
Forms (254b7-ci); his feelings as lover of the Forms leave him no
gubernatorial option. In both cases, the charioteer's resistance to the
demands of the black horse (at least for as long as that resistance
holds up) corresponds to the same outward behavior of the lover:
namely, self-control. The difference lies in motivation. The motiva-
tion of charioteer and white horse when together is the desire to do
what the law enjoins; to remain within the bounds of social propri-
ety. Even when the charioteer parts company with the noble horse,
the sense of staying within bounds does not vanish from his
thoughts, for he has a vision not of Beauty alone but of Beauty
together with Moderation (254b6-7). But it is a different sense. He
sees Beauty "on its holy seat" (254by). The bounds that feature in
this vision, then, are the bounds not of law but of the sacred. And
the act of attacking the boy is avoided not as something socially
forbidden but as something simply unbearable - a violation of all
that really matters.6* Nothing in the soul can resist the charioteer at
that moment of realization. But when debating propriety with the
black horse, he is untrue to the fullness of his motivation, and so
fails to assert himself.
This development explains what might otherwise seem odd: that
although the immediate effect of the charioteer's seizure is to pre-
vent an outrage of convention and law, the behavior to which it
will lead - as we know from the earlier description of the lover's
antics in the social world - will be anything but conventional. The
lover, we read, will for his beloved's sake "disdain the conventions
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 267
and niceties of which he previously made a show" (252a4-5); he
will give up family, friends, and wealth in order to be with his love.
And this behavior makes good sense when seen as the outcome of
the charioteer's shift of focus away from his team. In nonallegorical
terms, this would represent someone inclining to act in an other-
worldly fashion, as if not tied to a social and timebound existence;
ready to throw all else to the winds for the sake of his love.6* But
this is an inclination common to the true lover and the true
philosopher - notoriously so. Socrates here puts two cliches into a
bottle - the temporary insanity of the infatuated lover, the hopeless
impracticality of philosophers^ - and shakes them into a cocktail
of great strength and fizz. Simply put, his point is that love makes
philosophers of us all.
But only for a while. It is not as if the lover's memory of the
Beautiful itself somehow conveys him without effort to the peak of
the initiate's ascent in Diotima's Greater Mysteries. The soul-
shaking experience that Socrates describes, although it is not for all
(being limited to those who had a full vision of the Forms when
discarnate, and so do not rush to consummate their sexual desire but
hold back in reverence; 2506-251 a) is nevertheless for more than
just philosophic types. The "inspired" lover (255b6), despite the fact
that the initial vision of his charioteer transcends the motive of
social repute, is just as likely to end up leading with his beloved a
life that is merely honorable, rather than the philosophic life (256a-
e). The inspiration of love opens a potential path but does not ensure
that it will be followed; still less does it do the traveling.66 Nor does
Socrates tell us much about the subsequent journey for the philo-
sophic types.6? We learn that they, like other kinds of inspired lovers,
will be encouraged in their own development by their concern to
educate the beloved and bring out the character that they sense he
shares with them (252e-253c). But it is the honor-loving couple who
are described as pledged to one another and as regaining the wings of
the soul together, and for the sake of their love (256d-e). The philo-
sophic couple, who never consummate their sexual desire,68 live a
life "together in mind" (homonoetikon, 256bi), but a mind set on
their remaining always in control of themselves and on regaining
their wings for the sake of the good that awaits in the afterlife (the
vision of the Forms) rather than for the sake of love (256b). Socrates,
then, is not much more concerned in the Phaedrus than he was in
Cambridge Companions Online © Cambridge University Press, 2006
268 THE CAMBRIDGE COMPANION TO PLATO
the Symposium to answer our questions about the life of the philo-
sophic couple as it develops.6* For in the Symposium, the develop-
ment we witnessed was of the philosophic lover alone; while in the
Phaedrus, the focus is rather on the beginnings of love between
philosophers than on its development.
For this reason, also, the place of Beauty in the two dialogues is
different. Diotima's initiate comes to see Beauty just in itself; he
comes to see, I suggested, that there is such a thing as Beauty, inde-
pendent of what we find beautiful. The experience of the inspired
lover in the Phaedrus, by contrast, is to shuttle in memory between
the bodily beauty of the boy and the Beautiful itself. It is to be
awoken by an exemplar of Beauty to the conviction that there is
such a thing as Beauty. (The boy, too, comes to have this experience,
seeing the lover's face transformed by love, made beautiful by the
sight of beauty; 255b7-d3.)?° This is a conviction available to all
romantics, not just to philosophers. Hence Socrates' claim at 25od:
The images of Beauty shine brightest of all images of the Forms in
our world. And when that image of Beauty is human (contrast Socra-
tes' reaction to the beauty of landscape, 23ob2-C5 with disclaimer at
^3-5), it is further possible to see in it an image of oneself - as both
lover and boy do, in varying degrees of awareness (252d5-253C2,
25 5d3-d6). What happens next will depend on the kind of self that is
seen. Only the philosopher, for whom self-inquiry is paramount
(229e-23oa), will glimpse in Beauty's image a self that not only
reflects the Beautiful but is capable of reflecting upon it.
Bearing these contrasts between the two dialogues in mind, we
should refrain from the attempt to situate the inspired lover of the
Phaedrus at some precise point on the ascent undertaken by Dio-
tima's initiate.?1 Diotima deals with honor lovers and wisdom lovers
separately, in the Lesser and Greater Mysteries; Socrates in the Phae-
drus melds them into the figure of the inspired lover. This is further
evidence that Plato is not concerned to propound a comprehensive
and unified theory of love - if any were needed in addition to the
way in which we have seen love to be treated (that is, exploited) in
the two dialogues. In both cases, Plato takes one of love's cliches and
turns it to his metaphysical advantage. In the Symposium, the cliche
is "love promotes virtue." In the Phaedrus it is "love is wild." From
this source flow the differences between the two dialogues, their
limitations, and their achievements.
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 269
NOTES
1 We need not fear to translate the Greek eros as "love" (pace David
Halperin, "Platonic Eros and What Men Call Love/7 Ancient Philosophy
5(1985]: 161-3). Admittedly, eros is equivalent to our "love" only in
contexts where sexual desire is appropriate (not, then, between family
members; see K. J. Dover, Greek Homosexuality [Cambridge, Mass.:
1978], 42-54); but it is not - at least not in all contexts - simply equiva-
lent to "sexual desire." The experiences of eros described in the Sympo-
sium and Phaedrus are manifestly experiences of falling in, and being in,
love.
2 Commentaries in English on the Greek text include Robert Gregg Bury,
The Symposium of Plato, id ed. (Cambridge, 1932); and K. J. Dover,
Plato's Symposium (Cambridge, 1980). Stanley Rosen, Plato's Sympo-
sium, id ed. (New Haven, 1987), is a book-length study of the dialogue in
English.
3 Commentaries in English on the Greek text include W. H. Thompson,
The Phaedrus of Plato (London, 1868); G. J. de Vries, A Commentary on
the Phaedrus of Plato (Amsterdam, 1969); and C. J. Rowe, Plato: Phae-
drus (Warminster, 1986) (includes translation). R. Hackforth, Plato's
Phaedrus (Cambridge, 1952), is a translation with running commentary.
Book-length studies of the dialogue in English include Ronna Burger,
Plato's Phaedrus: A Defense of a Philosophic Art of Writing (University,
Ala., 1980); Charles L. Griswold, Jr., Self-Knowledge in Plato's Phaedrus
(New Haven, 1986); and G. R. F. Ferrari, Listening to the Cicadas: A
Study of Plato's Phaedrus (Cambridge, 1987).
4 On this dialogue, see, e.g., David K. Glidden, "The Lysis on Loving
One's Own," Classical Quarterly 31 (1981): 39-59; David B. Robinson,
"Plato's Lysis: The Structural Problem," Illinois Classical Studies 11
(1986): 63-83; Anthony W. Price, Love and Friendship in Plato and
Aristotle (Oxford, 1989), chap, i; and, for a different approach, David
Bolotin, Plato's Dialogue on Friendship (Ithaca, N.Y., 1979).
5 This is brought to our attention in the immediate context, at 177c:
Aristophanes' plays are "all about Dionysus and Aphrodite" (that is,
drinking and sex). The prosaic Greek for "sex" is not Aphrodite but ta
aphrodisia, "the things of Aphrodite."
6 Cf. Frangois Lasserre, La figure d'Eros dans la poesie grecque (Lausanne,
1946), 10—11.
7 Cf. Thomas Gould, Platonic Love (New York, 1963), 24; G. X. Santas,
Plato and Freud: Two Theories of Love (Oxford, 1988), 16.
8 Fragment 130 in Edger Lobel and Denys Page, eds., Poetarum Lesbiorum
fragmenta (Oxford, 1955). See, in general, Bruno Snell, The Discovery of
Cambridge Companions Online © Cambridge University Press, 2006
27O THE CAMBRIDGE COMPANION TO PLATO
the Mind in Greek Philosophy and Literature (New York, 1982), 52-60;
Lasserre, La figure d'Eros-, and Anne Carson, Eros the Bittersweet (Prince-
ton, 1986), esp. 3-9.
9 Agathon (at 196C4—7) mentions the common agreement that "no plea-
sure is stronger than Eros" only in order to prove that, since Eros is
stronger than any pleasure, he must be self-controlled in the extreme!
10 For a sense of their variety consult Bury, Symposium, lii-iv.
11 See section III of this chapter.
12 For a compilation of suggestions, see Bury, Symposium, xxii-xxiii;
Rosen, Symposium, 90-1. Note in this connection Aristophanes7 quip at
189a.
13 Cf. Meyer W. Isenberg, The Order of the Discourses in Plato's Sympo-
sium (Chicago, 1940), 60; and Diskin Clay, "The Tragic and Comic Poet
of the Symposium," in Essays in Ancient Greek Philosophy, ed. John P.
Anton and Anthony Preus (Albany, 1983), 2: 188-9. To object (as do, e.g.,
G. K. Plochmann, "Hiccups and Hangovers in the Symposium/' Buck-
nell Review 11 [1963]: io ; and W. K. C. Guthrie, A History of Greek
Phiposophy, vol. 4 [Cambridge, 1975], 382) that had Plato attached sig-
nificance to this order he could have so arranged the seating plan from
the start, is to miss the point. It is because Plato attaches significance to
this order that he has caused it to arrange and flag itself before our eyes.
14 "Is directed toward" rather than simply "is.;/ It is not as if Diotima's
doctrine would commit Plato to the belief that, say, the kind of "love"
(eras) manifested by the tyrant in Republic IX-namely, unrestrained
lust - renders him praiseworthy. Diotima engages in depth psychology.
The majority of lovers stand revealed by her as yearning for something
more ultimate than the apparent object of their desire. Their apparent
behavior can therefore be condemned, even while the object of their
ultimate yearning is praised. (The depth psychology to which Plato actu-
ally appeals in his account of the tyrant's love, however, depends on a
more elaborate theoretical framework than Diotima provides - that of
the tripartite soul. This I shall discuss when we come to the Phaedrus,
in section III.)
15 Phaedrus also briefly praises Eros, the god, for being among the oldest of
divinities (178b). Here he scrambles to justify the implication of his
initial complaint, that Eros's godhead is significant apart from his status
as abstract force - a claim that the transparently allegorical quotes from
Hesiod and Parmenides do little to support.
16 See in general Dover, Greek Homosexuality, esp. 202; Michel Foucault,
L'usage desplaisirs (Paris, 1984); David Halperin, One Hundred Years of
Homosexuality (New York, 1990); David Cohen, "Law, Society, and Ho-
mosexuality in Classical Athens," Past and Present 117 (1987): 3-21.
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 271
17 See I77d8-ei, i93b6-7; and for Agathon's beauty, I74a9, 2i2e8, Pit.
3i5d-e, and Aristophanes, Thesmophoriazusae, 191-2.
18 We know almost nothing about the classical Athenian attitude toward
Lesbianism, beyond what we might infer from the fact that this is our
only reference to it in the period, and from the treatment of women in
general.
19 Dover, Symposium, ad i92ai.
20 Agathon plays with its range, distinguishing Eros's beauty from his vir-
tue at i96b4-5, but dubbing kallos the undifferentiated object of such
varied arts as poetry, medicine, and archery (i97a6-b5). I will return to
the relation between the beautiful [kalon) and the good (agathon) in
Plato when considering the climax of Diotima's speech.
21 Michael C. Stokes, Plato's Socratic Conversations (Baltimore, 1986),
114-82, gives a detailed account of anticipations of Diotima in Aga-
thon's speech.
22 For analysis of its logic, see R. E. Allen, "A Note on the Elenchus of
Agathon: Symposium 199C-201C," Monist 50 (1966): 460-3; Martha C.
Nussbaum, The Fragility of Goodness: Luck and Ethics in Greek Trag-
edy and Philosophy (Cambridge, 1986), 177-9; and Price, Love and
Friendship, 18-20. They are perhaps taking the argument too seriously,
given its shameless use of a trick against Agathon, namely, attributing
the characteristic of lovers to a personified Love; this is a trick that
governed Agathon's speech, but he plainly did not employ it in earnest.
Cf. Michael J. O'Brien, " 'Becoming Immortal' in Plato's Symposium,"
in Greek Poetry and Philosophy, ed. D. E. Gerber (Chico, Cal., 1984), 192
n. 23. (Price, however, denies that Socrates' use of this device is trickery.)
23 Cf. Isenberg, Order of the Discourses, 38, 59; R. A. Markus, "The Dialec-
tic of Eros in Plato's Symposium," in Plato, vol. 2, ed. Gregory Vlastos
(Garden City, N.Y., 1971), 133.
24 Cf. note 14.
25 This we have seen to be the primary domain of these words. Like the
English "love," however, the Greek terms can also be used with an
inanimate object.
26 For discussion see Santas, Plato and Freud, 32-9.
27 This would be a truism, for the word translated as "happy" (eudaimon)
commonly denotes an objective rather than subjective state of the per-
son (see K. J. Dover, Greek Popular Morality in the Time of Plato and
Aristotle [Oxford, 1974], 174).
28 Here I follow Bury, Symposium, ad 206b; Santas, Plato and Freud, 34.
29 For critical discussion of this inference see Santas, Plato and Freud, 35-
6. The broader question, what relation Diotima's appeal to immortality
bears to the treatment of immortality in other dialogues, is extremely
Cambridge Companions Online © Cambridge University Press, 2006
272 THE CAMBRIDGE COMPANION TO PLATO
vexed. Those interested in pursuing it should begin with O'Brien, " 'Be-
coming Immortal.' "
30 Markus, "Dialectic of Eros/' 138-40, compares this new emphasis on
procreation to the abundant generosity of God in Christian theology. Yet
Diotima introduces it as a mark of mortality (hos thnetoi, 2o6e8). A. H.
Armstrong, in "Platonic Eros and Christian Agape," Downside Review
79(1961): 105—21, and "Platonic Love: A Reply to Professor Verdenius,"
Downside Review 82 (1964): 199-203, adopts toward the Phaedrus an
approach similar to Markus's toward the Symposium.
31 For her use of the language of the Eleusinian Mysteries here, see Bury,
Symposium, ad 210a; see also Thomas A. Szlezak, Platon und die
Schhftlichkeit der Philosophie (Berlin, 1985), 259 n. 25. The ritual at
Eleusis - the Greater Mysteries - was preceded, several months earlier,
by a required ritual of preparation at Agrai - the Lesser Mysteries.
32 Cf. Rep. X 595D9-C3, Phdr. 27807-62.
33 Cf. Francis M. Cornford, Principium Sapientiae (Cambridge, 1952), 85.
O'Brien (" 'Becoming immortal,' " 188-9) usefully compares other Pla-
tonic examples of the contrast between political and philosophical vir-
tue. The choice of activities to exemplify types of generic love at
205d- money-making, love of sports, philosophy - seems parallel to
the tripartite hierarchy of the ideal city in the Republic, with its pro-
ductive class, its athletic soldier-guardians, and its philosopher-kings.
34 Here I follow Leon Robin, Platon: Le Banquet, in Platon: Oeuvres com-
pletes, vol. 4, part 2 (Paris, 1966), xciii; and Price, Love and Friendship,
41.
35 Pace Terence Irwin, Plato's Moral Theory (Oxford, 1977), 167, who
treats the philosophic initiate as equivalent to those fertile in soul in the
Lesser Mysteries.
36 In view of 2i2a3~4 and Ti. 2904-5, the underlying thought here may be
that the beauty of the words follows from the beauty of the topic.
37 Cf. Rep. Ill 4O2c-e, where Glaucon - elsewhere in the work noted as an
aficionado of the beautiful in body (474d-475a; cf. 468b) and a man of
some culture (548d-e |-assents enthusiastically to the proposal that
beautiful traits of character combined with outward beauty make the
finest spectacle of all, and insists moreover that outer deficiency, unlike
inner, does not disqualify a boy in the eyes of the cultured person (the
mousikos, 402d8).
38 Cf. Rep. 538c and note 37.
39 The parallel between the two stages is marked in the Greek at 210C3 by
the particle au, "correspondingly."
40 My English here reflects a correspondence between the vocabulary of
the Greek text at 2iob6 and at 210c5-6.
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 273
41 Cf. Price, Love and Friendship, 40.
42 The matter is controversial. For discussion see Gregory Vlastos, "The
Individual as an Object of Love in Plato/' in Platonic Studies, 2d ed.
(Princeton, 1981), 33-5; Julius M. E. Moravcsik, "Reason and Eros in the
'Ascent'-passage of the Symposium," in Essays in Ancient Greek Phi-
losophy, ed. John P. Anton and G. L. Kustas (Albany, 1972), 1: 293; Irwin,
Plato's Moral Theory, 169, 323 n. 58; Santas, Plato and Freud, 42,- Price,
Love and Friendship, 47-9.
43 Harry Neumann, "Diotima's Concept of Love/' American Journal of
Philology 86 (1965): 44. Diotima substitutes in this role verbs of seeing
and contemplation (passim), recognizing (211C8), touching (2i2a4-s),
and "being with" (2iid8-2i2a2) (the last a sexual pun).
44 The text at 210C7 seems to say that the initiate is "led" to this new
stage; but because of an awkwardness in the construction, some trans-
late the term differently (e.g., Alexander Nehamas and Paul Woodruff,
Plato: Symposium [Indianapolis, 1989], ad loc), while some others
emend. The transmitted text and the translation above is defended by
Dover, Symposium, 155.
45 Bury, Symposium, ad 2iod.
46 For an account of the Forms that emphasizes their status as individuals,
see Richard D. Mohr, "Forms as Individuals: Unity, Being and Cognition
in Plato's Ideal Theory," Illinois Classical Studies 11 (1986): 113-28.
47 Cf. Bury, Symposium, xlvi; and pace R. Hackforth, "Immortality in
Plato's Symposium," Classical Review 64 (1950): 44; Neumann, "Dio-
tima's Concept," 42-3.
48 This is not to claim that its stages directly correspond to those of the
ascent to the Good in Book VII of the Republic.
49 See further Leon Robin, Theorie platonicienne de Vamour (Paris, 1908),
220-4; Santas, Plato and Freud, 41; Price, Love and Friendship, 16; F. C.
White, "Love and Beauty in Plato's Symposium," Journal of Hellenic
Studies 109 (1989): 149-57.
50 On Aristophanes cf. Isenberg, Order of the Discourses, 53.
51 Pace, Ulrich von Wilamowitz-Muellendorff, Platon, id ed. (Berlin, 1920),
169-76; Neumann, "Diotima's Concept," who argue that Plato wishes us
to see Diotima as a suspect, sophistic figure. For an account that includes
a full survey of scholarship on the figure of Diotima, see David Halperin,
"Why Is Diotima a Woman?" in One Hundred Years.
52 Cf. Apollodorus's behavior in the Phaedo (ii7d). That the prologue em-
phasizes the telling and retelling of the long-ago conversation at Aga-
thon's house - its being kept current - stands as a further illustration of
this danger.
53 Cf. Bury, Symposium, lx-lxii; Szlezak, Schriftlichkeit, 262-70.
Cambridge Companions Online © Cambridge University Press, 2006
274 T H E CAMBRIDGE COMPANION TO PLATO
54 Some scholars understand Alcibiades' speech not just as a supplement to
what has gone before but as a critique or subversion of it; see, e.g., H. G.
Wolz, "Philosophy as Drama: An Approach to Plato's Symposium," Phi-
losophy and Phenomenological Research 30 (1969-70): esp. 349-53;
John P. Anton, "The Secret of Plato's Symposium," Diotima 2 (1974):
27-47 ; Michael Gagarin, "Socrates' Hybris and Alcibiades' Failure,"
Phoenix 31 (1977): 22-37; Nussbaum, Fragility of Goodness, 165-99;
and the Lacanian analysis of John Brenkman, "The Other and the One:
Psychoanalysis, Reading, the Symposium," in Literature and Psycho-
analysis, ed. Shoshana Felman (Baltimore, 1982), 396-456, esp. 402,
430-2, 452. (For a literarily aware reading of Alcibiades' intervention
that takes the more traditional stance, see Helen Bacon, "Socrates
Crowned," Virginia Quarterly Review 35 [1959]: 415-30.) Nussbaum is
especially impressed (as most readers have been to some extent) by the
emphatic eccentricity of the figure of Socrates in this dialogue. But this
impression needs to be tempered by taking into account the character
and relation to Socrates of those who transmit the stories about him:
Apollodorus, Aristodemus, Alcibiades.
5 5 For analysis of the setting consult, e.g. (in addition to the studies men-
tioned in note 3), Anne Lebeck, "The Central Myth of Plato's Phae-
drus," Greek, Roman and Byzantine Studies 13 (1972): 280-3; A. Philip,
Recurrences thematiques et topologie dans le Phedre de Platon," Revue
de Metaphysique et de Morale 86 (1981): 452-76; Kenneth Dorter, "Im-
agery and Philosophy in Plato's Phaedrus," Journal of the History of
Philosophy 9 (i97i):28o-i.
56 Whether this is authentically by Lysias or a Platonic parody is disputed.
See K. J. Dover, Lysias and the Corpus Lysiacum (Berkeley, 1968), 69-71.
57 It will become apparent that I read the Phaedrus rather as turning a
different facet of love to the light of philosophy than as intended to
provide a needed supplement to the Symposium, and still less to remedy
a defect in it of which Plato had previously been unaware [pace Santas,
Plato and Freud, 64; Price, Love and Friendship, 55, 85). This is not to
deny that the Phaedrus may cast brighter illumination on certain points
than the Symposium (as is provided, e.g., by its tripartite analysis of the
soul, discussed below). But the claim could equally be justified vice
versa. Since I deny that Plato is out to provide a comprehensive theory of
love, I do not see the Phaedrus as written to fill the gaps.
58 Here I treat the two speeches as equivalent; more detailed analysis
would reveal a development between them - see the studies cited in
note 3.
59 Some scholars, however, argue that the wisdom-loving part of the soul is
in itself the rational desire for what is good (Charles H. Kahn, "Plato's
Cambridge Companions Online © Cambridge University Press, 2006
Platonic love 275
Theory of Desire/7 Review of Metaphysics 41 [1987]: 80), or consists of
rational desires for the overall good (Irwin, Plato's Moral Theory, 195).
60 Cf. note 14.
61 Dover, Greek Homosexuality, 44.
62 This point is reinforced in the Greek text by its use of the dual number
to describe their reactions.
63 Cf. Price, Love and Friendship, 81.
64 I am indebted to Myles Burnyeat, "The Passion of Reason in Plato's
Phaedrus" (unpublished manuscript), for the connection of thought
here.
65 For the former, see, e.g., in the poets, Carson, Eros the Bittersweet, 7 -
8; elsewhere in Plato, Smp. i83a-b, Rep. I 329di. For the latter, see
Phdr. 269e3-27oa8 (with Diogenes Laertius, II.6), Rep. 5i6e-5i7a, Tht.
66 Pace Thompson, Phaedrus, ad 249c; and Griswold, Self-Knowledge, 124,
who take Socrates to restrict inspired love to philosophic types. Hack-
forth vacillates [Phaedrus, 101). In Ferrari, Listening to the Cicadas, chap.
6, I myself confused the issue by using the term "philosophic lover/;
rather than "inspired lover."
67 Here I disagree with those scholars who believe that Socrates gives a
further snapshot of their life together in his later description of the
relationship between the dialectician and his pupil (e.g., Hackforth,
Phaedrus, 164; Griswold, Self-Knowledge, 130; and cf. Elizabeth Asmis,
"Psychagogia in Plato's Phaedrus," Illinois Classical Studies^ 11 [1986]:
164). See Ferrari, Listening to the Cicadas, 230, and note 71 below.
68 Is their experience therefore comparable to a Freudian "sublimation" of
sexual energy? Francis M. Cornford, "The Doctrine of Eros in Plato's
Symposium," in Plato, ed. Gregory Vlastos (Garden City, N.Y.: 1971),
2:128-9, points out an important contrast: Whereas sublimation in
Freud is the diversion of instincts originally sexual to a new, nonsexual
aim, in Plato "the self-moving energy of the human soul resides properly
in the highest part," so that "sublimation" in this case would be not
simply a diversion of energy but a return of it to its source. See also
Santas, Plato and Freud, 169-72; Anthony W. Price, "Plato and Freud,"
in The Person and the Human Mind: Issues in Ancient and Modern
Philosopy, ed. Christopher Hill (Oxford, 1990), 244-52.
69 Vlastos's "Individual as Object" sparked considerable response on this
topic, however. See, e.g., L. A. Kosman, "Platonic Love," in Facets of
Plato's Philosophy, ed. W. H. Werkmeister (Assen, 1976), 53-64; Nuss-
baum, Fragility of Goodness, 166-7; Ferrari, Listening to the Cicadas,
182-4; Price, Love and Friendship, 10-12, 97-102. (For further bibliogra-
phy see Halperin, "Platonic Eros," n. 13.)
Cambridge Companions Online © Cambridge University Press, 2006
276 THE CAMBRIDGE COMPANION TO PLATO
70 On the mutuality of this love, see David Halperin, " Plato and Erotic
Reciprocity/' Classical Antiquity 5 (1986): 60-80.
71 I suggest, however (and here I supplement and modify my account of
these matters in Ferrari, Listening to the Cicadas, chaps. 1 and 7), that
the displacement of attention that we found to drive that ascent does
indeed correspond to something in the Phaedrus - but to its overall
structure as a dialogue, rather than to what Socrates says specifically
about love. In effect, the action of the Phaedrus offers us an image of
that mechanism of displacement (although not an exemplification, for
Socrates and Phaedrus only play at being in love, e.g., 234d, 243c). An
interest in Lysias's speech (an exercise in style) is overtaken in Socrates7
second speech by an interest in its beautiful topic, love (25733-4), via his
intermediary first speech, in which he gives birth in the beautiful (the
lovely location, 230b, 238d, and Lysias's speech as reflected in Phae-
drus's shining face, 234d, 23739) to a speech primarily intended as "se-
duction" of Phaedrus (237aio-bi). Having shifted his focus from speech
prompted by beauty to the beauty of which it speaks, in the second part
of the dialogue Socrates further displaces attention to the beauty in the
speaking (258d), with his examination of the art of rhetoric. Compare
the ascent, in the Greater Mysteries, from the level of moral practices
and speeches aimed at individuals to that of types of knowledge, consid-
ered abstractly. (So here we move from a morally edifying speech, still
ostensibly addressed, like the predecessor, to a beautiful boy-see
25663-25732 - to an inquiry into a branch of knowledge, the knowledge
of speaking, conducted through the question-and-answer appropriate to
philosophy.) And it will turn out that the engine of development for a
true rhetorician is the same as that of the philosophic lover in the Sym-
posium: namely, his being more concerned about his topic than about its
seductive use (27368-27432, 278a5-b2). Perhaps, then, the ancient pro-
posal was right: The Phaedrus is about "the beautiful in its manifold
forms" (Iamblichus, cited and developed by Hermias, Hermiae Alexan-
drini in Platonis Phaedrum Scholia, ed. Paul Couvreur, [Hildesheim,
1971] 8-11). But the issue is highly controversial. In addition to the
studies mentioned in note 3, consult, e.g., C. f. Rowe, "The Argument
and Structure of Plato's Phaedrus," Proceedings of the Cambridge Philo-
logical Society 212 (1986): 106-25; Malcolm Heath, "The Unity of
Plato's Phaedrus," Oxford Studies in Ancient Philosophy 7(1989): 151-
74; and Asmis, "Psychagogia."
Cambridge Companions Online © Cambridge University Press, 2006
NICHOLAS P. WHITE
9 Plato's metaphysical
epistemology
i
For some time philosophers have thought of epistemology and meta-
physics as different branches of philosophy, investigating, respec-
tively, what can be known and the basic properties and nature of
what there is. It is hard, though, to see any genuine boundary here.
The issues irresistibly overlap. Certainly in Plato there is no such
divide. His views about what there is are largely controlled by ideas
about how knowledge can be accounted for, and his thinking about
what knowledge is takes its character from convictions about what
there is that is knowable. As a result his doctrines have a different
shape from characteristically modern ones.
Some earlier Platonic writings do have a somewhat modern look.
Socrates was notorious for having questioned whether he knew
much of anything, and for making people hesitant about their opin-
ions (Meno 8oc, 86b—c). Plato exploits this side of Socratic thought.
The namesake of the Euthyphro judges that an action of his is pious.
Socrates wonders whether Euthyphro ought to be confident about
that judgment, and tries to make him less so. Elsewhere Socrates
raises questions concerning his own judgments about which things
are beautiful (H. Ma. 286c). Such questions seem to suggest a general
policy of doubting, reminiscent to us of Descartes or of the various
programs of ancient skepticism. In Socrates' efforts to overcome
ignorance (Meno 86b-c) we might see a project of justifying beliefs
like that of typical contemporary epistemologists.
Plato's thinking, though, moves in a different direction, and starts
I am greatly indebted to Richard Kraut for extremely helpful comments on an earlier
draft of this paper.
^77
Cambridge Companions Online © Cambridge University Press, 2006
278 THE CAMBRIDGE COMPANION TO PLATO
from a different place, too. Questions concerning Euthyphro's judg-
ments about what is pious and Socrates' judgments about what is
beautiful arise from a very specific source, different from the sources
of skepticism in its ancient and modern forms. It is emphasized that
Euthyphro's judgment - that it is pious for him to prosecute his
father for murder - is shocking to most people. But Plato does not
infer simply that Euthyphro's capacities to judge such matters are
fallible. He focuses on a particular question: whether Euthyphro
knows what piety is. The unspoken suggestion is that the dispute
between Euthyphro and his critics is due simply to a disagreement
on that question, and that an incapacity correctly to resolve the
disagreement must be due to a failure to answer the question rightly.
At any rate Plato immediately turns to trying to solve the problem
by giving a definition of piety.
A modern epistemologist would be concerned with the basis of
our concept of piety, but he would also be worried about other
sources of uncertainty and disagreement over which actions are pi-
ous. If he thought that piety was an objective property, as Plato
evidently does (Euphr. iod- i ib), he might well think that even if we
were all sure of what it was, we might still, through deficiencies in
our powers of cognition, make mistakes and disagree about which
things possess it. When Descartes doubted whether he was sitting by
the fire, he did not focus on the question whether he understood
what a fire was or what it was to sit by one.
Thus, although Plato raises questions that are similar to the ones
that we now ask about the justification of beliefs, his questions
cause him to search for definitions, not to combat any sort of general
skepticism. The initial problem is whether or not something is F,
but the question quickly becomes What is F-ness?
Questions of the form, What is F-ness? are precipitated by other
issues besides those about which things are F. Notably there are
causal or quasi-causal judgments about how a thing comes to be F. In
the Protagoras and the Meno, it is how one comes to be virtuous,
and in the Laches (i89e-i9oc), how one becomes courageous. There
are also judgments about what results from being F. In Republic I, it
is whether being just is beneficial. Plato insists that to make these
judgments we have to know what P-ness is (Prt. 36oe-36ic ; Meno
71b, ioob-c,- La. i89e-i9oc ; Rep. 354a-b). So although his search
for definitions (logoi) is stimulated partly by questions similar to
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 279
modern epistemological ones, these other quasi-causal issues are at
least as important in precipitating it.
Once the search for definitions has begun, problems about the
entities that they deal with and our judgments involving them
dominate Plato's thinking on metaphysics and epistemology to-
gether. The search for definitions itself becomes less important in
his middle works than it was earlier. In the Phaedo it is hardly
present, and indeed he perhaps announces there that it no longer
occupies the same place in his method.1 At the end of Republic I he
says that in order to tell whether it is beneficial to its possessor we
must know what justice is (354a-b). Subsequently he gives an ac-
count of justice in the city (433a-d) and in the soul (44id-442b),
and uses these accounts to show that justice is indeed beneficial
(see esp. 58ob-c). On the other hand he refrains from stating a fully
general definition of justice, and explicitly holds that the account
of justice in an action is distinct from, though related to, the ac-
count of justice in a soul or city (443e-444a). Moreover his way of
arriving at his description of psychic and civic justice is different
from his procedure in the earlier dialogues, which was to present
proposed definitions and to test them against counterexamples and
other objections.
The story of Plato's metaphysics and epistemology in his middle
works is thus no longer the story of his views about definitions. This
is not to say that he had stopped believing in the possibility or
desirability of establishing them. The Republic says that the philoso-
pher and dialectician is the person who can produce the logos of the
being of each thing (ton logon hekastou . . . tes ousias, 534b). On the
other hand the proposing and testing of definitions plainly does not
have the same pivotal role in his method that it had before.
A central part, however, is played by his view that certain entities
exist that have come standardly to be called Forms, eide (and are also
called Ideas, ideal, though they are not mental entities, contrary to
the suggestion of the English word "idea"). These entities figure
prominently in the Phaedo, Symposium, and Republic, which are
the works I shall mainly treat here along with the Hippias Major and
the Timaeus (I believe that these also defend essentially the same
position).2 I would maintain, in fact, that fundamentally the same
views are defended in Plato's other works after the Republic, notably
the Theaetetus, Sophist, Parmenides, and Philebus, though in them
Cambridge Companions Online © Cambridge University Press, 2006
28O THE CAMBRIDGE COMPANION TO PLATO
he does propose clarifications and some adjustments of his basic
position. I do not believe that he gave to his views anything ap-
proaching a major overhaul.
The Forms are central to Plato's metaphysics and epistemology. So
is the distinction between them and the objects of perception in the
natural world around us. The contrast between these two sorts of
entities is involved in his main theses about what there is and what
can be known. Since this contrast is drawn in terms of both the
metaphysical and the epistemological status of each kind of entity,
his views about Forms and perceptible things fall simultaneously
under both metaphysics and epistemology.
II
A good way to organize one's thinking about Plato's views is to
begin from his only really explicit argument for the distinctness of
Forms from perceptible objects, Phaedo 74b-c.3 This argument con-
tains two premises, one about equality and the other about percepti-
ble things such as sticks and stones. Each premise formulates a
crucial fact, as Plato sees matters, about Forms and perceptibles,
respectively. By following out the lines of thought suggested by the
two premises, we can gain a synoptic view of his ideas about Forms,
perceptibles, and the differences between them.
The first premise concerns perceptible objects:
(A) Equal sticks and stones sometimes, being the same,
appear equal to one person and not to another.
As I understand him, Plato is alluding here to familiar facts about
perceptual perspective, in particular the fact that a pair of equal
objects will look equal to a person seeing them from one standpoint
and unequal to someone looking at them from somewhere else.
The other premise concerns equality as we ordinarily think of it.
This, Plato argues, turns out to be a Form distinct from perceptible
objects:
(B) The equals themselves (auta ta isa) have never
appeared to you unequal, nor equality [isotes] inequality.
Plato takes (A) and (B) together to show that equality is distinct
from any perceptible equal things. If equality were simply percepti-
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 281
ble things that are equal, he assumes, then to think of equality could
only be to perceive them.* But by (A), perceptible equals can appear
unequal. Now if equality were simply perceptible equals, then in-
equality would likewise be perceptible unequals. In that case equal-
ity would sometimes appear just as inequality does, since any equal
perceptible objects must appear unequal in some circumstances. But
by (B), equality never appears in such a way.
As we consider this argument a couple of points should be kept
firmly in mind. The first is that the argument is expressly directed at
someone who does not yet agree that equality is distinct from percep-
tibles, but who thinks instead that in some sense equality is, or "is
nothing but/' perceptible equal things.5 Such a person is willing to
assume, however, that equality does exist or "is something" (Phd.
64c, 6$d), and that the word "equality" is the "name" of it (io2a-b,
c). A more hardbitten opponent, of course, would have said that
there is no such thing as equality at all, but only perceptible equal
things, and that the word "equality" itself does not designate any-
thing whatsoever. Plato's actual opponent is milder than this. He
allows Plato the word "equality" as a designation (in a broad sense)
for something, and disputes with him only over whether "equality"
designates some perceptible thing or things, or something distinct
from them.
The second point to bear in mind is that because Plato's opponent
starts out disbelieving that equality is a nonperceptible thing, his
initial acceptance of (B) is supposed to reflect a naive judgment in-
volving the term "equality" as ordinarily used, not a use of the term
laden with elements of Plato's own metaphysical doctrine. The
same is true of the other term that Plato uses here for this entity,
"the equals themselves." Although Plato's use of "equality" and
"the equals themselves" as interchangeable shows something about
his doctrine, as we shall see (sections VII and XI), the argument is
not intended by itself to commit the person accepting it to a sophisti-
cated philosophical theory, but only to a relatively naive statement
about equality, namely, that it is not any perceptible thing.
Consider premise (A), the easier one to understand and accept.
Perceptible objects, Plato thinks, are invariably capable of present-
ing contrary appearances (cf. sections III and VI). They are necessar-
ily embedded in a world that allows them to be perceived from
different perspectives. Differences among these perspectives inevita-
Cambridge Companions Online © Cambridge University Press, 2006
282 THE CAMBRIDGE COMPANION TO PLATO
bly bring it about that a perceptible object can appear both one way
and the contrary way, in some cases simultaneously [Rep. 523b-c,
524d-525a). Plato does not commit himself here to saying that this
point holds for all predicates, but he does not exempt any predicates
either.6
Premise (B) seems far more problematical. Plato takes it that when
a person thinks of equality, it cannot appear to him that he is thinking
of inequality. Plato believes, moreover, that this is a feature of the
thing, equality, itself: that it is, when thought about, incapable of
presenting a certain sort of appearance, the appearance of inequality.
The objection arises, however, that how such an object appears is
not a feature of the object but of the way in which it is thought about,
notably of the terms by which it is designated and brought, so to
speak, into one's thinking. If equality can be referred to as "Plato's
second-least favorite political concept," and if someone can use that
phrase in the mistaken belief that Plato's second-least favorite politi-
cal concept is inequality, then, it might be argued, such a person
would be using the phrase to think about what is in fact equality
while nevertheless taking it to be inequality. If what such an object
"appears" to be is a function of the way in which it is thought of, and
if such objects can be freely thought of by means of designations that
reflect false identifications, like the one just given, then no object can
be unmistakable in the way that Plato here claims that equality is.
We must accordingly suppose that for Plato not just any way of
referring to a thing counts as relevant to (B). The case constructed in
the previous paragraph would not count as equality's "appearing" to
be inequality. Perhaps this is because in such a case equality is not
felt to "appear" (phainesthai) at all to the person who is thinking
about it. At any rate when equality does genuinely "appear" to a
person, Plato thinks, it is incapable of presenting the appearance of
inequality.
There is a reason why Plato was not deterred by the objection that
how equality appears is dependent on the way in which it is brought
into one's thinking. The reason is that the same does not seem to
hold for perceptible objects. This fact, he thought, shows that percep-
tibles are by their nature deceptive in a way in which entities like
equality are not.? Even when perceptible equal things are believed to
be equal, and are introduced into one's judgment as, say, "those
equal sticks over there," the tendency of the sticks to appear un-
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 283
equal from oblique perspectives is generally undiminished (see Rep.
6o2e-6o3a, where the same point is conveyed by a different exam-
ple). In some cases, to be sure, the belief that they are equal will
make a person adjust his thinking instantaneously to the oblique
perspective and so regard them as equal, just as one takes a circular
tabletop seen from an oblique angle to be in one sense circular-
looking. On the other hand a person can easily ignore this adjust-
ment, and realize that in another sense the sticks from many view-
points do look unequal and the tabletop does look elliptical. Nothing
like this, it would be argued, is true of equality or circularity. If one
introduces an item into one's thought as equality, it seems impossi-
ble to think of any way in which it then appears as inequality.
Plato's opponent will be unimpressed. He will argue that some-
thing other than equality can still be introduced into one's thinking
through a mistaken belief that it is inequality, so that in this way
equality will be intrinsically no less potentially "deceptive" than
perceptible equals are. And he will say that what makes it impossi-
ble to take something introduced into one's thoughts "as equality"
to be inequality is not the nature of the object itself, but a feature of
the term by which it has been introduced. I shall not try here to
determine whether this reply leaves Plato any ground for maintain-
ing that there is a difference between perceptible things and the
entities putatively referred to by terms like "equality." I have said
enough to show how his argument is motivated.
in
Taken in isolation, the argument at Phaedo 74-b-c seems to deal
with an epistemological matter, the difference between the kinds of
appearances that can be presented by perceptibles and the things
that Plato calls Forms. We saw the argument immediately lead, how-
ever, to a metaphysical claim about the natures of entities like equal-
ity, namely, that unlike perceptible things they are incapable of pre-
senting certain sorts of appearances. Beyond this, however, there
arises another metaphysical issue.
Immediately after this argument Plato says that perceptible equals
are only "deficiently" equal, and in general that perceptible Fs are
only "deficiently" F (Phd. 74d-75b). Some interpreters have taken
this to mean that no sensible equals are exactly equal but are only
Cambridge Companions Online © Cambridge University Press, 2006
284 THE CAMBRIDGE COMPANION TO PLATO
approximately so. Plato never argues for such a contention, however,
and it conflicts with much else of what he says.8 We can make better
sense of this claim of deficiency by taking it to follow directly from
what he says in this argument itself. Perceptible equals are only
deficiently equal precisely in that they inevitably present the appear-
ance of being unequal.
This idea may seem like a crude conflation of epistemological and
metaphysical theses. Why should something be the less F for appear-
ing sometimes non-Fl* But there is incontrovertible evidence that
Plato espouses a view that he formulates in this way. When we see
what he is up to, it will be clear that what seems like confusion is
actually the result of a quite different notion of reality and objectiv-
ity from the one that we expect and are used to (cf. sections VI and
VIII).
Republic 475c-48oa is the only other passage in which Plato
comes close to giving an explicit argument for saying that Forms are
distinct from perceptibles. At 479a-c he maintains that every percep-
tible thing that appears beautiful appears ugly also, and analogously
for "just," "double," "large," and "heavy."10 As soon as he says this,
however, he immediately infers that perceptibles "no more are than
they are not what one calls them" (b9-io). That is, a perceptible
thing no more is F than it is not F. This is a repetition of the conten-
tion in the Phaedo that perceptibles are only "deficiently" equal and
the like.
Extending this line of thought, Plato goes on to say that since
perceptible things no more "are than they are not," in that sense
they are "between being and not-being" (479C7). On this basis he
argues that perceptibles are not what "knowledge" [episteme] is con-
cerned with; rather they are what "opinion" or "belief" (doxa) is
about. Knowledge has to do with "being" or "what is," namely,
Forms, which are implied to be distinct from perceptibles on the
same ground as in Phaedo 74b-c, that is, by not presenting contrary
appearances.11
There will be more to say about knowledge and belief later (sec-
tion X). These passages show, however, that Plato interweaves a
certain sort of epistemological consideration with what seems to us
like a quite distinct ontological issue. By virtue of presenting con-
trary appearances, perceptible objects are said to possess properties
"deficiently" and to be between being and not-being, and it is in-
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 285
ferred that they are therefore the province of opinion rather than
knowledge. From a modern point of view, it looks as though facts
about what we can know and facts about the natures of certain sorts
of objects are being curiously intermingled.
IV
To try to understand this state of affairs, it will be helpful to turn to a
confusion that is often attributed to Plato, a confusion over rela-
tional notions. The grounds for this attribution can be illustrated
from Republic 479a-b and 5236-524a. There Plato speaks of large
and small, heavy and light, and hard and soft as "contraries/7 and
says that it is characteristic of perceptibles to present contrary ap-
pearances, sometimes simultaneously (523b-c, 524d-525a). The ob-
jection is nowadays made that in fact these terms are not contrary
but covertly relational, amounting respectively to "larger than/7
"smaller than/7 and so forth, and that if Plato had recognized this
fact he would have realized that there is no contrariety involved,
because being larger than one thing is not contrary to being smaller
than another. Commentators have even carried the attribution of
confusion to the point of supposing that at Phaedo 74b-c, Plato
endorses the intelligibility of the idea of something7s being equal but
not equal to anything.12
That Plato was so fully confused about relations is made ex-
tremely unlikely by the fact that he plainly says that terms like
"father/7 "brother/7 "master/7 "slave/7 and "knows77 differ from oth-
ers by being related to (pros) others (Smp. i99d-2ooa, Prm. 133c-
134a). Moreover, there is an explanation available for the passages in
the Republic that obviates the need to suppose any such confusion.
In the first place, as a philosophical matter it is a mistake to think
that there is any easy way of paraphrasing ordinary uses of "large77 or
"heavy77 or "hard77 in relational terms. A number of recent discus-
sions have shown that this is so even if we combine these words
with sortal terms, as is often recommended, into phrases like "large
flea77 and "small elephant."1* For example, the thought that "large
flea77 might be equivalent to "larger than most fleas/7 as is often
thought, runs afoul of the fact that if the population of fleas con-
tains, say, 100 fleas two millimeters long and 75 one millimeter
long, then what is undeniably a large flea will be larger than less
Cambridge Companions Online © Cambridge University Press, 2006
286 THE CAMBRIDGE COMPANION TO PLATO
than one-half of the population. Somewhat similar considerations
militate against the paraphrase "larger than the average flea." Con-
ceivably there is some paraphrase that escapes this sort of problem,
but if so it is not easy to discover, and Plato can hardly be accused of
a blunder for having overlooked it.
Secondly, if we attend to what we consciously have in mind when
we say that something is hard, for instance [Rep. 523e-524a), it
seems plain that we do not intend to ascribe a relation to it, but
rather a property that seems on reflection to be monadic. When we
call a piece of Parmesan cheese hard, we are not thinking of some
particular relatum and saying that the cheese is harder than it, nor
are we even thinking of any definite reference class of things with
which we are comparing it. Pieces of cheese, pieces of Parmesan
cheese, things on the kitchen table, and so forth, all might be candi-
dates, but it is artificial to say that a reference to any of them is
actually part of the content of the judgment. Whether we are right to
think of ourselves as making these nonrelational judgments, and
whether this sort of introspection is ever or in principle a valid
source of information about the content of our judgments, are not
questions that I shall treat here. That Plato, like many philosophers,
often uses such introspection is evident, and that is all that my
interpretative aims require.
Though the alleged confusion over relations has often been pro-
posed as an explanation of Plato's remarks about contraries at Repub-
lic 479a-b and 523e-524a, it never suited all of the examples that he
uses there. These include "good/7 "just," and "holy," which do seem
to have genuine contraries and do not much tempt us to paraphrase
them in relational terms.1* The best course is to recognize that Plato
held that all of the notions expressed by these terms are (unlike
those expressed by "brother," "master," etc.) nonrelational.
Plato's outlook is clearly exhibited at Phaedo io2b-c. He explic-
itly rejects an overtly relational sentence, "Simmias overtops [hypei-
echei) Socrates," because he says that it expresses the fact less well
than "Simmias has tallness in relation to Socrates' shortness." He
then goes on to speak casually of Simmias's tallness and Socrates'
shortness (iO2d7, e2). He obviously recognizes something relational
about attributions of tallness (note pros, "in regard, or relation, to,"
at 102b). Nevertheless he still thinks of what is attributed as a prop-
erty, not itself a relation.^
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 287
A similar attitude is in evidence in the Hippias Major. Plato says,
"The most beautiful pot is ugly, when grouped with girls" (289a4~5),
and "If someone compares the class of girls with the class of
gods, . . . won't the most beautiful girl appear ugly?" (a9-b3). What
is being ascribed here is the property, ugliness, not the relation,
being uglier than something.
But however natural such attributions of nonrelational properties
might have seemed to Plato, they certainly seem awkward in many
cases. For example, although it is possible, introspectively, to think
of heaviness or hardness as a property, it is more difficult to think of
largeness in that way. Moreover, if largeness is to be a property, what
property could it be? Finally there is the problem that explicitly
worries Plato himself: If largeness is a property and smallness is its
contrary, why do these properties seem to attach to one and the same
thing simultaneously (.Rep. 523b-c, 524d-525a)?
In spite of these difficulties, we need to understand that Plato did
regard these terms as expressing genuinely nonrelational properties -
as indeed they seem to introspection to be - and he worked out his
position on that basis.16 In so doing he was in effect working out a
distinctive conception of reality (or, as we might put it, objectivity),
that is, of what it is for something to be really (or objectively) F. Since
this conception is unfamiliar to modern philosophy, we have to work
carefully to discern its outlines (which will begin to appear more fully
by section VI).
Plato sometimes uses the example of beauty (which in the Sympo-
sium he forcefully denies is a purely subjective matter) to express
this conception. As we saw (in section IV), he thinks that when a pot
is grouped with girls, we should say that it is ugly [H. Ma. 289a).
Leaving aside other questions for the moment, we can see that the
notion of ugliness here is the notion of a nonrelational property. To
be ugly in this sense is just to be ugly - without reference to a par-
ticular context or a particular viewpoint or anything else. When we
call something ugly, then even though we are in a particular situa-
tion and are comparing the thing to other things, we do not refer to
the situation or the things in our judgment itself, and what we mean
to say about the thing is just that it is ugly, period. The same idea
Cambridge Companions Online © Cambridge University Press, 2006
288 THE CAMBRIDGE COMPANION TO PLATO
emerges in Symposium 2iod-2iia. The notion of beauty described
there is the notion of a thing's being beautiful nonrelatively to an-
other part of the thing or to a time or place or relation or to whom it
is beautiful or to what kind of thing it is or, indeed, to anything else.
Likewise at Republic 5236-524a. There Plato says that perception
apprehends largeness and smallness and the like "deficiently" (e7).
That is, the context in which a thing is viewed determines whether
it will appear large or small (52364-5). The same point is said to
apply to thickness and thinness and hardness and softness. Plato is
not denying that whether or not something appears hard, or the like,
depends on the context. The meaning or content of the judgment
that a thing is hard, on the other hand, seems to him to make no
reference to that context. The implication is that the notions in
question are notions of a thing's being, for example, hard, quite
independently of the context. The contrast is drawn with a finger,
which presents the appearance of being a finger quite apart from the
context in which it is observed (523cn-d3). The notion of hardness
in question is the notion of a thing's being hard in a sense into which
mention of this or that circumstance does not enter.
Let us turn to Plato's remarks about time and tense, which show
us another important way in which he thinks that the notions ex-
pressed by such terms are nonrelational. He frequently says that a
thing that we now say is F will later cease to be so, and in general
that perceptible things are subject to change (e.g., Phd. jSd-j^a, 79c,
8ob; Rep. 526e, 527b, 533b, 534a; Ti. 48e-49a, 5ie-52a). He also
says in the Timaeus that perceptibles, unlike Forms, involve "was"
and "will be," whereas Forms involve only "is" (37C-38b). And a
little later he says that if you have a piece of gold that is constantly
being remolded into different shapes, it is "by far the best with
regard to truth" to say that it is gold, rather than saying that it is a
triangle or some other shape (5oa-b).
Although it is often said that Plato's point is simply that if a thing
is not permanently or essentially F it is a mistake to say that it really
is F, this is an unsatisfactory explanation of his view. Why should we
have wished to reserve "is" to be equivalent to "is essentially (or
permanently)"? A philosopher could make such a terminological
stipulation, but if it is not to seem arbitrary there needs to be some
account of why he thinks that "is F" should not be attached to a
thing if it is only temporarily F.
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 289
The explanation of his thinking is straightforward if we recur to
the point that Plato derived from introspection of the meanings of
our terms (section IV). If I say that a thing is a triangle, what I
actually have in mind typically involves no reference to time at all.
True, the thing that I refer to seemingly was previously and subse-
quently will be nontriangular. Nevertheless what I consciously
think to say when I use the phrase "is triangular" is not "is triangu-
lar now" or "is triangular roundabout now" or "is triangular at 10:00
A.M. on such-and-such a date," or anything of that kind. Rather, the
consciously entertained meaning of "is triangular" abstracts from
and ignores all consideration of time. The point is not that it is felt
to be false to say that a thing is triangular when one knows that it is
only temporarily so. The point is that the introspectible notion ex-
pressed by the predicate incorporates no thought of time at all, so it
is inappropriately applied to objects that are in time.1?
The upshot is that the notions represented by predicates in general
do not incorporate elements of time or tense. A person could under-
stand these notions of being triangular and being equal even if he
had no conception of temporality. That is Plato's picture. Thus he
thinks of the demiurge in the Timaeus as comprehending all of the
Forms before time has been created along with the cosmos and its
ordered motion. The timelessness of the Forms is more than just a
matter of their being the same through all time (though Plato often
says that they are indeed always the same); it is that they lie "out-
side" of time in the "eternity" of which time is only an "imitation"
(37d-e).
VI
Although it is plain in a general way that Plato's view of these
notions involves abstracting from various sorts of relations to time,
context, and perspective, it is not easy to see exactly how this is
done. Here I shall try only to present some of the interpretative
issues that must be taken into account, and to give an outline of how
I think they should be handled.
Sometimes it looks as though the main thing that he is concerned
with is relativity to the perspective of a person who is either perceiv-
ing a thing or thinking about it. For example, at Republic 479a-b, as
I have said, Plato infers, from the fact that a perceptible thing can
Cambridge Companions Online © Cambridge University Press, 2006
29O THE CAMBRIDGE COMPANION TO PLATO
appear both beautiful and ugly, that it no more is than is not beauti-
ful. Here it looks as though he is contrasting this way of being
beautiful with something's somehow being beautiful independently
of the perspective from which a person considers it.
In other places it seems that the pertinent relativity is to some
feature of the circumstances in which the perceptible thing finds
itself, not to the way in which it is observed. This seems to be so in the
case of time. When a thing changes, it often has a characteristic at one
time and a contrary characteristic at another time, regardless of
whether anyone is observing or considering it.18 This sort of change
does not seem to arise simply from relativity to an observer.1*
In other cases the situation is ambiguous, and the two types of
relativity seem intertwined. At Symposium 2ioe-2iia Plato de-
scribes the notion of something "not [i] beautiful in one way and
ugly in another, nor [2] beautiful at one time but not in another, nor
[3] beautiful in relation to (pros) one thing but ugly in relation to
another, nor [4] beautiful here but ugly there, nor [5] beautiful to
some but ugly to others." Of these, (5) seems to have to do with
relativity to observer, and (1) might also, but (2)-(4) seem fairly
plainly to involve the other sort of relativity, to the circumstances of
the thing rather than to conditions of observation.
At Phaedo 97a-b, again, Plato seems to hold that sometimes
when two things are far apart from each other, each "is one" and
they are "not then two," but when they approach each other they
"become two/' though in other cases when something that "is one"
is split, it "becomes two." It would seem in fact, however, that — if
one is to talk in this way - things that are two are two regardless of
their distance from each other. Perhaps what Plato has in mind is
that their appearing two depends on its being possible to see them
together.20 Similarly at Hippias Major 289a-c, it is said that the
most beautiful girl will appear ugly when compared with a god.
Then Plato says that we should agree that the most beautiful girl is
ugly when considered in relation to gods. It is not made explicit
whether the girl's being ugly is supposed to be relative to the gods,
relative to the observer comparing her with gods, or relative to both.
To take the case of beauty in the Symposium as an example again,
the question is what notion of beauty Plato is referring to. Is it the
notion of a thing's being beautiful somehow independently of the
observer and the perspective that the observer occupies? Or is it the
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 291
notion of a thing's being beautiful independently of its circum-
stances (apart from those of its being observed or considered)? For us
nowadays it seems almost irresistible to think that it must be one or
the other of these, but that it cannot be both.
In view of the evidence just cited, however, it seems to me clearly
best to say that both kinds of independence are implicated in the
notions that Plato is trying to express, and that (at least as far as
present purposes are concerned) they are on a par with each other.
The notion of beauty associated in the Symposium with the Form of
the Beautiful, then, is the notion of a thing's being beautiful indepen-
dently of both the perspective from which it is considered (whether
by sense or through thought) and the circumstances in which it is
located. Nothing in the passages cited indicates that either of these
aspects of the matter should be omitted from the interpretation.
We can understand Plato's view as thus interpreted, and perhaps
sympathize with it to some extent, if we continue to use introspec-
tion to think about what we mean when we ascribe such properties.
When one says that a thing is beautiful, then - philosophical theo-
ries aside - one seems to mean that it has a certain property that
does not involve either the thing's being examined from a certain
standpoint or the circumstances surrounding it. That is, as far as
what one has in one's consciousness is concerned, one seems to
ascribe a property whose attaching to the thing is in a sense intrin-
sic, namely, in the sense that in the thought of the property there is
contained no idea of either the fact that one is ascribing it from a
certain perspective or the fact that the thing is placed in a certain
situation. Of course, one is occupying a certain perspective as one is
ascribing the property, and the thing is located in a certain situation.
But the humdrum and yet - for the interpretation of Plato - crucial
fact is that these two facts are not themselves incorporated into the
judgment that one makes. (Of course, there are other judgments in
which viewpoint and perspective are part of the content of what is
judged, but they are another matter.) This is clearly true, for better or
worse, of many of our ascriptions of beauty and justice and goodness,
and also for many of our ascriptions of hardness, heaviness, and
other such seeming properties. These facts enable us to have a clear
sense of why Plato says the things that he says about such judg-
ments and the notions that figure in them.
In outline, then, this is the conception of reality or objectivity that
Cambridge Companions Online © Cambridge University Press, 2006
292 THE CAMBRIDGE COMPANION TO PLATO
Plato develops (cf. section IV). The notion of being really F is the
notion of being F in a way that is independent of both the viewpoint
of the judger and the circumstances of the object. It is thus the
notion of being in a certain particular sense, as I have said, "intrinsi-
cally" F.21 That term, though, has enough uses in philosophy to
make its use here inadvisable and potentially misleading. The impor-
tant thing to emphasize, for now, is not that it is the notion of a
thing's being F "in and of itself/7 but what its being F thus is indepen-
dent of - namely, viewpoint and circumstances.22
For us these days there seems to be something strange about
Plato's combining into one idea these two kinds of independence.
The combining becomes more readily intelligible when we notice
the facts of introspection just described, which allow us a way of
thinking of the two kinds of independence as on a par. The combin-
ing also illustrates Plato's disinclination to separate epistemological
and metaphysical issues in the modern way. From an epistemologi-
cal point of view the role of perspective in a judgment can seem
paramount, whereas the role of the circumstances of the object
seems like a different kind of consideration, involving the nature of
the property that one is ascribing. But Plato does not divide the topic
up in this way.
VII
Plato associates these notions with his special entities, the Forms.
By "associate" I mean, for one thing, that he supposes that to under-
stand the notion of an F, as explained above, is to stand in a certain
cognitive relation to the Form of F. Saying what this cognitive rela-
tion might be is subject to numerous difficulties. Sometimes Plato
portrays it as involving an ability to give a definition, as we saw
(section I), which he seems to take as a specification or description
or analysis of a Form. At other times he talks, to some degree meta-
phorically, as if it were a kind of intellectual "seeing" of the Form
[Rep. 484C-d, 500c). Whatever the exact character and relationship
of these two ideas may be, the relation of mind (and particularly
reason) to Forms is the core of his view about what the understand-
ing of our notions comes to.
Something similar to the latter idea arose earlier (section II) in the
treatment of Phaedo 74b-c. If that argument is to make sense, we
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 293
saw, not all ways of bringing equality into one's judgment can count
as cases of equality's "appearing" one way or another. It seems that
the kind of "appearing" that Plato has in mind takes place when one
deliberately and consciously sets oneself to, as we might say, think
about equality and consider what is true of it. Doing that presum-
ably falls short of anything that anyone would call "intellectual
seeing," but it is all that we need to discuss here of what might have
led Plato to use that manner of speaking.^
Plato's association of the notions I have described with Forms is
affected by another issue that arises in Phaedo 74b-c. Premise (B)
receives two formulations, (1) "Equality has never appeared to you
inequality", and (2) "The equals themselves have never appeared to
you unequal," both seemingly presented as equivalent.
Formulation (1) seems to us relatively unproblematical, and so I
have exploited it here so far. Formulation (2), though, raises difficul-
ties connected with the matter of what is called "self-predication."%*
For (2) makes it sound as though equality can be described as itself
"equals," that is, as if it were things that are equal. This difficulty
has two aspects, one connected with the plural expression "equals
themselves," and the other arising from the peculiarity of saying
that equality is itself equal, regardless of how many things it might
be. The former aspect I shall here treat as unimportant, since Plato
soon switches to the singular "the equal itself" (74C4-5) and hardly
ever uses the plural in speaking of Forms.
It is clear that Plato's argument for the distinctness of Forms does
not itself require that the Form of F be in general F, and in view of
the oddity of this idea it has understandably seemed to many inter-
preters better not to attribute it to him.25 On the other hand there
are numerous passages in which he seems to use language by which
self-predication is very strongly suggested, including claims to the
effect that perceptible objects are "imitations" or "copies" of Forms,
which are "paradigms" for them (e.g., Ti. 28e, 49a; Rep. 540a). He
also engages in later works (especially the Parmenides and the Soph-
ist) in reasoning that might be taken, as it has been, as a criticism of
his own earlier self-predicationist view. On the other hand the Ti-
maeus, a late work, continues to treat Forms as paradigms and per-
ceptibles as copies of them.26
Probably the best construal of this state of affairs is to affirm that
Plato does believe that perceptible objects are in a sense imitations
Cambridge Companions Online © Cambridge University Press, 2006
294 T H E CAMBRIDGE COMPANION TO PLATO
of Forms, but to deny that he intends this view in the way in which
the self-predicationist interpretation takes it. Before we examine
how this might be so (section XI), it will be helpful to look at some of
his thinking about how the notions associated with Forms are in fact
applied to perceptibles.
VIII
If Plato believes that we have the notion of a thing's being F - equal,
beautiful, just, square, and so forth2? - independently of its circum-
stances and the point of view from which it is regarded, and if he
takes the understanding of such a notion to be the cognition of the
Form of F, the question arises how he thinks such notions can be
exemplified by perceptible objects. The best answer, I think, is to
develop and clarify the line of thought pursued by those who say
that according to him, perceptibles and Forms belong to different
"types" or "categories" and that, more importantly, the way in
which terms apply to perceptibles is different from the way in which
they apply to Forms.28
The notion associated with the Form of F is the notion of a thing's
being F in abstraction from or independently of viewpoint and cir-
cumstances. This emphatically does not mean that it is the notion
of a thing's being F in all circumstances, at all times, and to all
viewpoints. That would obviously be absurd. Rather, the idea is that
conceiving of a thing's being F in this way does not involve bringing
anything to mind about circumstances, times, or viewpoints at all.
Such matters are simply not part of the notion and not part of the
property that it expresses. For when we introspect the notion of
what we ascribe when we say that an object is F, we can see that we
have in mind the property of being F in a way that involves nothing
about such things.2* (When F is a relational term like "equal" or
"brother" - cf. section IV - there is of course a reference to a corre-
late object, but not to any further viewpoint or circumstance with
regard to which the relational property is thought of as attached to
the thing.)
On the other hand difficulties arise when we apply such notions to
perceptible things. Here it must be stressed that Plato thinks that we
do, in our ordinary judgments, try to apply these notions to percepti-
bles. One's judgment that a perceptible thing is beautiful does not
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 295
present itself to one's mind as the judgment that "This is beautiful
as viewed by me from here when it is in such-and-such circum-
stances." Rather, it is normally the judgment that "This is beauti-
ful/' just like that with nothing added. Plato takes it as simply an
undeniable fact that our perceptions stimulate us to make such judg-
ments. 3° Never does he hint that we should reconstrue our judg-
ments as really meaning "This thing is beautiful as viewed by
me. . . . " If he had done that, the entire structure of his doctrine
would have been different.*1 In particular, he would not have stuck
so persistently as he does to terms like "beautiful," "equal," "hard,"
and so on. Instead he would have focused explicitly on complex,
multirelational judgments. He would also have done the same if he
had proposed a conceptual revision, leading us to replace "beautiful"
and so forth with relational substitutes. But he suggests no such
strategy.
Nor does he suggest another revision that his terminology might
have allowed him. Along with the term "imitation" he uses the
term "participation" for the relation of exemplification that a percep-
tible thing may bear to a Form. If we call a perceptible thing F, then
we can say that it "participates" in P-ness. Plato never says, how-
ever, that "This is F," said of a perceptible thing, is equivalent to
"This participates in P-ness" or could be replaced by it. As I have
said, he thinks that the judgments that we actually try to make do
express what we mean by the former sort of sentence. He does not
entertain the prospect of a revamping of our way of thinking that
would eliminate them. He only expects us to understand wherein
they are inappropriate.
Plato thinks, then, that we unavoidably do try to apply to percep-
tibles the notions of being F viewpoint-independently and circum-
stance-independently (for brevity let me henceforth usually include
time-independence under circumstance-independence). When we
do, though, we face a problem. It strikes one in a certain situation
as right to call a thing beautiful in such a sense. But in another
situation or at another time it strikes one, or someone else, as right
to call it ugly. Thus the viewpoint, time, or situation affects the
seeming appropriateness of applying a term whose meaning, as we
think about it, seems to involve no reference to such things. Our
understanding of the terms that we apply, together with what our
perceptions make us inclined to say about perceptible objects, turn
Cambridge Companions Online © Cambridge University Press, 2006
296 THE CAMBRIDGE COMPANION TO PLATO
out to be a misleading guide to what we shall be inclined to say
about perceptible objects under other conditions. In this way Plato
thinks of perceptibles as only "deficiently" fitting the notions that
we try to use to describe them (cf. section III).
Plato's sense of the peculiarity of applying viewpoint- and
circumstance-independent notions to perceptible objects is con-
veyed by his remarks in Republic 479b-c, where he tries to explain
his contention that such objects are "between being and not-being"
(479b6-7). Alluding to the fact that a perceptible thing that appears
beautiful also appears ugly, he says that it is like things that "are
ambivalent," which "it is impossible to think of.. . as being or as
not being or as both or as neither" (479C3-5). Since he is using
"being" here to represent the predicative notion of "being F (e.g.,
beautiful)" (cf. section III), his thought is that where a perceptible
thing is concerned, there is something wrong with saying either
that it is beautiful, that is not beautiful, that it both is beautiful
and is not beautiful, and that it neither is beautiful nor is not
beautiful. This proposition makes sense as an expression of the
idea that the term is in a sense simply not applicable to the object
at all.
IX
It still has to be explained, though, why Plato does not describe
perceptibles in a different way, by finding a notion of a thing's being
really F that could apply to them. After all, we do think that sticks
can really be equal even though they often do not appear so. And
when we think this, we apply a sense of "being equal" in which no
notion of viewpoint seems to enter: The sticks simply are equal,
quite apart from any consideration of how they appear. We also, for
example, use a notion of being good under which a thing can be said
to be good, quite regardless of its circumstances, consequences, or
the like, and we sometimes apply this notion to perceptible things.
We need to understand why Plato did not do the same thing.
Two factors are relevant here. First, his way of abstracting from
circumstances is sweeping. In particular, his notion of being F in-
volves no reference to time (section V). He rejects, in effect, the idea
that most of what we say about perceptibles can be taken as relative
to t imes- tha t "is F" can amount to "is F at time £." Conceiving
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 297
simply of a thing's being F, he holds, does not involve thinking of its
being in time at all. Second, his view that we do actually entertain
and use these completely viewpoint- and circumstance-independent
notions makes it seem implausible to him that any other notion of
really being F could be legitimate. We might say that two sticks are
equal independently of perspective and circumstances, but that
would still leave us with their changeability. Plato would therefore
say that we still are not operating with the notion that we really
have in mind when we think of being equal, which is independent of
all circumstances, including that of time and tense.*2 So because he
believes that we have in mind such notions of being F, the kind of
notion of objectivity or reality that satisfies us and allows us to
apply it to physical objects would not seem to him to be a serious
option.33
This is one important reason why Plato's epistemological enter-
prise is so different from modern theories. He has no interest, for
instance, in the idea that a perceptible thing is F just in case it
appears F to a standard observer in standard conditions. One prob-
lem is that picking out an observer as standard would be arbitrary. 34
At Republic 58ie-583a, he says that the person who is familiar with
all pleasures is the one who can tell which ones are most pleasant,
but he never says that their being most pleasant simply consists in
their being so designated by such an observer, and he goes on to give
a metaphysical explanation of what being really pleasant does con-
sist in (583b-586e). Similarly, in the Theaetetus he says that an
expert is someone who can make judgments about how things are
(i7oa-c), but he never hints that the expert could be specified by the
conditions under which he forms his judgments.35
Plato's thinking is equally inhospitable to the idea that a percepti-
ble thing is F just in case its being taken to be F causally explains its
various appearances. This would be Russell's idea, for example,
when he holds that what makes us say that the penny is really round
is that its being round can be cited as the cause of its appearing
round under some conditions and elliptical under many others.36
Plato never broaches such an idea. 37 He would allow that a thing's
being F can be part of what causes it to present various appear-
ances^8 but there is no reason to believe that he would accept the
claim that F-ness just is the property of causing those appearances.
His rejection of such ideas as these fits with the fact that on his
Cambridge Companions Online © Cambridge University Press, 2006
298 THE CAMBRIDGE COMPANION TO PLATO
view, coming to conceive the notions that concern him is not, as
empiricists regard it, a matter of constructing or deriving them from
perceptual appearances. Although he says that perception stimu-
lates us to bring certain notions to consciousness, they are notions
that we already possess [Phd. 75a-d), through what he calls ''recol-
lection/' without reliance on perceptual appearances.39 There is
therefore no paradox for him here in the idea that the notions do not
properly fit perceptible things.
Plato's radical concept of reality - of things' really being F*° -
thus leads him away from any concept under which, in spite of
conflicting appearances and on the basis of inferences from them, we
could say that in the perceptible world things really are thus-and-so.
Reality for him is indissolubly linked to objects of a different sort,
the Forms.
In spite of his refusal to say that our notions properly apply to percep-
tible things, he nevertheless does not deny that we make judgments
about them (cf. section VIII). We have a native propensity to make
such judgments about perceptibles. Under certain conditions a per-
ceptible object will strike us as having a certain character like that of
a Form with which we already possess some sort of acquaintance
(Phd. 75b-c), and we will call the object by the "name" [eponymia)
that we have associated with that Form (io2a-d).4I But Plato does
not believe that these judgments are based on clear rational grounds.
For at this stage we do not possess a clear knowledge of the Forms
that we are in effect making use of, and we do not possess definitions
of them (cf. section I). We simply have feelings for certain similari-
ties between what we perceive and the notions that we already have
in mind, and they incline us to make certain judgments. We simply
make them, and many of them become customary (nomima, Rep.
479d3~4, 484d2). People who make these judgments uncritically do
not even recognize the existence of Forms that represent their mean-
ings (475b, 479a).
Plato allows a difference, though, between more and less defensi-
ble descriptions of perceptibles. The rulers in the Republic, for exam-
ple, after they have gained full knowledge of the Forms of Goodness
and Justice (540b, 534a-b), are better able to govern the city than
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 299
they would otherwise be. The crucial reason is that they now know
what it is for a city to be just (cf. section I). But they must also be
supposed to realize which actions on their part would help lead the
city in that direction. In Republic X the wise man whose son has
died is able to determine, among other things, how he may best
improve his condition (6o4C-d). In the Phaedrus Plato indicates that
a person might learn to predict how certain sorts of speeches affect
certain sorts of people (267//, esp. 270b, 277a-b). In fact, the allegory
of the cave at the beginning of Republic VII says that people can
"divine" the future by remembering what sorts of events have oc-
curred in sequence in the past (5i6c-d; cf. Tht. I7id-i72b, 178a-
179a, i86a-b).
On the other hand Plato believes that there are strict limits to the
effectiveness of such efforts. The allegory of the cave strongly sug-
gests that the regularities that can be observed are merely accidental
("which [shadows] have customarily [eiothei] gone past earlier and
later or at the same time," Rep. 5 i6cio-di), and indicates that using
them as a basis of prediction is comparable to soothsaying (apoman-
teuomenoi, di). Moreover, all the evidence seems to show that in
Plato's view, the perceptible world is not deterministic or fully pre-
dictable.*2 He seems to believe that a priori investigation of Forms
can yield general statements that roughly describe physical things.
In their perceptible applications these general statements admit ex-
ceptions, but they are much more secure than statements derived
simply from perceptual observation.^ Taken in their pure applica-
tion to Forms, however, they are exceptionless and certain.
At bottom, I think (though there is no space to explain the interpre-
tation fully here) Plato rests this view on the idea that the properties
that he associates with Forms are, in the way that we have seen, not
properly exemplified by perceptible things. If "fire," for example, in
a sense involving no relation to perspective or circumstances, could
be applied to perceptibles, then he thinks that other features would
attach to it by ironclad regularity. 44 But if the property of being fire
seems attributable only in certain circumstances and from certain
viewpoints, then the strict links to other features will be disrupted
and become only probabilistic. The less relational the attachment of
properties to a thing, so to speak, the more regular the linkage of the
properties to each other.
As I briefly noted earlier (in section III), Plato maintained that we
Cambridge Companions Online © Cambridge University Press, 2006
3OO THE CAMBRIDGE COMPANION TO PLATO
have knowledge [episteme) only of Forms, and only belief or opinion
(doxa) about perceptibles (Rep. 475-480; Ti. 48e, 5id-52a). To expli-
cate fully what he meant by saving this would require another essay.
The foregoing observations, however, give some sense of what is
involved. One element has to do with the possibility of well-founded
predictions: Certain regularities hold securely for Forms but do not
when one applies them to perceptibles.
The more fundamental element, however, has to do with the fact
that whatever application of a term one is inclined to make to a
perceptible object, it is guaranteed that other viewpoints and circum-
stances will yield a contrary inclination. If I am inclined to call
something hard here and now, I will be inclined to call it soft at
another points Nothing comparable is true of judgments about
Forms. Moreover, the application of the notion of hardness to a
perceptible is, as we have seen (in section VIII), a misapplication, in
that it is an attempt to attribute a perspective- and circumstance-
independent property to a thing that can be apprehended only from a
perspective and within particular circumstances.
In the context of present-day epistemology Plato's view on this
topic provokes puzzlement. He clearly is not concerned with whether
or not judgments about particular perceptibles can be adequately jus-
tified. It is therefore difficult to interpret his views within an outlook
that closely associates the concept of knowledge with a concept of the
justification of belief. Rather, he is focused on the thought that
perspective- and circumstance-independent notions that we have in
mind are not applicable to perceptibles, and that judgments that try to
apply those notions to sensibles are therefore never stable.
XI
The Form of F, Plato says, is the "paradigm" of which perceptible
things that we call F are "imitations" or "copies," and this fact
makes people ascribe a "self-predicationist" view to Plato (cf. sec-
tion VII). For two reasons, however, his thought is not straightfor-
wardly that the Form of F and perceptible things can all share the
feature of being F.
For one thing, he is not willing to say that any perceptible objects
are F in the way that the Form of F is. When we make judgments
about them, as I have said, we do intend to say that they are, but
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 301
our inclination to make any such judgments about them comes and
goes, and that fact shows that something is wrong with them (sec-
tion VIII). More importantly, however, his view is not that the
Form of F is a thing that itself is, predicatively, F in a viewpoint-
and circumstance-independent way. Rather, it is the notion, and
also the property (cf. note 24), of a thing's being F in that way. The
self-predicationist view would say that the Form of F is (predica-
tively) F. Plato's view is that the Form of F is what it is to be F
apart from viewpoint and circumstances.
It is not surprising that he sometimes writes as though those
features attach to the Form of F that instead properly would attach
to a thing of which the Form is the notion, that is, a thing that is F
independently of viewpoint and circumstances. After all, to con-
ceive of the Form is precisely to conceive of a thing's being F in that
way. I think, however, that this is not what he intends. When he
exhibits difficulties with this idea (as he does in the so-called Third
Man argument, of which two versions appear in Parmenides, at
I3ie-i32b and I32d-i33b), his aim, I think, is to make clear that he
does not subscribe to it, and that although he may sometimes have
failed to think and write clearly enough to avoid it completely, he
regards such lapses on his part as deviations from his fundamental
line of thought.
He often says that perceptible objects "partake" or "participate"
in Forms. As he recognizes, the idea is obscure [Phd. iood, Vim.
131-135), and his middle works do little to elucidate it. His later
works try to explain it, especially the Parmenides, the Sophist, and
the Philebus (though arguably they end up making matters more
obscure). He faces more than just the now-traditional problem of the
relation of particulars to universals. In addition, his notion of a
thing's being perspective- and circumstance-independently F, and
the relation of this notion to objects that can be apprehended only
from a particular perspective and within particular circumstances,
raise substantial difficulties of their own.
As I indicated earlier, Plato resists the idea that the circumstance-
and viewpoint-independent notion of F might ever be definable in
terms of a viewpoint- or circumstance-dependent notion - for exam-
ple, that "hard" might be defined in terms of "harder than" or "hard
compared to," or that "beautiful" might be explained in terms of
"beautiful to so-and-so" or "beautiful in such-and-such circum-
Cambridge Companions Online © Cambridge University Press, 2006
3O2 THE CAMBRIDGE COMPANION TO PLATO
stances/' In fact it is clear that his sympathies lie with the reverse
order of explanation. He thinks that our applications of terms to
perceptible objects are developed in some way on the basis of our
comprehension of the viewpoint- and circumstance-independent no-
tions. This way of thinking might seem natural on grounds of intro-
spection. After all, it seems intuitively easy to suppose that the
notion expressed by "harder" is derived from the notion of "hard"
plus some notion of greater or lesser degree, and that "beautiful to so-
and-so" is built notionally from "beautiful" along with a notion of
perspective. This might not be the logical order of explanation, but it
has something to be said for it, notably a naive sort of naturalness.
There are clear signs that Plato pursued this idea. The use that he is
reported to have made of the notion of "the greater and the less" (or
the "indefinite dyad") seems to me likely to have arisen from the aim
of deriving some of our notions from a notion of comparison such as I
have just alluded to.** Here again he may be said to have followed a
course that introspection in some ways recommends.
Other problems about the relation between Forms and percepti-
bles are intensely troubling to him, but there is no space to deal with
them here. One particularly severe one is his need, seemingly forced
on him by a variety of considerations, to say that just like percepti-
ble objects, Forms have features that we are inclined to ascribe to
them only in certain circumstances.*? Others arise from the need to
explain why it is that although perceptibles have features only
viewpoint-dependently, we nevertheless make the mistake of think-
ing otherwise. In the Republic he attempts such an explanation for
the notion of pleasantness (which s83b-587b construes as funda-
mentally noncomparative). The Timaeus contains a number of such
explanations for other notions.*8
XII
Plato's metaphysical epistemology is too complex to be easily
summed up. I have tried to show how thoroughly intertwined its
metaphysical and epistemological elements are, and how much it
differs from modern epistemological theories, while at the same
time giving an outline of it that stands on its own apart from
comparison to modern views. It seems to me that Plato's fundamen-
tal insight - that as they appear to introspection, many important
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 303
notions are of things' having features in a way independent of one's
own perspective and of the circumstances surrounding the objects
about which the judgment is made-is correct as far as it goes.
That is, Plato is correct about what our introspective capacities
actually tell us about these notions and the properties that they
purport to us to express. Whether Plato was right to develop this
insight as he did, on the other hand, is far more doubtful, as is the
question whether such introspection is a good philosophical guide.
The falsity of the insight would necessarily call his whole meta-
physics into question.
In spite of what introspection may indicate, it is highly question-
able whether the properties that we ascribe to perceptible physical
objects should be construed as actually viewpoint-, time-, and
circumstance-independent in the way Plato takes them to be. Per-
haps many of them, for example, should be taken as relations to
times. Others of them are perhaps subjective, that is, are relations to
observers or conditions of observation. Some may be comparatives,
as those who criticize Plato have maintained, and others seem to be
relational in other ways that he did not notice. Still others might be
"syncategorematic" - for example, many uses of "good" may be
equivalent to "good (as an) F."
These difficulties did not go unnoticed in antiquity. Aristotle main-
tained, for example, that Plato was wrong to think that we have a
notion of goodness that abstracts from all circumstances and things
to which it is applied, and took the same view about, for example, the
notion of being (Nichomachean Ethics 1.6). He also maintained that
Plato was mistaken in thinking that certain relational notions can be
understood in abstraction from their correlatives.49
A further problem arises for Plato from the fact that on certain
plausible "holistic" views, the notions we employ, and perhaps also
the properties they express, are each dependent for their identity on
the others. As Plato's view about Forms begins to be expounded, it
seems to tend toward a strict atomism.5° Elsewhere, though, he
seems to accept some sort of essential interrelation among Forms
that has the epistemological consequence that knowledge of certain
Forms, particularly the Good, is necessary for knowledge of the oth-
ers [Rep. 510b, 511b, 5i7b-c, 5i9C-d, 526c). This idea is alluded to
by Plato - though it is not clear to what conclusion - in later works,
especially the Theaetetus and the Sophist (e.g., Tht. 2o6c-2o8b), and
Cambridge Companions Online © Cambridge University Press, 2006
304 THE CAMBRIDGE COMPANION TO PLATO
was perhaps developed by others in Plato's Academy into a thorough-
going epistemological holism.*1
An interesting historical question, which I shall not try to answer
here, is whether Plato's distinctive approach to epistemological and
metaphysical problems, differing as it does from so much modern
epistemology and even from much ancient epistemology (notably
the Skeptics and also the Stoics and Epicureans), is due to the idio-
syncratic thinking of one philosopher, Plato, or whether it repre-
sents a general approach to these issues that was characteristic of
the philosophical outlook of the period. I incline toward the former
view, but it would be difficult to prove that it is correct.
NOTES
1 See Gareth B. Matthews and Thomas A. Blackson, " Causes in the
Phaedo," Synthese 79 (1989): 581-91.
2 I see no good reason not to regard the Hippias Major as genuine and as
reflecting views of Plato's as of the time of the Phaedo. See Paul
Woodruff, Plato: Hippias Major (Indianapolis, 1982), esp. 161-80.1 also
agree with the traditional dating of the Timaeus as a late work and with
the view that it espouses essentially the same epistemological and meta-
physical views as the middle works. For some considerations concerning
its relative date see Ian Mueller, "Joan Kung's Reading of Plato's Ti-
maeus," in Nature, Knowledge and Virtue: Essays in Memory of Joan
Kung, ed. Terry Penner and Richard Kraut, Apeiron 22 (1989): 1-27; and
William J. Prior, Unity and Development in Plato's Metaphysics (Lon-
don, 1985), 168-93.
3 The interpretation of this argument to be given here is defended more
fully in my "Forms and Sensibles: Phaedo 74B-Q" Philosophical Top-
ics 15 (1987): 197-2.14-
4 This presupposition raises difficult issues that cannot be fully treated
here. In a nutshell, the problem is that Plato is tacitly ruling out the possi-
bility that without there being such a nonperceptible entity as equality,
"thinking of equality" might consist in something other than the mere
perception of equal perceptible objects. For brief discussion see note 32.
5 For this way of characterizing the view of the opponents that Plato
envisions see Terry Penner, The Ascent From Nominalism (Dordrecht,
1987), 54, 60, 95-121.
6 There are many problems about just which predicates Plato takes to
correspond to Forms. For example, some interpreters think that, at least
in his early-middle period, there are Forms corresponding only to cer-
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 305
tain predicates. For one main version of this interpretation see G. E. L.
Owen, "A Proof in the Peri Ideon," Journal of Hellenic Studies 77 (1957):
103-11. Prm. 130b—e is sometimes taken to be evidence for such an
interpretation. I think that the passage should be explained in another
way, and that it is weak evidence in the face of the fact that in earlier
works Plato does not mention such a restriction. (A recurrent misread-
ing of Rep. 523a-525 is sometimes used to support the contention that
Plato does endorse the restriction there. Plato says there that some predi-
cates appear to apply to the same perceptible object simultaneously, and
that those predicates are the most apt to call forth the realization that
Forms are distinct from perceptibles. Some interpreters, starting from
Owen, mistake this for the quite different suggestion that only such
predicates have Forms corresponding to them.)
7 See Gregory Vlastos, "Degrees of Reality in Plato/' in New Essays on
Plato and Aristotle, ed. Renford Bambrough (London, 1965), 1-19.
Though the present interpretation is on this point much indebted to
Vlastos's and to Owen's (in "A Proof") from which Vlastos's interpreta-
tion is substantially derived, there is much that surrounds Vlastos's
account of this point that I would differ with, in particular the matters
treated in sections IV-VIII of this chapter (cf. note 21).
8 See Owen, "A Proof"; Vlastos, "Degrees of Reality/' 1-19; and Alexan-
der Nehamas, "Plato on the Imperfection of the Sensible World," Ameri-
can Philosophical Quarterly 12 (1975): 105-17.
9 For further discussion of the role of this problem in Plato, see Myles
Burnyeat, "Conflicting Appearances," Proceedings of the British Acad-
emy 65 (1979): 69-111.
10 Plato's treatment of "double" and "half" introduces special problems,-
cf. note 49.
11 There is no space here for an analysis of Plato's argument. The present
interpretation of its conclusion is the traditional one. The conclusion as
thus construed is treated later on as having been demonstrated (e.g., Rep.
507b, 484b, c-d, and 485b with 479a-b), and 475e-48oa seems to me the
only place where Plato could take the demonstration to be. It has been
maintained, however, that this and related Platonic arguments are ef-
forts to show, not that Forms are distinct from perceptible particulars,
but that some important notions (particularly ethical ones) cannot be
explained in terms of "observational" predicates. See esp. J. C. B.
Gosling, "Republic V: ta polla kala," Phronesis 5 (i960): 116-28; Ter-
ence Irwin, Plato's Moral Theory (Oxford, 1977), 147-8, 151-2; Gail
Fine, "Knowledge and Belief in Republic V," Archiv fur Geschichte der
Philosophie 60 (1978): 121-39; C. D. C. Reeve, Philosopher-Kings: The
Argument of Plato's Republic (Princeton, 1988), 58-71.
Cambridge Companions Online © Cambridge University Press, 2006
306 THE CAMBRIDGE COMPANION TO PLATO
12 See Owen, "A Proof/' i I O - I i. For further criticism of the idea that Plato
was confused about relations and more on the interpretation presented
here, see my "Perceptual and Objective Properties in Plato/' in Nature,
Knowledge and Virtue: Essays in Memors of Joan Kung, ed. Terry
Penner and Richard Kraut, Apeiron 22 (1989): 45-65, and references
therein.
13 See John Wallace, "Positive, Comparative, Superlative/' Journal of Phi-
losophy 69 (1972): 773-82; Samuel Wheeler, "Attributives and their
Modifiers/' Nous 6 (1972): 310-34; and Philip Kitcher, "Positive Under-
statement: The Logic of Attributive Adjectives," Journal of Philosophi-
cal Logic 7 (1978): 1-17.
14 In "A Proof," G. E. L. Owen suggested that we classify all of the terms
that Plato deals with here as "incomplete," requiring a filler to make
their sense fully explicit. This might seem to work for "good," provided
that it is taken to have to mean "good (as an) F," as some have claimed it
should be. On the other hand Plato nowhere espouses that claim, and
anyway there is no obvious way to apply it to "just" or "holy."
15 One way of thinking of the matter is to say that instead of thinking of
relations as polyadic and instantiated by ordered sets of objects, he
thinks of them as monadic properties that are each instantiated by a
single object, but are nevertheless in some sense tied to each other in
such a way that when one is instantiated by one thing, the other must be
instantiated by another. See Hector-Neri Castaneda, "Plato's Phaedo
Theory of Relations," Journal of Philosophical Logic 1 (1972): 467-80;
with White, "Perceptual and Objective Properties," esp. 46-7. See also
the treatment of "one" at Prm. i29c-d.
16 For us, though not for Plato, a natural way to explain these phenomena
is to say that such terms function somewhat as demonstratives do.
"That is the Acropolis" does not to introspection mean the same as
"The freestanding object most salient in the scene ahead is the Acropo-
lis," but nevertheless it is liable to the accepted as true just in cases
where the freestanding object most salient in the scene ahead of the
speaker and the hearer is the Acropolis. Very roughly, demonstratives do
their job not simply by conveying a certain content directly (particularly
not one that it consciously entertained), but by combining with the
context to cause the hearer to fix on the object that the speaker means to
say something about. A sentence with a term like "large" does its job,
likewise, not by itself expressing as part of its meaning a particular sort
of comparison to a particular reference class, but, again very roughly, by
collaborating with the context to cause the hearer to fix on both a refer-
ence class and a place in it that suits what the speaker wishes to say
about the object of which "large" is predicated. If different reference
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 307
classes are indicated simultaneously, sometimes the same thing can
acceptably be called both large and small.
17 I would also argue that in Plato's view, concerning an object, action, or
event in the past it can only be said that it was F, not that it is F (and
analogously for the future). For this reason it is inappropriate to apply
the term F to it in the particular time- and tense-independent sense that
he thinks we have in mind for our terms. (This is part of the more
complete story of why Ti. 37b-38c says that there is "was" and "will
be" for perceptibles but not for Forms. It is also the reason why Plato
usually emphasizes the perishability of perceptibles as much as their
changeability: of anything that previously existed we have to say that it
"was" F, even if it perhaps was F throughout its existence [cf. note 32].
Plato's point therefore does not require him to say - as he does not -
that it is false that a thing is F if it no longer exists.)
18 Someone might think that change in an object could be thought of as
generated merely out of a shift in the temporal perspective of an ob-
server. For someone might claim "X is F at tz but not-F at £2" just means
"X is F as viewed from tr but not-F as viewed from £2." Some passages in
Plato suggest such an idea, but there is no space to investigate them
here - and the claimed equivalence seems evidently false, since a thing
viewed from one perspective can appear to have changed or to be going
to change.
19 At Tht. 178-180, in arguing against the Protagorean relativist, Plato
relies on the assumption that the passage of time cannot be supposed to
take place only relative to the observer.
20 A related issue arises in connection with Tht. i54c-d.
21 Owen advanced the view that the Form of F is associated with the idea of
something's being F "unqualifiedly" (haplos) and kath' hauto. See "A
Proof," esp. 107-11. Owen's view was a significant advance on previous
interpretations, and the present way of understanding Plato is greatly
indebted to it and to its development by Vlastos in "Degrees of Reality"
(cf. note 7). However, Owen's view failed to recognize a number of impor-
tant things. One is that Plato's notion of being F is perfectly compatible
with a recognition that some terms are relational (cf. section IV and note
50). Another is that we must not confuse being F in this way with being
F in all ways or at all times, and that (cf. "A Proof," 110-11) the point of
Plato's notion is indeed to exclude relativity, not primarily contrariety.
A third is that Plato's notion combines independence of the two kinds of
"qualifications" highlighted here.
22 In other connections, though, a notion of a thing's being "intrinsically"
F would be apt. For example, Aristotle uses the term ousia for something
that is a "kath' hauto, or per se, being," in the sense of not depending (in
Cambridge Companions Online © Cambridge University Press, 2006
3O8 THE CAMBRIDGE COMPANION TO PLATO
a certain sense) on anything else for its being. This idea is closely related
to Plato's view that Forms are ousiai, and that the Form of F is the
notion of a thing's being F kath' hauto, i.e., "intrinsically" in the sense
described. (Note, though, that in this sense "intrinsically" is not equiva-
lent to "essentially.")
23 See my Plato on Knowledge and Reality (Indianapolis, 1976), 91, on
Plato's use of the metaphor of "seeing" Forms. For differing views on
how seriously Plato takes the idea of some kind of "nondiscursive"
cognition of Forms, see, e.g., ibid.; Gail Fine, "False Belief in the
Theaetetus," Phronesis 24 (1979): 70-80; Richard Sorabji, "Myths about
Non-Propositional Thought," in Language and Logos, ed. Malcolm
Schofield and Martha Nussbaum (Cambridge, 1982), 295-314.
24 The term was coined by Gregory Vlastos, "The Third Man Argument in
Plato's Parmenides," Philosophical Review 63 (1954): 319-49.
25 See esp. Penner, Ascent from Nominalism.
26 G. E. L. Owen, "The Place of the Timaeus in Plato's Dialogues," Classi-
cal Quarterly 3 (1953): 79-95. For some arguments against Owen's dat-
ing of the Timaeus in Plato's middle period, see the references in note 2
of this chapter.
27 Cf. note 6.
28 See, for example, Richard Patterson, Image and Reality in Plato's Meta-
physics (Indianapolis, 1985).
29 The attentive reader will have noticed that I have often oscillated be-
tween speaking of "properties" and "notions" or "concepts." Plato's talk
of Forms embraces both things, and for present purposes the differences
between them can be left aside.
30 This is not to say that the making of such a judgment is itself the
perception; cf. Tht. 184-186.
31 The closest he comes is his presentation of the "secret doctrine" of
Protagoras in the Theaetetus, but even here he does not stick consis-
tently to the linguistic revision (esp. at i7ia-c, as is often pointed out).
32 Perhaps some features of perceptible objects are timeless and tenseless.
For example it has been argued (for example by Irwin, Plato's Moral
Theory, 319-20) that a particular action that is good is good indepen-
dently of viewpoint, circumstances, and also time (because if it is good
at one time, it is good at all times). This matter requires extended exami-
nation. I would argue that Plato thinks that we are inclined to call a
particular action good, and the like, only by virtue of adopting a particu-
lar point of view. (I have also maintained that in his view, concerning an
action in the past it can only be said that it was good, whereas the notion
of being good is a tenseless notion; cf. note 17.)
33 In passing I earlier said that on Plato's view, if equality were perceptible
Cambridge Companions Online © Cambridge University Press, 2006
Plato's metaphysical epistemology 309
equals, then to think of equality could only be to perceive them (cf. note
4). We can now see, very sketchily, why this is so. For Plato to think that
we could gain the idea of the equality of perceptible objects by some-
thing other than merely perception, he would need to have the idea of
our somehow constructing, rather than merely apprehending by percep-
tion, the notion of a perceptible object's being really equal in spite of
sometimes appearing unequal. But this is, as we can now see, not an idea
that he has any interest in developing.
34 See Bertrand Russell, The Problems of Philosophy (London, 1912), chap.
2.
35 Indeed, he never even hints that such a strategy might be used against
the Protagorean relativism that is there being discussed.
36 Russell, Problems of Philosophy, chap. 4.
37 Sph. 247d-e proposes that to be is to have the capacity (dynamis) of
doing or suffering something. I do not think that Plato ever accepts an
analysis of any property other than being in terms simply of its capaci-
ties or effects.
38 Such propositions are presupposed by the accounts of perception in Ti.
6iff.
39 For present purposes it does not matter whether Plato conceives of this
process as genuinely one of recollection (see also Meno 80-86, 98-100).
The important thing is only that the notions be already in the mind, and
not put there by some process originating in perception.
40 This equivalence holds for present purposes (cf. section III). In a broader
treatment one would discuss the Sophist and its role in developing
Plato's notion of being.
41 The Cratylus discusses how this association is established. Cf. White,
Plato on Knowledge and Reality, chap. 6; and Bernard Williams, "Cra-
tylus' Theory of Names and its Refutation," in Language and Logos, ed.
Malcolm Schofield and Martha Nussbaum (Cambridge, 1982), 83-93.
42 Gregory Vlastos, "The Disorderly Motion in the Timaeus," Classical
Quarterly 33 (1939): 71-83; Glenn R. Morrow, "Necessity and Persua-
sion in Plato's Timaeus," Philosophical Review 59(1950): 147-64; Har-
old Cherniss, "The Sources of Evil according to Plato," Proceedings of
the American Philosophical Society 98 (1954): 23-30.
43 Cf. Alexander P. D. Mourelatos, "Plato's 'Real Astronomy': Republic
527d-53id," in Science and the Sciences in Plato, ed. John P. Anton
(Delmar, N.Y., 1980), 33-73-
44 See my "The Classification of Goods in Plato's Republic," Journal of the
History of Philosophy 22 (1984): 393-421.
45 If one does not understand what Plato is doing, it becomes easy to think
that he has confused two different kinds of problems that may affect a
Cambridge Companions Online © Cambridge University Press, 2006
3IO THE CAMBRIDGE COMPANION TO PLATO
tensed belief: (i) the fact that arguments may shake one's faith in it and
cause the holder of it to give it up, and (2) the fact that the circumstances
may change and cause the holder of it to give it up on that account. If I
believe that Socrates is just, I may change my mind because someone
convinces me that I have been mistaken; or Socrates may be corrupted
and I may accordingly give up my belief that he is just. To Plato it does not
matter that these two phenomena are different. They both illustrate
equally the problem that he sees in applying "is just/; to perceptibles,
which is that one's inclination to apply a given predicate will in fact - for
both sorts of reason - eventually somehow be countered by the inclina-
tion to apply the contrary.
46 See, for example, the passages collected by W. D. Ross as fr. 2 of the De
Bono in Ahstotelis Fragmenta Selecta (Oxford, 1955).
47 For example, I take it that according to the Sophist, being and not-being
are such features, and that a particular problem arises because being
itself can be said, in different relations, both to be and not to be.
48 For example, Ti. 6id-62b ("hot"), 62c-63e ("light" and "heavy").
49 See De Sophisticis Elenchis c. 31, 181^25-35, where Aristotle insists
that "double" either means nothing when detached from the phrase
"double of half" (26-28) or else means something different from what it
means in the phrase (33-35). Owen, "A Proof," n o , mistakenly con-
strues this as a general criticism of Plato for not understanding that
some notions are relational. Recall that "double" is treated in a particu-
larly problematical way at Rep. 479b (cf. note 10).
50 Note Plato's description of Forms as monoeides, Phd. 78d, 80b. I think
that this description is to be connected with the fact that each Form is
taken to be the Form of a single notion, F, detached from all others, as a
comparison with Smp. 2 i ia5-bi seems to confirm.
51 See W. D. Ross, Aristotle's Prior and Posterior Analytics, corr. ed. (Ox-
ford, 1957), 605, 659-660.
Cambridge Companions Online © Cambridge University Press, 2006
RICHARD KRAUT
10 The defense of justice in Plato's
Republic
In this essay I will try to identify and explain the fundamental argu-
ment of Plato's Republic for the astonishing thesis that justice is so
great a good that anyone who fully possesses it is better off, even in
the midst of severe misfortune, than a consummately unjust person
who enjoys the social rewards usually received by the just.1 Plato's
attempt to defend this remarkable claim is of course the unifying
thread of the dialogue, but his argument ranges so widely over di-
verse topics that it is difficult to see how it all fits together, and
anyone who attempts to state his argument must take a stand on
interpretive issues about which there is considerable scholarly con-
troversy.2 The dialogue's difficulty is increased by Plato's failure to
give any explicit justification for the complex moral equation he
boldly announces: Justice discounted by pain and dishonor is more
advantageous than injustice supplemented by the rewards of justice.
Even if he manages to show that justice is the greatest single good,
we are still left wondering whether its value is high enough to make
this equation come out right. My main thesis is that the theory of
Forms plays a crucial role in Plato's argument for that equation, but
that the precise way in which that theory contributes to his defense
of justice is difficult to recognize. It is hard to overcome a certain
blindness we have to one of Plato's principal theses - a blindness we
can find in one of Aristotle's criticisms of Plato's conception of the
good. My goal is not to show that Plato's theory is defensible against
I am grateful to audiences at Clark University, Johns Hopkins University, Northwest-
ern University, the University of Michigan, and Wayne State University for their
comments on earlier drafts of this essay. In addition, I profited from the criticism of
Christopher Bobonich, Sarah Broadie, Shelly Kagan, Ian Mueller, Constance Mein-
wald, and David Reeve.
Cambridge Companions Online © Cambridge University Press, 2006
312 THE CAMBRIDGE COMPANION TO PLATO
all objections, once we correct for the mistake Aristotle makes. But I
do think that there is something powerful in Plato's argument, and
by criticizing Aristotle I hope to bring this feature to light.
I said that I will focus on Plato's "fundamental" argument that justice
is in one's interest, but it might be wondered why any one argument
should be singled out in this way and given special attention. For on
the surface, the Republic seems to present four independent attempts
to support the conclusion that justice pays apart from its conse-
quences.3 First, at the end of Book IV, we learn that justice is a certain
harmonious arrangement of the parts of the soul. It is therefore re-
lated to the soul as health is related to the body, and since life is not
worth living if one's health is ruined, it is all the more important to
maintain the justice of one's soul (444C-445C). Second, in Book IX,
Plato compares the five types of people he has been portraying in the
middle books - the philosophical ruler, the timocrat, the oligarch,
the democrat, and the tyrant - and declares that the happiest of them
is the philosopher, since he exercises kingly rule over himself (580a-
c). Third, Book IX immediately proceeds to argue that the philosophi-
cal life has more pleasure than any other, since the philosopher is in
the best position to compare the various pleasures available to differ-
ent types of people and prefers philosophical pleasures to all others
(58oc-583a). And fourth, the pleasures of the philosophical life are
shown to be more real and therefore greater than the pleasures of any
other sort of life (583b-s88a).
Does Plato single out any one of these arguments as more funda-
mental than the others? It might be thought that his fourth
argument-the second of the two that concern pleasure-is the
one he thought most important, for he introduces it with the re-
mark that "this will be the greatest and supreme fall [of injustice]"
(megiston te kai kuhotaton ton ptomaton, 583b6-7). This could be
taken to mean that pleasure is the most important good in terms of
which to make the decision between justice and injustice, and that
the argument to come is the one that most fully reveals why jus-
tice is to be chosen over its opposite. But I think that such a
reading would give this argument far more significance than it
deserves, and that Plato's words can and should be given a different
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 313
interpretation. As I read the Republic, its fundamental argument in
defense of justice is the one that comes to a close in Book IX before
anything is said about how the just and unjust lives compare in
terms of pleasure. This is the argument that Plato develops at great-
est length, and if it is correct it makes a decisive case in favor of
the just life. It shows precisely what it is about justice that makes
it so worthwhile. By contrast, the two arguments that connect
justice and pleasure are merely meant to assure us that we do not
have to sacrifice the latter good in order to get the former. They add
to the attractiveness of the just life, but they are not by themselves
sufficient to show that justice is to be chosen over injustice, as is
the lengthier argument that precedes them.
Why should we read the Republic in this way, despite Plato's
statement that "the greatest and supreme fall" of injustice comes
with his final argument? The answer lies in the way he poses, in
Book II, the fundamental question to which the rest of the dialogue
is an answer. The thesis he there undertakes to prove is phrased in
various ways: It is better (ameinon) to be just than unjust (357bi);
justice must be welcomed for itself if one is to be blessed [makarios,
358a3); the common opinion that injustice is more profitable [lusi-
telein) must be refuted (360C8); we must decide whether the just
man is happier [eudaimonesteros) than the unjust (36id3);4 justice
by itself benefits [oninanai) someone who possesses it whereas injus-
tice harms (blaptein) him (367d3~4); we must determine the advan-
tages [opheliai) of justice and injustice (368C6). Plato does not give
any one of these phrases a special role to play in his argument, but
moves back and forth freely among them. And he surely must be
assuming that once the consummately just life has been shown to be
more advantageous, even in the midst of misfortune, than the con-
summately unjust life, then he has given decisive reason for choos-
ing the former over the latter.
Notice, however, that Plato never promises, in Book II, to show
that justice provides greater pleasures than does injustice, and never
even hints that he would have to defend this thesis in order to show
that we should choose the just life. This suggests that the question
whether the just or the unjust life has more pleasure will still be an
open one, even after the greater advantages of the just life have been
demonstrated. And of course, this suggestion is confirmed in Book
IX: Having shown that the just person is happiest, Plato thinks it
Cambridge Companions Online © Cambridge University Press, 2006
314 THE CAMBRIDGE COMPANION TO PLATO
requires further argument to show that the just person also has the
greatest pleasure. So, in order to accomplish the task Plato assigns
himself in the Republic it is both necessary and sufficient that he
show why justice is so much more advantageous than injustice. But
he never says or implies that if he can show that justice brings
greater pleasures, then that by itself will be a sufficient or a neces-
sary defense of justice. By supporting justice in terms of pleasure,
Plato is showing that there is even more reason to lead the just life
than we may have supposed. But the fundamental case for justice
has been made before the discussion of pleasure has begun.5
What then should we make of his statement that the "greatest and
supreme fall" for injustice occurs in the battle over pleasure? A
simple and plausible explanation of this phrase is provided by the
fact that at the end of his last argument Plato claims that the philoso-
pher's pleasure is 729 times greater than the tyrant's (587c). Whether
Plato is serious about this precise figure or not - and I am inclined to
think he is not - it provides an explanation of why he says that this
last argument gives injustice its greatest defeat.6 In no other argu-
ment had he tried to portray the gap between justice and injustice as
so great in magnitude. Once we realize that Plato's remark admits of
this interpretation, we can rest content with our earlier conclusion
that pleasure has a modest role to play in the overall scheme of the
Republic.
11
I will therefore set aside the two hedonic arguments Plato gives in
Book IX and concentrate entirely on the single complex defense of
justice that precedes them. But it might be thought that this material
contains two separate arguments, for by the end of Book IV Plato
already seems to have come to the conclusion that since justice is a
harmony of the soul comparable to physical health, it is far superior to
injustice.? We might therefore suppose that after Book IV Plato
launches on a second and independent defense of justice, one that
concludes in Book IX with the pronouncement that the life of the
philosophical ruler is happiest. But Plato himself makes it clear that
these two segments - Books II-IV on the one hand, Books V-IX on
the other - cannot be isolated from each other in this way. For at the
beginning of Book VIII we are told that the victorious pronouncement
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 315
of Book IV - that the best person and city had been found - was pre-
mature (543C7~544b3). This means that the argument of Book IV is
not complete after all, but is in some way strengthened by additional
material presented somewhere between Books V and IX. For by admit-
ting that Book IV did not yet discover who the best person is, Plato
indicates that he had not at that point presented a full enough picture
of the just life.8 It would therefore be a mistake to examine the argu-
ment of Books II-IV in isolation from later material as though they
were meant to provide a complete defense of justice.
Nonetheless, Plato clearly thinks that he has given at least a partial
defense of justice by the end of Book IV; the fact that he goes on to
strengthen the argument by giving a fuller picture of the just life does
not mean that by the end of Book IV we have no reason at all to think
that justice is superior to injustice. To understand the single argu-
ment that runs from Book II through Book IX, we must see why Plato
arrives at a preliminary conclusion in Book IV and how the additional
material that comes in later books strengthens that argument.*
To make progress on this interpretive question, let us begin with
an observation with which all scholars would agree: One of the
fundamental ideas that Plato puts forward in his defense of justice is
that we should look for a general theory of goodness. His proposal is
that when we say of a human body, or a human soul, or a political
community, that they are in good condition, there is some common
feature that we are referring to, and it is because they share this
common feature that they are properly called good.10 He expects his
audience to agree with him that the goodness of a body - health -
consists in a certain natural priority among various physical compo-
nents; and he appeals to this point to support his claim that one's
soul is in good condition if it too exhibits a certain order among its
components (444c-e).11 But the analogy between health and psychic
well-being is by itself only of limited value, because it does not tell
us anything about what sort of order we should try to achieve in the
soul. What Plato needs, if he is to give a stronger argument from
analogy, is a structure that has the same kind of components and can
exhibit the same kind of balance as the soul. He thinks he can
accomplish this by examining the question of what the best possible
city is, for he believes he can show that the tripartite structure of the
best political community corresponds to the structure of the human
soul.12 If he can convince us that these correspondences do exist, and
Cambridge Companions Online © Cambridge University Press, 2006
3 l 6 THE CAMBRIDGE COMPANION TO PLATO
if he can get us to agree that the city he describes is ideal, then he has
some basis for reaching the conclusion that the ideal type of person
is someone whose soul exhibits the same kind of order that is pos-
sessed by an ideal political community. ̂
But in Book IV Plato has not yet given us all of his arguments for
taking the political community he is describing as ideal. For one of
his main reasons for favoring the kind of city described in the Repub-
lic is that it alone is governed by individuals who have the wisdom
needed to rule well; and that kind of political expertise is only pre-
sented in Books VI-VII. This is one reason for saying that the argu-
ment from analogy presented at the end of Book IV is incomplete.
Furthermore, Plato has not yet said in Books II-IV everything he
wants to say about the kind of order that should be established in the
soul. He tells us that reason should rule and look after the well-being
of the rest of the soul, that spirit should be its ally, and that the
appetites should be kept in check (44ie-442a). But what is it for
reason to rule the soul? In what way can spirit help it? What would it
be for appetite to grow too large? Of course, Plato has already given
some content to these notions, for he has been describing the proper
education of these elements of the soul since the end of Book II, and
this gives us some sense of how they should be related to each other.
But that education has not yet been fully described; the most impor-
tant objects of study have still to be presented. When we find out
more about what reason must occupy itself with, we will have a
fuller idea of what it is for it to rule.1*
in
We must now turn to Books V through VII to see how Plato's depic-
tion of the philosophical life contributes to the argument that jus-
tice pays. We want to know what it is about this life that makes it so
much more worthwhile than any other,- and we must understand
how this new material is connected to the argument from analogy
that comes to a preliminary conclusion at the end of Book IV.
An answer to these questions must in some way or other appeal to
Plato's belief in Forms-those eternal, changeless, imperceptible,
and bodiless objects the understanding of which is the goal of the
philosopher's education.1* For the philosopher is defined as someone
whose passion for learning grows into a love of such abstract objects
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 317
as Beauty, Goodness, Justice, and so on (474C-476C). And as soon as
Plato introduces this conception of who the philosopher is, he lets
us know that it is precisely because of the philosopher's connection
with these abstract objects that the philosophical life is superior to
any other. Those who fail to recognize the existence of Forms have a
dreamlike kind of life, because they fail to realize that the corporeal
objects they perceive are only likenesses of other objects (476c-d).16
In a dream, we confusedly take the images of objects to be those very
objects. Plato's claim is that nonphilosophers make a similar mis-
take, because they think that the beautiful things they see are what
beauty really is; more generally, they equate the many observable
objects that are called by some general term, "A," with what A
really is.1? The philosophers are those who recognize that A is a
completely different sort of object, and so they rid themselves of a
systematic error that in some way disfigures the unphilosophical
life. This is of course the picture Plato draws in the parable of the
cave (5i4a-5i9d): Most of us are imprisoned in a dark underworld
because we gaze only on the shadows manipulated by others,- to free
ourselves from this situation requires a change in our conception of
what sorts of objects there are.
Plato's metaphysics is of course controversial, but our present
problem is to understand how it contributes to the defense of justice.
Suppose we accept for the sake of argument that at least these cen-
tral tenets of his metaphysics are correct: There are such abstract
objects as the Form of Justice, and to call acts or individuals or
citizens just is to say that they bear a certain relationship to this
Form. Calling an act just is comparable to calling an image in a
painting a tree: The image is not what a tree is, and it is correct to
speak of it as a tree only if this means that it bears a certain relation
to living trees; similarly, just acts, persons, and cities are not what
justice is, and it is correct to call them just only if this means that
they participate in the Form of Justice.
If we accept this theory, we avoid the errors of non-PlatonistS; we
recognize that a wider variety of objects exists than most people
realize, and that our words constantly refer to these objects. Even so,
we should still ask: Why would having this Platonic conception of
the world make our lives so much better than the lives of non-
Platonists? One possible answer Plato might give is that since knowl-
edge of reality is a great intrinsic good, a life in which we know the
Cambridge Companions Online © Cambridge University Press, 2006
3 l 8 THE CAMBRIDGE COMPANION TO PLATO
truth about what exists is far superior to one in which we remain
ignorant of the fundamental realities of the universe. But this strikes
me as a disappointing answer, and I will soon argue that Plato has a
better one. It is disappointing because it makes an assumption that
would be challenged by anyone who has doubts about the merits of
the philosophical life. To those who are not already philosophically
inclined, it is not at all obvious that knowledge of reality is by itself
a great intrinsic good. They can legitimately ask why it is worth-
while for us to add to our understanding of reality, if our failure to do
so would not impede our pursuit of worthwhile goods. Plato cannot
simply reply that knowledge is intrinsically worthwhile, apart from
any contribution it may make to the pursuit of other goals. That
would beg the question in favor of the philosophical life.
It might be thought that for Plato knowledge of the Forms is
valuable precisely because it is a means to some further goal. For
example, he might claim that unless we study the Form of Justice,
we are likely at some point to make errors in our judgment of which
acts, persons, or institutions are just; and when we make errors of
this sort, we will also make bad decisions about how to act. But if
this is Plato's argument, then he again begs the question. For we can
ask why it is so important to discover how to act justly in all situa-
tions. Of course, if acting justly is good for the agent, and knowledge
of the Forms is an indispensable means to this end, then one must
acquire that knowledge. But this argument merely assumes the the-
sis that Plato sets out to prove: that acting justly is a good for the
agent.
Perhaps he assumes that knowing the Forms is worthwhile not
merely as a means to action but because in coming to understand
the Forms we develop our capacity to reason.18 Human beings are
not just appetitive and emotional creatures,- we also have an innate
interest in learning, and if this aspect of our nature is not developed
our lives become narrow and impoverished. One problem with this
answer is that people differ widely in the degree of intellectual curi-
osity they possess, and the kinds of objects that satisfy their curios-
ity also differ widely. Those who have little or no bent for abstract
studies can satisfy their curiosity in simple ways, and again Plato
would be begging the question if he simply assumed that having an
easily satisfied appetite in matters of reasoning disqualifies one from
leading a good life. Furthermore, as Plato is aware, it is possible to
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 319
spend a great deal of one's time on intellectual matters without ever
arriving at the realization that the Forms exist. Those who study the
universe and seek to explain all phenomena without appealing to
Forms surely develop the reasoning side of their nature,- it is not
sheer emotion and appetite that leads them to their theories. Even
so, they are not leading the philosophical life, according to Plato's
narrow conception of philosophy, and so they don't have the best
kind of life. If he thinks that intellectuals who deny the existence of
Forms fail to develop their capacities and therefore fall short of happi-
ness, he owes his reader some argument for this thesis.
IV
I believe that Plato's answer to this question is staring us in the face,
but that we fail to recognize it because initially it strikes us as
doubtful or even unintelligible. My suggestion is that for Plato the
Forms are a good-in fact they are the greatest good there is.1? In
order to live well we must break away from the confining assump-
tion that the ordinary objects of pursuit-the pleasures, powers,
honors, and material goods that we ordinarily compete for - are the
only sorts of goods there are.20 We must transform our lives by recog-
nizing a radically different kind of good - the Forms - and we must
try to incorporate these objects into our lives by understanding,
loving, and imitating them, for they are incomparably superior to
any other kind of good we can have. This is why Plato thinks that
the philosopher is so much better off for having escaped the confines
of the dreamlike existence of the ordinary person: The objects with
which the philosopher is acquainted are far more worthy objects of
love than the typical objects of human passion. So Plato is not claim-
ing that it is intrinsically good to have a complete inventory of what
exists or that developing and satisfying our intellectual curiosity is
inherently worthwhile, regardless of the sorts of objects to which
our curiosity leads us. Rather, he takes the discovery of the Forms to
be momentous because they are the preeminent good we must pos-
sess in order to be happy, and he takes reason to be the most worth-
while capacity of our soul because it is only through reason that we
can possess the Forms. If there were nothing worthwhile outside of
ourselves for reason to discover, then a life devoted to reasoning
would lose its claim to superiority over other kinds of life.21
Cambridge Companions Online © Cambridge University Press, 2006
32O THE CAMBRIDGE COMPANION TO PLATO
The interpretation I am proposing has some resemblance to the
way Aristotle treats Plato's moral philosophy. According to Aris-
totle, we can discover what kind of life we should lead only by
determining which good or goods we should ultimately pursue. He
considers competing conceptions of this highest good and takes the
Platonist's answer to be that it is not some humdrum object of
pursuit like pleasure or virtue but is rather the Form of the Good.
Aristotle of course rejects this answer, but it is significant that he
takes the Platonist to be saying that a certain Form is the highest
good and should therefore play the role non-Platonists assign to
pleasure, honor, or virtue. So interpreted, the Platonist is not simply
saying that the Form of the Good is an indispensable means for
determining which among other objects are good; it itself is the chief
good.22 My interpretation is similar in that I take Plato to treat the
Forms in general as a preeminent good; the special role of ,the Form
of the Good will be discussed later.
At this point it might be asked whether the theory I am attribut-
ing to Plato is intelligible. For perhaps a Form is simply not the sort
of thing that a person can have or possess. Of course, a Form can be
studied and known, but studying something does not by itself confer
ownership. The moon, for example, might be a beautiful object wor-
thy of our study, but no one in his right mind would say that the
moon is a good he possesses by virtue of studying it. Similarly, the
claim that the Form of the Good is not the sort of thing that can be
possessed is one of Aristotle's many objections to the Platonist con-
ception of the good [N.E. iO96b35). Aristotle takes Plato to be saying
that the ultimate end is the Form of the Good, and objects that it is
disqualified from playing this role because it is not an object of the
right type. It might be thought that this objection is so powerful that
out of charity we should look for a different interpretation from the
one I am proposing.2^
But I think Aristotle's objection is weak. Of course it is true that if
we take the possession of a thing to be a matter of having property
rights to it, then studying the Form of the Good does not confer such
rights, and it is hard to understand what it would be to possess a Form.
But we can speak of having things even though we have no property
rights in them; for example, one can have friends without possessing
them. And we can easily understand someone who says that in order
to live a good life one must have friends. What it is to have a friend is
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 321
quite a different matter from what it is to possess a physical object; it
involves an emotional bond and activities characteristic of friend-
ship. What it is to have a certain good varies according to the kind of
good it is; different types of goods do not enter our lives in the same
way. And so the mere fact that a Form cannot be possessed (that is,
owned) gives us no reason to reject Plato's idea that if one bears a
certain relationship to Forms - a relationship that involves both emo-
tional attachment and intellectual understanding - then one's life
becomes more worthwhile precisely because one is connected in this
way with such valuable objects.
In fact, there are similarities between the way in which persons
can enter our lives and improve them and the way in which Plato
thinks we should be related to the Forms. We can easily understand
someone who says that one of the great privileges of his life is to
have known a certain eminent and inspiring person. Even if one is
not a close friend of such a person, one may have great love and
admiration for him, and one may take pleasure in studying his life.
That is the sort of relationship Plato thinks we should have with the
Forms - not on the grounds that loving and studying are good activi-
ties, whatever their objects, but on the grounds that the Forms are
the preeminent good and therefore our lives are vastly improved
when we come to know, love, and imitate them.
Suppose it is conceded that if the Forms are a good, then they are
the sorts of things that can improve our lives when we are properly
related to them. Nonetheless, it might still be asked whether we can
make sense of the idea that they are good. If someone says that water
is a good thing, we might be puzzled about what he has in mind, and
we might even be skeptical about whether water is the sort of thing
that can be good in itself (as opposed to a mere means).2* Similarly,
we might have doubts about Plato's Forms: How can such objects,
which are so different in kind from such mundane goods as health
and pleasure, be counted as good? And if he cannot convince us that
they are good, then of course he has no hope of persuading us that
they are vastly better than such ordinary goods as pleasure, health,
wealth, power, and so on.
For Plato's answer to our question, What is it to say of something
that it is a good thing? we might turn for help to his discussion of the
Form of the Good. But although he insists on the preeminence of
this Form, he does not say precisely what he takes goodness to be,- he
Cambridge Companions Online © Cambridge University Press, 2006
322 THE CAMBRIDGE COMPANION TO PLATO
simply says that it is not pleasure or knowledge (5osb-5o6e). There
is a marked contrast here between the fullness of his account of
what justice is and the thinness of his discussion of goodness. We
learn what it is to call a person, act, or city just, and we see the
feature that they all have in common, but Plato points to no com-
mon feature of all good things. So he does not take up the project of
showing that Forms are preeminent by stating what property good-
ness consists in and arguing that they exhibit that property more
fully than anything else.
Perhaps we can discover why Plato thinks of Forms as goods if we
focus on their distinguishing characteristics and ask which of them
Plato might put forward as points of superiority over other objects.
For example, he thinks that Forms are more real than corporeal
objects, and presumably he counts this as evidence of their superior-
ity in value.2* But this point will not take us as far as we need to go,
because he thinks that objects that are equally real can nonetheless
differ greatly in value. Consider two bodies, one of them healthy, the
other diseased: One is in better condition than the other, but Plato
never suggests that one of them must therefore be more real than the
other. Though Forms are more real than other types of objects, we
cannot treat differing degrees of reality as what in general constitute
differences in value.
But our example of the healthy and diseased bodies suggests an-
other line of reasoning: Plato equates health, the good condition of
the body, with a certain harmony among its elements; and he argues
that justice, the good condition of the soul, is also a certain kind of
harmony among its parts; and so the thought suggests itself that he
takes the goodness of anything of a certain kind to be the harmony
or proportion that is appropriate for things of that kind. According to
this suggestion, the goodness of Forms consists in the fact that they
possess a kind of harmony, balance, or proportion; and their superior-
ity to all other things consists in the fact that the kind of order they
possess gives them a higher degree of harmony than any other type
of object.26
Clearly Plato does think that the Forms exhibit the highest kind of
orderly arrangement. He says that the philosopher looks away from
the conflict-ridden affairs of human beings to things that are un-
changing and ordered [tetagmena, 500C2); by studying the divine
order [kosmos, C4) her soul becomes as orderly and divine as it is
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 323
possible for a human soul to be (c9-di). Even the beautiful patterns
exhibited in the night sky fall short of the harmonies present in true
shapes and numbers, since the corporeality of the stars makes devia-
tion inevitable, whereas the incorporeality of the Forms ensures that
the orderly patterns they exhibit will never deteriorate (529C7-
53ob4). But he does not say precisely what the orderliness of the
Forms consists in; bodies, souls, and political communities exhibit
order (and therefore goodness) when their parts or components are
related to each other in suitable ways, but we are not told whether
the Forms have parts or whether they achieve their order in some
other way. Perhaps this explains Plato's refusal to say what the Form
of the Good is (5o6d-e); though goodness simply is some kind of
harmony, he had not yet reached a firm grasp of what this harmony
is in the case of Forms, and so he could not put forward a general
characterization of harmony that would apply equally to the various
kinds of harmony exhibited by living bodies, souls, stars, and Forms.
But in any case, we can now see how Plato would try to address
doubts about whether Forms are the sorts of objects that can intelligi-
bly be called good. He would reply by appealing to his discussion of
politics, the soul, and health: In all of these cases, the goodness of a
thing consists in a kind of order; and so if the Forms can be shown to
have the kind of order that is appropriate for things of that kind, they
too will be good. And if they necessarily have a higher degree of
order than anything else, then they are the best goods there can be.2?
It may now be asked how any of this provides Plato with a defense of
the virtue of justice. Even if we see why he thinks that the philo-
sophical life is best, we still can ask why this should be regarded as a
defense of justice. Why is the philosopher the paradigm of the just
person? Part of Plato's reply, as I understand it, is as follows:28 When
the ideal state properly educates individuals to become philoso-
phers, their emotions and appetites are transformed in a way that
serves the philosophical life, and these affective states no longer
provide a strong impetus toward antisocial behavior, as they do
when they are left undisciplined (499e-5ooe). Someone who has
been fully prepared to love the orderly pattern of the Forms will be
free of the urge to seek worldly advantages over other human beings
Cambridge Companions Online © Cambridge University Press, 2006
324 THE CAMBRIDGE COMPANION TO PLATO
or to engage in the sort of illicit sexual activity to which people are
led by unchecked appetites. Furthermore, such a person is in the
best possible position to make wise political decisions; having under-
stood the Forms, she can see more clearly than others what needs to
be done in particular circumstances (5ood-5oia). One of the things
we look for, when we seek a paradigm of the just person, is someone
who has these intellectual and affective skills.2*
It is tempting to protest at this point that Plato is being extremely
naive. After all, we all know people who have impressive intellec-
tual abilities but who are hardly models of justice. And of course
there is nothing to prevent such individuals from recognizing the
existence of abstract objects, and even loving the contemplation of
the orderly pattern among such objects. Consider a Platonist mathe-
matician who occasionally gets drunk and indulges in other behav-
ior that conflicts with Plato's description of the just individual.
Aren't such individuals living refutations of Plato?
I believe not, for I don't take him to be making the implausibly
strong claim that the love of abstract objects by itself guarantees just
behavior or the emotional discipline that characterizes the just per-
son. Rather, his weaker and more plausible claim is that one will be
in the best position to lead a life dominated by the love of Forms if
one trains the nonrational components of one's soul to serve one's
love of philosophy. It is this weaker claim that lies behind his por-
trayal of the philosopher as the paradigm of human justice. By put-
ting oneself into the best position to lead the philosophical life, one
develops the intellectual and emotional skills that we look for in a
completely just person. The mere existence of unjust lovers of ab-
stract objects does not by itself refute Plato, for the issue is not
whether they exist but whether the psychological condition that
underlies their injustice makes them less able to profit from their
recognition of abstract objects. It might be argued, against Plato, that
sensuality, greed, and large appetites for food and drink make one all
the more able to understand and love the orderly realm of the Forms,
but it is far from obvious that this is so. He is not being unreasonable
in assuming that these emotional states are on the contrary obsta-
cles to the philosophical life.
We should recall, however, that Plato promises to do more than
merely show that justice is a great good. He has to show that it is a
greater good than injustice, so much so that even if the normal
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 325
consequences of justice and injustice are reversed it will nonetheless
be better to be just than unjust. The paradigm of justice must be
punished because he is thought to be unjust; and the paradigm of
injustice is to receive the honors and rewards because he appears to
be just. How can Plato show that even in this situation it is better to
be just?
The answer lies partly in the way he describes the situation of the
completely unjust person, that is, the tyrant. Such a person is al-
lowed to live out his fantasies of power and eroticism without re-
straint, and Plato's case against such a life is that this lack of re-
straint will inevitably exact a devastating psychological toll. When
erotic desires are allowed to grow to full strength, they become
impossible to satisfy; rather than leading to a life of peace and fulfill-
ment, they leave one with a chronic feeling of frustration (579d-e).
Similarly, tyrannical power inevitably gives rise to continual fear of
reprisals and an absence of trust in one's associates (576a, 579a-c).
The failure to impose any order on one's appetites makes one the
victim of frequent and disorganized internal demands (573d). So, in
order to achieve great power and intense sexual pleasure, the tyrant
must lead a chaotic life filled with anguish, fear, and frustration. No
one who reads this account of the tyrannical life could seriously
hold it up as a model of how human beings should live. When the
immoralist praises the life ruled by unrestrained desires for power
and pleasure, he simply fails to think through the consequences of
giving these desires free rein. He responds to something in human
nature, for Plato agrees that no one is completely free of the im-
pulses that the immoralist champions (57ib-572b). The presence of
these illicit urges seems to lend some credibility to the immoralist's
doubts about whether justice is a virtue, for the praise of immorality
answers to something within us. Plato's response to the immoralist
is that when we seriously consider the psychological consequences
of magnifying the power of our illicit urges, the life of maximal
injustice loses its appeal. This is something he thinks we will be
able to see without having the benefit of the theory of Forms; he
invokes the Forms because they are the objects around which the
best kind of human life must be built, but he makes no appeal to
these objects when he tries to convince us that the tyrannical life is
miserable.
Again, it is possible to protest that Plato's argument is naive. It
Cambridge Companions Online © Cambridge University Press, 2006
326 THE CAMBRIDGE COMPANION TO PLATO
seems to rest on the empirical assumption that anyone who pos-
sesses tyrannical power will also have sexual obsessions, and this
makes it easier for him to make such a life look unattractive. But in
fact such an empirical assumption is unwarranted: It is certainly
possible to tyrannize a community and hold all other passions in
check. 3° Here too, however, I think Plato is less vulnerable to criti-
cism than we might have thought. His portrait of the tyrant is not
meant to be an exceptionless empirical generalization about what
such individuals are like. Rather, he is developing the portrait of the
unjust life that is presented in Book II when Glaucon and Adeiman-
tus try to make such a life look attractive. According to their por-
trait, the unjust man can seduce any woman who appeals to him; he
can kill anyone he wants (36oa-c). Plato's idea is that if these fea-
tures of injustice capture its subrational appeal, then it is fair to
describe the paradigm of injustice as someone whose sexual appe-
tites and murderous tendencies are extreme. If that is how he is
proceeding, then it is irrelevant that in fact tyrants need not be
dominated by sexual appetite.
Plato's portrait of the tyrant makes it clear that his argument for
justice does not rest solely on the metaphysics of the middle books
and the political theory of the early books but also relies on various
assumptions about human psychology. Certain desires, if unchecked,
lead to the sorts of consequences - frustration, fear, pain - that every-
one tries to avoid and that no one regards as compatible with a fully
happy human life. What Plato is assuming is that the life of the com-
pletely just person is not marred by these same features. Fear, frustra-
tion, and chaos are not the price philosophers must inevitably pay for
having a love of the Forms and for giving this passion a dominant role
in their lives. On the contrary, those who are in the best position for
studying the Forms will have modest and therefore easily satisfied
appetites, and will be free of the competitive desire for power that
typically sets people at odds and destroys their tranquillity. So the
philosophical life will include the felt harmony of soul that everyone
can recognize and value, as well as the more complex kind of har-
mony that one can understand only through a philosophical investiga-
tion of the parts of the soul and of the metaphysical objects that enter
one's life when reason rules.
We can now see why Plato is confident that he can prove that
justice pays even when he allows the just person and the unjust
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 327
person to reverse their roles in Book II. Even if the just person is
mistakenly dishonored and punished, she will still be at peace with
herself; she will be free of the chaos and frustration that make the
life of the tyrant so repellent. In place of the great physical pain
imagined for the just person, the tyrant must endure great psycho-
logical pain. Neither is in an enviable condition, but there is a major
difference that Plato thinks counts decisively in favor of the just
person: her understanding and emotions gain her entrance into a
world of completely harmonious objects, and so she possesses the
greatest good there is. We have finally answered the question with
which we began: The consummately unjust person has troubles that
counterbalance the pain and dishonor imagined for the just person,
and if these were the only factors involved in their comparison, it
might be difficult to decide whose situation is worse; but once the
possession of the Forms is added to the just person's side of the
equation, the advantage lies with her, overwhelmingly so because of
the great worth of that nonsensible realms
VI
One important feature of Plato's theory has not yet been discussed,
and it is best brought to light by considering a well-known internal
difficulty in his argument. He says that the philosophers of the ideal
city must not be allowed to study the Forms without interruption,
but must instead return to the darkness of the cave and help adminis-
ter the political community (5i9d-52ib, 54oa-b). Why won't the
philosophers be tempted to resist this requirement, however just it
may be, since it seems to conflict with their self-interest I*2 After all,
life in the open air illuminated by the Form of the Good must be
better than life in the subterranean atmosphere in which one musi
rule the state. Won't the philosophers be strongly tempted to think
of ways in which they can escape such service? If so, they cannot be
held up as paragons of justice. Furthermore, this example seems to
show that justice does not always pay: If one could unjustly escape
service to the community and continue contemplating the Forms,
one would do what is best for oneself, but one would not act justly.
Plato is completely confident that the individuals he has trained
for the philosophical life will accept this requirement. After all, he
says, they are just, and the requirement is just (52oei). But why
Cambridge Companions Online © Cambridge University Press, 2006
328 THE CAMBRIDGE COMPANION TO PLATO
doesn't he see any problem for his theory here? Why doesn't it leap
to his eye that ruling is contrary to the philosopher's interests, so
that this feature of his ideal state presents a clear counterexample to
his thesis that justice pays? One possible answer to this question is
simply that Plato is willing to make exceptions to this generaliza-
tion. 33 But it is unlikely that he would restrict himself to the weak
claim that justice is usually in one's interests. It is more fruitful, I
think, to look at the problem in the reverse manner: Plato thinks
that ruling the state is a just requirement, and since he believes that
justice is always in one's interest, he must think that somehow it
does pay to rule the city. The question is how he could believe this.
He tells us at one point that when philosophers look to the harmo-
nious arrangement of the Forms, they develop a desire to imitate
that harmony in some way or other (500c). And then he adds that if
it becomes necessary for the philosophers to imitate the Forms by
molding human character in their likeness, they will be in an excel-
lent position to do this job well. So it is clear that when the philoso-
phers rule, they do not stop looking to or imitating the Forms.
Rather, their imitative activity is no longer merely contemplative;
instead, they start acting in a way that produces a harmony in the
city that is a likeness of the harmony of the Forms. Furthermore,
were they to refuse to rule, they would be allowing the disorder in
the city to increase. Were any single philosopher to shirk her respon-
sibilities, and let others do more than their fair share, then she
would be undermining a fair system of dividing responsibilities. The
order that would be appropriate to their situation would be under-
mined. And so failure to rule, whether in an individual philosopher
or in a group of them, would create a certain disharmony in the
world: Relationships that are appropriate among people would be
violated. And in creating this disharmony, the philosopher would in
one respect cease to imitate the Forms. She would gaze at the order
that is appropriate among Forms but would thereby upset an order
that is appropriate among human beings.
What this suggests is that Plato has the resources for showing that
justice is in one's interests even when it requires forgoing some
purely philosophical activity. What he must hold is that one's high-
est good is not always served by purely contemplating the Forms; 34
rather, one's highest good is to establish and maintain a certain
imitative relationship with the Forms, a relationship that is strained
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 329
or ruptured when one fails to do one's fair share in a just community.
The person who is willing to do her part in a just social order, and
whose willingness arises out of a full understanding of what justice
is, will see the community of which she is a part as an ordered
whole, a worldly counterpart to the otherworldly realm of abstract
objects she loves. When she acts justly and does her fair share, she
sees herself as participating in a social pattern that approximates the
harmony of the Forms, and she therefore takes her good to be served
by acting justly. In making this connection between social harmony
and the harmony of abstract objects, Plato offers an account of the
positive appeal that justice in human relationships should have for
us. We are - or should be - attracted to justice in human relation-
ships; when we act justly, we should do so not merely because of the
absence of such motives as greed, sensuality, and the desire to domi-
nate others. Rather, we should see something attractive about com-
munities and relationships in which each person does his or her
appropriate part, and we should be loathe to violate these relation-
ships because of our love of justice. If I have understood Plato cor-
rectly, he recognizes that justice as a relationship among human
beings can have this positive appeal.35
VII
I said at the beginning of this chapter that there is something power-
ful in Plato's argument that justice pays. What I have in mind is his
thesis that the goodness of human life depends heavily on our hav-
ing a close connection with something eminently worthwhile that
lies outside of ourselves. To live well one must be in the right psy-
chological condition, and that condition consists in a receptivity to
the valuable objects that exist independently of oneself. If one is
oblivious to these objects and devotes oneself above all to the acqui-
sition of power, or the accumulation of wealth, or the satisfaction of
erotic appetites, then one will not only become a danger to others
but one will fail to achieve one's own good. Psychological forces that
lead to injustice when they become powerful are forces that should
in any case be moderated for one's own good, for when they are too
strong they interfere with our ability to possess the most valuable
objects.
Even if we reject Plato's belief in Forms or his thesis that goodness
Cambridge Companions Online © Cambridge University Press, 2006
33O THE CAMBRIDGE COMPANION TO PLATO
consists in harmony, we should recognize that there are many differ-
ent ways of trying to sustain his attempt to connect the goodness of
human life with some goodness external to one's soul. Christianity
provides an obvious example, for it holds that the external good is
God and that no human life is worth leading unless God is somehow
present in it. Another example can be found in Romantic concep-
tions of nature, according to which a person who is cut off from the
beauty of the natural order has been excluded from his home and
must lead an alienated existence. We can even see some similarity
between Plato's theory and the idea that great works of art so enrich
human lives that the inability to respond to their beauty is a serious
impoverishment.
In this last case, the valuable objects are created by human beings,
but nonetheless it could be held that one's good consists in learning
how to understand and love these objects. Someone can reasonably
say that her life has been made better because she has come to love
one of the cultural products of her society - a great novel, for exam-
ple. This does not have to mean that the novel has taught her lessons
that have instrumental value or that it has brought forth psychologi-
cal capacities that would otherwise have lain dormant. It is intelligi-
ble to say that a relationship to a certain object - something beauti-
ful in nature, or some work of art, or a divinity - by itself makes
one's life better. And that seems to represent the way many people
view their lives, for it is difficult to sustain the belief that one's life
is worthwhile if one sees and feels no connection between oneself
and some greater object.
Plato would of course reject these alternatives to his theory: He
claims that the natural world for all its beauty is no model of perfec-
tion and that the works of poets are of lesser value still. Perhaps then
we should distinguish a weak from a strong form of Platonism:
Weak Platonism holds that the human good consists in having the
proper relationship to some valuable object external to oneself,
whether that object be a work of art, one's family or political commu-
nity, the natural world, or a divinity. Strong Platonism goes further
and holds that the valuable object in question must be some eternal
and unchanging realm. What is distinctive of Plato's own view, of
course, is that the objects in question are the Forms. But even if his
particular version of Platonism is rejected, it should be recognized
that some form of this doctrine, strong or weak, is deeply appealing
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 331
to many. Plato might be pleased and not at all surprised that
watered-down forms of Platonism have had such a long history.
NOTES
1 See Rep. 36oe-362c for the contrast between the just and unjust lives.
(All future page references will be to this dialogue, unless otherwise
noted.) It should be emphasized that Plato is not trying to show that it is
advantageous to act justly regardless of one's psychological condition.
His claim is that it is advantageous to be a just person.
2 I have learned most from these studies: Julia Annas, An Introduction to
Plato's Republic (Oxford, 1981); Terence Irwin, Plato's Moral Theory
(Oxford, 1977); C. D. C. Reeve, Philosopher-Kings: The Argument of
Plato's Republic (Princeton, 1988); Nicholas P. White, A Companion to
Plato's Republic (Indianapolis, 1979). Among older treatments still
worth consulting are R. C. Cross and A. D. Woozley, Plato's Republic: A
Philosophical Commentary (London, 1964); Horace W. B. Joseph, Essays
in Ancient and Modern Philosophy (Freeport, N.Y., 1971); N. R. Murphy,
The Interpretation of Plato's Republic (Oxford, 1951); Richard Nettle-
ship, Lectures on the Republic of Plato, id ed. (London, 1962).
3 Here I am setting aside the arguments of Book I of the dialogue and
concentrating entirely on the issue as it is reintroduced at the beginning
of Book II. Plato must have believed that the arguments of Book I were in
some way deficient; otherwise there would be no need to reopen the
question in Book II. Perhaps their deficiency lies principally in their sche-
matic nature: They need to be buttressed by political theory, metaphys-
ics, and psychology. An alternative reading is that in Book II Plato thinks
that the earlier arguments are entirely of the wrong sort. For this interpre-
tation, see Irwin, Plato's Moral Theory, 177-84; Reeve, Philosopher-
Kings, 3-24.1 also set aside the further considerations Plato mentions in
Book X at 612b//: These are the worldly and otherworldly rewards the just
can expect to receive. It is precisely these rewards that Plato agrees to
overlook when he promises in Book II to show that justice is in our
interest, apart from its consequences. It should be emphasized that Plato
thinks these rewards make the just life even more desirable. He agrees
that the just person who suffers the torments described at 36ie~362a
suffers a loss of well-being and is no paradigm of happiness. When he
refers to wealth and other "so-called goods'7 at 495a7, his refusal to call
them goods outright should be taken to mean that these ordinary objects
of pursuit are not in all circumstances good; he cannot hold the stronger
thesis that they are never good, for then the social rewards of justice
would be a matter of indifference. There has been considerable discussion
Cambridge Companions Online © Cambridge University Press, 2006
33^ THE CAMBRIDGE COMPANION TO PLATO
of what Plato means by saying that justice is good in itself. See Annas,
Introduction, chap. 3; Cross and Woozley, Plato's Republic, 66-g-, M. B.
Foster, "A Mistake of Plato's in the Republic," Mind 46 (1937): 386-93;
Irwin, Plato's Moral Theory, 184-91, 325-6; C. A. Kirwan, "Glaucon's
Challenge/' Phronesis 10(1965): 162-73; J.D.Mabbott, "Is Plato's Repub-
lic Utilitarian?" Mind 46 (1937): 468-74; David Sachs, "A Fallacy in
Plato's Republic," Philosophical Review 72 (1963): 141-58; Reeve,
Philosopher-Kings, 24-33; Nicholas P. White, "The Classification of
Goods in Plato's Republic," Journal of the History of Philosophy 22
(1984)- 393-42.1.
4 Readers of the Republic should bear in mind that Plato does not use
eudaimonia (often translated "happiness") and its cognates to refer to
the feeling of pleasure. For Plato, to seek one's own happiness is simply
to seek one's own advantage, and so to discover what happiness is one
must determine where a human being's true interests lie.
5 At 589C1-4 Plato distinguishes between praising justice for its advan-
tages and praising it for its pleasures (cf. 58ie7-582a2, 588a7-io). This
implies that the two arguments from pleasure in Book IX are not ad-
dressed to the issue of whether justice or injustice is more advantageous.
For an alternative reading, see J. C. B. Gosling and C. C. W. Taylor, The
Greeks on Pleasure (Oxford, 1982), 98-101; their interpretation is en-
dorsed by Reeve, Philosopher-Kings, 307 n. 33. For further discussion of
this alternative, see my review of Reeve's Philosopher-Kings in Political
Theory 18 (1990): 492—6.
6 For discussion of Plato's calculation, see Reeve, Philosopher-Kings, 150-
1. He argues that the correct figure should be 125.
7 Or, following Annas, Introduction, 168-9, w e might think that Plato is
arguing for two different conclusions: The earlier material is designed to
show that justice is good in itself, apart from happiness; whereas the
later material does try to link justice and happiness. But we should reject
her statement that "the notion of happiness has not occurred" in Book
IV. When Plato asks at 44467-445 a4 whether justice is more profitable
(lusitelei) than injustice, he is in effect asking whether the just person is
happier. As Book II shows, the thesis Plato is trying to prove can be
formulated in several terms that are treated equivalently. Annas's inter-
pretation was proposed earlier by Mabbott, "Is Plato's Republic Utilitar-
ian?," 62.
8 Other passages show that Plato does not take himself to have fully
revealed in Book IV what justice is: See 472b7 and 484a7-8.
9 For the contrary view - that Plato does not attempt to give any argu-
ment in Books II-IV for the thesis that justice is advantageous - see
Nicholos P. White, "The Ruler's Choice," Archiv fur Geschichte der
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 333
Philosophic 68 (1986): 34-41. But I think 44467-445by rules this out:
The interlocutors here agree that justice is advantageous and that injus-
tice is not; and surely they think that they have some reason for this
conclusion. I take 445bs-7 to mean that the conclusion has not been
supported as fully as possible, and that the fuller argument is now to
come. Despite this difference, White and I agree that Books II-IX should
be read as a single continuous argument in defense of justice.
10 This is of course a consequence of Plato's general principle that when-
ever we call a group of things by the same name there is something they
all have in common. See, for example, Meno J2b-c and Rep. 596a.
Plato's assumption that goodness is a single thing is attacked by Aris-
totle in the Nicomachean Ethics 1.6.
11 More fully, the argument is this: (1) Health is the preeminent good of the
body, in the sense that life is not worth living when one's body is com-
pletely lacking health. (2) What makes health so worthwhile is that it
involves a natural balance of elements - certain elements appropriately
dominate certain others. (3) Justice involves an analogous balance in the
soul. (4) Since justice has the same good-making characteristic as health,
it must be equally true that life is not worth living if one is greatly
deficient in justice. The crucial premise is (3), and to support it Plato
appeals to the analogy between city and soul. But even if Plato had
completely left aside the idea that health involves a balance, the main
argument from analogy of Books II-IV would still remain: What is best
for the polis is an internal balance, and so we should expect the same to
hold true of the individual. The appeal to healtn is an attempt to
strengthen the argument by adding one more case in which advantage
can be equated with proper balance.
12 For discussion of Plato's argument for the tripartition of the soul, see
John M. Cooper, "Plato's Theory of Human Motivation," History of
Philosophy Quarterly (1984): 3-21; Irwin, Plato's Moral Theory, 191-5;
Terry Penner, "Thought and Desire in Plato," in Plato, vol. 2, ed. Greg-
ory Vlastos (Garden City, N.Y., 1971), 96-118; Reeve, Philosopher-
Kings, 118-40.
13 Plato's strategy would fail if it were impossible to say anything about
what a good city is without first knowing what a good person is or what
human happiness is. Books II-IV try to convince us that we can discover
a good deal about how a political community should be organized, even
before we address the question of human virtue and happiness. For the
view that the argument of II—IV begs the question against Thrasyma-
chus by simply assuming at 427e-428a and 433a-435a that justice is a
virtue, see Michael C. Stokes, "Adeimantus in the Republic," in Law,
Justice and Method in Plato and Aristotle, ed. Spiro Panagiotou (Edmon-
Cambridge Companions Online © Cambridge University Press, 2006
334 T H E CAMBRIDGE COMPANION TO PLATO
ton, 1985). Stokes thinks that Plato is not really addressing his argument
to a radical critic of justice like Thrasymachus; rather, he is speaking to
Glaucon and Adeimantus, who are already half-convinced when the
argument begins. A similar view is defended by Reeve, Philosopher-
Kings, 33-42; contrast Martha C. Nussbaum, The Fragility of Goodness
(Cambridge, 1986), 155-6. I believe that Plato is trying to persuade
Thrasymachus (see 498d) and that he does not take his argument to beg
the question against him, but the issue requires more discussion than I
can give it here.
14 The limitations of Plato's argument as it develops from Book II to Book IV
are emphasized by John M. Cooper, "The Psychology of Justice in Plato/'
American Philosophical Quarterly 14(1977): 152-3; Irwin, Plato's Moral
Theory, 216-17; and White, "Ruler's Choice/' 39.
15 Among the most important passages characterizing the Forms are Phd.
65d-66a, 74D-C, 78c-8ob; Phdr. 247c; Rep. 477a-48oe; Smp. 2ioe-
21 ie ; Ti. 27d-28a, 38a, 52a-b; Phil. 59c. For a thorough examination of
Plato's reasons for postulating the existence of Forms, see Terry Penner,
The Ascent from Nominalism (Dordrecht, 1987).
16 For a lucid interpretation of this aspect of Plato's theory, see Richard
Patterson, Image and Reality in Plato's Metaphysics (Indianapolis,
1985).
17 See Penner, Ascent from Nominalism, 57-140.
18 See Irwin, Plato's Moral Theory, 236, for the claim that Plato's defense of
justice depends on the idea that we must develop all of our capacities.
19 The principal textual support for this reading derives from the many
passages in which Plato describes the Forms as the proper objects of
love: 476b, 480a, 484b, 49oa-b, 500c, 5Oid. They could not be such
unless they are good {Smp. 2O4d-2o6a). I am not claiming that according
to Plato each Form counts as a separate good; rather, it is the ordered
whole constituted by the Forms that is a good, although some of the
individual Forms (Goodness, Beauty, etc.) may by themselves be goods.
Of course, if my interpretation is to be an improvement over the ones
just considered, then Plato cannot simply assume that the Forms are a
great good. His argument for this claim will be discussed later. It might
be asked how the Forms can be the greatest good, since that distinction
is reserved for justice (36669). But there is no real conflict here. When
Plato says that justice is the greatest good, he does not mean that the
universe has no better object to show than a just human being; the
Forms are superior to this. He means rather that possessing justice is
better for us than possessing any other type of good; and this is compati-
ble with the claim that the Forms are the supreme objects. For on my
reading, being fully just and fully possessing the Forms are the same
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 335
psychological state, and so there is no issue about which state it is better
to be in.
20 It is widely recognized that according to Plato happiness consists in
possessing good things - a point he takes to need no argument. See Smp.
2O4e-2O5a. What is distinctive of my interpretation is the suggestion
that Plato defends the philosophical life (and therefore the life of con-
summate justice) by adding to the conventional list of goods.
21 The pattern of argument in the Philebus is similar: Reason is declared to
be a more important component of the good human life than pleasure
because it is more akin than pleasure to the good. Here, as in the Repub-
lic, something outside of human life is taken to be ideal, and those
elements of human life that most fully approach this ideal are to receive
priority.
22 This is why his discussion in N. E. 1.6 of the Platonic conception of the
Good is not out of place. Aristotle also considers the possibility that for
the Platonist the Good is not itself a desirable object but is instead a tool
for gaining the knowledge we need to make practical decisions. See
iO96b35-iO97a6. But this is an alternative to the main conception of
the Good that he considers in 1.6.
23 Thus G. X. Santas, "Aristotle's Criticism of Plato's Form of the Good:
Ethics without Metaphysics?" Philosophical Papers 18 (1989): 154. He
takes Aristotle to be obviously right that a Platonic Form is not the sort
of thing that can be possessed, and defends Plato by denying that his
theory makes any such claim. Instead, he takes Plato merely to be saying
that the Form of the Good must be known, the better to possess other
goods. A related view seems to be presupposed by Nussbaum, who takes
Plato to believe that "the bearers of value are activities/' See Fragility of
Goodness, 148. On this view, the Forms themselves cannot be "bearers
of value/' since they are not activities. Rather, they have value because
they are the objects of pure, stable, truth-discovering activity. See pp.
145-8.
24 See Paul Ziff, Semantic Analysis (Ithaca, N.Y., i960), 210, 216. His view
is that when we call something "good" we are saying that it "answers to
certain interests" (p. 117). Unless we are provided with further informa-
tion, it is not clear how water can meet this condition. Of course, on my
reading, Plato is not merely saying that the Forms answer to certain
interests. They are good quite apart from our interests, and because of
their great goodness it is in our interest to possess them.
25 The analogy of the cave (5i4a-5i7c) and the critique of artistic imita-
tion in Book X (see esp. 596a-597d) bring out this aspect of the theory
most fully. See too 477a, 478d, 479d. For discussion, see Gregory Vlastos,
"Degrees of Reality in Plato," in Platonic Studies, 2d ed. (Princeton,
Cambridge Companions Online © Cambridge University Press, 2006
336 THE CAMBRIDGE COMPANION TO PLATO
1981), 58-75. Vlastos holds that the Forms are fully real in two senses:
they have the highest degree of cognitive reliability, and they have a
kind of value that "transcends the usual specifications of value" (p. 64).
Of course, Plato cannot simply lay it down without argument that
Forms have this transcendent value, nor can he infer that they have it
merely because of their greater cognitive reliability.
26 So read, the arguments of Books II-IV and of V-IX are mutually support-
ing: The later material adds content and support to the thesis that justice
is a psychological harmony, and that thesis in turn supports the identifica-
tion of being in good condition with being harmoniously arranged.
27 Some support for this interpretation comes from the Philebus, since Plato
there appeals to measure and proportion to explain the nature of goodness
[Phil. 64d-e). Throughout the cosmos, and not merely in human affairs,
wherever limit is imposed on the disorder inherent in the unlimited, a
harmonious unification is achieved, and this harmony is what makes
things good. See Phil. 23c-26d. I take Plato to be saying that a thing of one
type is better than something of the same type if it has a greater degree of
the harmony appropriate for things of that type; and a thing of one type is
better than something of a different type if things of the first type can
achieve a higher degree of harmony than things of the second. Harmony is
for Plato a form of unification, and so on my view he connects goodness
and unity. Note his emphasis on unity as the greatest civic good: 462a-b;
cf. 422e-423c. On the role of unity in Plato's argument, see White, Com-
panion, 31, 38-40. For further discussion of the Form of the Good, see
Cooper, "Psychology of Justice," 154-5; Irwin, Plato's Moral Theory,
224-6; G. X. Santas, "The Form of the Good in Plato's Republic," in
Essays in Ancient Greek Philosophy, ed. John P. Anton and Anthony
Preus (Albany, 1983), 2:232-63; and Reeve, Philosopher-Kings, 81-95.
28 In section VI, I will discuss another part of Plato's answer: Some acts of
justice imitate the Forms.
29 For further discussion of the ways in which Plato's novel understanding
of justice is related to the ordinary Greek conception, see Gregory
Vlastos, "Justice and Happiness in the Republic/' in Platonic Studies, 26.
ed. (Princeton, 1981), 111-39. This is a response to Sachs, "A Fallacy in
Plato's Republic," which argues that these two conceptions are uncon-
nected. Some other responses to Sachs are Annas, Introduction, chap. 6;
Raphael Demos, "A Fallacy in Plato's Republic!" Philosophical Review
73 (1984): 395-8; Irwin, Plato's Moral Theory, 208-12; Richard Kraut,
"Reason and Justice in Plato's Republic," in Exegesis and Argument, ed.
E. N. Lee, Alexander P. D. Mourelatos, and R. M. Rorty (Assen, 1973),
207-24; and Reeve, Philosopher-Kings, chap. 5.
30 For this criticism of Plato, see Annas, Introduction, 304.
Cambridge Companions Online © Cambridge University Press, 2006
The defense of justice in the Republic 337
31 Here my reading differs from that of Nicholas P. White, "Happiness and
External Contingencies in Plato's Republic," in Moral Philosophy, ed.
William C. Starr and Richard C. Taylor (Milwaukee, 1989), 1-21. He
denies that Plato tries to defend the thesis (one that White labels "ab-
surd") that "every just person is at every moment better off than every
unjust person," regardless of differences in their good or bad fortune (p.
16). Rather, he takes Plato to hold the weaker view that justice "is the
best strategy" over the long run, and he thinks Plato's defense of this
thesis is not completed until Book X. On this reading, when the just
person is on the rack, he is at that point worse off then an unjust person
basking in undeserved glory. This commits Plato to the view that if the
just person dies on the rack, and so has no long run, then despite the fact
that he chose the best strategy, his life is worse than the lives of some
who are unjust.
32 The philosophers' motivation for ruling has received much discussion
but no consensus has emerged. For some of the conflicting views, see
Annas, Introduction, 266-71; Cooper, "Psychology of Justice," 155-7;
Irwin, Plato's Moral Theory, 242-3, 337-8; Richard Kraut, "Egoism,
Love and Political Office in Plato," Philosophical Review 82 (1973):
330-44; Reeve, Philosopher-King, 95, 197-204; White, Companion, 44 -
8, 189-96; White, "Ruler's Choice."
33 For the view that Plato makes an exception in this one case, see White,
"Ruler's Choice."
34 I do not believe that Plato ever claims or commits himself to the thesis
that the best human life is the one that has the greatest amount of purely
contemplative activity. What he does clearly hold is that such activity is
better than political activity (52oe-52ia) ; but this does not entail that
pure contemplation that creates injustice is more advantageous than
political activity that is justly required.
35 For a fuller presentation of the interpretation I have given in this sec-
tion, see my "Return to the Cave: Republic 519-521," in Proceedings of
the Boston Area Colloquium in Ancient Philosophy, vol. 7, ed. John J.
Cleary (forthcoming).
Cambridge Companions Online © Cambridge University Press, 2006
ELIZABETH ASMIS
11 Plato on poetic creativity
In Book X of the Republic, Plato expels the poetry of Homer and his
followers- "the poetry of pleasure/' as he calls it-from his ideal
state by observing that there is an ancient quarrel between philoso-
phy and poetry. At the same time, he expresses a willingness to put
aside the quarrel. His spokesman, Socrates, throws out a challenge:
If the poetry of pleasure or its defenders can show that it is "not only
pleasant, but also useful for cities and human life," they would
gladly receive it back (6o7a-e). Plato returns to this challenge in his
last work, the Laws. The tragic poets approach the lawmakers and
ask, May we bring our poetry to your city? The lawmakers reply that
they, the lawmakers, are "poets" too, rivals and competitors in mak-
ing the "most beautiful drama." Their drama is the state, an "imita-
tion of the most beautfiul and best life." If the tragedians can show
them dramas that agree with theirs, they will be allowed to perform;
otherwise not (8i7a-d).
In the Laws, Plato takes the more conciliatory stance of one who
admits rather than expels, but the quarrel persists. Only the type of
poetry that is politically correct is permitted; the rest is banished.
The reason is that poets and lawmakers are rivals in fashioning
human life. Both are at once "makers" (the etymological meaning of
poietai, "poets") and "imitators" of moral values; and in a well-
ordered society they must speak with one voice. This subordination
of poetry to politics has offended many readers of Plato from antiq-
uity to the present. Plato sees the poet primarily as a maker of ethics,
and this concern appears strangely one-sided. What makes his posi-
I am very grateful to Richard Kraut for his helpful suggestions on this paper.
338
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 339
tion especially jarring is that, like another famous literary moralist,
Tolstoy, he was himself a consummate literary artist. Yet Plato has a
much more complex view of poetry than his strict morality sug-
gests. Along with his censorship goes a far-ranging exploration of
poetic creativity. Trying out various approaches in different dia-
logues, Plato enters into a dialogue with himself; and the tensions
and variations in his own thinking illuminate many aspects of the
aesthetics of poetry. Plato's discussions are worth taking seriously,
even though some of his conclusions are repugnant - to himself in
part, as well as to his readers.
Plato's quarrel with poetry takes its start in the fact that Greek
poets had a crucial role in the creation and transmission of social
values.1 It was traditionally believed that poets, like prophets, were
inspired directly by the gods with wisdom about the human and
divine condition. It was the prerogative of poets to make known the
past, present, and future to their contemporaries and future genera-
tions by oral performances of their poems. Prose writings and books
did not become common until the fifth century B.C., and even then
the primary method of publishing a work was oral performance. The
poems were chanted or sung, usually to instrumental accompani-
ment, at gatherings that ranged from private affairs to celebrations
held by an entire community or region, such as the dramatic festi-
vals in honor of Dionysos. Most occasions had a religious setting,
and many poetic performances were a form of religious worship. The
audiovisual role of television in modern technological society pro-
vides a partial analogy to Greek oral culture. It differs in that it
usually lacks the immediate impact of an event that engages one's
most deeply held beliefs. To grasp the role of poetry in ancient
Greece, one might think of Hindu religious drama - in which gods
confront the audience directly in terrifying struggles between good
and evil - and gospel meetings, along with rock concerts, opera, and
television.
The values transmitted in poetry evolved continuously. While
many poems - most prominently those of Homer - were passed on
with little or no change from one generation to the next, poets and
performers were continually reinterpreting their past. Poets not only
preserved values, but also questioned and subverted the traditions
they inherited, and long before Plato's attack on poetry, there were
Cambridge Companions Online © Cambridge University Press, 2006
34O THE CAMBRIDGE COMPANION TO PLATO
poets who condemned poets. The first known critical attack on po-
etry was by the poet Xenophanes in the sixth century B.C. In the
same epic meter used by Homer and Hesiod, Xenophanes denounced
these poets for " attributing to the gods everything that is a shame
and reproach among humans - to steal, commit adultery, and de-
ceive one another."2 The quarrel continued with Heraclitus's stern
verdict that Homer and Archilochus " deserve to be thrown out of
the contests and beaten with the rod."3 Heraclitus used prose, but
his attack belongs to the same tradition of criticism as that of
Xenophanes. About the same time, Parmenides laid the foundations
of metaphysics and logic in a poem modeled in part on Homeric
epic. The dramatists of the fifth century quarreled with their poetic
predecessors no less vehemently than earlier poets did.
Plato's view of the quarrel between poetry and philosophy in-
volves a third group, the sophists. Their name, "wise men" [soph-
istai), which soon became a term of derision, shows that they consid-
ered themselves heirs and rivals to the poets. In Plato's Protagoras,
Protagoras (fl. c. 450 B.C), leader of the first generation of sophists,
proclaims that he was the first person to claim a place openly within
the tradition of Greek educators (316-317c). As heir to the poets, he
considers the most important part of education to be the criticism of
poetry (338e-339a), and he illustrates his contention by attacking a
well-known poem by Simonides. In their challenge to the poetic
tradition, the sophists used a new weapon, prose. Partly, they discov-
ered new possibilities of language in prose; partly, they attempted to
capture the power of poetry by modeling their prose on poetic usage.
One new use of prose was to engage the listener in an exchange of
questions and answers, with the aim of scoring a victory by forcing
the respondent to agree with whatever is proposed. Socrates' dialecti-
cal method is a development of this invention. The sophists were
also the first to teach methods of argument. Unlike the poets, they
claimed no authority for their teachings except their own "wis-
dom." They emphasized the practical utility of their teaching,
which they regarded as the culmination of a series of inventions
devised by humans for their own advancement.
Along with their new use of language, the sophists developed theo-
ries of language. We are fortunate that one of the few extant writings
of the sophists contains a brief theory of language - the first in the
Western tradition. In the Encomium of Helen, Gorgias (fl. c. 430 B.C.)
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 341
personifies language, logos, as a "great potentate, who with the tini-
est and least visible body achieves the most divine works. "* The
"works" created by language are, remarkably, its effects on others.
When logos is accompanied by persuasion, it "shapes the soul as it
wishes." Just as drugs can drive various humors from the body and
end either illness or life, so language can set up various emotions in
the soul, and "drug and bewitch with an evil persuasion."* As evi-
dence, Gorgias cites poetry and magical incantations as well as scien-
tific, forensic, and philosophical prose. Defining poetry as "language
with meter," he points out that by implanting intense fear, pity, or
yearning in the listener, it makes the soul "suffer an affection of its
own" at the fortunes and misfortunes of others.6 Elsewhere, Gorgias
singles out tragedy as a kind of deceit. In this case, the speaker is
justified in practicing the deceit, as the listener is wise for being
deceived.? In the Encomium, Gorgias explains that logos would not
have the power it does if there were not such a wide field of igno-
rance. As it was, we have only partial knowledge of the past, present,
and future,- language fills this gap by supplying fallible beliefs to the
soul.8
In this theory of language Gorgias classifies poetry as a subdivi-
sion of language, while extending its power to all of language. He
adopts a simple scheme of cause and effect: A message is sent from
speaker to a recipient who accepts it passively by a change in his
soul. The listener is momentarily put in the power of another, as
demonstrated most vividly by the effect of poetry and magical
spells.* The speaker controls the listener not by any insights that he
has, but by the language that bears his message. In general, what
language creates is neither knowledge nor proposals by themselves,
but beliefs and emotions imprinted in the souls of others. It is not an
instrument of learning, but of persuasion. This theory of language
fits within a general theory of the effect of the perceptible environ-
ment on a person. Just as language shapes the soul by being heard, so
objects of sight shape the soul by being seen. Examples adduced by
Gorgias are the terror produced by the sight of enemy soldiers as
well as the delight caused by paintings and statues.10 Artistically
fashioned objects, whether heard or seen, have the same kind of
impact on the soul as other objects of experience.
In his response to the poets and sophists, Plato sought to change
language into an instrument of investigation and moral reform. He
Cambridge Companions Online © Cambridge University Press, 2006
342 THE CAMBRIDGE COMPANION TO PLATO
aimed in short to replace the logos of the poets and sophists with the
logos of the philosophers - those who love wisdom, rather than those
who pretend to it. He generally draws a distinction between the disin-
terested creativity of the poets and the sophists' self-interested ma-
nipulation of language; but at times his criticism of the poets and the
sophists merge. In the Apology and Meno very briefly, and in the Ion
at length, Plato takes the additional view of the poet as a divinely
inspired individual and finds a flaw in it: Poets speak by divine inspira-
tion without knowing what they say. In the Apology, Socrates tells
that when he went to the poets to test the Delphic oracle, he found
that, although they said "many beautiful things," they were utterly
incapable of explaining what they said. Thereupon he concluded that
they composed "not by wisdom, but by some natural talent and inspi-
ration," like prophets.11 He suggests in the Meno that poets are in-
spired with correct beliefs by a god.12 Having such beliefs, he main-
tains, is not sufficient for having knowledge or being a teacher. In the
Ion Socrates extends the poet's inspiration to the performer and the
listener: All hang in a chain from an original divinity - poet first, then
performer, then listener - like successive iron rings from a magnet.
None of these human links has a "craft" (techne) because none has
any knowledge of what he is doing. ̂
This is a picture of innocence as well as ignorance. In both the
Apology and the Ion, however, ignorance is accompanied by delu-
sion. In the Apology, Socrates points out that poets think they are
wise about matters in which they are not. In the Ion, Plato exempli-
fies this delusion by the rhapsode Ion, who starts by claiming that he
knows everything that Homer does and ends with the ridiculous
assertion that he knows at least how to be a general. Ion is a false
teacher, Plato implies, because he claims to be able to explain what
Homer said, when he cannot.x* The link with deity, moreover, does
not preclude a poet's words from being false, as Socrates insinuates
when he requires that the correctness of poems be judged by an
expert. Nothing indeed prevents the poet, or any other inspired crea-
ture, from being as fatuously ill-informed as Ion. Although Ion bears
the brunt of Socrates' attack, Plato indicates that divine possession
is a bad reason to regard anyone - even "the best and most divine of
poets," Homer (530b 10) - as an authority.
The traditional link with deity appears very precarious, then, so
much so that what is commonly revered as divine inspiration threat-
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 343
ens to dissolve into an all-too-human talent.1* In the Ion Plato al-
ready suggests that the rhapsode does not wholly hang from a divine
source: Ion acknowledges that he pays great attention to the crowd,
to see whether they hang on his words; for otherwise he will lose
money (535c). In his concern for money and reputation as well as his
practice of poetic exegesis, Ion has much in common with the soph-
ists. It is not surprising, then, that in another early diaglogue, Gor-
gias, Socrates associates poetry with the rhetoric of the sophists.
The whole dialogue is an attack on Gorgias's theory that language is
a great power; and incidentally Socrates brings poetry into the at-
tack. Socrates does not indeed dispute Gorgias's contention that
language has great power over the listener; his whole attack rests on
a tacit acceptance of this view. Instead, he argues that those who use
language unjtistly have no power; for they lack the power to accom-
plish what they )really want - justice in their own souls. The rheto-
ric of the sophists, Socrates charges, is a pseudo-craft, a mere "sem-
blance" [eidolon] of the political craft of justice. It is a flattery of the
soul, just as cookery is a flattery of the body. Instead of seeking what
is best, it only seeks to gratify the crowd by taking their likes and
dislikes as a standard.16 Rhetoric shares this aim with poetry. For
when stripped of melody, meter, and rhythm, poetry is a form of
"public speaking" (demegoria, a term whose meaning verges into
"demagoguery")- a "rhetoric" of the theater-that aims to flatter
the populace.1?
In the Gorgias, Socrates subverts the claim of the sophists to have
a craft that improves human life. Correcting the sophistic theory of
language, he claims that poets, too, are adept only at pandering to
the crowd. This is an abrupt change from the qualified respect that
Socrates shows the poets in the other early dialogues. Socrates
builds up to his conclusion by an inductive chain in which he argues
that public musical performances, from flute playing and lyre play-
ing to choral productions, dithyrambic poetry, tragedy, and finally all
poetry, aim only to gratify the audience. In constructing this se-
quence, Socrates has the respondent agree that the "revered and
wondrous" poetry of tragedy (5O2bi) also aims at the pleasure rather
than the improvement of the spectators. The respondent's ready
assent is surprising in view of the traditional respect bestowed on
poets as divinely inspired teachers. But the response is not at all
implausible as a complaint against contemporary tragedy, as prac-
Cambridge Companions Online © Cambridge University Press, 2006
344 THE CAMBRIDGE COMPANION TO PLATO
ticed, for example, by Euripides.18 By a bold inductive leap, Socrates
extends the complaint against musical and dramatic performances
to all of poetry. As in the Ion, he does not attack the most respected
of all poets, Homer, directly. Instead, he attacks theatrical produc-
tions and infers that all poetry plays to the crowd.
Having broken the link with deity, Socrates suggests a different
reason why poetry, in common with rhetoric, has such great power
over the listener: The poet's logos shapes the soul by indulging the
listener's craving for pleasure. This logos has no autonomous power;
it is parasitic upon the desires of the listener, and the author's cre-
ativity is nothing but an adaptation of words to the beliefs of the
listener. Plato therefore undermines Gorgias's theory of language in
another way. While admitting that language has power over an-
other's soul, he proposes that this power is dependent on the condi-
tion of the listener's soul. If the listener is morally feeble, the ostensi-
ble power of the logos consists, paradoxically, in strengthening this
weakness; in short, it is utterly frail. With an allusion to Socrates'
trial, Plato has Socrates suggest that he may be the only person to
practice the true craft of politics (52id). His logos cannot but offend
the crowd since it aims at moral improvement.
The Symposium marks an important change from the early dia-
logues. Plato now uses his new theory of Forms to present poetry
more favorably than in any other dialogue. He subsumes poetic activ-
ity under the Form of beauty and makes love its motive force. Poetry
becomes a private concern-an act of communication between a
lover and his beloved. The whole theory is attributed to Diotima,
the prophetess who acts as Socrates' teacher. She begins her analysis
with a complex definition of love, eids, as an intermediary between
gods and human beings, and as a desire for the creation of an immor-
tal good in something beautiful.1* Drawing a distinction between
those who are creative in body and those who are creative in soul,
Diotima exemplifies the latter by poets and "inventor" craftsmen,
along with lawgivers. They are all creators of "prudence" (phronesis)
and other forms of "goodrtess" (arete). Being "divine" in soul, they
are pregnant with their offspring from an early age and give birth to
them when they come upon a beautiful soul joined to a beautiful
body. The act of procreation consists in an abundance of words about
goodness, spoken to the beloved with the aim of "educating" him to
be a good person and to "practice" (epitedeuein) the right way of life.
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 345
Lover and beloved then join in a close and stable union to rear what
has been created. Examples of such offspring are the poems of Ho-
mer and Hesiod and the laws of Lycurgus and Solon.20
From this explanation of intellectual creativity, Diotima proceeds
to what she calls the "perfect" mysteries (210a). She reveals an as-
cent from an attraction to beautiful bodies, to an appreciation of
beautiful souls and practices [epitedeumata), to a contemplation of
knowledge, and finally to a vision of beauty itself. The second main
stage consists in valuing beauty of soul much more highly than
beauty of body. Although Diotima does not assign a place to the
poets in this ascent, there are sufficient similarities in her account
to suggest that their endeavors may be mapped onto it. By valuing
psychic beauty jointly with bodily beauty and concerning them-
selves with correct practices, the poets appear to be ascending from
the first to the second main stage. Diotima's description of those
who have attained the second stage as "giving birth to and seeking
words that will make the young better" (210c) echoes her previous
description of the poets and others. The next main stage is the cre-
ation of philosophical discourse, as stimulated by the beauty of the
various kinds of knowledge. Finally, the philosophically strength-
ened person has a vision of beauty itself and produces no longer
"semblances (eidola) of goodness," but true goodness, constituting a
godlike immortality (212a). The poets therefore appear to occupy an
honorable position between the masses and the philosophers. Cre-
ators of "semblances of goodness," they are nonetheless advancing
along a path that leads to philosophy and true goodness.
Diotima offers a new interpretation of divine inspiration as the
stirring of moral insight in the soul. Teeming with these insights
from youth, the poet shapes and improves them in response to the
moral beauty of another. Poetic creation thus becomes a joint enter-
prise, nurtured by the sympathetic response of a privileged listener.
This is the opposite of the theatrical poetry denounced in the Gor-
gias. Very strikingly, Diotima does not draw a distinction between
the creation of a poem and the creation of moral goodness. In her
account, the poet is a creator of moral goodness and the poem serves
only as a means of conveying this goodness. This poetic ontology is
fundamental to Plato's whole conception of poetry: A poem is a
linguistic reflection, or image, or a psychic disposition. It is essen-
tially a moral rather than a linguistic construct; formulated in Ian-
Cambridge Companions Online © Cambridge University Press, 2006
346 THE CAMBRIDGE COMPANION TO PLATO
guage, it is realized by being imprinted in the soul of another. This
subordination of linguistic to moral form explains another oddity in
Diotima's story. We might expect her to associate beauty directly
with the poem; instead she associates it with the soul of the beloved.
The reason is that a poem's beauty is a response to the moral beauty
of the listener. The goodness of a poem is judged by reference to an
ideal beauty that is identical with moral goodness.
Plato is often thought to have lacked a notion of poetry as an expres-
sion of personal feelings and beliefs.21 Between his theory of imita-
tion, as developed in the Republic, and the traditional assumption of
divine inspiration, he seems to leave no place for self-expression in
poetry, as proposed so vigorously by the Romantics. Yet in the Sympo-
sium Diotima comes close to formulating a version of the notion of
art as self-expression. She views poetic creativity as an inner spring
that wells forth from the poet's soul and is continually replenished by
communion with another. Although Diotima explains this creativity
as a semidivine force, love, striving to attain a transcendental beauty,
it is an intensely personal endeavor, strengthened by an interpersonal
bond. As a search for moral goodness, poetic activity is inseparable
from self-searching and self-awareness. Like the other "inventors"
and lawgivers, the poet gives voice to his own aspirations as he at-
tempts to transcend his own mortal existence by union with another.
Whereas in the Gorgias Socrates took theatrical poetry as the para-
digm of all poetry, Diotima implicitly takes love poetry as her para-
digm. This is an unusual view, which goes against both ancient and
modern conceptions of the poet as a self-sufficient genius. Yet it
serves to explain another commonly recognized element in poetry,
the universality of its values. In Diotima's account, the poet's con-
cern with human values, like that of the legislator and others, takes
the form of a devotion to another human being, serving as an alter
ego; and from this basis it extends to the rest of humankind. This
approach to immortality allows Diotima to give approval to tradi-
tional poetry, beginning with Homer and Hesiod, and by implication
to such poets as the comic playwright Aristophanes and the trage-
dian Agathon, whose eulogies of love in the Symposium provide
vivid confirmation that their poetic creativity is inspired by love.
Love poetry, indeed, becomes the model of all intellectual creativ-
ity. All the discourses spoken at the banquet exemplify Diotima's
analysis of intellectual creativity and all may be mapped along
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 347
Diotima's ascent. Alcibiades fills in details about the highest form of
creativity, the philosopher's, in the crowning speech of the Sympo-
sium, his tribute to Socrates. He compares Socrates' words to carved
sileni - uncouth on the outside, but "most divine" on the inside,
"having very many representations of goodness in them" (222a).
These words have the power to bewitch and possess the listener, like
the music of Marsyas or the Sirens. As Alcibiades testifies about
himself, Socrates' words reduced him to a tearful frenzy of self-
recrimination; they filled him with a sense that his life was not
worth living as it was (2i5d-2i6c). Alcibiades transfers to Socrates
the power of enchantment traditionally assigned to poets, with the
important difference that Socrates' words incite his listener to look
into himself. Socrates practices in a much-heightened form what
Diotima's poets attempt to do.
It is easy to overlook Diotima's remarks on poetry in the Sympo-
sium in view of the much more extended treatment in the Republic.
In this dialogue, Plato considers poetry in the first place as a means of
educating children to be Guardians in his ideal state. Since Plato
believes that children's souls are especially malleable, he is especially
concerned with the impact of poetry on them; but he soon extends his
concern to adults.22 The public returns to replace the privileged lis-
tener. The whole discussion consists of two parts, whose relationship
has been much debated. Plato purges poetry in Books II and III; then
he returns to this purge in Book X by explaining in detail what is
wrong with the poetry that he eliminated. In the two parts, Plato
builds a powerful new theory of poetry as "imitation," mimesis.
In Books II and III, Socrates argues, first, that poets must present
the truth about the gods and heroes, who are to serve as models of
goodness. Second, poets must "imitate" only good individuals or
individuals engaged in good action, so that their listeners may in
turn imitate only good.23 By "imitation" (mimesis), Socrates means
"impersonation" (392d-394c). The poet "imitates" another when-
ever he speaks the words of a character in direct speech, as though he
were that character. By contrast, the poet "narrates" whenever he
reports in his own person what a character is doing or saying. As
moral educator, the poet must imitate only morally good speech,
and he must narrate all the rest. For the poet's experience becomes
the listener's, by a similar transfer of experience as is assumed in the
/on's image of the iron rings. If the listener persists in imitating the
Cambridge Companions Online © Cambridge University Press, 2006
348 THE CAMBRIDGE COMPANION TO PLATO
same kinds of characters from childhood, he ends up by having the
same moral configuration (395d).
Earlier in Book II, Socrates' healthy city, the "city of pigs," had
become isolated and feverish by the admission of imitators, such as
painters, poets, rhapsodes, actors, dancers, and others (373b). Subse-
quently Socrates purges this city,- and his first and main concern is
poetry. Like a physician, he reconstitutes poetry as a health-giving
drug. Plato agrees with Gorgias that poetry has the power both to
heal and to poison the soul, and that it is especially effective because
it can make a person assume the identity of another. In his analysis,
which has the dry detachment of a clinical dissection, Socrates asso-
ciates this special power with a particular type of diction, "imita-
tion." It poses a special danger because it removes the distance of
personal judgment. The right kind of poetry, therefore, will have
only a small amount of imitation, consisting only in the imitation of
the good, and it will have much narration (396c). This requirement
automatically eliminates traditional tragedy and comedy, as Socra-
tes' interlocutor recognizes. As Socrates hints, it also eliminates
Homeric epic (394d). After determining the right proportion of imita-
tion and narration, Socrates completes his pharmaceutical activity
by prescribing melodies and rhythms that suit the content. Just as
the language must mark the difference between good and bad, so the
melodies and rhythms must reinforce this difference by resembling
the simplicity and restraint of morally good habits. With its kernel
of imitation, the new poetry is carefully designed to confer a maxi-
mum of moral benefit by providing an experience that simulates
that of a good person as closely as possible.
Just as the physician of so many of his examples does not hesitate
to drive out corruption by the most violent and painful means, Socra-
tes purifies the city by expelling the offending poets, who also give
the most pleasure, and forcing the rest to comply with the laws. The
poets, he decrees, "must be directed and forced to implant the image
(eikona) of a good moral habit in their poems, or not compose poems
in our midst" (401b). Socrates' violence reaches a shocking climax in
his banishment of the poet who can imitate anyone and anything.
He proclaims the expulsion in a resounding sentence: If "a man who
is able to become every kind of individual (pantodapos) because of
his wisdom and to imitate all things" were to come to our city,
wishing to perform his poems, we would honor him as "sacred
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 349
(hieron), marvelous, and giving pleasure/' but say that "there is no
such man in our city nor is it lawful that he should come to be, and
we would send him off to another city, pouring myrrh on his head
and crowning him with a chaplet" (398a). There is no such man in
the new city because none is allowed to assume more than one role,
especially not a combination of virtue and vice. Reverentially, Socra-
tes does not name the man; but everything points to Homer. The
"sacred" poet, though revered, is a blight upon the city; and he is
sent off, consecrated to a god like a scapegoat.2*
From the purge of poetry, Socrates moves to a general cleansing of
the city and a general theory of the arts (4oia-d). Not only must
poets be forced to create images of goodness, but all artists and
craftsmen must do likewise. Paintings, weavings, embroideries,
houses, and all equipment whatsoever must express beauty and
seemliness, so that the young will receive this seemliness into their
souls. Still showing the concerns of a physician, Socrates compares
this aesthetic environment to a healthy place, in which the inhabit-
ants are caressed by gentle breezes from their surroundings. He as-
sumes that sounds, shapes, and indeed all sensory objects-both
natural and artificial — have qualities that resemble moral qualities,
and that these qualities have a direct influence on the moral habits
of the perceiver. Not only the beliefs conveyed by language, but also
sensory stimuli that do not engage one's cognitive faculties at all,
shape the moral habits of the soul. Like Gorgias, Plato assumes that
sensory objects, like language, shape the soul directly. But his asso-
ciation of moral character with sensory objects is new; and it is
highly questionable. It is a far step from demanding that the lan-
guage, music, and rhythms of poems must suit the content to claim-
ing that sounds and other sensory objects express or resemble moral
character. In Plato's general aesthetics, the whole sensory environ-
ment is an image, or "iconic" symbol, of moral goodness or bad-
ness.2* Unsatisfactory as this claim is in its context, it anticipates
the metaphysical system that will be proposed subsequently in the
Republic, according to which the sensible world is ultimately an
image of the Form of the Good. The sensible world contains images
of goodness; and the human craftman must emulate the divine
craftsman by creating images of goodness of his own.
In the Republic Book X this true aesthetics fades from sight, so
much so that Plato has been thought to abandon it altogether. This
Cambridge Companions Online © Cambridge University Press, 2006
35O THE CAMBRIDGE COMPANION TO PLATO
book, which contains Plato's most searching analysis of what poetry
is, raises serious problems of interpretation. The first problem is:
What and how much poetry does Plato expel? In particular, what is
the meaning of Socrates' introductory statement that they were right,
in founding the city, "not to admit in any way any [part of poetry] that
is mimetic [mimetike)."16 This announcement is clearly a reference
to the purge of poetry in Book III. But in Book III Plato certainly did not
ban all poetic mimesis. Is there a discrepancy then? Some scholars
have argued that there is no clash, whereas many others have argued
that Plato has shifted position. The decisive point in favor of the first
interpretation, it seems to me, is that "mimetic" can mean not only
"imitating," but also "imitative" in the strong sense of "given to
imitation," with the connotation of "indiscriminately imitative" or
"all-mimetic." As others have pointed out, this strong sense has al-
ready been prepared in Book III.2? There is, then, neither a terminologi-
cal nor a substantive clash with the earlier discussion. For "mimetic"
poetry is not just poetry that imitates, it is poetry that imitates any-
thing at all. In Book III Socrates expelled the poet who is indiscrimi-
nately mimetic - in short, the "mimetic" poet; and in Book X he
defends this expulsion.
This is not to deny that in Book X Plato shifts away from the earlier
discussion both in his terminology and in his proposals. He returns to
the problem of poetry because the metaphysics and psychology that
he developed in the meantime provide a new justification for the
expulsion of mimetic poetry; and this new investigation is accompa-
nied by a new use of terms. In keeping with his use of "mimetic" to
imply "indiscriminately imitative," Plato now uses the terms "imita-
tion" (mimesis) and "imitator" (mimetes) in the new senses of "indis-
criminate imitation" and "indiscriminate imitator." At the same
time he adds a new dimension to the meaning of there terms: "mi-
metic" poetry, like all "imitation," is not only indiscriminately imita-
tive, but also thoroughly imitative - imitative to the core - because
it is at two removes from genuine creation. As an "imitation" of the
manifold world of human action, it is at the farthest distance from the
creation of genuine goodness. Although it might be argued that this
definition applies to all poetry, Plato carefully restricts his analysis to
the indiscriminately imitative poetry that he banished in Book III.
The first restriction occurs in the introductory phrase "any [part of
poetry] that is mimetic" (595a), with the implication that some is not
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 351
"mimetic." Subsequently, Socrates repeatedly singles out tragedy,
with its leading practitioner Homer, as exemplifying mimetic poetry.
When he reiterates his order of banishment in Book X, he describes
the poetry that he expels as the "poetry of pleasure and mimesis."28
Along with this expulsion, Socrates proposes to keep "hymns to gods
and praises of the good" (6o7a4). Just as he argued in Book III, the
poetry that he permits is a celebration of divine and human goodness;
and nothing contradicts the conclusion of Book III that this poetry
requires a correct mixture of impersonation - mimesis in the earlier,
narrower sense — and narration.
The second main problem is: What precisely is the new sense of
"imitation"? Plato develops his answer in three stages, in which he
successively reduces mimetic poetry into an object of detestation.
First, he defines "imitation" in general; second, he shows that the
poetry of Homer and his successors fits this definition,- and third, as
the climax of the entire argument, he shows that Homer and all
other imitative poets corrupt the listener with their poems and there-
fore deserve to be banished.2* In all three stages, Plato bases his
conclusions on an analogy between the painter and the poet; and a
major challenge to interpretation is to determine the relevance of
this comparison. Since the analogy works differently in different
stages of the argument, it is useful to consider each stage in turn,
while keeping in mind the overall structure of the argument.
Plato builds his new definition of "imitation" directly on the con-
clusions of Book III. There he banished the poet who "is able to
become every sort of individual because of his wisdom and to imi-
tate all things." Plato now explains this "wisdom" (sophia). It is no
wisdom really, but fakery, because it is just like taking a mirror and
reflecting all things in it. The person who looks as though he can
make all things is a "marvelous sophist" (sophistes, 596di), but he
only seems "all-wise" (passophos) to the ignorant (598d3-4). He
looks like a "maker" (poietes)} in fact, he is an "imitator," not a real
maker or craftsman, but a pseudo-maker of pseudo-creations. Plato
brings in his metaphysics of Forms, together with the painter, the
carpenter, and God himself, in order to reduce this amazingly pro-
lific "maker" into a mere shadow of a makers
Apart from one reference to the "maker of tragedy" (597e6), the
"poet" (poietes) is present in this first stage of the argument only
implicitly - though very emphatically - as "maker." The "imitator"
Cambridge Companions Online © Cambridge University Press, 2006
352. THE CAMBRIDGE COMPANION TO PLATO
is represented by the painter, who makes all things by making ap-
pearances of them, just like the person with the mirror. He imitates
objects made by human craftsmen, such as the bed made by the
carpenter. Above the carpenter is God, maker of the Form of the bed,
on which the carpenter models his creations. The painter's imita-
tions are at two removes from real being, the Forms; and the painter
is at two removes from the one truly genuine maker and craftman,
God. To this basic scheme Plato adds a refinement that will be of
crucial importance later (597e-598d). The painter does not imitate
the objects of human craftsmanship as they are, but as they appear.
For example, he imitates the bed as it appears from the side or the
front, not as it is. His imitation, therefore, and all imitation in gen-
eral, is the imitation of an appearance, not of things as they are in
this world. In short, an imitation is a "semblance (eidolon) that is far
from the truth," and the imitator knows nothing about any of the
crafts that he imitates.
Plato's mirror simile has had an overwhelming influence on the
interpretation of his aesthetics and on aesthetic theory in general. It
stands as a compelling symbol of the view that it is the job of the
artist to copy nature. But Plato's use of the simile needs to be inter-
preted with some care. In the first place, Plato thinks that copying
things of the sensible world is a perversion of the poet's function; it
is what the mimetic poet does. In the second place, the mimetic poet
does not aim to give faithful representations of the sensible world:
he gives impressions of it. The commonplace dichotomy between
mirroring and self-expression in the acts does not really fit Plato's
use of the mirror simile. As some of the Romantic spokesmen for an
expressive theory of art themselves pointed out, there is no incom-
patibility between imitation and self-expression. In the Republic
Plato combines the two approaches by requiring both external mod-
els and a personal response. Just as the painter renders aspects of
reality as they appear to him, so the poet renders his own impres-
sions of reality. These appearances prompt the ignorant to take them
for the real thing; but they are nonetheless distortions produced by
the imitator's own view of the world. Even though the brief account
of poetry in the Symposium comes much closer to the Romantic
notion of self-expression, the artist in the Republic also draws on his
own inner resources - in particular, as will become clearer, his emo-
tional inclinations - to give an interpretation of the external world.
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 353
Plato fills in gradually in the course of the discussion how the poet's
impressions correspond to those of the painter; and this elaboration
shows that the analogy between painting and poetry is not as ill-
conceived as has sometimes been thought. If we wish to keep the
simile of the mirror, it turns out that the mirror is the poet's own
soul: The external world is refracted by the poet's soul, not cast back
as a faithful reproduction.
Plato's hierarchy of being has also proved misleading. It might
seem that the poet can escape the shadowy unreality of his pseudo-
craft by moving higher on the scale, either to an imitation of the
Forms themselves or, at least, to a faithful replica of sensible reality.
The first alternative is barred by the argument, which is presented
later, that if the poet did model his creations on the Forms, then he
would be a maker of human conduct in this world, taking his place
among the craftsmen of this world; he would not devote himself to
making "semblances" of goodness (599a-6ooe). This argument has
been perceived as a serious difficulty for the view that in Book X
Plato leaves open the possiblity of a morally beneficial craft of po-
etry, as proposed in Books II and III; and we shall return to this
problem. The second alternative is also barred. As Collingwood has
argued, since imitations and sensible things are distinct categories of
being, there is no way in which the imitator can attain the truthful-
ness of his models1 Just as the carpenter cannot recreate a Form,
neither can the imitator recreate a sensible object. To suppose other-
wise would be, let us say, like supposing that Picasso can recreate a
woman by combining various aspects of her face and body in a single
picture. It might be objected that painters can make more or less
realistic copies. But a photographic type of realism is what Plato
most decries, as the most perfect illusionism. Although Plato indi-
cates in the third stage of his argument that a poet would err less if
he did approximate sensible reality more closely, he nowhere pro-
poses that a poet should model his work more closely on the actual
world. His reason is that hanging on sensible reality, without inde-
pendent guidelines, can only produce further distortions of reality. *2
In the second stage of his argument, Plato launches his direct
attack against poetry by reducing Homer and his followers from the
position of educators of the Greeks to the level of ignorant imitators.
There are those, he mentions, who assign a knowledge of all crafts to
Homer (598d-e); he will not address this claim. What he will investi-
Cambridge Companions Online © Cambridge University Press, 2006
354 T H E CAMBRIDGE COMPANION TO PLATO
gate is whether Homer had any knowledge of "the greatest and most
beautiful things" about which he attempts to speak, that is, "about
wars, generalship and the administration of cities, and about the
education of a human being" (599C-d). His first reference is an allu-
sion to the /022.33 The second part of his statement is an allusion to
Diotima's claim in the Symposium that Homer and other poets are
among those who "educate" men and lead them to right "practices."
Diotima proposed that "by far the greatest and most beautiful part of
prudence concerns the administration of cities and households"
(2O9a5~7); and she associated the poets, as exemplified by Homer
and Hesiod, with the lawgivers, in particular Lycurgus and Solon,
and with inventive craftsmen in general, as creators of prudence and
other forms of goodness. Plato now expels Homer and the poets from
this group.
With a point-by-point refutation of Diotima's account in the Sym-
posium, Socrates argues that Homer did not know what "practices"
[epitedeumata) make humans better either as a community or as
private individuals.34 Homer was not a good lawgiver like Solon or
Lycurgus, or an innovator in the crafts such as Thales,- and he was
not a private educator. In particular, his "companion" Creophylus
would provide a laughable example of an educated disciple (6oob6-
9). This obscure reference to Creophylus fills in a gap in Diotima's
account by supplying an example of someone allegedly loved and
educated by Homer. Socrates also brings Hesiod into his attack by
suggesting that Homer's and Hesiod's contemporaries would hardly
have let them roam around singing if they really had been able to
help others become good. His conclusion that all "poetic persons
starting with Homer" are "imitators of semblances of goodness"
(6ooe4~5) assigns a new meaning to an expression used in the Sym-
posium. There all except those who had reached the very pinnacle of
enlightenment were destined to create only "semblances of good-
ness." Now Homer and the other poets are singled out as creating
"semblances" that are fakes.
In this part of the argument, then, Plato corrects the positive view
of poetry that he presented in the Symposium. He now drives a
wedge between Diotima's view of poetic creativity and the "perfect"
mystery of the upward ascent. Instead of progressing toward knowl-
edge, poets are exposed as cultivating ignorance. Poetic education is
a fraud because poets don't know what they are talking about. This
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 355
is a further development of a view sketched in the Apology and Ion.
But Plato also keeps in mind the possibility that he canvassed in the
Meno, that correct belief may be just as good a guide as knowledge.
"So as not to leave things half said/' Socrates goes on, mimetic poets
also lack the right belief that comes from taking instructions from a
user (6oic-6o2a). Socrates resorts once again to the analogy of the
painter and now pairs him with a different type of craftsman, the
maker of reins and bridles. Although the latter lacks knowledge, he
makes a good product by acquiring correct belief from a user who
has knowledge. By contrast, the painter of reins and bridles neither
has knowledge himself nor acquires correct belief by associating
with a user who knows. Similarly, the mimetic poet has neither
knowledge nor correct belief. In his ignorance, he imitates what
appears beautiful to the ignorant masses (602b), as suggested previ-
ously in the Gorgias.
The first two stages of the argument prepare the third, which
culminates in the reiteration of the expulsion order, now addressed
explicitly against Homer and his successors.^ So far, Socrates has
shown that Homer and the others have nothing but a semblance of
wisdom. Now he shows that this semblance is a corruption of the
soul. By revealing the moral ugliness of traditional poetry and its
power to corrupt even the best citizens, he reduces it to the lowest
level of abomination, so that it must surely be purged. For the last
time, Socrates takes the painter as an analogue of the poet. Just as
the painter creates impressions that are accepted uncritically by the
beholder without any attempt at calculation, so all imitation ap-
peals to the nonrational, worthless part of our soul (6o2c-6o3b).
This use of the painter analogy is perhaps the weakest part of Plato's
entire argument. But it has the merit of signaling that the response
to a work of art differs fundamentally from the response to a real-life
situation. The beholder of a painting does not measure the impres-
sions before him against reality; he is seduced by appearances to
accept them as they are, without bothering to calculate how close
they are to the real thing. Plato touches on the notion that a work of
art is valued for its own sake, insulated from real-life concerns by a
suspension of disbelief. He spurns this outlook on art because he
thinks that make-believe is harmful unless it agrees with a transcen-
dent reality.
After depreciating the painter, Socrates goes on to show what the
Cambridge Companions Online © Cambridge University Press, 2006
356 THE CAMBRIDGE COMPANION TO PLATO
reader has been waiting for all along, an explanation of what sort of
appearances are produced by the mimetic poet. The mimetic poet
imitates humans engaged in action and thinking, with attendant
pain or pleasure, that they are faring well or badly (603c). Moreover,
he tends to imitate the release of emotions, such as grief, laughter,
lust, and anger, instead of the control of the emotions by reason. He
has a "natural" propensity to the emotional part of the soul, for it is
easy to imitate. This part is also easy to appreciate, especially by the
crowd in the theater. Therefore, in order to win fame among the
many, the imitative poet directs all his efforts to it (6o4e-6o5a).
In case the mirror simile, along with the examples of the painter,
carpenter, and bridle maker, misled the reader into thinking that the
poet creates word pictures, as it were, of objects such as couches and
bridles, he is now set straight. The poet imitates moral goodness and
badness, as shown in human actions, beliefs, and feelings. The depic-
tion of Achilles driving a chariot or running around the plain of Troy,
for example, is incidental to the imitation of his moral character.
Corresponding, therefore, to the Form of the couch or table are the
Forms of the virtues - justice, moderation, courage, and wisdom.*6
The mimetic poet creates nothing but distortions of these qualities,
since he looks only to what humans actually do and is drawn, more-
over, to the opposite of virtuous behavior - the disorderly rule of the
emotions. It would be bad enough if he merely imitated human con-
duct as it is; instead, he imitates it as it appears to the worthless part
of his and everyone else's soul. Instead of showing us Oedipus, let us
say, bringing order to the city by his wisdom, he dwells on the agony of
his downfall. Oedipus's success would be too boring to most people.
As in the Gorgias, Plato condemns poetry as theatrical demagoguery,
aiming to please the crowd by indulging its ignoble desires. He also
takes back the view of poetic creativity that Diotima sponsored in the
Symposium. Substituting base terms for Diotima's language of love
and goodness, Socrates claims that all imitation "consorts with and is
a companion (hetaira) and friend (phile) to something far from pru-
dence in us for the purpose of nothing healthy or true"; in short,
"being worthless, it associates with something worthless and creates
worthless things. "^ In contrast with Diotima's poet, who is united by
love to a beautiful soul and creates prudence, the mimetic poet prosti-
tutes himself in the service of the worthless part of the human soul
and creates nothing but worthlessness.
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 357
In many respects, Plato's mimetic poet stands for what many of us
value most in poetry. He creates his poems in response to the specta-
cle of human action about him. Immersed in this world, he derives
all his inspiration from it, ranging over it with a full engagement of
his emotions, entrancing others as he is entranced by life. Plato casts
out this poet on the ground that his emotional intensity "feeds"
(6o6b-d) and strengthens the emotional part of the listener's soul,
like a malignant cancer, while weakening the rational part. In con-
flict with the good lawgiver, the mimetic poet "implies a bad consti-
tution privately within the soul of each person" (605b). Worst of all,
this poetry is so powerful that it corrupts not only the many, but the
better sort (with very few exceptions) who ordinarily try to control
their emotions by reason (6o5C-6o6d). For they are seduced into
gaining pleasure by letting go of their emotions, with the excuse that
it is not disgraceful to share the emotions of someone else.
There are many ways of resisting Plato's diagnosis of the effects of
mimetic poetry while accepting much of his analysis of what the
poet does. Aristotle takes the view that the emotionality of poetry
cleanses instead of corrupts. Without denying the power of poetry to
shape moral habits, we might look for an escape in Plato's analogy of
the painter. In creating his illusions, one might argue, the painter
does in fact carefully measure appearance against reality. In fore-
grounding one aspect of the table or couch, for example, he calcu-
lates how this aspect is related to the whole and indicates this rela-
tionship to the beholder. This sort of measuring is wholly intrinsic
to the work of art; there is no reality apart from that of the artistic
object itself. Similarly the poet, even though he may emphasize the
emotions, sets them in the wider context of a moral order. Oedipus's
suffering is the more acute because of his search for wisdom; and
when the theatergoer suffers with him, his suffering is a measure of
his recognition of the nobility of Oedipus's character. By opposing
the emotions to rational insight, Plato closes off this way of defend-
ing poetry as a genuinely creative endeavor.
In Plato's view, how can the poet escape degradation? This ques-
tion takes us back to our first problem. What room is left for the type
of poet who implants "the image of a good moral habit" in his
poems, as proposed in Book III? It makes little difference whether
Plato exempts a part of poetry from his attack in Book X if his
analysis precludes there being such a part. Plato, I suggest, has
Cambridge Companions Online © Cambridge University Press, 2006
358 THE CAMBRIDGE COMPANION TO PLATO
worked an answer into his attack. The politically correct poet does
not indeed look to the Forms; for if he did, he would be a creator of
actual goodness in human beings - a lawgiver in fact. But it is open
to him to take correct beliefs from the lawgiver, the user of his
poems. Nor need one look for such instruction. Socrates' entire
analysis of poetry in Books II and III is nothing other than an elabo-
rate rule book, devised by the founders of the city for the instruction
of poets. Having correct beliefs, the poet is raised to the level of a
craftsman, like the carpenter, the bridle maker, and all the other
nonphilosophical craftsmen in the new city. Like the rest, the poet
has the position of serving the lawgiver. Instead of imitating humans
as they are or appear to be, he creates images of humans as they
should be, by taking directions from the lawgiver, who looks toward
the Forms. It is very doubtful whether this type of poet, coerced and
working according to rule, could have any appeal except to the law-
giver's educational zeal. Plato's own analysis of the mimetic poet as
exuberantly - even though morbidly - creative highlights this diffi-
culty. In the Republic, Plato presents just two possibilities: the poet
follows his own inclinations and corrupts, or he is coerced into
improving others. Nor does the painter or any other artist fare any
better. Like all the other craftsmen in the new city, they must create
a moral environment for the citizens; and they cannot achieve this
by being enamored of the sights and sounds of this world.
Plato reconsiders the place of poetry in human life in the Phae-
drus. As though dissatisfied with his coercion of poetry in the Repub-
lic, Socrates proclaims that no one can become a good poet unless
inspired with divine madness. In a brief, traditional definition of
poetry that is reminiscent of the Symposium, Socrates explains that
possession by the Muses gets hold of a "tender" soul and "educates"
later generations (Phdr. 245a). Poetic possession is one of four types
of divine madness, along with prophecy, ritual, and love (eras); love
is the best madness of all. Since divine madness is responsible for
the "greatest goods" to humans (244a), we might expect poetry to
have some role in the soul's return to heaven. But Socrates explains
this ascent as motivated entirely by love. Moreover, he places the
"poetic person" surprisingly low in his ranking of lives. The best life
is that of the "philosopher or lover of beauty or a musical and erotic
person (philosophou e philokalou e mousikou tinos kai erotikou)."
There follow the lawful king or military commander, the politician
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 359
or businessman, and the trainer or physician. Fifth is the life of the
prophet or conductor of rites; sixth is the "poetic person [poietikos]"
or anyone else concerned with mimesis'; seventh is the craftsman or
farmer,- eighth is the sophist or demagogue; and ninth and last is the
tryant (248d-e). The "poetic person" is lowest of the four types of
divinely inspired persons, with the prophet and conductor of rites
just above him. Divine inspiration is no more a guarantee of enlight-
enment than it is in the Apology or Ion.
Whereas in the Symposium Plato included poetry among the
works of love, he now demotes it by distinguishing it from love as a
separate kind of madness. As suggested by the conjunction of poetry
with mimesis, this demotion appears indebted to the Republic. At
the same time, Plato softens the position of the Republic by ranking
the mimetic poet one step above the manual worker. Divine inspira-
tion, it seems, assures the mimetic poet a place above ordinary crafts-
men and two ranks above the sophist by infusing some educational
value into his poems. But it is not nearly sufficient to raise him to
the level of the "erotic" person, the truly "musical" person who is
inspired by the Muses with the only genuine beautiful discourse,
that of the philosopher.
In the Phaedrus, then, Plato drives another wedge between tradi-
tional poetry and philosophy. At the same time, he proposes a trans-
formation of all types of discourse - in particular, poetry, political
rhetoric, and legal language - into philosophical discourse. The
main topic of the Phaedrus, like that of the Gorgias, is rhetoric. But
"rhetoric" is now taken explicitly in the broad, etymological sense
of "art of speaking" in general, so as to subsume public speaking.
Defining "rhetoric" as a "leading of the soul (psychagogia) by
words" in both private and public (261a), Socrates gradually trans-
forms the basic meaning of psychagogia, "conjuring" and "enchant-
ment," which fits the sophistic view of language, into the sense of
"leading the soul" to the truth, which fits philosophical discourse.
In the course of Socrates' examination, rhetoric becomes the art of
dialectic or "discussion." It requires, first of all, the ability to define
and divide the subject matter correctly, and, further, a knowledge of
the types of souls and their occurrences, together with a knowledge
of the kinds of language, so that the speaker may adapt his language
to the soul of each listener. Socrates exemplifies this genuine art of
rhetoric throughout the Phaedrus. Using various kinds of language,
Cambridge Companions Online © Cambridge University Press, 2006
36O THE CAMBRIDGE COMPANION TO PLATO
including a "mythic hymn" adorned with "poetic" words, he leads
his companion, Phaedrus, to an understanding of the proper use of
language.*8 Genuine rhetoric may include prose speeches and poetry,
but it is essentially an act of communication between two individu-
als. The main reason that Plato rejects writing as serious use of
language is that he thinks it precludes this communication.
Plato's new, comprehensive theory of language is accompanied by a
new theory of love. Revising Diotima's view of intellectual creativity
in the Symposium, Plato now proposes that only the philosophical
use of language is genuinely an act of love, aimed at enlightening the
beloved as well as oneself. Poetry without philosophy is without love
or insight; but it can be transformed into genuinely creative dis-
course. At the end of the Phaedrus, Socrates asks Phaedrus to relay a
message from the Muses: Their message to Lysias and any other
speechmaker, as well as to Homer and any other poet and to Solon and
any other writer of laws, is that if they know the truth about what
they composed, and can defend what they wrote by speaking about it,
while showing that what they wrote is worthless, then they deserve
the name of "philosopher" rather than the name that corresponds to
their compositions - that is, "speechwriter," "poet," or "lawgiver"
(278b-e). Plato's own dialogues may be regarded as attempts to exem-
plify this use of language. We may call them poetry as a tribute to
Plato's literary skills But from Plato's point of view it would be more
accurate to regard them as adumbrations or "semblances" of how all
sorts of language - poetic, political, legal, and the rest - may be
transformed into philosophical discourse.
In the Phaedrus Plato adds to his range of portraits of the poet the
limiting case of the composer of poetry who is truly inspired. This is
no longer a poet, but a philosopher. His compositions are but mo-
ments of thought, frozen in language and worthless in themselves.
The serious part of his endeavors consists in the discussion that
supports and supplements his compositions. Like islands in a sea of
reflection, the compositions are merely stages on the way to a dis-
tant goal. Although Plato withholds from this type of composer the
title of "poet," he presents a highly appealing model of a poet's goal.
For this creator of poetry is neither fettered by laws nor blinded by
appearances. What makes this model of poetry particularly attrac-
tive is that although Plato considers the linguistic construct worth-
less in itself, he assigns to it a special merit of its own - organic
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 361
unity. Like any composition, a poem must have its parts fitted into a
harmonious whole, with middles and extremities, like the body of
an animal (264c, 268c-d). This organization is determined not only
by the subject matter, but also by the needs of the listener. While
organic unity may be regarded as an aesthetic property - and Aris-
totle would later treat it as such by making it the focus of his analy-
sis of poetry - Plato subordinates it to a moral purpose. The unity of
the composition reflects both the truth of what is said and the moral
aspirations of the listener and speaker. The completeness of the text
contrasts with the incompleteness of the striving that inspires it,
and so signals its own incompleteness. In emphasizing the worthless-
ness of the composition, Plato warns the reader not to be seduced by
its merit as an artistic creation into accepting it as a definitive state-
ment. The cult of texts, oral or written, is alien to Plato.
One does not want to leave the last word on poetry to the Laws,
where Plato reduces the poet once more to a servant of the law-
maker. The old Athenian who has replaced Socrates as Plato's chief
spokesman suggests that the discussion that he and his companions
have had about the laws is a kind of "poetry": It is, indeed, the most
suitable of all poems and prose works for children to hear and teach-
ers to approve (81 ic-e). In this rivalry with the poets, the lawmakers
will surely lose if we appoint as judge the Socrates of the Phaedrus.
NOTES
1 In his pioneering study Preface to Plato (Cambridge, Mass.: Harvard
University Press, 1963), Eric Havelock emphasizes the importance of
oral poetic teaching in Greek society.
2 D.K., 21 B 11.
3 D.K., 2 2 B 4 2 .
4 D.K., 82 B 11, 8.
5 D.K., 82 B 11, 13-14.
6 D.K., 82 B 11, 9.
7 D.K., 82 B 23.
8 D.K., 82 B 11, 11.
9 Gorgias sought to recreate the enchantment of poetry in his own prose
by using balanced clauses and sound pattern to emulate the rhythms of
poetry. See Jacqueline de Romilly, Magic and Rhetoric in Ancient
Greece (Cambridge, Mass.: Harvard University Press, 1975), 8-11.
10 D.K., 82 B 11, 15—19.
Cambridge Companions Online © Cambridge University Press, 2006
362 THE CAMBRIDGE COMPANION TO PLATO
11 Ap. 22a-c.
12 Meno 99C-d.
13 Ion 533d-534d, 535e~536d.
14 Ion prides himself not only on being able to recite Homer's poems, but
also on being able to explain his meaning better than anyone else (Ion
53ob-d).
15 See further Paul Woodruff, "What Could Go Wrong with Inspiration?
Why Plato's Poets Fail," in Plato on Beauty, Wisdom, and the Arts, ed.
Julius Moravcsik and Philip Temko (Totowa, N.J.: Rowman & Little-
field, 1982), 137-50.
16 Grg. 464b-465d, sooe-5Oic; the term eidolon occurs at 463d2 and e4.
17 Grg. 5oid-503b, with the term demegoria at 502c 12 and d2.
18 In the Frogs (1009-10), Aristophanes has Euripides say that poets are
admired because they make the citizens better. Aristophanes here con-
victs Euripides with his own words: Euripides subverts traditional moral-
ity, even though, like the sophists, he claims to make the citizens better.
Iris Murdoch points out that "like all puritans Plato hates the theatre"
(in The Fire and the Sun [Oxford: Clarendon Press, 1977], 13).
19 See esp. Smp. 2o6e-2O7a.
20 Smp. 2O9ai-e4. There is no need to emend theios ("divine," 2O9bi),
which occurs in all the manuscripts. This entire section deals with what
A. W. Price calls "educative pederasty" in Love and Friendship in Plato
and Aristotle (Oxford: Clarendon Press, 1989), 27-9. Touton at 2O9a9
refers to hon at 20934, that is, to the entire group of psychically creative
individuals. The problems raised by Price and K. J. Dover (ed., Plato's
Symposium [Cambridge: Cambridge University Press, 1980], 151-2)
about poetry and legislation dissolve when these are taken as the two
main examples of psychic, that is, educational, creativity.
21 M. H. Abrams discusses the distinction between imitative and expres-
sive theories of art in his highly influential book The Mirror and the
Lamp (New York: Oxford University Press, 1953).
22 Rep. 377a-b, 378d-e, 380C1, 387b4.
23 At 396c5-e2, Socrates specifies that the poet must imitate good charac-
ters most of all when they act prudently and less when they err, and that
the poet will imitate unworthy characters only briefly when they do
something worthwhile. Socrates does allow that "for fun" [paidias
charin, 39662) a poet may occasionally imitate someone unworthy.
24 On the purification (katharmos) of a city by the expulsion of a human
scapegoat (pharmakos), see Walter Burkert, Greek Religion, tr. John
Faffan (Cambridge, Mass.: Harvard University Press, 1985), 82-4. In one
text, the scapegoat is "decked in boughs and sacred vestments" before
being chased away. Socrates uses diakathairein and kathairein at Rep.
Cambridge Companions Online © Cambridge University Press, 2006
Plato on poetic creativity 363
39965 and 8. In his far-ranging study "Plato's Pharmacy," in Dissemina-
tion, tr. Barbara Johnson (Chicago: University of Chicago Press 1981),
esp. 134, Jacques Derrida suggests that Socrates is a scapegoat, phar-
makos, as well as pharmakeus ("sorcerer," see esp. Smp. 2O3d8), ex-
pelled from Athens by poison. Homer's banishment may be viewed as an
ironical counterpart of Socrates' execution.
25 See further Bernard Bosanquet, A History of Aesthetic (London: Swan
Sonnenschein, 1892), esp. 49.
26 Rep. 595a5: TO nrjdâ fj jraoa&exeadai auxfjg 6or\ \II\IX\TIKY\.
27 Elizabeth Belfiore, in "A Theory of Imitation in Plato's Republic," Trans-
actions of the American Philological Association 114 (1984): 126-7,
argues for this meaning, which was first proposed by V. Menza in "Po-
etry and the Techne Theory," Ph.D. dissertation Johns Hopkins Univer-
sity, 1972), 161-3. Menza (132, 161-2) reasonably takes the occurrence
of HLfiT]tix6g at 395a2 as an implicit definition of the term. Following
Menza, Belfiore translates [nn-TjTixri as "versatile imitation." G. R. R
Ferrari also adopts this interpretation in his chapter "Plato and Poetry"
in The Cambridge History of Literary Criticism, ed. George A. Kennedy
(Cambridge: Cambridge University Press, 1989), 124-5.
28 Rep. 607C5. At 6o7a5, he calls this poetry the "sweetened Muse"
(f|&uan6vr|v Movaav). At 6ooe4-6, Socrates claims that "all poetic per-
sons (poietikous), starting with Homer, are imitators of semblances of
goodness" and the rest; cf. 6oia4. This has often been interpreted as a
general claim about "poets." But it can be taken only as a claim about
the poets of Greek tradition. Moreover, although poietikos may be used
as a synonym of poietes, "poet," Plato appears to use the term (in the
Republic at least) in the same way as "mimetic," with the special conno-
tation of "given to poetry," that is, "given to imitations"; see esp. 6o7a2,
where Homer is described as "most poetic" (poietikotaton).
29 The three stages are 595C7~598d6, 598d7-6o2bn, and 6o2Ci-6o8b2.
30 As Harold Cherniss, "On Plato's Republic X 597B," American fournal of
Philology 53 (1932): 233—42, has argued, Plato is not here revising his
theory of Forms by making God their creator, but brings in God merely
for the sake of the analogy. Socrates himself qualifies this stratagem by
the tentative "we might say" at 597b6.
31 R. G. Collingwood, "Plato's Philosophy of Art," Mind 34 (1925): esp.
157-9. For a different view, see Alexander Nehamas, "Plato on Imitation
and Poetry in Republic 10," in Plato on Beauty, Wisdom, and the Arts,
ed. Moravcsik and Temko (Totowa, N.J.: Rowman & Littlefield, 1982),
esp. 60-3.
32 In the Sophist, Plato develops his ideas on imitation further by dividing
mimetike into (a) the making of images (eikastike), which consists in
Cambridge Companions Online © Cambridge University Press, 2006
364 THE CAMBRIDGE C O M P A N I O N TO PLATO
preserving the proportions of the "paradigm," and (b) the making of
appearances [phantastike], which changes the proportions of the para-
digm so as to make it appear beautiful (235b-36c). In the Republic,
mimetike does not include (a), which may be taken as a philosophical
descendant of the craft of making an "image" (eikona) of goodness, as
proposed in Book III of the Republic. In the Sophist Plato explores the
connection between poetry and sophistry by casting the sophist in turn
as an imitator of appearances. The poet does not appear explicitly, but
fits the category of the simpleminded imitator who wrongly thinks she
knows; by contrast, the sophist is a dissembling imitator (268a).
33 See esp. Ion 536ei-3, where Ion claims that Homer speaks well about
"everything whatsoever."
34 The Symposium is generally thought to be close in date of composition
to the Republic. It seems to me that the much more elaborate account of
the Forms and of the ascent to knowledge in the Republic is subsequent
to the account in the Symposium, and that Socrates' criticism of Homer
and all traditional poetry in Book X is additional evidence for dating the
Symposium before the completion of the Republic.
35 Julia Annas, "Plato on the Triviality of Literature," in Plato on Beauty,
Wisdom, and the Arts, ed. Moravcsik and Temko (Totowa, N.J.: Row-
man & Littlefield, 1982), 1-28, argues that there is a serious discontinu-
ity in the argument; for Plato regards poetry as trivial in everything
except the very last part of the argument (6o5c-6o8b), where it suddenly
appears as a danger to mankind.
36 See further M. Pabst Battin, "Plato on True and False Poetry," Journal of
Aesthetics and Art Criticism 36 (1977): 163-74.
37 Rep. 6o3ai2-b4: JTOQQW 6' au qpQovr|ae(og ovxi xcp ev rjiiTv 7EQOOO|UA.ET xe
xai exaiQa xai qpiXt] eaxlv era' ot>6evi vyiel oi>6' aA^deT. . . . <$>avh(\ ago.
(pauXa) ovyyiyvo\ieyr\ cpaOXa yevvq.r\ jujir|Tixifj. Cf. 6osa9-b2.
38 Phdr. 265CI and 257a5; cf. 24161-3. See further Elizabeth Asmis,
"Psychagogia in Plato's Phaedrus," Illinois Studies in Classical Philol-
ogy 11 (1986): 153-72.
39 Martha Nussbaum suggests in The Fragility of Goodness (Cambridge:
Cambridge University Press, 1986), 227, that the Phaedrus may be the
first example of the "philosophical poetry" now proposed by Plato.
Cambridge Companions Online © Cambridge University Press, 2006
CONSTANCE C. MEINWALD
12 Good-bye to the Third Man
Plato's commitment to what has been called "self-predication/' that
is, to sentences of the form
Bravery is brave
The Large is large
is one of the most evident and characteristic features of his work.
This commitment figures in dialogues of all three periods, and is so
far from optional as to be at the foundation of Platonism.1 Yet these
sentences immediately produce negative reactions in us.2 The first
one displayed above, for example, seems clearly false. It seems to be
attributing a feature to Bravery that it could not have-we can
hardly imagine it performing deeds of valor or bearing up under
adversity. The characteristic claim that the Large is large without
qualification seems ludicrously to take its subject as another thing
like an elephant, only bigger. The presence of claims of this type in
Plato's text can make us feel that his way of thinking is not merely
foreign to us, but seriously confused as well. Aristotle's rude and
dismissive outburst "So good-bye to the Forms. For they are non-
sense'^ seems about right.
Our background suspicion that the Platonic Forms may not really
be respectable makes the Parmenides especially intriguing to us. For
This paper explores some applications of the main innovation that emerged from the
Parmenides on the interpretation I developed in my book Plato's Parmenides (New
York, 1991). Thanks to Richard Kraut for giving me this opportunity to address read-
ers who may not wish to concern themselves with all the issues treated in this book. I
am extremely grateful to Charles Chastain, Dorothy Grover, Wolfgang-Rainer Mann,
Pamela Meinwald, and Mary a Schechtman for being "test readers" of drafts of this
essay.
365
Cambridge Companions Online © Cambridge University Press, 2006
366 THE CAMBRIDGE COMPANION TO PLATO
that dialogue owes its fame to the presence, in its first part, of an
exchange between a young Socrates and a venerable Parmenides.
The utterances of Socrates are reminiscent of statements that are
widely regarded as constituting Plato's theory of Forms, as presented
in the Republic and the Phaedo. Yet here, when questioned by
Parmenides, Socrates fails repeatedly to uphold his views, and falls
into perplexity. This passage has seemed to generations of readers to
show a sensitivity to the kinds of problems they themselves associ-
ated with Platonism (including but not confined to confusion over
self-predication). Did Plato come to hold the view that Platonism
was misguided, or was he able to react in some positive way?
The natural starting point for answering this question is to study
the rest of the dialogue to see whether it addresses the problems
raised in the first part. Plato himself indicates that this approach is
correct: He makes Parmenides tell Socrates that the reason he has
gotten into trouble is that he has posited his Forms too early, before
having "exercised"; the second part of the dialogue then consists of a
demonstration of the exercise recommended.* Since Plato meant the
second part of the dialogue to bear on the problems of the first, we
must understand the new exercise it contains if we wish to assess
Plato's response to the famous problems.
Yet this approach is not generally taken. The common types of
approach that are taken are two. At one extreme we find a large body
of work seeking to discover Plato's response to selected problems
from the first part of the dialogue by careful study of that part alone. (I
will discuss this approach and give references for it in the next sec-
tion, in the context of considering the function of the passage.) And at
the other extreme we find an approach that jumps out of the
Parmenides entirely, trying to project Plato's reactions to its prob-
lems from what he says in other works. Harold Cherniss is perhaps
the most famous representative of this group.5 But the claims
Cherniss made in connection with what is sometimes called the
Third Bed Argument in the Republic (alternatively the "Third
Couch") have not convinced many philosophers. The conjectures
Cherniss based on the appearance of the language of model and copy
in the Timaeus seem now to be receiving renewed expression.6 But
we should note that this type of reasoning leaves itself open to reserva-
tions of three kinds. First, even if Plato continued to make characteris-
tic claims, this does not by itself show that he had in hand definite and
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 367
adequate solutions for the problems associated with them.? Second,
we can only be confident that Plato went forward with a positive
response to the problems if we see him building the theory of Forms
after he articulated the problems. Thus attempts to reason from pas-
sages in the Timaeus become involved in the controversy (given un-
fortunate animus by the magisterial sarcasms of Owen and Cherniss)8
over the dating of the Timaeus. Third, passages from other dialogues
do not contain evidence that Plato thought they had anything to do
with the problems of the Parmenides^
For these reasons then, it is clearly undesirable to have to rely solely
on projections from other works in forming one's view of Plato's
response to the problems of the Parmenides. This brings us back to
the second part of the dialogue - the natural place to which Plato
himself indicates we should look. There is in fact a very good reason
why scholars have not been applying its lessons to the famous prob-
lems: there has been no agreement about what its lessons are.10 Thus
J. L. Ackrill in 1984 called the second part of the Parmenides "Plato's
most intractable text," and David Bostock in his 1988 book on the
Theaetetus writes, "Now of course one can never tell, with any argu-
ment in the second part of the Parmenides, what Plato himself
thought of it."11 Indeed, the surface strangeness of this text can make
it look unlikely to yield results that could be useful: Many of its
arguments are seemingly so bad as to be embarrassing, and they are
arranged so as systematically to produce apparently contradictory
conclusions. Yet as the longest stretch (almost thirty Stephanus
pages) of uninterrupted argument in the Platonic corpus it ought to be
of considerable philosophical importance. In fact, it does deserve our
attention. Its superficial eccentricity has delayed recognition of what
is really a sound and congenial character. Having approached the
arguments systematically and read them in light of Parmenides'
methodological remarks, I believe that the second part of the Par-
menides is intelligible after all. A positive and crucial innovation - a
distinction between two kinds of predication - emerges, with whose
help we can recognize the exercise to consist of good arguments to
conclusions not contradictory after all.
The recognition of this innovation provides us with new appara-
tus for dealing with the problems of the first part of the dialogue,
now with improved prospects of recovering Plato's response. For the
present essay I have selected two important problems: the notorious
Cambridge Companions Online © Cambridge University Press, 2006
368 THE CAMBRIDGE COMPANION TO PLATO
Third Man and the one Plato himself called the "greatest difficulty."
I will start with an initial treatment of the problems, turn to an
exposition of what I take to be the main innovation of the second
part of the dialogue, and finally apply this innovation to the selected
problems. It will turn out that Plato continues to uphold self-
predication sentences. But he develops a use for them on which their
truth conditions are entirely different from the ones we have been
giving them.12
I. THE PROBLEMS INTRODUCED
It will be useful to start with some general consideration of the first
part of the dialogue and its function. This text (if we break off before
the description of the new exercise, perhaps in I35d3, just after
Parmenides has congratulated Socrates on his efforts) has the appear-
ance of a work complete in itself; in particular it resembles the
canonical Socratic dialogue (that is, it resembles Plato's own early
productions). For there is a general pattern there. Some philosophi-
cally interesting subject comes up. One of the persons present can be
expected to be an expert on this subject. This person enters into
conversation with Socrates about the subject matter of his supposed
expertise, answering a series of questions. By dialogue's end, the
interlocutor has revealed that he is not in a position to uphold his
views: His confusion is such that he has not managed to avoid con-
tradicting himself.
The twist in our dialogue is that Socrates (here a youth) is the
interlocutor, while the venerable Parmenides is the questioner. Soc-
rates presents himself as an expert on Forms by his aggressive criti-
cism of Parmenides' Eleatic comrade, Zeno. For Socrates' criticisms
are made from the standpoint of a view relying crucially on asser-
tions about Forms. Yet these Forms are not things that everyone
knows about, but are rather special theoretical entities. Someone
who uses controversial assertions about special theoretical entities
as a basis for attacks on others ought to be an expert on the relevant
theory, so Socrates ought to understand Forms. But notoriously,
when Parmenides questions him further about his views on Forms,
Socrates falls repeatedly into difficulties, and admits his perplexity.
The resemblance of the first part of the Parmenides to an early
dialogue taken in its entirety perhaps gives it the air of something to
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 369
be studied by itself. And certainly the obscurity of the rest of the
dialogue adds to the attraction of concentrating one's attention on
this more accessible portion of the text. In fact, a great deal of very
careful work has been done to analyze particular arguments from
this passage.J3 My purpose is not to offer detailed discussion of any
such work. But I think that consideration of certain presuppositions
shared by many of these interpretations will help to organize prelimi-
nary discussion of the passage. For this purpose I will start by identi-
fying what I take the characteristic pattern of these interpretations
to be. That pattern combines concentration on the first part of the
dialogue with the beliefs that each argument in the passage should
be treated as a reductio ad absurdum, and that study of these argu-
ments will allow us to determine whether Plato knew what to reject
(and, if he did know, what it was).1*
Despite the attractions of this approach, recognition has been
growing recently that it is inappropriate.1* First of all, one cannot
take this approach without ignoring the explicit indication of the
transitional passage (135C8-137C3) that the exercise forming the sec-
ond part of the dialogue is relevant to handling the problems of the
first part. As I mentioned earlier, Parmenides there tells Socrates
that he won't be able to get things right until he has done a certain
exercise, and is prevailed upon to demonstrate that exercise. The
second part of the dialogue is just this demonstration. We can spell
out the implications of this connection between the two parts of the
dialogue in terms either of our approach to the passage or of our
understanding of Plato's development. In terms of our approach to
the passage, the connection indicates that we should not, after all,
treat the first part of the dialogue in isolation,- despite its resem-
blance to an early dialogue it cannot be regarded as self-contained. In
terms of Plato's development, the connection between the two parts
of the dialogue means that we should not derive our account of
Plato's development from analysis of the first part's problems only;
since he presents the dialectical exercise as relevant to handling
those problems, it is only fair to try to understand that exercise.16
A second consideration involves what one might call the logical
import of the passage. As already noted, we have here, as in many a
Socratic elenchus, a person who tries repeatedly to sustain conversa-
tion on some favorite subject matter but ends each time in admit-
ting that he has contradicted himself. This result shows, at a mini-
Cambridge Companions Online © Cambridge University Press, 2006
37O THE CAMBRIDGE COMPANION TO PLATO
mum, that the person does not have knowledge of the subject in
question - if he did (given Plato's strong conception of knowledge)
he would be able to avoid contradiction. Yet while it is clear, if the
interlocutor admits that he contradicts himself, that something has
gone wrong, this by itself does not tell us what the source of the
problem is. For, as we would put it, any one of the premises the
interlocutor has been willing to accept may have been false, or he
may, starting from true premises, have made some illicit inference
from them.
There is controversy between scholars of the Socratic elenchus
concerning whether, when repeated over time, that procedure might
not have positive results, and concerning what the status of those
results would be.1? However, a difference between our exchange and
those making up Socrates' program of moral inquiry renders the
difficult question of so-called elenctic confirmation irrelevant for
our purposes. This difference is that the program of moral inquiry
consisted in a large number of conversations with a variety of inter-
locutors taking place over a long period, while the present exchange
is clearly a unique occasion, and meant to have its effect as such.
Let us return, then, to consideration of the case in which someone
fails the elenchus on a single occasion. We have noted that the fact
that the interlocutor contradicts himself does not automatically
prove, of any specific premise, that it is false. However, sometimes
we are faced with an argument that has this purpose, and so it is
worthwhile to consider what facts can be used in deciding when to
take an argument in this way. It seems that the paradigmatic reduc-
tio works as follows: It makes explicit all of its premises, so that we
can see that all but one are already known to be true, while that one
is marked out as vulnerable. It then proceeds by explicit and irre-
proachable reasoning to derive an unacceptable conclusion. This
clearly indicates that we should reject the targeted premise. Of
course, many actual examples of reductio arguments are more ca-
sual than the paradigmatic reductio as I have described it. But, while
they may neglect to make all their premises and all the reasoning
from them explicit, their basic strategy requires that they make it
possible for us to identify which premise is supposed to be rejected,
and that they also make possible sufficient identification of what
else is involved so that the target premise can be seen to be the most
vulnerable element.
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 371
If we now consider our passage with this description of the effective
reductio in hand, we can see how far the passage is from matching the
description. (The claims of this paragraph will be confirmed when we
turn to discussion of the individual arguments.) For one thing, often
the very premises that reductio-oriented readers wish to reject are
unexpressed. A premise that does not appear in the text can a fortiori
not be marked as the target of the exercise. And more importantly, the
arguments are underspecified in a strong sense: Not only does the text
often not set out enough premises for the announced conclusion to
follow, but there is just not enough information from which to deter-
mine exactly what we are supposed to understand as completing the
arguments. And different ways of completing the arguments are not
just trivially different. (The variety of formulations of the Third Man
Argument that have been produced by careful interpreters is a sign of
the extent to which that argument is underspecified, while the heat of
their disagreement with each other indicates that the different formu-
lations differ importantly.)
In short, all too often in our passage, the text does not effectively
target a determinate premise for destruction, nor does it give us a
sufficient sense of what else is involved to ensure that some putative
target premise is indeed the most vulnerable element. Yet an author
providing a reductio must do both these things. Plato is so far from
having produced arguments here that follow the reductio strategy
effectively that we must doubt whether he can have intended the
passage to function in that way.
Let us now draw together the two general observations we have
made on Socrates' falling into difficulties. (1) Our final response to the
issues raised here is to be determined by our understanding of the
second part of the dialogue. (2) The arguments that appear in this
passage are extremely underspecified. It seems to me that (1) indicates
that the purpose of the first part of the dialogue is introductory: It
motivates us to work at the difficult developments of the second part
of the dialogue, given that we have some interest in the concerns of
the first part. (One might compare Book I of the Republic, which
clearly has a purpose of this kind.) Given this purpose, (2) no longer
appears to be a weakness in composition, rendering the arguments
strangely ineffective in their task of proving certain claims to be false.
It rather serves to help in the characterization of Socrates. His getting
into trouble on the basis of sketchy arguments indicates something
Cambridge Companions Online © Cambridge University Press, 2006
THE CAMBRIDGE COMPANION TO PLATO
important about his personal level of expertise: that he is inexpert.
This is of course compatible with the harsh evaluation that he holds
determinate beliefs that are provably false. But it is equally compati-
ble with the milder evaluation that he is not yet able to bring to bear
the doctrines or specifications that would allow him to avoid trouble.
According to this milder evaluation, his notions simply require fur-
ther explication and understanding. Since Plato has refrained from
writing the sort of passage that would force on us the harsh evaluation
of Socrates, there is some hope that the milder one is what he in-
tended. And this hope is nourished by the remarks Parmenides makes
at i33b6-9, I35a7-b2, I35b5-C3, I35c8-d3, and I35d8-e4. In these
remarks, Parmenides commends Socrates for his interest in Forms
and his eager impulse toward arguments, says that an able person
could deal even with the "greatest difficulty" that arises for Form
theory, and announces that Forms are necessary if one is not to de-
stroy thought. That is, even Parmenides, the poser of the problems,
endorses Socrates' program.
What this passage shows about Socrates is therefore that he is not
yet an adequate exponent of the theory of Forms. Because of the
unmistakable resemblance of Socrates' views here to those ex-
pressed by the Socrates of the middle dialogues, this passage has
traditionally been regarded as a comment by Plato on the status of
the so-called middle theory, but considerable disagreement has cen-
tered on the content of Plato's comment. My study of this portrait of
Socrates suggests that we ought to regard Plato as telling us that his
middle-period works do not contain a fully and adequately devel-
oped theory of Forms.
Thus, as I see it, his care in being guided by Plato's text has led
Gregory Vlastos rightly to coin his famous phrase that our passage is
a "record of honest perplexity."18 But while Vlastos's concentration
on the first part of the dialogue led him to attribute the perplexity to
Plato as he wrote the Parmenides, I believe the second part of the
dialogue shows Plato himself to hold more adequate views than does
the character Socrates. The immaturity of Socrates (at around
twenty, he is significantly younger than he was in the preceding
works) indicates that the Platonism he offers is itself somewhat
immature.1?
I believe, then, that the overall purpose of the passage we have
been discussing is to prepare interest for the hard work that lies
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 373
ahead by showing that Plato's famous middle-period presentation of
Forms is insufficiently developed. (I do not wish to take a stand on
what Plato's private views were during the middle period; I attribute
insufficient development only to the views as they appear in the
dialogues of that time. I do not know how to decide between the
positions [a] that Plato's own views were not more developed, and
[b] that Plato had adequate views whose exposition he considered
out of place given the subject matter and purpose of the works in
question.) To find out whether further development will involve
rejection of any of the basic tenets, or will simply require handling
them in a more sophisticated way, we must come to understand the
second part of the dialogue. But before doing that, let us look at our
two selected passages so as to get some feelings for the issues that
are in play.
The argument that appears at I3ie8//has had a career quite indepen-
dently of the rest of the Parmenides; it has been a subject of enor-
mous interest to Plato's rivals and admirers alike since antiquity,
when Aristotle popularized the Third Man as a crucial problem for
Platonism.20 Nevertheless Plato's presentation is brief and without
fanfare, bringing foward only the following consideration:
Large things must have some one thing in common [the
Large].
The Large and the other large things now require to have
something new in common, by which all of them will
appear large.
This gives rise not only to a "Third Large," but is supposed to be
reiterated to yield an unending series of Forms,- Socrates regards
such a result as unacceptable.
Clearly, what we have in the text is not enough fully to specify the
argument in question. But we can see that some version of the claim
that the Large is large must play a role here. And as soon as this
becomes apparent, the idea that Socrates' basic views may be viable
might appear quixotic; the claim in quesion here seems simply false.
By the end of this essay we will see that Plato believed with justifica-
tion that the situation was more complicated than that. At this stage
Cambridge Companions Online © Cambridge University Press, 2006
374 T H E CAMBRIDGE COMPANION TO PLATO
Socrates glimpses (rightly) that the self-predication sentence must
express an important truth, but in his immaturity he misinterprets
the sentence, and so gets into trouble. (As we could put it, he acqui-
esces in illicit inference as a result of a mistake in semantics.) It is
crucial for the understanding of Plato's philosophy to come to under-
stand that the form of words guarantees that each such sentence
expresses some truth, and that Plato is the heir of a program that
gave a central explanatory role to sentences of this type.
To take the first of these points first, we must start from the
circumstance that expressions of the form "the Large," "the Beauti-
ful/' "the Just," can be used in Greek to refer to two very different
kinds of things. "The Just," for example, can refer on the one hand
to something that happens to be just (or to whatever does), and on
the other, to what it is about these things that is just. Similarly,
"the Beautiful" could be used of vases or of Helen, but could also
be used to refer to what is beautiful about these things. Abstract
nouns like "justice" and "beauty" come to be used increasingly in
Plato's time as a way of being unambiguous in one's reference to
the second kind of thing; Plato himself uses both forms of words
extensively.
What is relevant for present purposes comes from thinking about
the fact that in describing the second kind of use we employed
phrases like "what is just about just things," "what it is about Helen
that is beautiful."21 In these phrases "just" and "beautiful" are al-
ready being predicated. This suffices to guarantee that
The Just is just [or: Justice is just]
and
The Beautiful is beautiful [or: Beauty is beautiful]
must hold. They do no more than repeat the predications we ac-
cepted within the relative clauses glossing our subject terms. To
interpret these sentences may be nontrivial, but the point is that
even without making up one's mind about how they are to be under-
stood, one can see that they must express some truths. (It is the fact
that a competent speaker of Greek automatically knows that sen-
tences so framed must express truths that accounts for Protagoras's
acceptance of "Justice is just" and "Peity is pious" at Protagoras
33OC2-e2. He certainly does not have a Platonist metaphysics and is
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 375
not accepting these sentences as an expression of his devotion to
Form theory.)
The other background we need to fill in now is the work of the pre-
Socratic natural philosopher Anaxagoras.22 Anaxagoras developed a
theory to account for the phenomena of generation, destruction, and
change in face of the Eleatic proscription against attributing any
change or not-being to what is. (In what follows I will use "change" in
a wider sense to cover generation and destruction as well.) He was one
of a generation of pluralists who explained what we ordinarily think
of as change in terms of rearrangement of elements that are eternal
and unchanging, except in position. Such a strategy involves seeing
what we ordinarily think of as individuals - the trees and horses, the
fish and lakes of our daily world - as composites. For Anaxagoras,
these familiar objects are composed of shares or portions of certain
basic stuffs, such things as the Hot, the Cold, the Bright, the Dark,
Gold, Bark, Wood, Blood, Bone, and so on for a very long list. These
shares stand to the composite objects in a simple relation: We can
think of them as physical ingredients. Thus if the lake becomes warm
we are to think of it as getting an increased share of the Hot; its
continuing wetness is due to its having a long-standing and large
portion of the Wet. Anaxagoras thought of the basic constituents as
having the very qualities they contributed to composites. Thus the
Hot - the totality of heat in the world - was itself hot. This gave a
very easy explanation of how the basic constituents accounted for the
features of the observable world: Since they had the relevant qualities
themselves, so did the shares of them that composites got; that is, the
basic stuffs were in a position to endow participants with portions of
their own qualities. To resume our example, when the lake warms up
it gets an additional share of the Hot; the share is itself hot and this
accounts satisfactorily for the change in the lake.
It will by now have become evident in what way Anaxagoras's
work prefigured Plato's. Both undertake to explain everything about
the derived entities in terms of their getting shares of, that is, partici-
pating in, the fundamental ones. (While Anaxagoras uses the same
language of participation as Plato does, notice that for the pre-
Socratic there is no mystery about what is: It is having a portion of
some stuff as an ingredient or physical constituent of onself.)
Readers may already have recognized (as Plato did before us) that
while this approach works well in some cases, it cannot just be
Cambridge Companions Online © Cambridge University Press, 2006
376 THE CAMBRIDGE COMPANION TO PLATO
generalized straightforwardly. One can hardly claim that when some-
one becomes beautiful this is because of the mere physical addition
of a share of some elemental stuff, the Beautiful, qualitatively identi-
cal with what the Manhattan skyline loses as it becomes overbuilt.
Similarly, something's largeness is not adequately conceived of as a
transferable physical ingredient. While one might respond to this
recognition simply by throwing up one's hands, Plato more optimis-
tically sought to retain the basic explanatory scheme. That is, he
retained the central tenets of the Anaxagorean theory:
Things are X because they participate in the X
The X is itself X
despite recognition that in general (and in particular for many of the
qualities that interested him) an interpretation in terms of material
transfer of portions of ingredients themselves having the properties
in question will not work. What Plato had to do was to find a model
for participation other than getting a physical share of a thing, and
find the interpretation (guaranteed as we have seen by their linguis-
tic form to exist) on which the self-predication sentences are true.
* * * * *
Though the Third Man is often seen as the argument against Socrates,
Plato clearly did not see it as the only serious difficulty, since he made
Parmenides refer to another one as the "greatest" and placed this
greatest difficulty prominently, at the end of the series.^ What we see
here is that Socrates does not have fine enough control of his belief in
the special status of Forms to prevent that belief's committing him
ultimately to the irrelevance of his Forms to the world around us.
The difficulty takes its starting point from the conjunction of the
claim that since Forms are kath' heauta (literally, "by or in relation
to themselves") they cannot be in us, with the observation that
Forms associated with relations have their being in relation to other
Forms, and not in relation to the things around us, while the things
around us are related to other things around us, and not to the
Forms. To take an example more congenial to our sensibilities than
the ones that appear in the text, we are the siblings of each other, not
of the Forms; nor do the Forms have us as their relatives, for they are
relatives only to each other.
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 377
To see how the difficulty develops, we can follow the sample
argument given in the text in terms of Knowledge and its special
branches (not identified more particularly), and see how the argu-
ment would go in the case of Arithmetic. In that case, the claim
about the patterns of relations yields:
Arithmetic knows the Numbers2* (and presumably not
anything around us).
The knowledge of this world knows numerous collections
of objects around us (and presumably not the Numbers).
Now since the knowledge of this world (which we might have) does
not know the Numbers, and since we are obviously not Arithmetic
itself, nor do we have it among or in us (by the claim that the status
of Forms prevents their being in us).
We do not know the Numbers.
A fortiori we are not in a position to apply knowledge of the
Numbers in order to derive our knowledge of numerous collections
of objects around us. Moreover, there is no such application of Arith-
metic to the sensible world. For, by the claim about the pattern of
relations, only we and the things around us, and not Arithmetic, are
related to sensible objects. Given the explicitly made point that the
particular branches of knowledge know the particular Forms, we can
see that the availability of this type of argument will prevent us
from knowing any of the "things that are," and will also prevent any
of the Forms from explaining our world at all.
The stress in setting up this problem on the claim that Forms are
what they are only in relation to Forms, while we are what we are
only in relation to other sensibles, indicates that it will bear on this
problem if we come to see that there is a way in which sensible
particulars have their being in relation to Forms. Socrates now
thinks he cannot posit such a relation without thereby making
Forms degenerate into just more mundane things around us.
We can conclude our initial treatment of the problems with a
summary. Socrates, starting from well-motivated claims, somehow
gets into difficulty, showing that he is unable competently to deal
with the entities he is so eager to introduce. Plato meant this to
indicate the limitations of his earlier presentation of his own theory
Cambridge Companions Online © Cambridge University Press, 2006
378 THE CAMBRIDGE COMPANION TO PLATO
of Forms. But he suggested the second part of the Parmenides would
help with further successful development of the theory, and we have
seen no reason to consider this impossible.
II. PLATO'S INNOVATION
It is now time to turn to the second part of the dialogue. I believe
that Plato so composed that exercise as to lead us to recognize a
distinction between two kinds of predication, marked in the Par-
menides by the phrases "in relation to itself" [pros heauto) and "in
relation to the others" [pros ta alia). These phrases belong to the
kind of use of the Greek preposition pros ("in relation to") in which
a sentence of the form
A is B in relation to [pros) C
indicates that some relation unnamed in the sentence is relevant to
A's being B. In cases of this kind, the context provides information
that allows identification of the relation in question. In the Par-
menides, the in-relation-to qualifications indicate the relations that
ground each of two kinds of predication.^ In this way, they mark a
difference in the way in which B can be predicated of A. Thus, the
difference between what holds of a subject in relation to itself and
what holds of the same subject in relation to the other is not simply
due to the distinction between the others and the subject. It derives
more fundamentally from the fact that a different relation is in-
volved in each kind of case. A predication of a subject in relation to
itself holds in virtue of a relation internal to the subject's own na-
ture, and can so be employed to reveal the structure of that nature. A
predication in relation to the others by contrast concerns its sub-
ject's display of some feature, which Plato takes to be conformable
in general to something other - namely the nature associated with
that feature.
To prepare now for a precise specification of the first of these two
kinds of predication, let us consider the sort of genus-species tree
familiar from the Linnaean classification system. To illustrate the
idea, we might imagine a tree showing the Animals. We can imagine
dividing Animal into Vertebrate and Invertebrate, dividing Verte-
brate in turn into Mammal and so on, and continuing with such
divisions through Feline and Cat, to produce at last such infimae
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 379
species as Persian Cat.26 In the Sophist, Statesman, and Philebus
Plato devotes a great deal of attention to such trees, discussing ex-
plicitly the methodology of constructing them, as well as providing
numerous examples. In such a tree, a kind A appears either directly
below or far below another kind B if what it is to be an A is to be a B
with a certain differentia (or series of differentiae) added.2? That is,
the natures of A's and B's are so related that being a B is part of what
it is to be an A.
In any such case, B can be truly predicated of A (or of the A) in
relation to itself, and so can A, and so can any of the differentiae D.
The idea here is that this kind of predication is grounded in the
structure of the nature in question: A's nature is what it is to be (an)
A - that is, (a) B with . . . with D, and it is in virtue of this that the
predications hold. (I will sometimes use the phrase "tree predica-
tion" for this type.)
It may be helpful at this point to take some examples of true tree
predications. We will get sentences like:
The Just is virtuous.
Triangularity is three-sided.
Dancing moves.
The Just is just.
It is clear that such sentences come out true in Plato's work, as well
as fitting our characterization of predication of a subject in relation
to itself.
The Just is virtuous
holds because of the relationship between the natures associated
with its subject and predicate terms: Being virtuous is part of what it
is to be just. Or we can describe the predication as holding because
Justice is a kind of Virtue. If we assume that to be a triangle is to be a
three-sided plane figure (i.e., that Triangle is the species of the genus
Plane Figure that has the differentia Three-Sided), then
Triangularity is three-sided
holds too. We can also see that
Dancing moves
Cambridge Companions Online © Cambridge University Press, 2006
380 THE CAMBRIDGE COMPANION TO PLATO
is a true tree predication, since Motion figures in the account of
what Dancing is. Finally,
The Just is just
turns out to exemplify the limit case of predication of a subject in
relation to itself: It is uninformative but safe. Thus self-predication
sentences can be used to make true tree predications, though not all
tree predications are of the form: The A is A.18
Predication of a subject in relation to itself takes some explaining
because we ourselves are not in the habit of making such predica-
tions. Predication in relation to the others is much easier to under-
stand, because this is the category into which Plato would put our
own common or garden predications. (Thus I will sometimes call
these "ordinary" or "everyday" predications.). For example:
Aristides is just
Northern Dancer is a horse
The Triangle is intelligible
and in general all sentences which we could describe nontechnically
as concerning the displays of features by individuals count in our
present terminology as predications in relation to the others. This is
because Plato characteristically holds that Justice, Horsehood, and
Intelligibility have crucial roles to play in the states of affairs in
question here: Nothing can be just without Justice having some-
thing to do with it, and so on. Fuller formulations accordingly would
run:
Aristides is just in relation to Justice.
Northern Dancer is a horse in relation to Horsehood.
The Triangle is intelligible in relation to Intelligibility.
These formulations make explicit the bearing of a relation to some-
thing that is in general other than the subject of the sentence, and
this makes apposite the name "in relation to the others." Thus
whenever a sentence:
A is B
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 381
concerns the display of features by an individual, we classify it as a
predication in relation to the others.
The crucial distinction of the Parmenides, which I have just expli-
cated, coincides with the distinction between the kath' hauto and
pros allo uses of "is" in the Sophist. These were described in Mi-
chael Frede's Prddikation und Existenzaussage.2* His contribution
to the present volume (Chapter 13) provides a shorter exposition in
English. Section III summarizes his interpretation of what he there
calls the "first" and "second" uses of "is" in the Sophist. These
correspond respectively to the Parmenides' predication in relation to
itself ("tree" predication) and predication in relation to the others
("ordinary" predication). Perhaps it will be helpful to mention now a
minor point where our accounts differ: Frede's account of the Soph-
ist associates only the second use of "is" with participation, while
the Parmenides seems to me to speak of participation in connection
with both kinds of predication.
While my original purpose was simply to understand the Par-
menides itself, part of the interest of the interpretation that I have
developed lies in its strengthening of the evidence we now have that
Plato made such a distinction and considered it of wide import.
Extended discussion of the Sophist would take us out of the bounds
set for this essay,- nevertheless, it may be useful to make four points
very briefly. First and most obviously, the basic fact that we can
make independent cases from two different dialogues for Plato's use
of this distinction is significant. Second, the circumstance that Plato
uses slightly different language for the distinction in the two texts is
typical of his compositional style, which generally relies on use of
ordinary language in such a way as to make technical points (rather
than introducing and adhering rigidly to special technical terms). I
believe we should regard the occurrence of a unique pair of phrases
as neither necessary nor sufficient for the operation of any one dis-
tinction. Thus we should not feel any worry over the circumstance
that the Academy sometimes in other contexts uses phrases superfi-
cially similar to those of the Sophist to mark the familiar (and differ-
ent) distinction between relational and nonrelational terms.3° Third,
the starring role of the distinction in the Parmenides can explain
why Plato relied on the distinction at Sophist 255C12-13 not only
Cambridge Companions Online © Cambridge University Press, 2006
382 THE CAMBRIDGE COMPANION TO PLATO
without explaining it, but without even drawing our attention to its
importance: there was no need to do these things, because of the
fanfare he had already given the distinction when he introduced it in
the Parmenides. Fourth, if, as now appears, the distinction is intro-
duced as being of great systematic importance in the Parmenides, it
would naturally be ready to be used as needed subsequently, so that
there is no reason to require its later application to be limited to
Sophist 255.31
We can now make some observations about Plato's distinction that
will be useful later. First, we can note briefly that the implicit bear-
ing of one member of the pair "in relation to itself "/"in relation to
the others" on any form of words
A is B
is independent of whether B is a relational predicate. Because our
examples have not yet explored how this works, it may be helpful to
consider a relational case now. For example,
Lady Lufton is kind
can be expanded to show to whom she stands in the kindness rela-
tion, thus:
Lady Lufton is kind to Fanny.
From Plato's point of view, this assertion, telling us as it does about
Lady Lufton's display of a feature, is a predication in relation to the
others; more specifically, it is a predication in relation to Kindness.
Thus we can write most fully:
Lady Lufton is kind to Fanny in relation to Kindness.
The relation to Kindness is of course not the kindness relation but
the relation (unnamed in the sentence) of conforming to, which was
introduced above.
To conclude our exploration of Plato's innovation before going on
to apply it, I will show that certain sentences can be used to make
predications of either of the two types. The force these assertions
have on particular occasions of use, and hence their truth conditions
and sometimes their truth-values as well, depend on the kind of use.
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 383
Sentences whose subject terms name Forms can be used to make
predications of either kind. That is, one can employ the form of
words:
The A is B
on some occasions to make tree predications, and on different occa-
sions to make ordinary ones. The force of the tree predication is:
Being B is part of what it is to be A. By contrast, the ordinary predica-
tion has the force: The Form A displays the feature associated with
the word "B." These need not have the same truth value, and they
always have different truth conditions. This will be easy to see if we
take some examples. The tree predication
Cathood has vertebrae
is true: Having vertebrae is part of what it is to be a cat. But the
ordinary predication that could be made by the same form of words
is ludicrous. That is, as an ordinary predication
Cathood has vertebrae
is plainly false; only sensible animals that inhabit the world around
us display the feature in question.
The difference between the two uses of the sentences is especially
obvious in this case because they have different truth-values. Let us
now consider a case that will show the more subtle difference that
still obtains even when the truth-values are the same.
The Just is not curved
could be used to make a true predication of either kind, but these
would still not be the same assertion; they would not have the same
force. As a tree predication the sentence holds in virtue of the fact that
being curved is not part of what it is to be just. As an ordinary predica-
tion the sentence holds in virtue of the distinct fact that the Form, the
Just, is not a curved object, that is, does not display curvature.
Now that we have seen that a form of words
The A is B
can be used to make predications of each of our two kinds, we will
have an important interpretative choice whenever we come across a
sentence of this form. We will have to be guided by our sense of the
Cambridge Companions Online © Cambridge University Press, 2006
384 THE CAMBRIDGE COMPANION TO PLATO
argumentative context in determining the force on a given occasion
of any such sentence.
III. THE INNOVATION APPLIED
We are now ready to return to the problems of the first part of the
dialogue. But before considering them individually, let us consider
the situation in a more general way. The views of the Socrates of the
first part of the dialogue have always reminded readers of certain
passages from the great middle-period masterpieces, perhaps espe-
cially the Republic and the Phaedo. For convenience, I will call the
position produced by concretizing the suggestions of those passages
in the most simpleminded way " 'Platonism.' " The purpose of the
scare quotes is of course to mark the fact that I question whether
Plato himself had any enduring commitment to this position. There
is, however, no doubt that many people have thought that he did.
Intuitively, the most bizarre feature of 'Tlatonism" was that it
thought of Beauty as the single most beautiful thing, of Largeness as
doing its job by outclassing all other objects in size, and so on. That is,
"Platonism" was supposed to be a view that believed in entities that
managed feats of superinstantiation that ought to be impossible.*2
And it cast these entities in the functional role of properties. Beauty,
for example (the single most beautiful thing), was supposed to be
somehow the common thing among a group of sensible beautiful indi-
viduals. (This is of course why the Forms have a series of names of the
form "Beauty," "Justice," "Largeness," as well as "the Beautiful,"
"the Just," and "the Large.") Thus, as an anachronistic reader might
put it, "Platonism" makes the ridiculous mistake of thinking that
properties do their job by having the very properties they are. The su-
perexemplification theory of Forms seems obviously to be a mistake.
Let us see how our investigation of the vital innovation of the
Parmenides is connected with all this. Clearly, the superexempli-
fication view results naturally in taking sentences of the form
Bravery is brave
to be doing the same kind of thing, or describing the same kind of
state of affairs, as those like
Achilles is brave.
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 385
That is, the superexemplification view assimilates the crucial self-
predication sentences to everyday true predications in relation to
the others.
Earlier, when discussing the situation of the immature Socrates, I
developed the thought that while self-predication sentences may be
trivially false on one reading, Plato had good reason for wanting to
find another reading on which they are true. The distinction between
kinds of predication can now be seen to be the distinction between
these two readings. That is, for example, the ordinary predication
Bravery is brave
may well be false. If being brave is a matter of behaving in a certain
way in fearsome circumstances (or indeed is any condition on people
or their behavior), then Bravery does not seem to be the kind of thing
that could be brave in relation to the others (i.e., that could display
the feature in question). Nevertheless,
Bravery is brave
can still be true when it is made as a tree predication. As we have
seen, a single form of words can change its truth-value, depending on
which kind of predication it is being used to make. Self-predication
sentences will always be true when they are used to make tree
predications. It was the failure of the immature Socrates to recog-
nize that this reading was the one to go for that led him as I put it to
misinterpret his own theory.
Let us now consider the application of Plato's innovation to our
selected problems. It applies straightforwardly to the Third Man
Argument.
Large things must have some one thing in common [the
Large]
is in itself not problematic; Plato can continue to analyze this in
terms of the large things being related to a single Form, the Large. We
noted before that, while the argument is seriously underspecified, it
relies on some version of the crucial claim
The Large is large
Cambridge Companions Online © Cambridge University Press, 2006
386 THE CAMBRIDGE COMPANION TO PLATO
in order to reach the threatening conclusion
The Large and the other large things now require to have
something new in common, by which all of them will
appear large.
Indeed, the production of new Larges depends crucially not just on
the claim being made that the Large itself is large, but on that
claim's being treated in the same way that
Mont Blanc is large
would be. To begin with, the Large itself and the original group of
visible large things are treated as being large in the same way. This
induces the notion that we have a new group of large things whose
display of a common feature must now be analyzed in the same way
the display of the common feature of the original group was. If this is
taken to require the introduction of a new Form, a regress is started.
And the regress will be vicious given the purpose of Forms. Each
Form purports to be the single thing that grounds and explains the
predications it is invoked in connection with and should therefore
not yield to an unending series of further Forms.
But now that we have exercised, we can see immediately that
there are two different predications the single form of words
The Large is large
could be used to make. It is important to Plato to maintain the tree
predication. But we are now clear that that predication does not
claim that the Large itself is large in the same way that the original
groups of large things is. It therefore does not force on us a new group
of large things whose display of a common feature requires us to
crank up our machinery again and produce a new Form.
It may be helpful to consider also the example of Man.
Man is man
and
Man has vertebrae
are ridiculous if we read them as being the same sort of assertion as
Socrates has vertebrae.
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 387
However,
Man has vertebrae
does express a truth that conveys part of the structure of the world,
namely that having vertebrae is part of what it is to be a man. Since
he takes them to express the real structure of the world, it will
always be important to Plato to maintain the tree predications. But
the crucial point is to realize that he has an interpretation of these
important sentences on which they make no claims about the
Forms' exhibiting features. The Parmenides has emerged as showing
conclusively that Plato does not suppose each property to do its job
by having the property that it is. Since his support of the self-
predication sentence does not require him to take Man itself as an
additional member of the group that displays the feature common to
men, and as requiring a new Form to explain the display of this new
group, there will be no regress. Plato's metaphysics can say good-bye
to the Third Man.
The "greatest difficulty" will now appear to be no difficulty at all; it
too admits a straightforward application of our distinction. The easi-
est way into the difficulty will be to return to the particular version
of it I developed using the case of Arithmetic. In this case, the claim
about the pattern of relations yielded
Arithmetic knows the Numbers (and presumably not
anything around us)
and
The knowledge of this world knows numerous collections
of objects around us (and presumably not the Numbers).
Then since this world's knowledge did not know the Numbers,
and since we were obviously not Arithmetic itself, nor could we
have it in us, it followed that
We do not know the Numbers.
A fortiori we could not be in a position to apply knowledge of the
Numbers in order to derive our knowledge of numerous collections
of objects around us. And moreover, by the claim about the pattern
Cambridge Companions Online © Cambridge University Press, 2006
388 THE CAMBRIDGE COMPANION TO PLATO
of relations, there could in principle be no such application of Arith-
metic. For by that claim, only we and the things around us, not
Arithmetic, could be related to sensible objects.
Let us proceed by collecting the responses that we are now able to
make to some of the statements that play a crucial role in generating
this result. As noted above,
Arithmetic knows the Numbers
formerly had always a strange ring; we felt a little unsure what it
meant, perhaps even embarrassed by it, and were accustomed to
peoples' hurrying it by with the thought that it is the sort of thing
Plato presumably did believe. Now we have an interpretation on
which it can be asserted unproblematically, for we can make the
following tree predications without embarrassment:
Arithmetic knows the Numbers^
and
It is not the case that Arithmetic knows the things around us.
Further, we can certainly say
It is not the case that knowing the Numbers can be
predicated of our knowledge in relation to itself
and
It is not the case that knowing the Numbers can be
predicated of us in relation to ourselves.
Plato would regard the first two of these four statements as being true
in virtue of the fact that the correct account of Arithmetic is that it is
knowledge of the Numbers and not anything in terms of things
around us (in this sense Plato is indeed a Platonist). And the last pair
of assertions also holds: We and our knowledge are not the kind of
things that figure in the structures represented by tree predications.
We now come to the crucial point. None of this gives us grounds
for rejecting the ordinary truth
We know the Numbers.
This is of course true as a predication in relation to the others, and
the claims we have just accepted are not at all incompatible with it.
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 389
Thus the crucial inference that the difficulty needs to make at this
point (to: We do not know the Numbers) cannot now be made.
Earlier I pointed out the emphasis in this problem on the claim
that Forms have their being only in relation to other Forms, while
we are what we are only in relation to other sensibles. The fact that
sensible particulars are indeed what they are in relation to the Forms
is thus relevant to this problem. Forms do have their being in rela-
tion to forms - the nature of Siblinghood is not given in terms of us.
And we are of course the siblings of each other and not the Sibling
itself. But let us now consider this claim that we are siblings of each
other more closely. Because it concerns our display of features, Plato
will regard it as a predication in relation to the others. (This is the
basic Platonist move of regarding sensibles as dependent on the
Forms.) That means the most fully specified way of putting it is
We are siblings of each other in relation to the Sibling.
Of course here, the relation to the Form is not the sibling relation.
Thus, it can be true both that we are siblings of each other and not of
the Form, and that we are what we are (namely, siblings) in relation
to the Form. This is ultimately why the Form is not competing with
our blood relatives.
The basic innovation of the second part of the Parmenides has
applied directly to these two problems (and indeed, as I have argued
elsewhere, enables an aspiring Platonist to handle all the problems
of the first part of the dialogue). This gives us good grounds for an
answer to the question concerning Plato's state of mind. He was able
after all to respond to the problems, avoiding difficulty and finding
satisfactory interpretations for the claims characteristic of his pro-
gram. In conclusion, let us step back to get a broader view.
IV. THREE STORIES OF PLATO'S DEVELOPMENT
A very unhappy story of Plato's career has been increasingly out of
fashion in the last half-century. (Though, as is often the case with
old fashions, it continued to be followed by people out of the circle
of the trendsetters.) In this most unhappy story, Plato started writing
with a gracious compliment to his master. The high literary achieve-
ment of the middle-period works coincided with a philosophical
high point: a heady and confident time of glorious dogmatism. Then,
Cambridge Companions Online © Cambridge University Press, 2006
39O THE CAMBRIDGE COMPANION TO PLATO
after a major crisis in which he attacked and actually destroyed the
theory that was his masterpiece, Plato spent his last years in exten-
sive critical activity. His now-failing literary powers produced the
late dialogues as a record of this barren final period.
A second and completely opposite story has been in vogue more
recently. (R. E. Allen, writing in 1965, said of this view that it "[had]
grown increasingly prominent in recent years, and [was] liable to
become still more prominent in future."34 By now the heyday of this
view has passed, and we find Malcolm Schofield in 1990 announcing
that "the paradigm . . . is becoming frayed at the edges . . . and the
search for alternative paradigms is under way.")35 This second kind
of interpretation, which received perhaps its greatest impetus from
Gilbert Ryle, regards the theory of Forms of the middle period as a
hopelessly flawed creation, whose hopelessness was realized by
Plato himself in the Parmenides. He was then in a position to do
some good philosophy in the late period. This story is in a way
happier, but the attribution to Plato of a middle theory that can only
be nonsense is a problem. Moreover, it seems to me unsatisfying in
its separation of the passages expressive of the greatest motivation
and excitement from the program of the dialogues it regards as con-
taining Plato's enduring philosophical contributions. There is some-
thing in Allen's complaint that this is "the portrait of a man who
abandoned a voyage of discovery for essays in county cartography. "3*
Opposite as they are, these stories have something crucial in com-
mon: that the Parmenides records Plato's realization of the unvia-
bility of a determinate theory of Forms contained in the middle dia-
logues, and so ushers in a late period whose program would have to be
entirely different from that of the middle works. My study of the
Parmenides indicates that we need not accept this. We can rather see
our dialogue as indicating where the underdetermined middle-period
view of Forms required further development, and as contributing to-
ward that development a key innovation that should allow unim-
peded natural growth of the theory in the late period. This makes
available a story of Plato's career that could avoid attributing total
nonsense to him at any period, would tell of a fruitful late period
instead of a barren one, and would not cut the final investigations off
from the exciting motivating passages of the middle period.
I offer the following as a third story. In the early dialogues Plato
showed, following Socrates, that people who might have been ex-
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 391
pected to have knowledge on various matters turned out not to. They
were revealed in this condition in a series of elenctic confrontations,
by being unable to sustain discussion on the subjects of their sup-
posed expertise without falling into contradiction. The middle works
then presented Plato's own theories of these matters, including jus-
tice, love, the soul, and rhetoric. These theories were associated with
some metaphysical remarks, and in effect Plato claimed that the fail-
ures of others resulted from their ignorance of metaphysics.
But the extreme brevity of the passages devoted to metaphysics in
the middle dialogues indicates that laying down dogmatically the
tenets of a mature theory is not their main task.*? Thus I regard
these passages as indicating the motivations and outlines of views
that it is not their purpose fully to develop. Indeed, I believe these
passages underdetermine the "theory" to be attributed to their au-
thor. It can come as a surprise on rereading these passages to see how
much more specific are the doctrines on sensibles to which Plato
commits himself than are those on Forms. The language of the mid-
dle dialogues has some tendency to suggest - but is not sufficient to
demonstrate - that Plato was a "Platonist."
I believe that Plato composed the first part of the Parmenides in
order to exhibit where his middle-period description of Forms
needed development. Our view of Plato's ability to sustain that de-
velopment need no longer derive from scrutiny of this passage alone.
Nor need we rely solely on complicated conjectures involving the
evidence of other dialogues. The Parmenides as a whole gives the
best possible evidence for Plato's response to the problems it intro-
duces. If I am right or even on the right track, the dialogue shows
that his response was successful. As the late period began, the theory
of Forms was in new leaf.
NOTES
1 The term " self -predication" is used in slightly different ways by differ-
ent scholars. I use it to pick out sentences of the form I gave, with no
stipulation about their interpretation. We can see examples of Plato's
commitment at Prt. .33002-62 (early), Phd. 100C4-6 (middle), and Sph.
258b9-C3 (late).
2 Negative opinions about self-predication have been frequently expressed;
I here offer a sampling. Sir David Ross pointed out the self-predications at
Cambridge Companions Online © Cambridge University Press, 2006
THE CAMBRIDGE COMPANION TO PLATO
33oc2-e2 in the Protagoras as an example where a "mistake occurs in its
crudest form" (Plato's Theory of Ideas [Oxford, 1951], 88). C. C. W. Taylor
in his commentary on the same passage reports that "the confusion
amounts to a failure to distinguish between being an attribute and having
it" [Plato's Protagoras [Oxford, 1976], 112). And Gregory Vlastos, in his
editor's Introduction to Plato: A Collection of Critical Essays, vol. 1,
Metaphysics and Epistemology (Garden City, N.Y., 1970), tells us that
"had Plato known the Russellian paradox, he would have seen instantly
the absurd consequences of "The F is F' for most values of F" (p. 2).
3 Posterior Analytics S^SL^I-^^.
4 By "the second part of the dialogue" I understand from 13764 to the end.
5 Cherniss discusses whether Plato thought the Third Man Argument
fatal to his theory in Aristotle's Criticism of Plato and the Academy
(Baltimore, 1944), 293-300. The broader context is his treatment of
Aristotle's use of the argument, starting on p. 289.
6 To take two examples from the present volume, in section III of editor
Richard Kraut's Chapter 1 and section VII of Nicholas P. White's Chap-
ter 9.
7 So A. Wedberg, "The Theory of Ideas," in Plato, ed. Vlastos, 1: 44 n. 20.
8 G. E. L. Owen, "The Place of the Timaeus in Plato's Dialogues," Classi-
cal Quarterly 3 (1953): 79-95; H. F. Cherniss, "The Relation of the
Timaeus to Plato's Later Dialogues," American Journal of Philology 78
(1957): 225-66.
9 The single exception to Plato's silence in later dialogues about problems
from the first part of the Parmenides is the reprise of the one-many
problem in the Philebus (14C1//). But this is only one from a long list of
problems, and exactly how and even where the Philebus might solve it
is inexplicit.
10 Of course, individuals have worked on the second part of the Par-
menides over the centuries. We have treatments of the dialogue as a
whole as well as papers on the second part alone. From antiquity a large
portion still survives of the neo-Platonist commentary of Proclus (now
translated into English by Glenn R. Morrow and John M. Dillon as
Proclus' Commentary on Plato's Parmenides [Princeton, 1987]). Contri-
butions made in our century include the Bude edition of Auguste Dies
[Platon: Oeuvres completes, vol. 8 [Paris, 1923]); Max Wundt, Platons
Parmenides (Stuttgart, 1935); Francis M. Cornford, Plato and Par-
menides (London: 1939); Gilbert Ryle, "Plato's Parmenides," Mind 48
(1939): 129-51, 302-25; G. E.L.Owen, "Notes on Ryle's Plato," in Ryle:
A Collection of Critical Essays, ed. Oscar P. Wood and George Pitcher
(Garden City, N.Y., 1970), 341-72; R. E. Allen, Plato's Parmenides (Min-
neapolis, 1983); Mitchell Miller, Jr., Plato's Parmenides (Princeton,
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 393
1986); and Kenneth M. Sayre, Plato's Late Ontology (Princeton, 1983).
But the approaches of these authors, and their results, vary enormously,
and their combined efforts led to no consensus among philosophers and
classicists generally.
11 Ackrill's assessment appears in his memorial piece "Gwilym Ellis Lane
Owen," Proceedings of the British Academy 70 (1984): 493. The Bostock
remark is at Plato's Theaetetus (Oxford, 1988), 214.
12 As Stephen Menn has pointed out to me, it is certainly no novelty to say
that for Plato the Form F is F of itself while particulars are F in some
other way, and to claim that this bears on the Third Man. In this sense
my treatment is of a familiar kind. What makes it fresh is that it goes on,
guided by the text of the Parmenides, to spell out the force of this type of
locution.
13 An enormous vogue for this sort of thing was established by Gregory
Vlastos, "The Third Man Argument in Plato's Parmenides," Philosophi-
cal Review 63 (1954): 319-49. Other notable contributions are those of
Wilfred Sellars, "Vlastos and the Third Man/' Philosophical Review 64
(1955): 405-37; P. T. Geach, "The Third Man Again/' Philosophical
Reivew 65 (1956): 72-82; Colin Strang, "Plato and the Third Man,"
Proceedings of the Aristotelian Society, supp. vol. 37 (1963): 147-64;
Gregory Vlastos, "Plato's Third Man' Argument (Parm. 132A1-B2):
Text and Logic," Philosophical Quarterly 19 (1969): 289-301, which has
a useful bibliography of articles published on its subject; S. Marc Cohen,
"The Logic of the Third Man," Philosophical Review 80 (1971): 448-75;
and Sandra Peterson, "A Reasonable Self-Predication Premise for the
Third Man Argument," Philosophical Review 82 (1973): 451-70. While
the Third Man has received the lion's share of the attention, one author
who has recognized that the "greatest difficulty" deserves sustained
work is Sandra Peterson, in "The Greatest Difficulty for Plato's Theory
of Forms: the Unknowability Argument of Parmenides 133C-134C," Ar-
chiv fur Geschichte der Philosophie 63 (1981): 1-16.
14 Although authors I group together here as following a single pattern all
agree that the arguments they discuss are, so to speak, trying to follow
the reductio strategy, there is important disagreement among them on
the question whether Plato knew what the trouble was, or whether he
remained unable to see this and even unable to formulate his argument
correctly. But I group them together because Gregory Vlastos (the lead-
ing exponent of the latter opinion) takes for granted that we should
approach (an argument from) the passage by making an explicit and
formal reconstruction of the argument, and by diagnosing the trouble.
15 Kenneth Sayre (from whose treatment in Plato's Late Ontology I have
learned the most), R. E. Allen in his Plato's Parmenides, and M. Miller
Cambridge Companions Online © Cambridge University Press, 2006
394 T H E CAMBRIDGE COMPANION TO PLATO
in his book of the same name have recognized this point. But because of
the prestige of the authors who followed the pattern described above and
the inclusion of their papers in influential collections, I suspect that the
pattern is still dominant. Partly for this reason and partly because my
detailed views are different from those of the other authors, I will go on
to offer my own discussion.
16 It has been pointed out to me that those who have worked on the first
part of the dialogue in isolation might not endorse the claim that it can
be fully understood without reference to the second part of the dialogue.
They may have been proceeding on the sensible plan of starting by
getting clear about a manageable chunk of text. But this still seems to
me to leave their work open to the charge that its approach is inappropri-
ate. Moreover, whether or not the authors of these papers believed that
approaching these arguments in isolation was the ideal way to handle
Plato's text, their articles have had the effect of leading many readers to
suppose that it is - since it is the procedure of so much influential work.
17 Terence Irwin's discussion in Plato's Moral Theory (Oxford, 1977) has
played a major role in the controversy. (But perhaps unfortunately for
present purposes, it is scattered throughout the book.) An important
recent pair of contributions is the exchange between Gregory Vlastos
and Richard Kraut in Oxford Studies in Ancient Philosophy 1 (1983):
27-58, 59-7O.
18 Vlastos, "The Third Man Argument in the Parmenides," as reprinted in
Studies in Plato's Metaphysics, ed. R. E. Allen (London, 1965), 254.
19 I27bi-C5 tells us that at the time of their meeting, Parmenides was
about sixty-five, Zeno nearly forty, and Socrates very young.
20 See the commentary of Alexander of Aphrodisias, in Metaphysica
84.21//. I follow the established practice of calling Plato's argument the
Third Man Argument even though that name does not match his formu-
lation. No differences between the Man and the Large are relevant for
our purposes. For an interesting discussion of the difference between the
two for Aristotle and the significance of his formulating the argument in
terms of Man, see Alan Code, "On the Origins of Some Aristotelian
Theses about Predication," in How Things Are: Studies in Predication
and the History of Philosophy, ed. J. Bogen and J. E. McGuire (Dordrecht,
1985), 104-10.
21 I take the content of this paragraph from a seminar of Michael Frede's at
Princeton.
22 On Anaxagoras's physical theory in connection with Plato, see David f.
Furley, The Greek Cosmologists (Cambridge, 1987), 1: 45-8, 61-3, 65-
70, 171-3; J. Brentlinger, "Incomplete Predicates and the Two-World
Theory of the Phaedo," Phronesis 17 (1972): 61-79; and David J. Furley,
Cambridge Companions Online © Cambridge University Press, 2006
Good-bye to the Third Man 395
"Anaxagoras in Response to Parmenides," Canadian Journal of Philoso-
phy, supp. vol. 2 (1976): 61—85.
23 Perhaps a comment on the sense in which this difficulty can be the
greatest is also in order here. It certainly does not strike many now as
harder to deal with than the Third Man. Plato may mean not that this
problem is the hardest to handle, but rather that this difficulty, if not
handled, involves the worst result. (K. Sayre suggests this, Plato's Late
Ontology, 34-6.) For the difficulty, while it initially seems to be con-
cerned only with Forms associated with relations (as Mastership is asso-
ciated with the mastery relation), ultimately leads to the consequence
that no Forms can do their basic job of explaining the sensible world and
grounding our knowledge of it.
24 Perhaps even more than "The Large is large/' this type of claim strikes
us as foreign. On the only interpretation that we find natural, it is
grotesquely false: so much so that puzzlement should arise about why
Plato represents it as acceptable. As with "The Large is large," it will
turn out that Plato distinguishes two interpretations of this sentence:
On one (our natural reading) he can join us in rejecting it, but on the
other we can join him in finding it acceptable.
25 The development of this interpretation of the ciualifications is the cen-
tral chap. 3 of my Plato's Parmenides, to which I refer readers interested
in more than the bare assertions of the present essay.
26 I base this example on the discussion in J. C. B. Gosling's commentary,
Plato: Philebus (Oxford, 1975), 156//. Obviously it sketches the tree in
question only partially. Note that this sketch assumes, as is generally
done, that the trees do not go down as far as sensible individuals or
features unique to them.
27 The"kinds" of late dialogues are Forms. The circumstance that we re-
spond to Plato's late treatment of eide by abandoning the translation
"Forms" and preferring "kinds" or "species" should not mislead us into
thinking that Plato has changed the subject.
28 Sometimes in what follows, I will remind readers of how tree predica-
tions function by speaking somewhat loosely of one nature's being part
of another, and so on. I of course do not mean to exclude the limit case in
which the natures are identical, but it is too cumbersome always to be
mentioning it.
29 Prudikation und Existenzaussage, Hypomnemata 18 (Gottingen, 1967).
Frede's interpretation is often assimilated to the one advanced by G. E.
L. Owen in "Plato," in Plato, ed. Vlastos, 1:223-67; and elsewhere.
While both authors have encouraged this through footnotes expressing
agreement, it is of great importance to recognize that the agreement is
primarily a matter of their both rejecting the view that the Sophist
Cambridge Companions Online © Cambridge University Press, 2006
396 THE CAMBRIDGE COMPANION TO PLATO
distinguishes between existence and a two-place notion. It is a mistake
to assume that all the details of their positive interpretations will agree.
In fact Owen constantly puts his view in terms of a distinction between
identity and predication, which (while he does not explicate it fully)
does not seem to coincide with Frede's distinction. It is telling that
Owen illustrates his distinction in "Aristotle on the Snares of Ontol-
ogy/' in New Essays on Plato and Aristotle, ed. Renford Bambrough
(London, 1965), 71, with the sentences "Arrowby is mayor of Margate"
(identity) and "Arrowby is idle" (predication). In Frede's terms neither of
these sentence illustrates the kath'hauto use of "is."
30 Vlastos regards the latter as "the" distinction associated with this lan-
guage, in "An Ambiguity in the Sophist," in Platonic Studies, id ed.
(Princeton, 1981), 288-90 n. 44.
31 Thus we can answer the objection of Vlastos ("Ambiguity in the Soph-
ist," n. 44) that even if Frede's characterization of the 255 distinction
were correct we would have no right to use it outside 255, even at 256.
32 At least in many cases, Largeness seems problematic. And examples can
be multiplied. For example, this "theory" seems committed to taking
Manyness as the single (!) most multitudinous thing.
33 This is equivalent to: Arithmetic is knowledge of the Numbers. We can
imagine characterizing branches of knowledge according to their ob-
jects, so that we give accounts like: Grammar is knowledge of language.
On this scheme, Arithmetic is the Form of knowledge that deals with
numbers. (Cf. Peterson, "The Greatest Difficulty for Plato's Theory of
Forms," 4-6).
34 R. E. Allen, "Introduction," in Studies in Plato's Metaphysics, ed. R. E.
Allen (London, 1965), ix-x.
35 Malcolm Schofield, "Editor's Notes," Phronesis 35 (1990): 327.
36 Allen, "Introduction," xi.
37 Here I am gratified to find myself in agreement with fowett, whose
assessment seems to me to apply today as well as it did in the nine-
teenth century when he wrote: "Plato's doctrine of ideas has attained an
imaginary clearness and definiteness which is not to be found in his own
writings. The popular account of them is partly derived from one or two
passages in his Dialogues interpreted without regard to their poetical
environment" (The Dialogues of Plato, 4th ed. [Oxford, 1953], 2: 13).
Cambridge Companions Online © Cambridge University Press, 2006
MICHAEL FREDE
13 Plato's Sophist on false
statements1
In Plato's dialogue the Sophist, the main interlocutors, the Eleatic
Stranger and Theaetetus, are trying to determine the nature of the
sophist. Given that the phenomenon of the sophistic movement is so
many-faceted and somewhat amorphous, it is not surprising that the
first attempts in the dialogue to get a grip on the elusive reality under-
lying this phenomenon turn out to be not particularly successful,
since they at best capture some superficial feature of the sophist.
These features are recapitulated at 23ic8-e7.2 Then, in 232a]://, a
renewed attempt is made to capture the sophist; this attempt seems
to go more to the heart of the matter, but runs into difficulties whose
resolution occupies the remainder of the dialogue. The suggestion is
that the sophist has a remarkable ability to represent things in a way
that makes this representation, the sophist's statement about things,
appear and seem to be true, though, in fact, it is not. This raises a
series of difficulties, first alluded to in 235d2, then again in 236C9//,
and spelled out in considerable detail in 2$6d()ff. These problems the
sophist will exploit to the fullest to reject the characterization sug-
gested and thus again to elude capture (cf. 239C9//., 241^3). The diffi-
culties, in brief, amount to this: There are problems about the very
possibility of false statements. For a statement, in order to be a state-
ment at all, has to manage to say something, that is, there has to be
something that gets said by it. But both in ordinary Greek and in the
language of Greek philosophers a false statement is one that says
what is not (or: what is not being).3 Yet what is not being does not
seem to be something that is there to get said. Hence it would seem
that there is nothing that gets said by a false statement. But in this
case it fails to be a statement. So it seems that there can be no false
statements. But, if there is a problem about the possibility of false
397
Cambridge Companions Online © Cambridge University Press, 2006
398 THE CAMBRIDGE COMPANION TO PLATO
statements, there is - a fortiori - also a problem about the possibility
of false beliefs (of false doxa), about something's seeming (dokein) to
somebody, but not being a certain way. And if there is a problem about
the possibility of false beliefs, there also is a question concerning the
possibility of false appearances. Indeed, there is a problem about ap-
pearances as such; given that they are not the real, the true thing
itself, but just an appearance of it, they in some way do lack reality
and, in some special sense, which is difficult to pin down, truth (cf.
236ei-2, 239C9-240C6).
In the Sophist Plato tries to deal with most of these problems. But
he does focus on the central problem of false statements. Here he
primarily tries to show that and how there can be such a thing as a
false statement, and in the course of this he tries to clear up the
confusions that give rise to doubts about the very possibility of false
statements. His view seems to be that these confusions have at least
two sources. First, they rest on a misunderstanding of the negative
particle "not" in "not being"; because of this misunderstanding one
tends to think that what is not, or what is not being, is nothing
whatsoever and hence not something that, for example, could get
said in a statement. Secondly, there is considerable confusion about
what a statement is. Hence one fails to realize that the truth or
falsehood of a statement is a matter of what gets said in the sense of
what gets said about, or predicated of, a subject. Once we realize
this, and once we understand how the expression "not being" is to
be construed, Plato thinks, we also see that it is entirely unprob-
lematic to say that what gets said by a false statement is something
that is not. It is something that is perfectly real, it just happens to be
something that is not (true) in the case of the particular subject in
question that it gets said of.
Given this diagnosis of the problem, Plato proceeds in two stages.
He first (241C7-259C4) tries to show that it is unproblematic to say
that there is something that is not. He then (259C5//) turns to state-
ments to show in which way it is unproblematic to say of a state-
ment that it asserts something that is not. In this way he tries to
accomplish the task he had set himself at the outset when he said
(236e4-237ai): "For how one should put the matter when one says
or thinks that there really are falsehoods and, in uttering this, not to
get involved in contradiction, that, Theaetetus, is altogether diffi-
cult." We should note that the aim Plato sets himself in a way is a
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 399
rather modest one; it is not to solve all difficulties one might want,
or be able, to raise concerning false statements, but to find a coher-
ent way of thinking about them such that, thought of in this way,
they no longer seem to pose a problem.
The discussion of what it is to be something that is not, that is, to
be something that is not being, takes up a much larger space,
though, than the discussion of what it is to be a false statement. At
least in part this is so because Plato thinks that the notion of some-
thing that is, of a being, is no less puzzling than the notion of some-
thing that is not, and that the two puzzles are related. So the prob-
lems about not being, on Plato's view, are not just due to problems
about the proper understanding of the function of "not" in " . . . is
not being/' they are also due to problems about the proper under-
standing of being. What is more, the problems about being stand in
the way of properly understanding what a statement is. Hence when
Plato finally at 24ib6ff actually sets out to refute Parmenides and to
show that there is something that is not, he does so by first calling
into question our understanding of "being" (cf. 243C2-5, 25065//),
which turns out to be as problematic as "not being." He then
(25ia5//) turns to at least a partial resolution of the problems about
being before he returns to the problem of not being. Hence the con-
siderable overall length of the discussion of how we should under-
stand the phrase "not being" or "what is not."
Instead of discussing all this, though, I want in what follows to
focus on the discussion of false statements. Hence I will, only very
briefly, comment on the remarks about being, and, in somewhat
more detail, consider the remarks about what it is to be not being, to
the extent that this seems necessary to understand Plato's resolu-
tion of the difficulty concerning false statements.
I. THE PROBLEM OF BEING
The problem about being, in a nutshell, seems to be this. Suppose
we follow the philosophers in their attempt to determine and iden-
tify what is to count as a being (cf. 242C5-6); and suppose we
decide in the end that we have to recognize as being whatever is in
motion and whatever is at rest, and that these two classes exhaust
what there is (cf. 249cio-d4). There still is a problem about what
being is (249d6//). Though it is true that whatever is is in motion or
Cambridge Companions Online © Cambridge University Press, 2006
4OO THE CAMBRIDGE COMPANION TO PLATO
is at rest, being itself is neither in motion nor at rest. Neither to be
in motion nor to be at rest is what it is to be; and hence what is, as
such, in itself, by itself, is neither in motion nor at rest. It is by
itself just what it is to be. But if it is neither in motion nor at rest,
it does not seem to be a being (25oci-d5).
To see the solution to this problem, we have to see how each thing
can be said to be lots of things, not just what it is by, or in, itself (if it is
the kind of thing that is something by itself), but also other things
that it is not by itself, but by standing in the appropriate relation to
something else. Thus being, of itself, is just whatever it is that it is to
be. But this does not prevent it from being at rest, or from being in
motion, by standing in the appropriate relation to rest, or to motion.
What makes it difficult to see this is that this problem is entangled
with a problem about statements. There is the view of some of Plato's
contemporaries (perhaps, e.g., of Antisthenes) that it does not make
sense to say of something that it is something else, to say of some-
thing that it is something that it is not. It is fine, they seem to argue,
to call a man "man" and what is good "good"; but how can one say of
a man that he is good, if a man is not what is good, but something else
(cf. 25ia5//)?. This involves a misunderstanding of what statements
are. To make a statement is not just a matter of calling a thing by its
own specific name. It is rather a matter, as Plato will point out later
[161&1—6), of naming something so as then to go on to say something
about it. But this failure to understand what statements are, to under-
stand how something can be said to be lots of things and can be called
by many different names (cf. 25^5-6), gets aggravated by, as it in
turn aggravates, a failure to understand being.
The crucial point to understand here is a point Plato makes in a
much-debated passage in 255C12//. The being that we attribute to
things is of two kinds (cf. 255d4~5). Some of the things we say
something is, it is by itself; other things we say something is, it just
is with reference to something else, it is by standing in the appropri-
ate relation to something else. Thus Socrates is or is a being, for
instance, in being white. But white is not something Socrates is by
himself; it is something he only is by being appropriately related to
something else, namely the color white. He only is a being in this
particular way, or respect, namely in being white, by standing in a
certain relation to something else, namely color. He is white, not by
being this feature, but by having this feature. He is white, as we may
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 401
say, by "participation" in something else. The color, on the other
hand, is said to be white, not by participating in, by having, this
feature, but by being it. Similarly the color is a color, not by having
this sort of feature, but by being this sort of feature. Hence it is not
just white, but also a color, by itself. On the other hand, it is differ-
ent from the color pink. And though there is a sense in which it is
different from the color pink by being the color white, this is not the
relevant sense here. It is not part of being the color white not to be
the color pink. So the color white is different from pink by being
appropriately related to something else, namely to difference. And
so, quite generally, the color white is a being in two quite different
ways. It is a being by being whatever it is by itself, for example,
white and a color; it also is a being by being appropriately related to
other things, such as difference, so as to be, for example, different
from pink. Once we understand that being takes these two forms,
we also understand how it is possible that we can tell a thing not
only by its specific name, but by many names, how, for instance, we
can say of the color white not only that it is white, but also that it is
a being, that it is different from pink, and that it is identical with
itself. Moreover, we can explain how being itself can be in motion or
at rest, though of itself it is neither.*
But, what is more, we begin to see how the solution of the prob-
lem concerning being sheds light on the problem concerning not
being. It lies in the very nature of being that whatever is, is many
things that it is not, namely whatever it is with reference to some-
thing else.
This interpretation crucially rests on the assumption that Plato in
255C12-13 distinguishes two uses of " . . . is . . . ."5 Since I want to
assume that Plato in what follows continues to rely on this distinc-
tion, but since this interpretation of255Ci2-i3 has been challenged,
I want to make a few remarks in its defense. It has been attacked, for
instance, by David Bostock in his "Plato on 'is not/ " Oxford Stud-
ies in Ancient Philosophy 2 (1984): 89. To begin with, it should be
noted that the distinction is not supposed to be the distinction of
two senses of the incomplete use of " . . . is . . . , " let alone the dis-
tinction between the "is" of identity and the "ordinary" copulative,
predicative "is." What speaks decisively against this is that Plato
recognizes just one idea of being and that he talks throughout the
dialogue as if this one idea was involved both in saying that not
Cambridge Companions Online © Cambridge University Press, 2006
4O2 THE CAMBRIDGE COMPANION TO PLATO
being is, namely not being (258bn-c4), and in saying that some-
thing is, namely different from something (263b! i). It is one and the
same being (255d5) that we are attributing to something in both
cases. Moreover, the two uses are such that ". . . is . . ." in its first
use could not be replaced by "is identical with" without changing
the meaning. To say that man is a rational animal is not to say that
man is identical with a rational animal. To say that man is a verte-
brate or that white is a color clearly is not to make an identity
statement, though it is a case of saying what man is of himself or
what white is of itself.
It also would be a mistake to think, as Bostock seems to suppose,
that the distinction of the two uses is supposed to be a grammatical
or logical distinction, if by this we mean a distinction that can be
made independently of the metaphysics we rely on. Thus, contrary
to what Bostock (p. 92) seems to think, "Socrates is a man" and
"Socrates is the man in the corner" for Plato clearly are cases, not
of the first, but of the second use of ". . . is . . .," given that Socra-
tes, on Plato's view, is not a human being in himself, but only by
participation in something else, namely the form of a man. But
though Socrates and man, on this view, are two different items of
which the first participates in the second, we are not even tempted
to think that "Socrates is a man" means that Socrates is different
from man, or that Socrates participates in man. Hence we also
should not be tempted to think that "not being is not being" means
that not being is identical with not being. Indeed, it is not the case
that the identity of X and Y constitutes either a necessary or a
sufficient condition for the truth of "X is Y" in the first use of
". . . is . . . . " It does not constitute a necessary condition since
white of itself is a color. And it does not constitute a sufficient
condition, given that, for example, "The same is the same (i.e.,
with itself)" should be a case of the second use of ". . . is . . .," as
"The different is the same (i.e., with itself)" clearly is. This also
allows us to distinguish different kinds of self-predication and to
claim that the kind of self-predication Plato had been interested in
all along, and continues to hold on to, is the one that innocuously
involves the first use of " . . . is . . . . "
With this unfortunately rather brief and sketchy account of the
resolution of the problem about being, let us turn to how Plato tries
to deal with the problem of not being.
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 403
II. THE PROBLEM OF NOT BEING
The resolution of the difficulties concerning what is not begins at
255e8. It clearly falls into four parts:
1. 255e8-257ai2: Plato shows that things, indeed the form of
being itself, can be said to be not being.
2. 257bi-257C3: Plato tries to show that the difficulties con-
cerning what is not have their source in misconstrual of the
word "not" in the phrase "not being."
3. 257C4-258C5: Plato tries to show not only that there are
things that are not, but what the nature of not being is.
4. 258C6-259C4: Plato gives a summary that leads to the discus-
sion of statements.
Let us first consider 255e8-257ai2. In the preceding section Plato
had shown that there are five distinct genera or forms of particular
importance: being, motion, rest, the same, and the different. In
25568// he singles out motion and argues that motion, being differ-
ent from rest, the same, the different, and being, is not rest, is not
the same, is not different,6 and hence also, pah ratione, is not being
(cf. 256cio-dio). For this he relies on the single fact that if X is
different from Y we can say that X is not Y. From this, in 256di 1-12,
he draws the following inference: "Hence, of necessity, not being is
in the case of motion and all other forms." The following lines
(256di2-e4) make it clear how this inference is to be understood.
Given that not only motion, but all other forms (except, of course,
being itself) are different from being, it will be true of all of them
that they are not being. So "something that is not being," far from
being a phrase that cannot possibly be applied to anything correctly,
does at least characterize all forms other than being. We should also
note in passing the language of the conclusion; Plato seems to be
using an expression like "F (or: F-ness) is in the case of a" if it is true
that a is F. Underlying this language there seems to be the notion
that one way for F to be is for there to be some a that is F.
In 256e6-7 Plato draws a further conclusion from the argument
that begins with 25 5e8: "Hence, with reference to each form there is
much which is being, but an immense amount which is not being."
Here we seem to have the same manner of speaking that we noted in
Cambridge Companions Online © Cambridge University Press, 2006
404 THE CAMBRIDGE COMPANION TO PLATO
the preceding conclusion, except that we now not only talk about
something that is, or is being, with reference to a given subject, but
also about items that are not, or are not being, with reference to a
given subject. It seems that this language is to be understood in the
following way: just as F-ness is said to be in the case of a if a is F, so
P-ness is said not to be, to be something that is not, or to be not
being, with reference to a, if a is not F. At least on this assumption
we can see how the conclusion would follow from the preceding
argument. There are lots of things that motion is not, for example all
the other forms. And what is true of motion is, of course, true of all
other forms. So it is true with reference to each form, that there is an
immense number of things that it is not, or is not being. So here is a
second apparently quite problematic way in which even forms can
be said to be not being. It is not only that each particular form is not
the form of being, it also is the case that any other form is not this
particular form and that, in that sense, it is not being with reference
to any form other than it.
Finally, at 257ai//, Plato shows that even the form of being itself
can be said not to be or to be something that is not, namely all the
things that are different from it. From the way Plato talks here it is
also clear that he regards something's not being F as a further way in
which something is not, that is to say as something that is not being.
For he says [2$ja.4—s) "Being, hence, is not in as many ways as there
are other things." If, with this in mind, we return to 255e8//, we
notice that Plato already there, when he argued that motion is not
rest, the same, the different, or being, apparently had taken these as
ways in which motion is not or is not being. For in each case he had
paired off the negative statement about motion with a positive state-
ment that was supposed to show that motion is, as by being differ-
ent, or the same, or being. Thus there is yet a further way in which
things are not, or are not being, namely just by not being this, that,
or other.
Let us summarize the results of this section. There are various
ways in which we can say of something that it is not being that seem
entirely innocuous. It is generally accepted that if X and Y are differ-
ent, we can say "X is not Y." Hence, since all other things are differ-
ent from the form of being, each of them can be said to be not being.
This seems to be unproblematic. But there are more interesting
ways in which something can be said to be not being. For X not to be
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 405
Y is for X in a way not to be, namely Y; at the same time this also is
for Y in a way not to be, namely with reference to, or in the case of,
X. But that X is something that in this way is not, obviously does not
mean that X is nothing. It is something, for example different from
Y Indeed, it would not have this way of being something that is not,
namely Y, unless it was being, namely different from Y. Similarly,
that Y is something that is not in this way obviously does not mean
that Y is nothing. Y, as we have shown, is lots of things, for example
different from X. So there are here two further ways in which some-
thing can be said to be not being: (i) the way an X can be said not to
be insofar as it is not some Y, and (ii) the way a Y can be said not to be
insofar as it is not with reference to some X. Obviously, the second
way is just the converse of the first. In any case, it is clear that there
is an entirely unmysterious way in which there are things that are
not being.
With this we can turn to the next section, 257bi-257C3, in which
Plato tries to explain where people got confused when they came to
think that there is no such thing as what is not. The claim is that
they came to think "what is not" or "not being" must refer to the
contrary of what is. What is, is something of which it is true that it is
in some respect. Instead of realizing that what is not, by contrast, is
something of which it is not true that it is in this respect, they
assumed that it is something of which it is not true that it is in any
respect. But this understanding is not justified by the use of expres-
sions of the form "not X." What is meant by expressions of this form
is not something that somehow is contrary to X.
This much, perhaps, is rather uncontroversial. But interpreters
have had great difficulty with Plato's own positive characterization
of the use of expressions of the form "not X" and in particular his
elucidation of the expression "not being." To understand Plato's
remarks, we have to keep in mind the close connection between
these remarks and the preceding passage. Plato begins with a remark
about "not being" (257b3~4). He says, "Whenever we talk of what is
not being, we do not, it seems, talk of something contrary to being,
but only of something different." The qualification "it seems" indi-
cates that the remark is to be understood in light of what precedes.
But what is meant also becomes clearer when we look at the next
sentence (257b6-7): "When, e.g., we speak of something as 'not big,'
do we then seem to you to indicate by this phrase the small rather
Cambridge Companions Online © Cambridge University Press, 2006
406 THE CAMBRIDGE COMPANION TO PLATO
than the equal?" As is clear from the fact that this is presented as an
elucidation of the preceding sentence (cf. Theaetetus's question in
257b5 to which this is a response), but also from the "e.g." [hoion],
talking of something as not big is supposed to be a case of talking of
something as not being. It seems immediately obvious in which way
it is a case of talking about something as not being: It is a case of
talking about something as not being in a certain way, namely as not
being big. And that this is the correct way to understand the text is
supported by the following considerations. In light of 25 5e8-257ai2,
we should assume that the not being we talk about in 257b3 is either
(i) the not being of something which is not the form of being, or (ii),
more generally, the not being of something that is not something or
other, or (iii), finally, the converse of (ii), the not being of something
that is not with reference to something or other. Now it is quite true
that something that is not being insofar as it is not the form of being,
is not the contrary of being, but just different from it. But what is not
big is not an example of what is not being by not being the form of
being. Nor is the not big a straightforward example of what is not
being with reference to something or other. Plato here does not seem
to think of a case in which it would be false to say of something that
it is not big, but rather of a case in which it would be true to say of
something that it is not big, and he seems to be saying about this
case that to say of something that it is not big is not to say that it is
small. Hence it seems that the not being Plato is talking about here
in 257b3 is the not being of something that fails to be something or
other, for example to be big. And this fits the fact that this was the
kind of not being Plato had last talked of in the preceding passage,
when he had said that even the form of being itself in many ways is
not being, namely in not being all the things that differ from it.
Moreover, it is true that what is not being in that way, for example
by being not big, is not the contrary of what is being in this way, but
simply different. For what is not big is not contrary to what is big,
but merely different from what is big. This is clear because, though
one way of being different from what is big (relative to something) is
being small (relatively), another way of being different from what is
big is to be of equal size. Quite generally, then, "not X" applies to
something that merely differs from what is X.
Now the crucial difficulty is how Plato can assume, as I have just
taken for granted, that the kind of not being that is involved in
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 407
something's not being big, or in something's being not big (Plato
does not seem to distinguish the two) is the same as the kind of not
being that is involved in motion's not being rest, the same, the
different, being, or any other form. Interpreters are agreed that Plato
in 257b3 or in 257b6 shifts to a different kind of case. Up to this
point he has been talking about cases in which something straight-
forwardly is different from something else, more specifically about
cases in which a form is different from some other form. The feature
smallness, for instance, is different from the feature bigness. And
thus smallness, we can say, is not bigness, or, if we grant Plato his
language, the small is not the big, or even smallness is not big. But at
257b3 Plato begins to talk of what is not being, for example about
what is not big, as if this at least covered the case of what fails to be
big, not by being different from the feature bigness, but by failing to
have it. And this seems to be a radically different kind of case. That
Plato moves, without warning, from one kind of case to the other,
might make us suspect that he is just confused. This is what Bostock
is arguing. Less charitably we might think that Plato is cheating. But
most interpreters have tried to be charitable and, moreover, to find
some way to free Plato of the charge of confusion. Yet it is difficult to
see how we can avoid attributing some confusion to Plato.
The problem cannot be solved by claiming that Plato in 257b3//is
just making a point about the use of "not" in expressions like "not
big," quite generally in expressions of the form "not F/f and hence
also in the expression "not being," namely the point that "not" does
not signify contrariety. It is true enough that he does make a point
about the use of "not" in expressions of the form "not F," but it also
seems to be true that, in making this point, he believes he has made
a point about what is not something or other. It is misleading to say,
as Owen does ("Plato on Not-Being," 232, 237, 238), that Plato here
offers an analysis of expressions like "not big" and tries to explain
"not being" in analogy to them. Talking of something as not big is,
at the same time, treated as a case of talking of something as not
being rather than as a mere analogue of it.
The difficulty precisely is that Plato moves from talking about
motion's not being rest to something's not being big, as if the not
being in both cases were the same kind of not being. And this does,
indeed, seem to be his considered view. For when in 257C4// he
moves on to explain what the nature of not being is, he specifies
Cambridge Companions Online © Cambridge University Press, 2006
408 THE CAMBRIDGE COMPANION TO PLATO
(258an//) one nature that supposedly is involved in all cases of not
being we have been considering, that is to say, on anybody's interpre-
tation, both in cases of simple nonidentity and in cases of what we
would regard as ordinary negative predication. It is also clear from
this that Plato cannot mean to solve the problem by distinguishing
two senses of ". . . is not . . ." or "not being." He must assume that
there is one sense of " . . . is n o t . . . " involved both in "motion is not
rest" and in "Theaetetus is not flying." The problem obviously is
how he can assume this.
We make this a hopeless task if we think of Plato as moving from
negations of identity to considering falsehood in predicative state-
ments, as Owen does (p. 237). To begin with, the move in question
here is not the move to falsehood in predicative statements, but
rather the intermediate move to, as it seems to us, negative predica-
tive statements, a move Owen tellingly glosses rather too quickly
(pp. 237-8). But, more importantly, it is a mistake to assume that
Plato thinks of the earlier statements simply as negations of iden-
tity. If he did, the task would be hopeless. He rather must think that
if X and Y are not identical and we thus say that X is not Y, we are
not, in saying this, denying the identity of X and Y, but are attribut-
ing not being to X. And he must equally assume that to say of a
small thing that it is not big is to attribute the very same not being
to it; it is to say of it that it is not, namely big.
Nevertheless it seems that Plato must want to make some distinc-
tion here. For consider the following. The small is different from the
big. Hence the small is not the big. Plato allows himself to move from
this to "The small is not big" (presumably relying on the fact that this
is the denial of "The small is big" in the use of " . . . is . . . " in which
this is true if "The small by itself is big" or "The small by its very
nature is big" is true). So we have both "The small is not big" and
"This (a small thing) is not big." There seems to be a clear difference
between the two statements. The first seems to deny that something
is a certain feature, the second that something has a certain feature.
What means does Plato have to locate this difference?
Even without getting into the details of the subsequent section
257C4//, we can note already here that Plato does not identify not
being with difference, but with a particular form or kind of differ-
ence, with "a part of the different." Hence he cannot assume that
"The small is not the big" or "The small is not big" means, or
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 409
should be analyzed as, "Smallness is different from bigness." I take
it that he thinks it should be analyzed as "The small is different
from what is big." And I also take it that he thinks that "This (a
small thing) is not big" should similarly be analyzed as "This is
different from what is big." This explains why he thinks that there is
just one account for the use of "not," that there is no ambiguity in
"big" or "the big," that there is no ambiguity in "difference" or in
" . . . is not . . . ." But we can also readily see how Plato can, if he
wants to, distinguish the two kinds of cases we want him to distin-
guish. He can do so by distinguishing the two uses of " . . . is . . . " in
". . .is big."
In a third section (257C4//), then, Plato tries to show that what is
not or is not being, far from being nothing at all and unthinkable, far
from being an impossible and illegitimate subject of discourse, con-
stitutes a definite, specifiable kind. At the same time we are sup-
posed to see that, given the way this kind is constituted, what is not
is as real as what is. Take what is not beautiful. It is constituted, to
begin with, by difference. Difference is something that is. But differ-
ence always is difference from something, just as knowledge is
knowledge of something. Thus there is difference from the beauti-
ful, that is, from what is beautiful. It sets off a class of things,
namely all those things that are not beautiful, over against another
class of things, namely all those things that are beautiful. Since
difference is perfectly real and the beautiful is perfectly real, differ-
ence from the beautiful, and hence being not beautiful, is perfectly
real and unproblematic, as real as the beautiful. So in this sense the
not beautiful constitutes an unproblematic class of things, no less
real than the beautiful. Similarly with not being. It involves a differ-
ence, and, more specifically, a difference from what is or is being in a
certain way. There is nothing mysterious about this. And this spe-
cific difference sets off a class of things, namely all those things that
are not in the same way, over against another class of things, namely
all those things that are in this way. To be not being just is to be
something that is set off from what is in a certain way, something
that is different from what is being in this way. In this sense not
being is as real as being and hence a nature of its own.
One may be worried here about the phrase "in a certain way"
which I have introduced into the account. It reflects the fact that
being for Plato always is a matter of being something or other. And
Cambridge Companions Online © Cambridge University Press, 2006
4IO THE CAMBRIDGE COMPANION TO PLATO
correspondingly not being always is a matter of not being in a certain
regard, respect, or way. There is no such thing as unqualified not
being. This seems to be captured by Plato in the summary that
follows, where he talks of not being as the part of the different that is
set over against a particular kind of being [pros to on hekaston,
258ei).7 So to be not being is to be something that is different from
something that is in a certain way, in a certain regard.
What is clear now, as a result of the discussion of this section, is
something that was not clear in the previous section about the use of
"not," and that we, at best, could have gotten out of that section by a
prejudicial interpretation of the words that " 'not' put in front signi-
fies one of the things which are other than the name which follows
it, or rather, which are other than the things which are designated by
the name which is applied to them and which follows the negation"
(257bio-C2). What is clear now is that Plato understands "not be-
ing" and " . . . is n o t . . . " in such a way that it covers the case in
which something fails to be something in that it fails to have a
certain feature. It is clear now that "not being" is intended to cover
the case of what we would call ordinary negative predication. Given
all this, we have no difficulty in understanding what Plato has to say
when we turn to the next section, 258C7//, the recapitulation of the
argument. But this recapitulation also is puzzling and, indeed, must
be confusing, if one has not been able to see how Plato takes the
cases discussed in the first section to be cases of not being precisely
in the sense at issue here, that is to say at issue in statements like
"Theaetetus is not flying." For in ^S9^sff Plato explains again how
unproblematic it is to say of something that it is not or that it is not
being. Take the form of the different; it is different from being;
hence it is not being (259a6-bi). The form of being is different from
all other forms; hence it, too, is not, namely all these other forms
(259bi-5). Similarly each of the other forms is different from the
rest; hence each of the other forms in many ways is not (25^5-6).
Now it is true that this shows that there is a use of " . . . is not" or
" . . . is not being" that is innocuous, but the question is whether this
is going to help us much if we want to understand how statements
can be false. All along, down to the end of the summary, Plato is
relying on the fact that if X and Y are different we can say that X is
not Y. And on the basis of this he has thought that we are justified in
saying that X is not or that X is not being, namely Y. He also has
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 411
relied on the fact that in the case in which Y is the form of being
itself, it will be true immediately of anything other than being that
it is not being.
Now if one believes that the negative statements in the first section
(255e8-257ai2) are non-identity statements, that they have to be
analyzed as stating of some form X that it is different from some
form Y, then this summary must be very confusing. For the sense of
"X is not Y" that we need in order to understand false statements
clearly is not the sense of "X is different from Y." For if it is false that
some particular object a is beautiful, we want to say something to
the effect that "a is beautiful" is false because a, in fact, is not
beautiful. And a's not being beautiful is not a matter of a's being
different from beauty or the beautiful. It would be different from
beauty even if it were beautiful by participation in beauty. So the
account of not being that we need for false statements has to be
more complex than an account according to which to say that X is
not Y is just to say that X is different from Y So, if the summary is
supposed to give us the result that we will need to explain false
statements, it seems that it does not even come near to giving us
what we need, as along as we hold on to the assumption that Plato
regarded the negative statements in the first section as non-identity
statements.
What this, I take it, shows is that it was wrong all along to assume
that Plato takes statements like "The different is not the same" to
be non-identity statements to be understood in the sense of "The
different is different from the same." He takes them to be attributing
not being to the subject in the same sense in which he takes "a is not
beautiful" to attribute not being to a. But even if one sees this, it has
to be granted that it is puzzling that Plato in the summary returns to
the cases of not being that do not seem worrisome and that, in any
case, we are not worried about if we are worried about false state-
ments. Perhaps the explanation is that Plato in the summary wants
to emphasize again how unproblematic it is to talk of something as
not being, and that to talk of something that is not is not to talk of
something that is nothing at all.
For our purposes we may sum up the result of the discussion so far
in the following way. Plato thinks that there is a use of " . . . is
not . . ." in which X can be said to be not Y if X is different from what
is Y, where both the form Y-ness and whatever participates in this
Cambridge Companions Online © Cambridge University Press, 2006
412 THE CAMBRIDGE COMPANION TO PLATO
form count as something that is Y Correspondingly he finds it
unproblematic to claim that X is something that is not, for instance
in not being Y He, correspondingly, introduces (25666-7) a converse
use of " . . . is not. . . ." If X is not Y, Y-ness can be said to be not with
reference to X. And this, correspondingly, is to say that Y-ness is
different from what is with reference to X. And this again is to be
understood in such a way that both difference and difference from
the same count as what is with reference to difference, though in
different ways. This, Plato thinks, suffices to understand the not
being involved in false statements. But to understand the way in
which false statements involve not being we have to have a better
understanding of statements. And so this is the topic he turns to
next.
III. FALSE STATEMENTS
Now the crucial move Plato makes to arrive at a more adequate
understanding of statements is to point out that to make a state-
ment we have to do two things: (1) identify an item we mean to say
something about, and (2) specify something we mean to say about it.
Hence a statement will minimally consist of two parts, a part identi-
fying a subject of discourse and a part by means of which something
gets said about the subject. That a statement manages to single out a
subject is a condition for having a statement in the first place. Hence
its truth or falsehood is a matter of what then gets said about this
subject; that is, the locus of truth or falsehood, as it were, is not the
statement as a whole, but the predicative or stating part of it. To put
the matter differently: To make a true statement is to say something
about something that is true of that something; correspondingly, to
make a false statement is to say something about something that is
false about, that is not true of, that something. So the problem about
a false statement is not that it is about nothing. It is about whatever
the subject expression names as the subject of the statement. Now
the fact that something is true of a given subject, of course, does not
mean that it is true of any subject. It will, at least as a rule, be false of
some subjects. Conversely, what is false of a given subject will not,
at least as a rule, be false of all subjects, but be true of something. So
there is no problem about what gets attributed in a false statement
as such, either. If there is a problem at all, it must be a problem about
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 413
attributing it to a subject it is not true of. But this now does not seem
to be problematic anymore, either. For to say something false about
something, that is, to say something about something that is not
true of it, now just seems to be a matter of saying something that is a
perfectly good thing to say about something, except that it happens
to be different from what is true about this particular subject. So if
we know what it is for a statement to be about something, if we
know what it is to be true of or about something, and if we know
what it is to be different from what is something, we should have no
problem understanding how there can be false statements. Or we
can put the matter thus: Once we have seen that to be what is not is
just a certain way of being different, and if we know what it is to be a
statement and what it is to be what is not, we should have no
problem seeing the possibility of false statements. This, in a nut-
shell, seems to be Plato's solution.
But let us look at the details of Plato's discussion concerning
statements more closely. Central to it is the claim that a statement
minimally has two parts, a name (onoma) and a verb (rhema), as
Plato identifies the two kinds of parts (262C4//). The function of the
name is to name, to refer to, something. But it is only by adding a
verb that we "get somewhere," as Plato puts it rather vaguely
(262d4), that we can be said to say (legein) something (262d5),
whence also the resulting complex expression is called a logos
(262d5-6). Obviously there are all sorts of problems here about the
identification and characterization of the two kinds of expressions
that minimally constitute a statement. Thus, if, as seems likely,
"name" (in the sense of the ancient grammarians) and "verb" here
are supposed to refer to the respective word-classes, we, strictly
speaking, only get a characterization of an irrelevant subclass of
statements, whereas it seems that Plato is aiming at a characteriza-
tion of simple (i.e., nonmolecular) statements quite generally and
really is looking for syntactical categories. The semantic character-
ization of the two kinds of expressions seems inadequate (cf. 262a3-
7), however we interpret their classification, whether by word-class
or by syntactical category. But whatever the difficulties and the prob-
lems may be, this much seems to be right - and noting it seems to
constitute a major advance - that simple statements are constituted
by two parts with radically different functions, one part whose func-
tion is to name, refer to, to identify a subject, and another part by
Cambridge Companions Online © Cambridge University Press, 2006
414 THE CAMBRIDGE COMPANION TO PLATO
means of which we say something, state something, predicate some-
thing of or about the subject.
Having clarified this, Plato turns to two features of statements of
which he obviously supposes that they need to be kept strictly apart:
(i) they are statements about something, and (2) they have, as he
puts it, a certain quality, that is, they are true or false (26264//). A
surprising amount of attention is given to the first feature (cf.
262en-263aio, 263C5-12). It is made clear that these not only are
two different features, but that the first is independent of the second.
In order to have a statement that is true or false we first of all have to
have a part of the statement that manages to specify a subject, and
which subject it does specify, at least in principle, is settled indepen-
dently of what gets said about this subject and, a fortiori, of its truth
or falsehood. Plato quite pointedly lets the Eleatic Stranger settle the
question of reference for the sample statements discussed before he
lets him go on to consider their truth or falsehood.
Some of the relevance of the care and detail with which this is
discussed becomes apparent, if we turn to the historical background.
As we noted above there were some unclarity and confusion about
the object of the verb "to say" (legein). This could lead to the view
that what a statement said was what a statement was about. Thus,
for example, Euthydemus, in the dialogue of his name (Euthd.
283e9-284ai), asks Ctesippus: If one says something false, "does
one do so in saying the thing the statement is about?" And Ctesip-
pus answers in the affirmative. From this Euthydemus infers that
there cannot be such a thing as making a false statement. For the
statement must be about something that is, whereas what gets said
by a false statement is something that is not. To clear up this puzzle
we need to be clear that what a statement is about is something
independent of its being true or false, and that what gets said about
this subject is another thing, and that it is this latter thing that is
true or false.
There is another historical puzzle, or perhaps rather set of puzzles,
that Plato seems to have in mind here. There was the view, which
Aristotle repeatedly attributes to Antisthenes, that there is no such
thing as statements contradicting each other (Met. io24bi6ff; Top.
io4b2o//). In fact, in the Metaphysics Aristotle links this view to the
view that there can be no false statements. Antisthenes' view, ac-
cording to Aristotle's testimony in the Metaphysics (io24b32-4),
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 415
was that each thing has its unique logos or statement, which identi-
fies or spells it out. This has always reminded scholars of the view
expounded at the beginning of the last section of the Theaetetus,
according to which, if a statement can be made about something at
all, it must be its own statement, a statement proper and peculiar to
it (cf., e.g., Tht. 2O2a6-8). And it also reminds one of the view at-
tacked in the Sophist itself (251 as//), according to which each thing
should be addressed only by its own name, and not by the name of
something else, so that we should not call, for example, an object
"white," since "white" is the name of a color and an object is not a
color. But, however this may be, we can see how, if each thing has its
own statement and all statements have to be the statement of some
one thing, contradiction will be impossible. For, of two apparently
contradictory or even just contrary statements only one can be the
statement of the thing in question. In this case the other statement
will fail to be a statement of the thing in question, and thus will be a
true statement about something else or will fail to be a statement of
anything, and hence will also fail to be false.
In the Euthydemus, 285djff, we get a somewhat different version
of this antilogia argument, which here is said to be quite common
and attributed to associates of Protagoras or even earlier dialecti-
cians (286C2-3). Here it is argued that for two persons to contradict
each other they could not be producing the statement of the same
thing, which would be the same statement and hence not yield any
disagreement (286a4~7). Nor would there be any disagreement if
neither of them produced the statement of the thing in question
(286a7-b3). So if there is to be even the appearance of a disagree-
ment, it must be the case that one person is producing the statement
of the thing in question, whereas the other produces a statement
that disagrees, that is in conflict with the first statement. But in that
case the second person must be producing a statement of something
else, in which case there is no contradiction. Or he fails to produce
the statement of anything, in which case he does not manage to say
anything, let alone to contradict the first speaker (286b3-6). In what
follows Socrates takes this, too, to be an argument concerning the
possibility of false statements. It may be noted in passing that this
and similar arguments somewhat gain in plausibility if they are
associated with certain metaphysical views, for example the denial
of nonsubstantial change.
Cambridge Companions Online © Cambridge University Press, 2006
416 THE CAMBRIDGE COMPANION TO PLATO
Now one reason why one may suspect that Plato here, too, is
thinking of a version of this argument is, apart from its relevance
and appropriateness, a striking linguistic detail in the passage we are
considering: Instead of just using the language of "The statement is
about X [peri)," for example, "about you," it also talks of "the state-
ment of X" and "X's statement," e.g., "the statement of you" and
"your statement" (cf. 262e6, ei4; 2650.4, as, a$, cj), sometimes com-
bining both ways of talking (cf. 2632.4, as, a9-io), without in these
cases maintaining a definite order that would allow us to say that
one way of speaking was supposed to elucidate the other (cf. "about
me and mine" in 26$as and "mine and about me" in 263a9~io).
Given this, given that the language of "about" is perfectly clear, and
given that the language in terms of possessive pronouns is neither
ordinary nor natural, it is difficult not to see it in an allusion to the
way of thinking about statements underlying the antilogia argu-
ment and, indeed, to the antilogia argument itself. The point, then,
would be that what a statement is about and what gets said about it
in the statement are two things to be distinguished, which in fact
may be, and normally are, different even if the statement is true.
Thus there may be conflicting statements about the same thing, and
one of them may be true and the other false. Nor does it follow from
the fact that what gets said by the false statement is false, and
something that is not, that what the statement is about is something
that is not, let alone that it is about nothing.
It also becomes clear what we have to say about a view that is
related to Antisthenes' position or to the one reflected in the
Euthydemus, but that in one regard significantly differs from them.
Antisthenes' view was that each thing at best has one statement. At
this point in the Sophist we have already argued that "Each thing
has many statements," to stay with the language of Antisthenes and
the sophists (cf. Aristotle, Met. 102^32//). But somebody might still
want to hold on to the view that there are no conflicting or false
statements, that an apparently conflicting or false statement in real-
ity is a true statement about a different subject. Thus, if Socrates is
healthy, somebody might take the view that this was a statement
about Socrates, but more precisely about a healthy Socrates, and that
the statement "Socrates is ill" was not in conflict with it, let alone
false, since it was about a different Socrates, namely one who was ill.
Note that in 263C7 the Eleatic Stranger assures himself that Theaete-
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 417
tus is going to grant that the statement "Theaetetus is flying" is
about himself who actually is sitting (cf. 26 ̂ ai), rather than about
somebody else. To take such a view, again, among other things, is to
fail to see that it is one thing for a statement to be about a certain
subject and quite another thing for this subject to have something
said about it that is true or false of it; it is a failure to realize that
what the statement is about in theory is settled independently of
what then gets said about it, whether it is true or not.
So, to conclude our review of Plato's discussion of this feature of
statements, Plato is quite willing to grant that a statement cannot be
a statement about nothing, that a statement has to be about some-
thing (263C9-11). But he resists, as relying on some confusion about
statements, any move to argue that false statements as such are
about what is not and hence about nothing. Having a better under-
standing of statements and having a better grasp of what it is to be
something that is not, we now are in a position to understand pre-
cisely in what sense a false statement is a statement that states
something that is not, and this in such a way that we can see that
this is entirely unproblematic.
We can thus, finally, turn to the discussion of the crucial feature,
the quality of statements, their being true or false (cf. 263a! i-d5).
To begin with, though, a brief comment on the term "quality" here
(cf. 26265, e9; 26$aii, e2). The mere language, relying on the famil-
iar contrast between the what it is, or essence, of a something (cf.
26064-5, 263C2) and the what it is like, suggests that a statement is
a statement independently of its being true or its being false. A
statement is a statement by (1) managing to specify a subject and (2)
saying something about this subject. Once these two conditions are
satisfied we have a statement, and the question whether the state-
ment is true or false only arises, Plato, in using this language, is
claiming, once both conditions are satisfied. Moreover, he is claim-
ing that to understand truth and falsehood we have to focus on what
gets said in the sense of what gets said about the subject, that is to
say we have to focus on the predicate.
Plato first turns to the (ex hypothesi) true statement that Theaete-
tus is sitting, but considers it quite generally as a true statement
about Theaetetus. What he seems to aim at is a general characteriza-
tion of true statements. In any case, he characterizes the statement
about Theaetetus in the following way (263b4~5): legei . . . ta onta
Cambridge Companions Online © Cambridge University Press, 2006
4 l 8 THE CAMBRIDGE COMPANION TO PLATO
hos estin pen sou. This, taken by itself, is ambiguous in various
ways. Given the ambiguity of hos, it might mean that a true state-
ment says things that are as they are, or that a true statement says of
things that are that they are. The parallel uses of hos in 26^9,
263CI1, and 263CI2 suggest that the latter is meant. There is a further
ambiguity, depending on whether onta refers to predicates that are
affirmatively true of the subject or whether the claim is supposed to
cover all predications, whether affirmative or negative. Given that
this seems to be intended as a characterization of at least simple true
statements in general, we should assume that onta is to be taken in
the latter sense. There is yet a further question, namely whether peri
sou goes with onta or estin. A comparison with 263b! 1 shows that it
certainly goes with onta, though possibly with both. Given this
disambiguation, the claim seems to be that a true statement says of
what in fact is, namely about, or with reference to, the subject, that
it is. The phrase "what is about, or with reference to, X" obviously is
a bit of quasi-technical language that needs some elucidation. We
readily recognize something like the converse use of ". . . is . . ."
which we discussed earlier in connection with 256e6, a passage
Plato himself seems to be referring to a few lines further down
(263bn-i2) to explain his language here. I say "something like,"
though, because there seems to be a slight difference. If we assume
that Plato is trying to explain here the truth of true statements in
general, and not just the truth of affirmative statements, we have to
assume that he now allows "F" in "F is something that is with
reference to a" to be itself of the form "not G." Plato's view about
true statements, then, seems to be this. Corresponding to the set of
true simple statements about a, there will be a set of F's that are
with reference to a. And a true statement will be one that says of
such an F that is with reference to a that it is, or that it is with
reference to a.
Consider "Theaetetus is sitting." For this to be a statement in the
first place it has to be about something, it has to manage to refer to,
to name something, which then we can go on to say something
about. It does so in referring to Theaetetus. For Theaetetus is some-
thing that is there to be talked about. So there is no problem in this
regard. There also is no problem about sitting. There is such a thing
as sitting. One can give a coherent account of what it is, and (let us
assume) a complete account of the world would be impossible with-
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 419
out some reference to sitting. Moreover it is clear that there is such a
thing as sitting insofar as there are things that are sitting, and so also
in this regard sitting is something that is, namely with reference to
them. So there is no problem, as far as sitting is concerned. The only
question is whether sitting is something that is with reference to
Theaetetus, or - to put the matter the other way around - whether
Theaetetus is sitting. And the claim is that "Theaetetus is sitting" is
true precisely if sitting is something that is with reference to
Theaetetus.
But if all this is clear, we should have no problems with false
statements, either. Take "Theaetetus is flying." There is no problem
about what the statement is about. It is about Theaetetus, and he is
something that is. There is no problem, either, about what gets said
about the subject, that is, about flying. There is such a thing as
flying. One can give a coherent account of it. A complete description
of the world will have to make some reference to it. There surely is
such a thing as flying, insofar as there are plenty of things that are
flying. The only question is whether flying is something that is with
reference to Theaetetus. If it is not, then the claim is false. But this
now is not a problem, either, given what we have said earlier about
not being, or what it is to be not. That flying is not with reference to
Theaetetus just means that flying happens to be different from what
is with reference to Theaetetus, that is, flying is not one of the Fs
that are with reference to Theaetetus. But this does not mean that
flying is nothing at all. We have already seen that it is something
that is, and this in more than one way. Indeed, its not being in the
case of Theaetetus itself is just another regard in which it is some-
thing that is, namely different from whatever is with reference to
Theaetetus.
But, given that this is the point that Plato has been working up to
so carefully from 236c onward, let us consider in detail how he
himself now resolves the question of false statements and how he
deals with this particular example. He turns to the false statement
in 26$by. This is what he says: ho de depseudes hetera ton onton. It
is clear that this sentence is elliptical and has to be understood as
parallel to the corresponding "men" clause about the true statement
"Theaetetus is sitting" in 263b4~5. Thus understood, it will be ren-
dered in the following way: The false statement then speaks of
(legei) things other than those that are. For the claim about true
Cambridge Companions Online © Cambridge University Press, 2006
42O THE CAMBRIDGE COMPANION TO PLATO
statements had been that they say of those things that are that they
are. There is a slight difficulty here: It holds of true statements in
general that they say of those things that are that they are; but when
we are talking about a particular statement, as we are here, it,
strictly speaking, only says of one of the things that are that it is, and
not, as Plato puts it, of the things that are that they are. But this
should not particularly worry us. Plato could have said in 26^3-4.
that the true statement says of something that is that it is (on ti hos
esti). But he wants to get a reference to the whole class of things that
are, relative to a given subject, into the characterization of the true
statement, as this will be needed to get an adequate characterization
of the false statement. This corresponds to the need for a universal
quantifier in a proper characterization, first, of the use of ". . . is
not. . ." along Plato's lines, and then of falsehood, a need several
commentators rightly have insisted on.8 Only thus can Plato say
that the false statement says, speaks of, something other than any of
the things that are, that is, something other than any of the things
that are in relation to the given subject. For it is clear that it will not
do simply to say of a false statement that it speaks of something
other than something that is. To be false it has to speak of something
other than any of the things that are, namely with reference to the
given subject. Moreover, it is clear, given the parallel to 263b4~5,
that 26$bj has to be understood in this sense: The false statement
says of something other than whatever is in relation to a given
subject that it is, namely in relation to that subject. Hence Plato can
move on to claim in 26^9: "Hence it says of what is not that it is."
Obviously, all he does here is (a) to supply the "that it is" that had to
be understood with 263b/ from 263b3~4, and (b) to move from
"other than whatever is;; to "what is not." That latter move is cov-
ered by our earlier explanation of how, for example, what is not
beautiful or big is just what is other than whatever is beautiful or
big. Correspondingly what is other than whatever is being is not
being.
We now have allowed ourselves for the first time in the dialogue
since the problem arose to say that a false statement states, says,
speaks of (legei) what is not. But we also know in what sense this has
to be understood, and why, thus understood, it does not pose any
problems. To say what is not is to say of something other than
whatever is in relation to a given subject that it is. To do away with
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 421
any residual qualms we may have, Plato in 263bn-i2 goes on to
explain the phrase "what is not" as it is used in 263b7. There is a
minor textual problem here that hardly affects what Plato means to
say. The first word in 263bi 1 is given by the manuscripts as ontos. If
we follow the transmitted reading, Plato would be explaining that,
though a false statement says, or talks of, what is not, what it is
talking of is something that really (ontos) is, namely different. To be
more precise, he would be explaining that the false statement says
about the given subject something that really is, namely different.
We would have to understand this in the following sense: The false
statement says about the given subject something that really is,
namely different from whatever is with reference to this subject. But
we could have Plato say something that comes much closer to this,
if, instead of the received text, we followed Cornarius's conjecture,
adopted by all modern editors, and read onton for ontos. We would
get closer still, if we conjectured ton onton.* But whichever text we
adopt, the point Plato is trying to make is clear enough. To say what
is not is to say something that is not altogether nothing, but some-
thing that is; in fact, it can only be called "not being" insofar as it is,
namely different from what is with reference to the given subject.
And at this point (263b! 1-12) Plato reminds us of our earlier finding
that with reference to everything there is much that is and much
that is not. Given the language he must be referring back to 256e6-
7. But this has constituted a major problem for scholars that we need
to look at in some detail.
Plato's thought here in 263b seems to be the following. Take the
false statement that Theaetetus is flying. It says of Theaetetus some-
thing that is not with regard to him, namely flying. It presents, talks
of, something that is not, namely flying, as if it were in relation to
Theaetetus. This is supposed to be unproblematic, because we have
already seen earlier that in relation to anything there are lots of
things that are not. And so flying is just one of those things that are
not with reference to Theaetetus. And this is what makes the state-
ment that Theaetetus is flying false; it presents flying as something
that is with reference to Theaetetus, when, in fact, it is not, when, in
fact, flying is different from whatever is with reference to Theaete-
tus (cf. 263di-2), or, to put the matter yet differently, when, in fact,
Theaetetus is not flying.
But if we look back at the claim in 256e6-7, to which Plato seems
Cambridge Companions Online © Cambridge University Press, 2006
422 THE CAMBRIDGE COMPANION TO PLATO
to be referring in 263bn- i2 , it was arrived at by considering cases
in which X and Y are different, in which we hence can say that X is
not Y, and, moreover, hence can say that Y is not being with refer-
ence to X. And, given that for any X there are lots of things X is
different from and that hence X is not, we were able to say that with
reference to everything there are lots of things that are not being
with reference to it. But this, it is thought, does not help here, since,
though it is true enough that Theaetetus and flying are not the same
thing, that Theaetetus is different from flying and hence that
Theaetetus is not flying, this is not the sense we need to explain why
it is false to say that Theaetetus is flying. For even if it were true that
Theaetetus is flying, it still would be the case that Theaetetus and
flying are not the same thing and that hence in that sense Theaete-
tus is not flying. So this cannot be what explains the falsehood of
"Theaetetus is flying."
Given their understanding of 25 6e6-7 and of the preceding sec-
tion, given in particular their assumption that in this section state-
ments of the form "X is not Y" express non-identity statements,
commentators also are rightly puzzled why Plato here in 263b does
not refer to the later part of the discussion of not being, where - on
anybody's interpretation - Plato tries to come to terms with what
we would regard as ordinary negative predication (cf. McDowell,
"Falsehood and Not-being," 122//; cf. Bostock, "Plato on 'Is not/ "
i n ) . It should be clear by now that Plato's reference back to 256e6-
7 only makes sense, because Plato all along did not understand the
statements of the form "X is not Y" in that section as non-identity
statements, not as statements of difference, but as statements of not
being, that is to say as statements of a particular way of being differ-
ent. And Plato even there must have thought that this way of being
different was exhibited not only in those cases in which two forms
are different from each other, but also in those cases in which some-
thing fails to have a certain feature. So already 25666-7 has to be
understood as involving the use of " . . . is n o t . . . " that Plato needs
to rely on in 263b. What the subsequent discussion, 25-jhiff, added
was an understanding of this use that allowed one to see that it also
covered the case of negative predication. It took the subsequent
discussion to determine the nature of the very not being whose
being we had ascertained by 256e6-7. This is what allows Plato in
263b to refer back to 25666-7, rather than to, say, 258b. So, though
Cambridge Companions Online © Cambridge University Press, 2006
The Sophist on false statements 423
there are details of the text that require further clarification, and
though it seems that even with further clarification we would wish
Plato to have been clearer and more precise on certain points, the
outline of his argument and his general position are reasonably clear
and do not seem to be vitiated by confusion.
IV. CONCLUSION
In fact one thing that is striking about the Sophist, in comparison to
the earlier dialogues, is its "dogmatic" and systematic character. It
sets out carefully constructing a series of puzzles, aporiai. In this
respect its first half resembles the early dialogues or even its immedi-
ate predecessor, the Theaetetus. But then it turns toward a resolu-
tion of these aporiai. In this regard the procedure of the dialogue
reminds one of the methodological principle Aristotle sometimes
refers to and follows, the principle that on a given subject matter we
first of all have to see clearly the aporiai involved before we can
proceed to an adequate account of the matter, which proves its ade-
quacy in part by its ability both to account for and to resolve the
aporiai (cf. De An. I, 2, 4O3b2o-2i; Met. Bi, 995a27//). And the
Sophist proceeds to resolve these difficulties in a very systematic
and almost technical way. By careful analysis it tries to isolate and to
settle an issue definitively. In this regard it does stand out among all
of Plato's dialogues. And because of this it also is more readily acces-
sible to interpretation. If, nevertheless, we do have difficulties with
this text, it is in good part because in his day Plato was dealing with
almost entirely unexplored issues for whose discussion even the
most rudimentary concepts were missing. Seen in this light, Plato's
solution of the difficulty presented by false statements is a singular
achievement.
NOTES
I am particularly indebted to the following papers: G. E. L. Owen, "Plato
on Not-Being/7 in Plato, edited by Gregory Vlastos (Garden City, N.Y.,
1970), 1: 223-67; John McDowell, "Falsehood and Not-being in Plato's
Sophist," in Language and Logos, ed. Malcolm Schofield and Martha
Nussbaum (Cambridge, 1983), 123; David Bostock, "Plato on Is not/ "
Oxford Studies in Ancient Philosophy 2 (1984): 89-119.
Cambridge Companions Online © Cambridge University Press, 2006
424 THE CAMBRIDGE COMPANION TO PLATO
2 I quote the text in Dies's edition in the Bude series; its line-numbers at
times differ from those of Burnet's text.
3 In Greek: me on; in what follows, I will render this and ouk on indiscrimi-
nately by "what is not" or "what is not being" or even just "not being."
4 For a different interpretation of the problem, see Jean Roberts, "The Prob-
lem about Being in the Sophist/' History of Philosophy Quarterly 3
(1988): 229-43.
5 I developed and argued for this interpretation at great length in Prddika-
tion und Existenzaussage, Hypomnemata 18 (Gottingen, 1967), 12-36. It
was then, following a suggestion by R. Albritton, adopted by Owen in
"Plato on Not-being"; Owen had originally taken a different view of the
passage (cf. Journal of Hellenic Studies 20 (1957): 107 n. 25). Cf. Owen's
first footnote to the reprint of "Plato on Not-being" in Owen, Logic,
Science and Dialectic, ed. Martha Nussbaum (Ithaca, N.Y., 1986), 104.
6 There are technical reasons why Plato here moves back and forth be-
tween "different" and "the different." As I noted earlier, there is, accord-
ing to Plato, a (self-predicational) way of being different, and generally of
being F, such that difference or the different is different in this way (and
generally F-ness or the F is F)} thus not to be different in this way is not to
be the different, just as being different in this way is to be the different.
7 Translators tend to construe hekaston with morion; wrongly, as the pre-
ceding part of the sentence shows. Difference is distributed among all
things that are with reference to each other, i.e., it is distributed among
things that are not beautiful with reference to things that are beautiful,
among things that are not big with reference to the things that are big,
etc.; the pros to on hekaston clearly picks up the pros allela.
8 Cf. David Wiggins, "Sentence Meaning, Negation, and Plato's Problem of
Not-being," in Plato, ed. Gregory Vlastos (Garden City, N.Y., 1970), 1:
299; McDowell, "Falsehood and Not-being," 123; Bostock, "Plato on Is
not/ "113.
9 There are yet other possibilities, e.g., ontos de ge onta hetera ton onton
peri sou.
Cambridge Companions Online © Cambridge University Press, 2006
DOROTHEA FREDE
14 Disintegration and restoration:
Pleasure and pain in Plato's
Philebus
Each of us will be trying to prove some condition or state of the soul
to be the one that can render life happy for all human beings. - You
that it is pleasure, we that it is knowledge.
[Phil, i id)
Although the main topic of the Philebus, the rivalry between plea-
sure and knowledge as candidates for the dignity of the highest good
in human life, is a familiar one from the early Socratic dialogues on,
for the wider congregation of Plato's admirers the Philebus to this
very day remains largely terra incognita. It is regarded as one of the
late and difficult dialogues, an area for the specialist who has mas-
tered the intricacies of the late Platonic doctrine that we find more
alluded to than explained in the Parmenides, the Theaetetus, and
the Sophist. What frightens the student of Plato's ethics off the
territory is most of all the long "dialectico-metaphysical preface" of
the Philebus (i4c-3ib). For the first quarter of the dialogue is filled
with a rather complex discussion of dialectical procedure, dealing
with "the one and many," and with a new kind of ontological classifi-
cation that is, at least at first sight, more bewildering than enlighten-
ing and may exhaust the reader's patience before he has even pene-
trated to the lengthy discussion of different sorts of pleasures that
starts at 31b and fills most of the rest of the dialogue.
The specialist, on the other hand, usually confines herself to the
first part of the Philebus, which contains the discussion of the "dia-
lectical procedure" and the suggested new ontological classification;
I am grateful to Richard Kraut for his help, which far exceeded an editor's concerns.
His penetrating questions forced me to clarify the argument at various crucial points,
and his correction of my English saved me from many awkward phrases.
425
Cambridge Companions Online © Cambridge University Press, 2006
426 THE CAMBRIDGE COMPANION TO PLATO
she is not amused by the subsequent long-winded discussion of plea-
sure and pain and of their shortcomings, since it seems not to con-
tribute anything to our understanding of the metaphysical first part.
The suspicion has sometimes been entertained that, Plato's claims
to the contrary notwithstanding, the dialectical-metaphysical part is
only loosely connected with the discussion of hedonism vs. intellec-
tualism and that it may have been used as a preface because Plato
saw the need to put his thoughts on paper somewhere, without
caring too much about its integration into the text.1
Our situation with respect to the dialogue is therefore rather para-
doxical: The "general" reader cannot and the specialist will not deal
with it as a whole. As a consequence the Philebus has remained
largely excluded from the discussion of Plato's philosophy, even
though it is acknowledged that it may contribute significantly to our
understanding of the possible revisions in the philosopher's later
thinking. This omission is all the more paradoxical, since at least a
bird's-eye view of the dialogue (one that avoids the nitty-gritty of the
detailed analysis) will show that Plato himself never loses sight of
his aim. The arbitration between knowledge and pleasure is the
express motive throughout, not only in the long discussion of differ-
ent kinds of pleasure and knowledge and in their final ranking (31b-
67b), but also in the "metaphysical preface." An overview of the
dialogue as a whole and of this metaphysical preface in particular
will therefore have to serve as an introduction to the central topic of
this essay, the nature of pleasure and pain. As will gradually emerge,
the long preface is in fact quite indispensable for a proper understand-
ing of the determination of the nature of pleasure and pain, as well as
of their evaluation on the scale of goods.
For economy's sake this essay will be mostly expository, and omit
a critical discussion of Plato's position on the nature of pleasure, its
drawbacks and advantages. It will also have to refrain from an en-
gagement with many of the important textual and philosophical
problems raised in the extensive secondary literature.2
I. THE STRUCTURE AND CONTENT OF THE
DIALOGUE AS A WHOLE
The set-up of the discussion is quite straightforward. We find Socra-
tes engaged in the defense of the superiority of knowledge against a
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 427
group of uncompromising hedonists who regard only a life of plea-
sure as worth living (na-i2b). At the outset Philebus, the lazy
beauty of the day, hands over the discussion to his follower and
admirer Protarchus, who remains Socrates' partner for the rest of the
dialogue. Through a comparison of the nature and benefits of plea-
sure and knowledge respectively (3ib-66d), Socrates finally wins his
case and convinces Protarchus. For it turns out that in the final
ranking of goods only certain kinds of pleasures, the true and pure
ones, are accepted as good at all, and even they end on the fifth and
last place on the scale of goods, clearly outstripped by reason and
knowledge (65a-67b).
This conversion of the hedonist proceeds in stages. At first Socrates
has to insist, against his opponent's resistance, on the need for a
differentiation between the various kinds of pleasures and knowledge
that can only be achieved through an orderly treatment of their unity
and plurality. This claim allows Socrates to elaborate on the appropri-
ate method of dialectic. He enjoins that any science has to start out
with a proper "collection" of the genus; it must then provide a com-
plete and numerically exact division into subgenera and species, be-
fore it can finally admit the unlimited multitude of individual in-
stantiations. The procedure thus incorporates both limit [peras) and
the unlimited (apeiron): without such a methodological procedure no
satisfactory knowledge in any discipline can be reached.̂ If the spe-
cialist's appetite is whetted by this introduction to the method of
dialectic, which is nowhere as elaborate as in this passage of the
Philebus, it is soon frustrated, however. For Plato lets Socrates dis-
miss the dialectical method ("the gift of the gods to mankind," 16c)
almost as soon as he has introduced it, because it turns out not to be
necessary for the pursuit of their particular topic. Instead, a "sudden
memory" has belatedly come to Socrates that allows him to dispense
with the dialectical treatment of all kinds of pleasure and knowledge
respectively (20b). According to Socrates, neither pleasure nor knowl-
edge taken by themselves can fill the bill of what makes a human life
good by procuring our happiness, since the good must be "perfect,
sufficient and worthy of choice." Neither pleasure nor knowledge
passes this test. For nobody could want a life of pleasure without any
kind of reason, nor a life of reason without a morsel of pleasure. Only a
mixed life that contains both pleasure and reason is sufficient in that
respect and must therefore be superior to the life of the hedonists as
Cambridge Companions Online © Cambridge University Press, 2006
428 THE CAMBRIDGE COMPANION TO PLATO
well as that of the "intellectualists." So the partner's task now turns
on the question whether either pleasure or reason is more closely
related to what is responsible for the goodness of the "combined life,"
and therefore deserves at least the second prize on the scale of goods
(2id-23b).
This investigation of whether pleasure or knowledge is more akin
to the good forces Socrates to start with a new distinction or division
of "all there is now in the universe" into four kinds (23C-27C): the
unlimited, limit, their mixture, and the cause of the mixture (23c-d).
This new ontology is exemplified by a few cases that make clear what
he has in mind. In nature there are countless things that are "unlim-
ited" in the sense that they in themselves have no definite degree (as
we would say). These are entities that shift on a continuous scale,
such as what is hotter and colder, faster and slower, and so forth; they
possess no definite nature until a definite and stable "limit" or mea-
sure is imposed on them. Such a limit is attained, according to Plato,
when they reach a stable or harmonious state, which therefore repre-
sents a mixture of the unlimited and the limit. As instances of such
stable mixtures he refers to good weather, health, strength, and
beauty (24b, 25c, 26b). It seems that for Plato not just any degree in
the continuum of opposites will do, but that only harmonious and
stable combinations deserve the title of a "mixture of limit and the
unlimited." For only such mixtures display the appropriate measure.
The fourth class that is added is that of the causes of the generation of
such "measured" (exnmetron) mixtures (26e).
This ontological distinction at first sight seems to relate directly
to the dialectical method. For it presupposes, as Socrates asserts,
partly new devices, but partly the "same ones as they used before"
(23b). But if the specialist in Platonic metaphysics now hopes that
the ontological division of "all there is" with the help of the con-
cepts of limit, the unlimited, their mixture, and the cause of such
mixtures will shed further light on the nature of Plato's late dialectic
and metaphysics, this hope will at best be partially fulfilled. The
"same devices," unfortunately, seem to be the same only in name.
For in the dialectical method "limit" had referred to the unity of the
genus and to the number of its subgenera and of the species con-
tained in it, while the "unlimited" comprised the innumerably
many different instances. Now, by contrast, the "unlimited" are
continua of opposite qualities, the hotter and colder, while "limit" is
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 429
the proportion that is necessary to turn the continua into harmo-
nious mixtures of these opposites. Why does Plato even pretend that
the function of "limit" and "unlimited" in the two passages is the
same (23b-c)? This question has not only considerably hampered
the interpretation of the Philebus itself/ but also vexed scholars,
who see in this passage the missing link between Plato's late dialec-
tic and metaphysics, as we have it in the dialogues, and the report on
Plato in Aristotle's Metaphysics, in which "limit" and "unlimited"
are crucial concepts.* No attempt will be made here to settle the
issue how, precisely, these two passages are related to each other,
because this would require an extensive discussion of the text and of
the ample literature; it would prevent progress on our own topic. We
will rather have to focus on the question of what the two passages
contribute to what follows in the dialogue itself, to the clarification
of the nature of pleasure and knowledge.
What made at least a brief summary of this fourfold division neces-
sary here is the fact that Plato actually uses it in the subsequent
discussion, and that he comes back to it explicitly in his final settle-
ment of the arbitration between pleasure and knowledge. As to its
immediate purpose, the fourfold division allows Socrates to deter-
mine the genera that contain pleasure and reason: Pleasure belongs
to the unlimited (the "more and less"), while reason-as Socrates
establishes by an elaborate argument that depicts human reason as a
descendant of the cosmic divine reason - belongs to the fourth class
of "all there is," to the class of the causes of successful mixtures
(27e-3ib).
The subsumption of pleasure under the class of unlimited, the
more and less, turns out to be the first step toward its "degradation."
It allows Socrates to urge the view that pleasures by themselves are
in need of moderation and harmonization. If they are good, their
goodness does not come from their own generic nature but is condi-
tional on something besides pleasure itself. "So we have to search
for something besides its unlimited character that would bestow on
pleasure a share of the good" (28a). What that condition is will gradu-
ally become apparent once different kinds of pleasures and pains
have been defined and their origin explained: Pleasures arise in liv-
ing beings that are in a state of harmony, the mixed class. When such
a harmonious state is disrupted, the creature experiences pain,- when
it is restored, pleasure ensues. So the goodness of the different kinds
Cambridge Companions Online © Cambridge University Press, 2006
43O THE CAMBRIDGE COMPANION TO PLATO
of pleasures clearly depends on the kinds of restorations they are;
and this is indeed the topic that Socrates pursues in the long middle
part of the dialogue, which is designed to sort out "true and false"
pleasures (3ib-55c). We will have to turn to Socrates' critique of
hedonism in detail later to see what justifies the claim that pleasure
and pain can be "true or false."
Evenhandedness in the comparison of pleasure and knowledge in
their kinship to the good demands also a critique of reason and
knowledge. It gets rather short shrift (55C-59b), but Socrates man-
ages to come up with a constructive critique: Some of the intellec-
tual disciplines are "truer," that is, purer, than others; pure and
exact sciences are better than applied sciences, and dialectic is the
purest of all. This critique of both pleasure and knowledge allows
Socrates to return to his original set of questions: What is the happy
mixed life, what is responsible for its goodness, and how are pleasure
and reason related to what makes the good life good (59b)? The final
answer that crowns Protarchus's conversion from hedonism to a
mitigated "intellectualism" brings no surprises: Only a mixed life
that compromises both pleasure and knowledge can be regarded as
the happy life. But while this life must comprise all intellectual
disciplines, even the less pure and exact ones, because they are neces-
sary for a successful life under earthly circumstances, only a very
restricted set of pleasures is acceptable for the good life. Besides the
true and pure pleasures Socrates includes "the pleasures of health
and of temperance and all those that commit themselves to virtue"
(63c), that is, those that lead to health and harmony of soul and body.
In the final ranking of goods Plato seems to be even more restric-
tive. A comparison of reason and pleasure with the threefold good,
the combination of truth, beauty, and proportion, yields the follow-
ing scale (66a-c): First come measure and proportion (limit), second
the harmonious mixtures (the limited), third rank goes to reason and
intelligence (its cause), fourth place to "the soul's own properties, to
knowledge, the arts and true judgment" (the limited in a less pure
condition), and fifth and last place is assigned to true and pure plea-
sures (the unlimited sanctioned by limitation).6
The justification of this final ranking of goods should not exercise
us here. It is important to note, however, that Plato has completed
his project with the help of the fourfold division of "all things that
are now in the universe." If one wonders why reason does not obtain
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 431
first place (especially in view of the repeated exaltation of the divine
reason [12012s] throughout the dialogue), it should be kept in mind
that this scale of goods is tailored toward the explanation of the good
human life. Plato is here not talking about the good as such, but
about the factors responsible for making our life good. This is not to
say that the human good can be or is treated in the Philebus in
isolation from the overall cosmic good; the ranking on the scale of
human goods leaves open, however, what relationship and ranking
Plato would envisage for proportion and the divine reason them-
selves. The focus on the human good also explains the separation of
reason and intelligence from the other kinds of human intellectual
capacities, as well as the admission of right judgment (doxa) and
pure pleasures.7 To describe the rationale for the final scale more
concretely: First comes the ingredient (proportion) that makes the
good life good, second comes the harmonious life itself, third comes
the cause in us that allows us to aim for proportion and harmony,
fourth the sciences and arts we have to acquire for the successful
guidance of a "measured" life, and finally come the pure pleasures
that enrich the imperfect human life without detriment to truth,
proportion, and beauty. With this final ranking Socrates has fulfilled
the promise he made at the beginning of the discussion, to arbitrate
in the contest between pleasure and intelligence. As this brief over-
view has shown, Plato has never lost sight of his own project, but
carried out as promised and made use of the dialectical and ontologi-
cal distinctions where appropriate.
Before we can take a closer look at the details of the discussion of
the nature of pleasure and pain, the different kinds of disintegrations
and restorations themselves, and the critique that emerges in this
discussion, we have to raise a problem that affects our evaluation of
the dialogue as a whole. Why does Plato here return to a question
that he had left largely untouched after the Republic, the question of
the good life and the rivalry between reason and pleasure? And why
does he see the necessity to go into a long-winded ontological expla-
nation of the nature of pleasure again, after he seems to have come
to a satisfactory conclusion about it in Book IX of the Republic2.
Connected with this last question is also our irritation with the
form of the dialogue. What is disturbing is the fact that we seem to
have a normal Socratic discussion in front of us, in what seems to be
one of Plato's latest works.8 Prima facie the Philebus does not seem
Cambridge Companions Online © Cambridge University Press, 2006
432- THE CAMBRIDGE COMPANION TO PLATO
to display any of the dramatic features of the other late dialogues.
Socrates himself is guiding the discussion. There is no man from
Elea, as there is in the Parmenides, the Sophist, and the Statesman.
Nor is there a Pythagorean taking over from Socrates, as there is in
the Timaeus. And we do not find ourselves removed from Athens to
Crete, under the guidance of an unnamed Stranger from Athens, as
we do in the Laws. Nor do we find any such elaborate introduction
as presented in some of the middle or earlier late dialogues, such as
the Symposium, the Parmenides, and the Theaetetus. We are instead
back in Athens, plunged right into the middle of a lively dispute
between a Socrates and interlocutors not much different from the
ones we are used to in the "Socratic" dialogues.
There are, to be sure, some differences, apart form doctrinal inno-
vations. Socrates does not plead ignorance anywhere, elenchus is
confined to the first part where Protarchus confesses his reduction
to speechlessness (2id), and the discussion ends with positive re-
sults. But the Philebus shares such traits, after all, with other middle
dialogues, like the Phaedo and the Republic. It also shares with
them the liveliness of the discussion, a concern for drama not found
in the other late works. Protarchus, Socrates' main interlocutor, is
not at all reduced to the ayes and nays of the respondents in the
second part of the Parmenides, the Sophist, and the Statesman; he
plays an active role to the very end of the discussion. There is true
antagonism, wit - including sexual innuendoes - and a gradual con-
version.* If one looks at the dramatic form, it is almost as if Plato
wanted to prove to the world that even in his ripe old age he was
quite capable of composing lively dialogues, and also that Socrates
had by no means been forgotten!
Since such speculations on Plato's personal motives for his rever-
sion to the form of his Socratic dialogues must remain idle, a more
promising explanation would seem to lie in the topic, the discussion
of the nature of pleasure as a competitor with reason for the honor of
the highest good. Plato would have found it difficult to summon
either an Eleatic or a Pythagorean as the leader of such a discussion
and there were good reasons to let Socrates take up the challenge.
Plato, it seems, reverted to the earlier form of Socratic dialogues
because he wanted to reopen a question that had been a major con-
cern of the earlier Platonic Socrates, the debate on the nature and
evaluation of pleasure. As a closer analysis of the long discussion of
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 433
pleasure in the Philebus will show, there is indeed an abundance of
doubtlessly intentional allusions to these earlier discussions of plea-
sure, as we find them in the Protagoras, the Gorgias, the Phaedo,
and the Republic. The Philebus, then, reverts to the earlier form of
Socratic dialogues because Plato regarded the question of the con-
cept of pleasure and its role as a good in human life as unfinished
business. This will remain the working hypothesis for the rest of
this essay. But what, precisely, prompted this reversion, especially in
view of the fact that some of Socrates7 claims in the middle part of
the Philebus seem mere repetitions of what is contained in those
earlier works? It remains to be seen whether they are in fact repeti-
tions and not rather improvements on these earlier attempts. To see
what improvements may have seemed necessary we have to take a
look at Plato's earlier treatment of pleasure. At the same time, this
will help us to clarify why Plato subjects pleasure to such protracted
scrutiny in the Philebus, and thereby also shed some light on the
need for the dialectical preface.10
11. REVIEW OF PLATO'S EARLIER THEORIES OF
PLEASURE
The first extensive treatment of hedonism is found in the Prota-
goras, and by general consent this treatment is at the same time the
most puzzling one (35ib-358a).11 For Socrates proposes without
comment the hedonistic position that pleasure is "the good" and
confines himself to an "intellectualization" of hedonism, with the
simple argument that even the hedonist needs reasoning because
there must be an "art of measuring" pleasures and pains when a
decision is necessary (356d). Socrates argues that persons who claim
that they have been "overwhelmed by passion" when they give in to
the temptation of an imminent pleasure at the cost of a later pain are
wrong in their account. What really happens is that they "mismea-
sure" the pleasures: They deem the present pleasure (e.g., of eating a
lavish dinner) as much larger than the later ensuing pain (whether
stomachache or anguish over weight gain). So there is no such thing
as "reason being overcome by passion." Reason remains the master
throughout; it merely makes a mistake: It falls prey to an optical
illusion concerning the relative size of the present pleasure and the
later pain.
Cambridge Companions Online © Cambridge University Press, 2006
434 T H E CAMBRIDGE COMPANION TO PLATO
Plato very likely never accepted such an intellectual hedonism for
himself. He could have used any other standard of the good and
adduced the same measuring argument to show that reason must
always be the supreme arbiter. It may be one of the ironical points
aimed at in the Protagoras that the sophist, who nobly refuses to
accept unmitigated hedonism, does not manage to extract himself
from Socrates' argument. The very fact that the definition of virtue
as the art of measuring pleasure and pain is not exploited any further
and that the dialogue ends in perplexity should warn us not to take it
as Plato's own definition or as his intended solution. But whatever
may be the point of the argument in the Protagoras, Plato cannot
have been satisfied with it as a sufficient critique of hedonism that
even here reason remains supreme. Nevertheless, that goods of any
kind must (among other things) be measurable and therefore open to
rational control is a point Plato kept in mind over years, as is clear
from the fact that the question of measuring pleasure reappears in
the Philebus. The possibility of mismeasurement turns up as consti-
tuting one of the many "falsities of pleasures" that Socrates dis-
cusses at great length. Plato, it seems, never forgot the "art of mea-
suring pleasure and pain" introduced in the Protagoras.
The Gorgias takes quite a different road of attack. Callicles is an
unmitigated hedonist (as Protagoras was not) who professes not to be
open to Socratic intimidation (cf. his great speech in 482C-486C). He
will not be cajoled into admitting any difference among pleasures
that would import anything like the need for the rule of reason over
them. He wants them all! Even the threat of moral turpitude is at
first not effective, nor is the demonstration that he will be con-
demned to the Sisyphean labors of an eternal hunt for satisfaction
(493a-b). But Socrates catches Callicles in the end. Such unseemly
pleasures as that of scratching when itching, especially when one
thinks of a catamite's degenerate pleasures, or the pleasure of the
coward who runs away in battle, are too much for this aristocrat. His
defeat is sealed when he concedes that there must be better and
worse pleasures and thus a rational "master art of life" that discrimi-
nates between them (501a).
Interestingly, no such appeal to moral taste is made in the
Philebus at all, although the example of the pleasure of scratching is
taken up again and supplemented with even more vivid descriptions
("leaping, kicking, color changes, distortion of features, shouting
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 435
around like a madman," 45d-47a). The critique in the Philebus,
however, does not exploit our aesthetic revulsion or moral scruples,-
there is no ad hominem argument here. Instead, such pleasures are
judged from a scientific, almost medical point of view: They are
really mixtures of pleasure and pain, and constitute unwholesome
excitements.12 Once again we see that Plato has not forgotten his
earlier treatment of pleasure; he echoes his old examples but trans-
forms the doctrine.
In no dialogue does Socrates show himself as much of an antihe-
donist as he does in the Phaedo (64d-69e). Pleasure here seems to be
entirely confined to the body. It is depicted as one of the encum-
brances and disturbances that the body imposes on the soul as long
as the soul is confined to that prison. The philosopher is gladly
preparing for death because this means release from the shackles of
the body and from the distractions of the mundane pleasures, which
allow at best for vulgar prudence in the hunt for more pleasures
(68e). Although it is not impossible that Plato meant to leave room
for separate pleasures of the mind, since he calls the philosopher the
"lover of wisdom" (erastes, 66c), nothing is made here of any higher
pleasures for the philosopher.
The Republic represents quite a change of mind by introducing a
systematic differentiation among pleasures, in the sense that there
are true and pure and good pleasures of the rational part of the soul,
and at least relatively good ones in the other two parts, provided that
they follow the rule of reason and do not submit to tyrannical and
licentious pleasures (58od-588a). The idea of assigning different
kinds of pleasures to each of the three parts of the soul may well have
been an afterthought of Plato's, for it is not introduced until Book IX,
where he adds it as one more point in his proof of the enormous
superiority of the life of the paragon of justice, the philosopher.13 Free
from all tyrannical fetters and fears, the philosopher enjoys not only
superior pleasures, but also more pleasures than all others (in fact he
lives 729 times more pleasantly than the tyrant! 587e).
The passage constitutes quite an advance on Plato's side because
it permits him to justify a systematic differentiation among plea-
sures and to formulate general criteria for their evaluation. Pleasures
are called (a) "motions of the soul" (kinesis, 583c) and (b) "fillings of
a lack" (plerosis, 585b). These two accounts are not in rivalry, as
witnessed by the fact that Plato moves freely from the one to the
Cambridge Companions Online © Cambridge University Press, 2006
436 THE CAMBRIDGE COMPANION TO PLATO
other, but they are obviously supposed to complement each other.
What Plato tries to capture by the two terms seems to be the follow-
ing. The conception of pleasure as a "motion" allows him to distin-
guish between the kinds of pleasures that are mere liberations from
pain ("motions from below") and the true pleasures that proceed
from a neutral or middle state "upward" to sheer pleasure (584b-e).
The conception of "filling" or "fulfillment" focuses on the content
of the pleasures and justifies the discrimination among things the
person is filled with: The pleasure of learning is a filling with true,
reliable, and good things, while other fillings have neither true nor
clear nor fully real objects (585a-586b).
The combination of the two accounts permits Plato to demonstrate
the great superiority of the philosopher's pleasures over everyone
else's. They involve no liberation from pain but are motions toward
the true "above"; they are fillings with real and true being. So the
account serves the purpose it is supposed to serve in Republic IX. It is,
however, an account that relies on a metaphorical use of the crucial
terms ("filling," "moving truly upward") and on a mixture of the
metaphors that does not bear too close a scrutiny. It remains unclear,
for instance, whether the liberations from pain are pleasures at all, or
whether (as the epithet "illusory" suggests) they are only semblances
of pleasures.1* This would work for the motion metaphor with its
distinction between three levels (the above, middle, and below) and
do away with the mistake of treating the middle state of rest as a
pleasure (583c-584a); but it would have the consequence that libera-
tions from pain are not pleasures at all. The filling metaphor does not
allow such a distinction of three levels, since there is no middle state
discernible in this case, nor does it explain why such fillings are
actually "mixed with pain" and do not merely have pain as their
precondition. Plato could of course try to claim that assuagements of
pain, as in hunger and thirst, are not real fillings, but he clearly does
not want to go this far. Instead, he only draws the conclusion that
such pleasures are fillings with what is less real, true, and pure, and
winds himself through a long series of constructions that exploit the
comparative (585b-e) so as to avoid a clear pronouncement about the
exact status of such pleasures. So he ends up with the contrast be-
tween one genuine and two bastard pleasures for the three parts of the
soul (587c), and leaves the reader with the puzzle of how, precisely,
this bastard state is to be interpreted.
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 437
A total discrediting of all but a few pure true and unmixed plea-
sures would not really serve Plato's purpose in the Republic, which
is to demonstrate the superiority of the philosopher's pleasures over
all others. The denial of the existence of real pleasures for the lower
two parts of the soul would be rather an overkill; not only would it
be a self-defeating denial of the phenomena, it would make nonsense
of the calculation of the superiority of the philosopher's pleasures
(587d-e); one can compare and multiply only what exists and is, in
principle, comparable. So Plato must have regarded the "bastardiza-
tion" of the common pleasures as a proper compromise. But he
could not have regarded this as a final and satisfactory treatment of
the concepts of pleasure and pain themselves because it clearly
leaves us with the uncertainty whether liberations from pains are
pleasures or not. Whether Plato did regard it as satisfactory when he
wrote the Republic and reversed himself only much later must re-
main a moot point. He may well have realized then and there that
more needed to be said about the problem.
The Philebus, I propose, brings order into this diffuse picture pre-
sented by Republic IX and takes up suggestions from earlier dialogues
as well, and this task explains the reappearance of Socrates and the
return to the Socratic method of inquiry.1* The whole problematic is
an eminently Socratic concern. But the Philebus does not only dis-
cuss pleasure in a more orderly fashion. It does so on the basis of a new
ontology of pleasure and pain. And therefore Plato prefaces the re-
sumption of his old problem with the exposition of the dialectic
method and the new division of all beings. Although the dialectical
procedure is not officially pursued after its introduction, the new
systematic procedure does in the end lead to a satisfactory definition
of pleasure and pain and to equally satisfactory criteria for a distinc-
tion between different kinds of pleasures and knowledge.
III. PLEASURE AND PAIN IN THE PHILEBUS
This much should be uncontroversial: The dialectic approach, as
sketched in 16c-17a, would demand a definition of pleasure as a
unitary phenomenon, a complete subdivision of the different kinds
of pleasures, and it would also have to include an explanation of how
many species of pleasures there are, and how they differ. The same
procedure would have to be applied to pleasure's rival, knowledge.
Cambridge Companions Online © Cambridge University Press, 2006
438 THE CAMBRIDGE COMPANION TO PLATO
That Socrates, thanks to a sudden "inspiration/' conies across an
easier way to decide the question whether pleasure or reason is the
highest good, should increase rather than settle our curiosity about
why he can dispense with the dialectical method after he had as-
serted so emphatically that without this work "you will count for
nothing and amount to nothing" (i7e). "Divine assistance" allows
him to omit this strenuous enterprise, but this explanation must
leave us less than satisfied.
The student of the divisions (dihairesis) in the Sophist (219a-
232a; 264b-268d) and in the Statesman (258b-268d; 274e-end) will
probably regard this shortcut in the Philebus with unmitigated re-
lief, in view of those long and still far-from-complete (!) exercises
devoted to the divisions of all arts. But it is not just the shudder
provoked by the thought of the ordeal of a complete division of all
pleasures and all kinds of knowledge that justifies our gratitude for
Socrates' parsimony in the Philebus. There are good reasons to
doubt that a completion of dialectical divisions, ending with two
Porphyrean "Christmas trees" of all pleasures on one side and all
intellectual disciplines on the other, would be of any real use. What
could Socrates' next step be? A comparison of the two trees, branch
by branch, to see which one is better, or a quantitative analysis as to
which of the two has a larger number of better branches? But what
could be the criteria for such a piecemeal comparison or an overall
reckoning?16
If a complete division would be unmanageable and of no use any-
way, we are back with our old problem, to explain the purpose of the
introduction of dialectic in the first place. Did Plato have a negative
end in view, namely to prevent the possible misunderstanding that
the long discussion of pleasures and pains and of the different kinds of
knowledge respectively are meant to represent a proper dihairesis2.
There may, of course, be something to this suggestion. A better expla-
nation emerges, however, if we take note of the fact that Socrates does
make use of the method of dialectic at least in part, even though he
does not come up with a neat tree of divisions, nor with a numerically
exact account of the species of pleasures or knowledge that could
claim completeness of dihairesis.
If Socrates suggests that his new ontological distinction uses at
least "in part" the "arms" provided by the dialectical method be-
cause he makes use of the terms "limit" and "unlimited," although
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 439
in a quite different sense, this is rather a (consciously) misleading
depiction (23c). It is not the equivocal use of these two terms that
justifies the claim that he uses "some of the old armament/' but the
fact that he actually follows some of the prescriptions of dialectic! In
establishing the "four kinds of things that exist now in the uni-
verse/7 he proceeds quite in agreement with the dialectical method,
as far as he goes. For he establishes the unity of the genus with care
(25a: "for whatever is dispersed and split up into a multitude we
must try to work out its unifying nature as far as we can see, if you
remember"). Collection and a certain amount of division (even if not
divisions into all subgenera and ultimate species) are the tasks Socra-
tes pursues in his discussion of the four kinds of beings (25d5; 26d).
The immediate advantage of this metaphysical approach is obvi-
ous: at least pleasure can now find its unmistakable place on the
ontological map. It belongs to one of the four types of onta ("be-
ings"). We are told what its genus is, what constitutes the unity of
the genus, and also how pleasure differs from other kinds of apeira.
All unlimited entities consist in a continued flux (24d), so they are
motions. These motions terminate when order is imposed; this im-
position of a limit leads to the establishment of a definite being, the
successful mixture (genesis eis ousian, 26d). Pleasures are motions
of "filling" that are terminated by such an imposed order. At first
sight the achievement of this preliminary determination of pleasure
as a kind of boundless motion or filling may seem small. But in fact
it is quite a step forward. All perceived restorative processes are now
real pleasures,- there can no longer be any concern with semblances
of pleasures or bastard pleasures. Furthermore, things in motion are
now included among the onta, they are no longer of such dubious
status that one cannot even refer to them in any definite way, as we
find it, for instance, in the Theaetetus, where objects in motion
cannot be grasped by any firm notion at all and cannot even be called
"somethings."1?
If the introduction of a special genus for things in flux represents a
change of mind on Plato's side, so that change and changeable things
can now be included among the things that are, and can be subjects of
knowledge, this change was, of course, prepared by the inclusion of
motion on the list of the Forms in Parmenides (i29d-e) and on the list
of the "most important kinds" in the Sophist (254c-d). The Philebus
is, then, not unique in giving generation and change a definite place
Cambridge Companions Online © Cambridge University Press, 2006
44O THE CAMBRIDGE COMPANION TO PLATO
among the things that are. But only in the Philebus does Plato give a
specific explanation of what constitutes the unity of their genus, how
they differ from the other genera that there are - that is, what consti-
tutes the unity of the genus of the unlimited and the more and less,
and what particular subgenus of all apeira it is that contains pleasure
and pain. For this division allows Socrates to proceed to a more pre-
cise definition of pleasure and pain. "When the natural combination
of limit and unlimitedness that forms a live organism is destroyed,
this destruction is pain, while the return towards its own nature, this
general restoration, is pleasure" (32b).
This definition, as we will see, is sufficient to explain the nature
of all pleasures of both soul and body, and it forms the basis of the
critique of pleasure that is to follow. Further reflection on this defini-
tion will show why pleasure cannot be the good as such but can at
best be a conditional good. Properly understood, all pleasures presup-
pose a disintegration, a disturbance of the harmonious integrity (the
good mixture). Welcome as such a restoration must be, freedom
from such disruptions would clearly be a much better option. This is
indeed the conclusion Plato draws immediately: there is a third
condition, besides pleasure and pain, and "perhaps there would be
nothing absurd if this life turns out to be the most godlike" (33b).18
The third state of integrity replaces the state of rest that Plato in
the Republic had placed "in between" the "truly upward" motion of
"genuine" pleasure and the "bastard" pleasure of liberation from
pain. This third state is here no longer a second-best, nor is it placed
between pleasure and pain,- it is now the end point of all restorations
and therefore above both pleasure and pain. It turns out to be the best
of all states to attain. If it is not here declared to be the winner in the
competition for the good in human life, that is because it is not
possible for us to remain permanently in a state of harmony. "We
necessarily are always experiencing one or the other [i.e., pleasure or
pain], as the wise men say. For everything is in an eternal flux, up-
ward, and downward" (43a). The reason for Socrates' pliability earlier
in the Philebus in favor of a life that contains some pleasure becomes
apparent now: It is humankind's constantly shifting state that
makes pleasures desirable, if not all pleasures, then at least some of
them. Pleasure is clearly a remedial good; as a motion toward fulfill-
ment it is desirable for all those who suffer from a deficiency, and we
all are always in fact subject to some deficiency or other!
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 441
Plato has thus discarded the Republic's troublesome distinction
between motions that are pleasures but "not quite real ones": All
motions that are restorations are real pleasures. The "trick argu-
ment" of the Republic that pleasure may be mere liberation from
pain, and pain loss of pleasure, does not work anymore; it had
prompted Plato there to distinguish between real pleasures, bastard
pleasures, and the state of rest between them (584a-b). In the
Philebus pleasure and pain each have their own reality. How, pre-
cisely, they are related to each other, and how they are sometimes
confused with each other (as well as with the third, the pleasure- and
pain-free state of integrity) will be discussed later when we look at
different points of criticism in Plato's evaluation of pleasures and
pains.
The general definition of pleasure as restoration of natural integ-
rity is soon modified in the Philebus by an important qualification
that appears necessary when Socrates tries to introduce the further
subdivision between pleasures of the soul and pleasures of the body.
Strictly speaking, pleasures and pains are never a matter of the body
alone, because only those disturbances in the body are pains and
those restorations pleasures that are perceived by the soul (33C-d;
35b).J9 This leads to the definition of the so-called pleasures of the
body as perceptions. They are to be distinguished from the pleasures
that the soul can experience all by itself. These are specified, at least
at first, as the pleasures that the soul has originally experienced with
the body and can then reexperience all by itself through memory.
Only incidentally does Plato indicate that such recollections may
include other kinds of experiences, like those of learning (cf. 34b,
mathematos).
Plato then adds a further significant feature of the mechanism that
is in place even in the case of the most simple pleasures and pains.
That there is the "perception of a disintegration" is not enough to
explain what goes on when we experience hunger or thirst. In each
case the soul does not merely notice the emphasis of the body, but it
desires at the same time the respective "filling" and thus employs
memory. Desire is therefore not a matter of the body at all; it is a pain
of the soul that brings it in touch with the appropriate filling. And
such pains and pleasures turn out to be necessary for the preservation
of every animal (35c-d). The fact that the soul can anticipate plea-
sures that it has experienced before with the body explains then why
Cambridge Companions Online © Cambridge University Press, 2006
44^ THE CAMBRIDGE COMPANION TO PLATO
it can, in particular situations, experience a rather complex "syn-
drome" of pleasures and pains. Being in pain because of some loss, it
can anticipate the coming restoration and enjoy it in the certainty of
this expectation.20 Alternatively, it may realize the hopelessness of
its situation and therefore be in "twofold pain" (35e-36c).
Plato's careful analysis of the physiological and psychological
mechanism that underlies even the most simple-looking pleasure or
pain and that allows for combinations of pleasures and pains is not an
end in itself. It prepares the ground for the extensive critique of plea-
sure that is based on the result of this preliminary analysis (36c: "let
us apply the results of our investigation of these affections. . . ."(.The
critique fastens on the content of pleasures and turns first to pleasures
that are free from disturbances by the body, the pleasures of expecta-
tion or anticipation. To argue that pleasures can involve expectations
is to pave the way to their evaluation; for these pleasures of anticipat-
ing future goods are the means by which Socrates introduces the
possibility of attributing truth and falsehood to pleasures and pains.
An expectation can obviously be mistaken, he suggests,- therefore
pleasures of anticipation can be false.
In spite of his careful step-by-step preparation of the troublesome
subject, Socrates encounters tenacious resistance from Protarchus,
who has been a cooperative partner until now: "How could there be
false pleasures and pains?" (36c). This dramatic change shows that
Plato does not want to smuggle in his controversial conception of
false pleasures by exploiting the malleability of the half-converted
hedonist. When he introduces it he does so by drawing our full
attention to its controversial nature: Protarchus keeps up his resis-
tance and forces Socrates to give a detailed defense (36c~4ia).
IV. THE FALSITY OF PLEASURES AND PAINS
Protarchus is not the only one to give vent to skepticism in the face of
Socrates' contention that there can be truth and falsity in pleasures
and pains. Many scholars nowadays agree with Protarchus.21 One of
the main difficulties is that Plato uses "truth" and "falsity" in differ-
ent senses in the long ensuing discussion. As Gosling put it in his
commentary, "it seems impossible to acquit Plato of the charge of
rank equivocation."22 That there is equivocation seems indeed unde-
niable. In the different parts of the long critique of the pleasures,
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 443
"truth" and "falsity" are obviously used in different senses. But a
close look will show that this equivocation is not anything Plato
wants to deny or be acquitted of. It seems to be a quite conscious and
explicit move on his part to use truth and falsehood in a loose way
whose precise meaning depends on the different specifications under
discussion. His overall intention in the ensuing discussion of false
pleasures is to show that there are many different ways in which there
can be something wrong with them, so as to make good on his claim
that pleasures are only a "sometime-good." There are, roughly, four
different ways in which pleasures and pains can be "false" in this
loose, commonsense meaning of "falsity." We will follow Plato
through the text and see what kinds of falsity he has in mind, and
whether his critique is valid or not. He discusses the following differ-
ent cases, which we will evaluate by looking at a prototype:
1. Falsity of pleasures and pains as propositional attitudes
(36c-4ib) - the Dairymaid's pleasures.
2. Falsity of overrated pleasures and pains (4ib-42c) -Esau's
pleasures.
3. Falsity in the identification of pleasure with freedom from
pain (42c-44d) - the Ascetic's pleasures.
4. Falsity of pleasures intrinisically mixed with pains (44d-
5oe) - Calliclean pleasures.
The definition of pleasure as restoration of a natural state of
harmony allows Socrates to proceed with a critique of pleasure that
fastens on several possible shortcomings intrinsic to such restora-
tions - with the exception of (3), where pleasure is not a restoration
at all but is confused with the state of harmony itself. The basic
idea in this critique is a very simple one: If all pleasures are restora-
tions or "fillings of a lack," then our evaluations must turn on the
object and the conditions of different restorations, as well as on the
question whether there is in fact a restoration. To express the same
point in the more picturesque language of "filling," the question is
whether there is a filling, what the filling is a filling with, and how
the filling is achieved. There is a wide range of possible distur-
bances of soul and body in the sense suggested by Socrates. Any
kind of perceived lack or deficiency can be the source of pleasures
as the replenishments of such lacks.2^ In order to provide some
basis for a critical evaluation of the goodness of the different kinds
Cambridge Companions Online © Cambridge University Press, 2006
444 T H E CAMBRIDGE COMPANION TO PLATO
of pleasures, Socrates has to show what different types of shortcom-
ings there can be. This explains the length of the middle part of the
Philebus, the (to many readers tedious) exposition of the pleasures
of soul and body.
Plato's definition of pleasure as a perceived filling or restoration is
designed to cover all kinds of pleasures; it precludes the identifica-
tion of pleasure with a simple indistinct feeling that merely accom-
panies these processes or results from them. Pleasure is for him not
an epiphenomenon in the sense that it is an unspecific kind of
"glow" that supervenes on experiences and could arise from any
source. A supervenient state of excitement or elation would not, as
such, be open to any further evaluation: In that case it does not
matter where the pleasure comes from (push-pin or poetry); all that
matters is that we are pleased. Such an epiphenomenal view, it
seems, was initially presupposed by Protarchus when he tried to
reject any kind of differentiation among pleasures ("the pleasures
come from different sources; but they are not at all opposed to each
other," nd). If Protarchus is later still restive at some points of
Socrates' critique of pleasure, this shows that he is not quite ready to
give in to Socrates' new conception of the nature of pleasure or that
he has not quite realized the full extent of the implications of this
new conception. His main objections are ventured against the first
kind of falsity of pleasures; it is also the one that has received the
most flak from contemporary critics. We will therefore turn to the
"Dairymaid's pleasures" first.
(i) The question why Plato thinks that there are false pleasures in
the literal or "propositional" sense of falsity has been much debated
in recent years.2* How could pleasures be false, except in a rather
metaphorical sense of "false" that signifies no more than that there
is something wrong with them? As the elaborate analysis at the
beginning of the discussion of false pleasures and pains shows,
Plato's conception presupposes that not all pleasures are simple con-
ditions of the body or the mere awareness of the soul of what is going
on in the body. Of course I cannot be wrong about my pain of "feel-
ing emptied out" nor about my actual pleasure of "feeling filled."
But these simple kinds of "fillings" are not what interests Plato for
the most part. He is much rather concerned with the role that the
mind plays in many of our experiences. The first step in the ascrip-
tion of propositional content to pleasure is his claim that the soul
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 445
can, thanks to memory, experience the same kinds of things all by
itself that it has once experienced with the body; so it can anticipate
future pleasures as real (36C-42C). Such pleasures are not mere feel-
ings, nor are they daydreams. They are hopeful (or fearful) states that
anticipate future states of affairs as facts; they are therefore what
contemporary philosophers call "propositional attitudes" and can be
true or false in the same sense as other such states.2*
Although Plato has, of course, no such terminology at his disposal,
his elaborate explanations designed to show that some pleasures and
pains are mental processes like beliefs, allow us to attribute such a
theory to him. Our experiences are logoi (statements) "written into
the soul as into a book/' often supplemented by pictures (eikones),
which can be actualized by the soul itself (38e-39b). If the soul
entertains pleasant expectations or visions about the future as accu-
rate depictions of the future, then these logoi or images are true or
false, as happens in the case of the Dairymaid. She anticipates the
pleasures of a golden future that will come from selling her milk at a
good price and further prudent investments. The hapless Dairy-
maid's pleasures turn out to have been false anticipations: One
wrong step makes all her pleasant calculations null and void.
To make Plato's theory of true or false pleasures acceptable, we
have to agree to two conditions: that there are expectations of future
states of affairs (as in the case of the Dairymaid's depiction of a
golden future) and that such states of affairs are experienced as fill-
ings of a lack. It is the Dairymaid's poverty that makes her entertain
and enjoy such high hopes. If she was only daydreaming, enjoying
the mere thought or the picturing without putting any trust in the
anticipated state of affairs, or if she were in no need of such riches,
there would be no disappointment, no painful farewell to chicken,
piglet, calf, and cow.
Is it plausible to assume that the soul can "fill itself" with pleas-
ant logoi or pictures? There is no reason to reject such "fillings,"
once we accept Plato's story of how the soul builds up its stock of
interpretations of the world in a silent dialogue (38c). The existence
of such a stock of experiences in the soul makes it possible to project
a similar experience into the future, an experience that is quite
independent of actual physical pleasures and pains. If such mental
actualizations are pleasant anticipations of the future, and the future
turns out to be the way it was envisaged, then the pleasures have
Cambridge Companions Online © Cambridge University Press, 2006
446 THE CAMBRIDGE COMPANION TO PLATO
been true; if it does not turn out in the way envisaged, then the
pleasures have been false.26 So the gist of the argument is that some
pleasures are actually logoi or pictures,- they do not merely follow or
accompany them. Hopes, that is, pleasant expectations, are logoi
(4oa6-7), and so are the corresponding pictures,- some of them are
clearly false, and such seems to be the case in Socrates' example of
the man who often "sees himself as rich" and envisages all the
pleasures that result from it. These pleasures often turn out to be
empty hopes, as Plato indicates when he adds the qualification that
only persons who are beloved by the gods are usually right in their
anticipations (39e-4ob). Had the gods been favorable to our poor
Dairymaid, her pleasures of anticipation would have been true. As it
was, she entertained vain hopes, that is, false pleasures.2?
That Plato focuses on true or false anticipations does not mean that
only future pleasures (or pains, when the anticipations are of future
ills) can be true or false. If they stand in the center of his discussion,
that is because in their case it is most obvious that we enjoy the
propositional content and not the "thing" in question. If Plato had
concentrated on examples in the present or the past, it would have
been much more difficult to point out that a logos plays an important
role in our pleasures. That there are such false propositional pleasures
in the present or the past becomes obvious if we look at some exam-
ples: Sophocles' Electra "falsely" grieves over her brother's death;
Clytemnestra "falsely" enjoys her relief. Both grieve over or enjoy a
logos, not the actual state of affairs, or more precisely, they grieve over
or enjoy a state of affairs under a given false description. What is
crucial here is that we often enjoy what we enjoy only under one
description and not under another. Some persons can only enjoy a
present if they know that it was expensive; some of our friends will
not enjoy our company if they know that we are visiting them only
because other plans have fallen through; the spectators enjoy an excit-
ing move in a game only if their own team accomplished it, and so on.
The logoi and pictures we carry in our souls thus play a powerful role
in determining whether we enjoy what we experience, how we enjoy
it, and whether it is a thing that ought to be enjoyed.28
(2) That pleasures can have propositional content makes it possi-
ble to evaluate, even morally evaluate, the pleasures we enjoy; it
also makes it possible to compare, to measure, and to experience
pleasures as having a certain size or worth. This seems to be the
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 447
rationale for introducing the "false," that is, disproportionate, plea-
sures that we may here call Esau's pleasures.
If pleasures can be logoi, then it makes sense to revive Socrates'
old idea in the Protagoras of measuring pleasures and pains against
each other. Driven by hunger (the pain of emptiness and the desire
for filling), Esau was induced to overrate the worthwhileness of fill-
ing himself with a dish of lentils to the point where he thought the
pleasure was worth the price of his primogeniture, that is, the future
pain of its loss. Plato compares this situation to an optical distor-
tion. Just as we may misjudge the size of what we see from far away
in comparison to what we see close up, so we may be misled in our
appreciation of the "size" of pleasures and pains (4id). The size of
the hunger seen from nearby creates enormous desire for replenish-
ment with food, so that Esau takes inordinate pleasure in his repast,
falsely enjoying the experience as well worth the future pain of his
loss.
If the pleasure itself is experienced as something quantifiable,
then this pleasure as enjoyed can be false, and Plato is right that
"more is enjoyed" than is actually warranted. So enjoying some-
thing as an "enormous pleasure" would then be false, while enjoy-
ing it as a medium-size or as a small pleasure would have been
true. If indeed we are willing to grant Plato that we sometimes
make it part of our pleasure that it is of a certain magnitude (enjoy-
ing an A+ or a B- pleasure), we will also agree with him that to
the degree to which our estimate of the pleasure was not justified
we enjoyed a wrong "portion" of it, that is, more than is actually
"there." Such a hedonic calculus makes sense if we can experience
pleasures and pains as having a certain quantum or worth. Experi-
ences need not be as remote as Esau's case may look to us. We
often do not enjoy a $25 dinner that we might well have enjoyed if
we had eaten it as a $7.50 dinner. So not only does pleasure have its
price, we enjoy it as having a price. And just as one may pay a
"false" (i.e., inflated) price for some good, so one may enjoy a false
or inflated pleasure, that is, enjoy it as if it were worth more than it
actually is.
Plato himself gives only a short treatment to the problem, and we
may surmise that the reason for mentioning it at all is not only that
such cases occur in everyday life, when we ask ourselves whether
some pleasures are "worth the pain," but because he wants to give
Cambridge Companions Online © Cambridge University Press, 2006
448 THE CAMBRIDGE COMPANION TO PLATO
the old argument of the Protagoras its place in his final and defini-
tive discussion of pleasure. So the "art of measuring pain and plea-
sure" is vindicated as one of the means to sort out worthwhile plea-
sures and pains.
(3) After the overinflated pleasures, Plato takes up "even falser
pleasures" (42c-44d) - pleasures so false that they are not pleasures
at all. It is at first sight surprising that Plato mentions the neutral
state and its distinction from pleasure and pain here once again, in
spite of the fact that he had made it clear before how rest differs from
both pleasure and pain (32d-33c). There is, of course, a colloquial
use of "false" that Plato can exploit here: A false Rembrandt is no
Rembrandt, and false friends are not friends. But there are more
serious reasons than the existence of this faqon deparler to include a
discussion of what is not a false pleasure but a false conception of
pleasure (44a). First of all, Socrates had agreed earlier, at least tenta-
tively, that it is impossible for us to believe that we are pleased when
we are not (36c). This is now corrected: It is indeed possible to
mistake a state of undisturbedness for a state of pleasure.2* Secondly,
Plato seems to answer an unnamed Ascetic who proposed such a
conception of pleasure. This is not the place to speculate who the
"difficult person" might have been whose harshness made him, as
Plato claims in a rare ascription of psychological motivation, "refuse
to acknowledge anything healthy in pleasure, even to the point that
he regarded its very attractiveness itself as witchcraft rather than
true pleasure" (44c-d).3°
The temptation to confuse the state of rest with pleasure or pain
had been one of Plato's concerns in the passage in the Republic
discussed earlier. If he wants to straighten out these difficulties in
the Philebus he has to give precise reasons why contentment cannot
be a pleasure, even though he himself regards it as the best possible
state. It cannot be regarded as a pleasure because it is not the restora-
tion of a lack. To stress this point is of great importance for Plato's
attempt to refute philosophical hedonism, and this must be the rea-
son why he returns to the difference several times in the dialogue.
Socrates therefore does not agree with this anonymous curmudgeon.
At the same time the Ascetic's complaints against the "unsound"
character of pleasure as such serve Socrates as a welcome occasion
to sort out what makes certain pleasures unsound, so he uses the
Ascetic as a "guide." The pursuit of this question will eventually
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 449
lead to the correct determination of true, in the sense of unblem-
ished, pleasures.
(4) Unsound as a whole are what I call the Calliclean pleasures,
made famous by Callicles' fervent plea for a life of unimpeded pur-
suit of ever-new excitement. These are here called the pleasures of
excess (44e-5oe). According to Socrates the Ascetic's denial that
there is anything sound in what ordinary people call pleasure is the
result of his focus on pleasures in their most intense form, when our
bodies are in a diseased state. For such a diseased state intensifies
our awareness of all physiological changes, and release from a state
of distress is therefore felt most keenly. "Do they not feel greater
deprivations, and also greater pleasures after their replenishment"
(45b). Such pleasures result not only from a diseased body, but
equally result from a diseased soul ("a vicious state of soul and
body/7 45e). Plato's own diagnosis of such "overheated" pleasures
differs from that of the Ascetic, who denies them the status as plea-
sures at all and recognizes only freedom from pain as pleasurable.
For Plato such experiences remain pleasures, but they are not true
pleasures because they are inextricably mixed with pain: "the bitter-
sweet condition that first causes irritation and later on turns into
wild excitement" (46d). So they are false in the sense in which we
colloquially call impure gold false gold.
Depending on the kind of diseased state, the pain involved may be
small or large, and the release may be felt more or less strongly. If
grave irritation is the preponderant state and the release insignifi-
cant, we call the whole state "pain"; if there is only a mild irritation
that is totally outweighed by wild excitement of release, we call the
overall phenomenon "pleasure." The funny description of this in-
flamed state leaves no doubt that these are the Aphrodisian plea-
sures without which neither Callicles nor Philebus find life worth
living (47a). Neither of the two conditions is a true, unmixed plea-
sure or pain.
If Plato's description of these incensed pleasures of excess reminds
us strongly of the discussion of Callicles' pleasures of the superior
man in the Gorgias, there is a major difference in Plato's critique in
the Philebus. We are not asked to reject the pleasures because there is
something unseemly about them, in the way Socrates forced Callicles
to scorn the catamite's pleasures, because even ruthless Callicles was
not unscrupulous enough to claim that such a life was really the good
Cambridge Companions Online © Cambridge University Press, 2006
45O THE CAMBRIDGE COMPANION TO PLATO
life, worthy of a gentleman. In the Philebus it is regarded as a suffi-
cient critique of pleasure that such frolickings are necessarily mixed
with pain, and that this is in fact so in a "vast number of cases." Plato,
as stated earlier, seems to have exchanged his earlier aesthetic-moral
criteria of evaluation for scientific-medical ones.*1 He does not have
to shock a recalcitrant opponent into submission: The unsoundness
of soul and body alone is a sufficient criterion for the judgment of such
pleasures. The drastic depiction of the mad pleasures of excess is here
only dramatic by-work. If Socrates does not refrain from a well-aimed
stab at Philebus's morality (46b), this does not serve any purpose in
the argument itself; it seems rather designed to show how far
Protarchus has moved away from his previous unconditional hedonis-
tic stance ("Your description fits exactly the preconceptions of the
common run of people, Socrates," 47b).
The analysis of the mixed pains and pleasures of the body, amus-
ing as it is to read, is not the end of Plato's critique. The ensuing
introduction of the mixed pleasures of the soul by itself seems to be
the real focus of his discussion of mixed pleasures, the discussion of
the emotions and their conditions (47d-5oc). Here Plato is truly
innovative: All our passionate affections turn out to be mixtures of
pleasures and pains, and we call them one or the other because there
is a preponderance of one or the other, just as there was in the
conditions of the body. He insists that all our painful emotions, such
as rage, longing, lamentation, love, jealousy, and envy, are depriva-
tions of some sort or other that contain a portion of pleasure. That
Plato should come to such a negative evaluation of the emotions is
no surprise: They are all motions of the soul, disturbances or restora-
tions of the harmony of contentment.
Unfortunately, Plato does not give a more detailed elucidation of
the kinds of pleasures that he sees contained in all of these affec-
tions, with the exception of rage; here he refers to Homer as his
witness that rage contains a certain sweetness, "sweeter than soft-
flowing honey." The sweetness in rage is most likely the sweetness
of anticipated revenge,- and anticipation of the opposite state (47c) is
probably also the explanation for the mixed nature of all other emo-
tions. As disturbances or deficiencies, they contain a desire for the
opposite state, and the logoi or images we possess will make us
envisage their fulfillment. All longing will contain an element of
anticipated assuagement, lamentation of consolation, love of sue-
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 451
cess with the loved, jealousy and envy of satisfaction in seeing one-
self triumphant over the other. Plato clearly sees the same emo-
tional ambiguity at work in drama, both in tragedy, where there is
"laughter mixed with weeping" (48a), and in comedy, where our
laughter is conditioned by ill will against those we laugh at.32
Plato's claim that there is a fundamental ambiguity at work in the
emotions concerned with comedy and tragedy seems to refer both to
the personae in the drama and to the audience. He confines himself,
once again, to an all-too-short summary of his views, but it is well
worth speculating what his reasons are. Comic amusement is often
regarded as one of the most innocent pleasures. By working out that
it is in fact based on envy or malice, Plato may intend to show that
there are no innocent emotions. Laughter in comedy, so he points
out, is a kind of rejoicing at our neighbor's misfortune;^ it is laugh-
ter at his ignorance, when he prides himself falsely on his beauty,
wealth, or smartness. Plato's underlying assumption is that without
an inherent resentment against our neighbor we would not find such
a display of his folly amusing. So even the innocent-looking comic
amusement presupposes a kind of moral lack or disintegration: with-
out the ill will (a negative feeling or need) no such "filling" as the
spectacle of the other's foolishness would give us any pleasure. That
Schadenfreude is not the purest pleasure, contrary to what a Ger-
man proverb claims, everyone will know who has ever analyzed the
edge that is contained in such feelings.
Plato is certainly the first to put the double-sidedness of our emo-
tions into full relief. If he is right, all our emotions involve a kind of
moral lack and are the sign of an unsound state of our soul. It must
be a lack or disturbance, because otherwise we would be at peace
with others and ourselves. It must be moral, because such enjoy-
ments are pleasures with a propositional content: We enjoy the spec-
tacle of our neighbor making a fool of himself, so long as he is weak
(because he cannot revenge himself on us), while such a display of
ignorance in a powerful neighbor causes fear in us, because she is
capable of taking revenge. Having such content, our emotions are
subject to moral judgment: There is unjust envy, as there is just
rejoicing (49c-d). More importantly for Plato's theory, in none of
these cases can the simple fact that we are enjoying ourselves prove
that such pleasures are good, or that they are true, that is, pure,
pleasures. If he is right, the pursuit of the excitement of cultural
Cambridge Companions Online © Cambridge University Press, 2006
452- THE CAMBRIDGE COMPANION TO PLATO
pleasures or pains in drama or comedy is to be regarded with dis-
trust, because they may further incense our already disharmonious
souls.34
It would have been interesting to see Plato spin out these rather
fascinating remarks about the ambiguity of the emotions into a full-
fledged new critique of the educational influence of tragedy and com-
edy. No mention is made of any positive influence of the arts on our
emotions, or of a profitable employment of their "restorative pow-
ers": It is quite conceivable that Aristotle's much-debated theory
about the "purification" of our affections was influenced by the dis-
cussion of the impurity of our emotions in the Philebus.^
V. TRUE PLEASURES
After the long discussion of the manifold shortcomings of pleasures
there is a relatively short treatment of the true pleasures (5^-55).
The "truth" of these pleasures is as equivocal as their "falsity" had
been. If pleasures are to attain truth they must be unmixed with
pain and have the appropriate size, and there must not be any false-
hood involved in the object of the pleasure. True to the definition
that all pleasure is the filling of a lack or the restoration of a defi-
ciency, only such fillings can be regarded as true pleasures that are
based on an "unfelt lack" ("imperceptible and painless," 51b), while
the restoration is perceptible and pleasant. Of the sensory pleasures
only very few fulfill these conditions, such as the pleasures of smell
or of beautiful sights and sounds. Even among these Plato imposes
further rigid limitations concerning their content: Only those are
acceptable as pure pleasures whose beauty is not relative to any-
thing else and does not depend on particular conditions. Such are
the pure colors or the pure lines of geometry; but not the colors or
shapes of any particular objects, for their beauty is determined by
the purpose the object serves or by the connection it stands in, like
the beautiful sounds in a melody that may be out of place or in the
wrong mode and thus ugly. Therefore it is not surprising to see that,
of the many necessary pleasures of restoration, in the end only a few
have any chance to get into the exalted rank of things that deserve
to be put on the scale of goods. Especially among the sensory plea-
sures, the purity conditions (given the fact that there are so few
patches of pure white and so few instances of unblemished A-
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 453
naturals) ensure that only those instances of pleasures that are small
in amount, great in purity, and unmixed with pain possess the right
"moderation" (52c).
This stricture applies even to the pleasures of learning. They hold
a much more modest place here than the intellectual pleasures had
obtained in Republic IX; for they seem to have lost a lot of the luster
assigned there to the philosopher's glorified pleasures. There is no
trace now of a calculation that makes them come out exponentially
superior to all others. The reason for the much more subdued role of
intellectual pleasures in the Philebus cannot lie only in the point
that Plato does not here maintain a privileged position for the "first
part" of the soul;^6 it must also be a consequence of the fact that
among all intellectual activities only learning can be a pleasure at
all. Being true to his definition of pleasure as the filling of a lack,
Plato can no longer accept any other pleasures of the mind, such as
pleasures of "contemplating reality" (Rep. 58ie). The wise man does
not need any filling but lives, as far as humanly possible, in a state of
self-sufficiency and contentment. Plato is not very explicit about
this new limitation of the pleasures of the mind, but learning is the
only intellectual pleasure he mentions, and he takes pains to assure
us that this pleasure is based on a painless lack (5 ie). That he has no
use for pleasure in philosophical contemplation itself is due to the
fact that he does not make the distinction between process and
activity that allowed Aristotle to grant pleasure to all of our success-
ful activities, especially to the philosopher's.^
It may look at first sight rather peculiar that Plato treats anything
that we do not "have" as a lack, so that there is a "painless lack" of
beautiful sounds, sights, or smells when we happen not to see, hear,
or smell anything. The peculiarity vanishes if we realize that this is
his ultimate explanation of why human beings would and could not
choose a life without pleasure, even though a life of total imperturb-
ability would, in principle, be "more divine." Such a life is simply
not open to human beings: There are lots of things we do not have,
know, hear, feel, or see, and that will enrich our existence even if we
do not have any perceived need for them. We are born as needy
creatures, and as Plato's reference to the possibility of a "painless
loss" of knowledge (52a-b) reminds us, a state of completion once
attained need not remain such; everything that is not strictly eternal
needs constant maintenance and restoration, even knowledge. Hu-
Cambridge Companions Online © Cambridge University Press, 2006
454 T H E CAMBRIDGE COMPANION TO PLATO
man beings never live in permanent possession of the good, in a state
of continued perfect self-sufficiency (6oc).38
But if pleasure is a restoration that is necessary for a fulfilled life
and enriches it, why, so one may feel prompted to ask, is it not
regarded as a good per set This is the question Socrates turns to in
his conclusion of the long critique of hedonism. The point is a
purely ontological one. Even the true and pure pleasures remain
processes that lead to the restoration of being [ousia), but they never
have being in themselves nor a permanent immutable natures As
processes they will always be inferior to the end that terminates
them, that "for the sake of which" they happen. Even the best kind
of generation is only good relative to the being that is its end, but it
is not desirable in and for itself. Good pleasures are good because of
the harmonious state they lead to; this end imposes limit on the
pleasures and justifies their place on the lowest rung of the final
scale of goods. But though such pleasures of generation or restora-
tion are necessary and even good for us, it would be foolish to claim
that it is the pursuit of fillings that makes our life worth living
rather than the state of fulfillment itself. It would be a confusion of
means and ends (54e).
We have to refrain here from a critical analysis of Plato's treat-
ment of the different disciplines of knowledge, their relative purity
and truth, which fulfills Socrates' promise (22c) to subject knowl-
edge to the same critical scrutiny as pleasure. The distinction be-
tween pure sciences and applied sciences (55c-59b) would deserve
an essay of its own, one that would also have to deal with the
difficult question of Plato's conception of dialectic as the "insight
into true reality" (59d), which is more hinted at than discussed in
the Philebus. The existence of these hints seems to speak against a
"revisionist" interpretation/0 and for the assumption that Plato still
holds in the main the same metaphysical doctrine that we find in
dialogues of the earlier late period. There is no support for the claim
that separate Forms have been given up, especially in view of the
fact that he refers to the eternally selfsame as the object of real
knowledge, in contradistinction to the changeable objects of belief
(59a-e). Plato's silence as to the precise nature and origin of the
harmonious limit - the truth, proportion, and beauty that make all
things good - does not permit us to go beyond these hints and to
import any further-reaching new metaphysical doctrine. In can be
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 455
no accident that he remains uncommitted in the passage where he
sums up the difficulties of the theory of the Forms (isa-c).*1
Leaving these problems open for future scholarly debate, we
should in concluding return again to the initial question of why
Plato lets Socrates conduct this investigation. As our discussion has
shown, Plato has arrived at a conception of pleasure that differs
significantly from his earlier one, in spite of undeniable continu-
ities. The fourfold ontological distinction seems tailored for his de-
velopment of this new conception. So why did he resurrect Socrates
as the protagonist in a discussion that turns out to be quite remote
from the earlier treament of pleasure, both in its means and in its
ends?
It had been suggested above that it must be the topic, the concern
with pleasure, that prompted Plato to let Socrates conduct the discus-
sion once again. But this cannot be the whole story. There is indeed
more of a continuity than the new ontological framework lets one
suspect. If pleasure is now defined as a "coming into being" and as
something that is as such neither good nor bad, then this constitutes
the development of ideas that have been advocated by an earlier Socra-
tes, even if he himself did not extend this earlier doctrine to give
pleasure its proper ontological niche. It was the Socrates of the Sympo-
sium who had described the philosopher as the "mighty daimon," the
son of Poverty and Plentiful, whose needy state and love for wisdom
urges him on in search of perfection (Smp. 2O2e//). The best kind of
pleasure, one may supply from the perspective of the Philebus, is the
kind that consists in the process of the eternal hunt for knowledge
envisaged in the Symposium. It is this need for fulfillment that drives
the soul in the Phaedrus up and around in pursuit of its deity, and the
Socrates of that dialogue is no longer averse to seeing something
divine in such a "manic" endeavor of the soul [Phdr. 244//).
To deal with such processes is not the dialectician's business, for
he is the perfectly wise man whose concern is "being, true reality
and the eternally self-same" (58a); it is the the concern of the seeker
for completion, whose representative is the poor, unshod, but relent-
less and indefatigable Socrates portrayed in the Symposium, or the
concern of the rueful singer of a palinode in favor of love in the
Phaedrus. The progress toward perfection is what the best kind of
pleasure in human life represents, and that is why a human life
would not be desirable without pleasure: Insofar as it is human it is
Cambridge Companions Online © Cambridge University Press, 2006
456 THE CAMBRIDGE COMPANION TO PLATO
incomplete and needs the constant tending and completion that the
Socratic philo-sopher (although not the Platonic dialectician) craves
for.
This is not the only Socratic element in the Philebus, however.
The "medical" character of the treatment of pleasure and pain has
been stressed before. It is hardly an accident that so much emphasis
is put on the conception of pleasure as the restoration of a naturally
harmonious state. Health and harmony of soul and body had been
the hallmarks of the conception of goodness advocated by the Socra-
tes of the earlier dialogues, such as the Gorgias and the Republic. In
the Philebus pleasure (in its best form) finally finds its place in this
ontological scheme as the restoration of the natural harmony of soul
and body. This shows that the medical analogy that justified Socra-
tes' criticism of his contemporaries' morality has never been given
up; it has been redescribed in new terms that tie together being and
becoming, the imposing of limits on the unlimited. Socrates' doc-
tors, his horse and dog trainers who strive for excellence in their
subjects, have not been forgotten.**
Critics of the coldness of Plato's conception of love, which seems
to take so little notice of the essence of our passionate commit-
ments, will probably not see any great improvement in the revision
of the conception of pleasure in the Philebus A* Does the new "medi-
cal" treatment of pleasure as restoration and the question of the
soundness of our emotions in general not display the same kind of
disrespect toward what we regard as essential about ourselves as
human beings? Plato, the philosopher whose striving for perfection
constituted the basis of his lifelong attachment to Socrates, would
not deny that love and pleasure make us human; he came to see that
and why they cannot and should not be eradicated but even deserve
to be cultivated. This is just what the Philebus was designed to
confirm; our needy state is precisely what makes us human, but it is
also what makes us all too human. It is a necessary ingredient of our
mortal condition.
NOTES
i For a discussion of the problems of coherence, cf. esp. the commentary
by J. C. B. Gosling, Plato: Philebus (Oxford, 1975), ix-xxi, notes ad loc,
and Epilogue, 226-8.
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 457
2 No more than a few references to the literature can be given here: Helpful
are the introduction and notes in R. G. Bury, The Philebus of Plato (Cam-
bridge, 1897); R. Hackforth, Plato's Examination of Pleasure (The Phile-
bus) (Cambridge, 1945); and the commentary in Gosling, Plato: Philebus.
For further discussion see Hans-Georg Gadamer, Platos Dialektische
Ethik — PhdnomenologischeInterpretation zum Philebus (Leipzig, 1931);
W. D. Ross, Plato's Theory of Ideas, id ed. (Oxford, 1953); I. M. Crombie,
An Examination of Plato's Doctrines, 2 vols. (London, 1962, 1963); R.
Shiner, Knowledge and Reality in Plato's Philebus (Assen, 1974); J. C. B.
Gosling and C. C. W. Taylor, The Greeks on Pleasure (Oxford, 1982); and
most recently, C. Hampton, Pleasure, Knowledge, and Being: An Analy-
sis of Plato's Philebus (Albany, 1990). All of these works contain exten-
sive bibliographies and discussions of the relevant literature. For a discus-
sion of the dialectical-metaphysical part, see esp. G. Striker, Per as und
Apeiron: Das Problem der Formen in Platons Philebus, Hypomnemata
30 (Gottingen, 1970); K. Sayre, Plato's Late Ontology: A Riddle Resolved
(Princeton, 1983); and R. M. Dancy, "The One, The Many, and the Forms:
Philebus i sb i -8 , " Ancient Philosophy 4 (19841:160-93.
3 The model of such a division is that of the art of writing: You cannot claim
mastery until you not only know that"letter" is the genus that unifies all
sounds, but also know all the subgenera (vowels and consonants) and the
sub-subgenera of semiconsonants, consonants, and mutes. Further divi-
sion renders the species of the letters; it ends with the unlimited multi-
tude of actual instantiations of letters. Music is the other example Plato
discusses. Because of the complexity of the systems of Greek scales (classi-
cal Western music contains nothing that is comparable) we have to omit
the discussion of music here. Cf. Gosling, Plato: Philebus, 155-81.
4 Cf. esp. Striker, Peras und Apeiron; and Gosling, Plato: Philebus, 185-
206. The unity of the Philebus is discussed by O. Letwin, "Interpreting
the Philebus," Phronesis 26 (1981): 187-206. It is also the main concern
of Hampton, Pleasure, Knowledge, and Being.
5 The debate about the connection between the Philebus and Aristotle has
not come to rest since Henry Jackson first tried to establish it; see H.
Jackson, "Plato's Later Theory of Ideas/' Journal of Philology 10 (1882):
253-98. The most recent contributions have been made by Sayre, Plato's
Late Ontology, 84//, 133//; and Hampton, Pleasure, Knowledge, and Be-
ing, 1-7, 95-101. Since the terminology of peras and apeiron is not used
elsewhere in Plato, the Philebus must be the work where this doctrine
manifests itself, if Aristotle is not either missing the mark or referring to
the arcane source of Plato's oral teaching, the "unwritten doctrine." So far
attempts to use these passages to solve the riddle of Plato's late ontology
have not led to satisfactory results.
Cambridge Companions Online © Cambridge University Press, 2006
458 THE CAMBRIDGE COMPANION TO PLATO
6 The formulation in the text (66c-d) does not definitely settle the ques-
tion whether Plato regards this list of goods as closed or whether there is
a possible sixth class of less pure and true pleasures, as Hackforth sug-
gests in his commentary (Plato's Examination, 140 n. 3). If Plato is as
restrictive as he seems, the pleasures admitted as necessary to the good
life would only be remedial goods (63c, e.g., gymnastic and the painful
exercises to acquire virtue). If this is the solution to the problem, then
not all acceptable ingredients of the good human life are on the list of
goods because they do not satisfy the "truth, beauty, proportion" test.
7 The question why reason and intelligence (nous and phronesis) form a
class separate from science, art, and true judgment (episteme, techne,
and doxa orthe, 66b-c) can only be touched here: The latter are acquired
skills that allow for different degrees of purity, while the former are the
capacities that make such acquisitions possible in the first place. For the
importance of the focus on the human good, see Gosling, Plato: Phile-
bus, 132-3, 224-6.
8 A late date for the Philebus is supported by stylometry as well as reasons
of doctrine. See H. Thesleff, Studies in Platonic Chronology (Helsinki,
1982), esp. 198-200. For a relatively earlier date, cf. R. A. H. Waterfield,
"The Place of the Philebus in Plato's Dialogue/' Phronesis 25 (1980):
270-305; and the introduction to his translation, Plato: Philebus (Har-
monds worth, 1982).
9 Protarchus is at first a convinced hedonist. He is obviously a friend and
admirer of Philebus, doing his best to please him. After a tense and almost
hostile beginning (see i3b-d), Protarchus, son of Callias (19b), disciple of
Gorgias (58a), turns into an attentive and cooperative interlocutor, a re-
formable version of Callicles (with true knowledge, good will, and frank-
ness; see Grg. 487a). He changes his allegiance without coercion or obse-
quiousness and ends up a Socratic rather than a hedonist.
10 This can only be a very brief summary. For further discussion and
references see Gosling and Taylor, Greeks on Pleasure, chaps. 3-6; and
my "Rumpelstiltskin's Pleasures: True and False Pleasures in Plato's
Philebus," Phronesis 30 (19851:151-80, esp. 151-61.
11 Whether the Protagoras in fact temporally precedes the Gorgias is not of
any great importance here. It is assumed here because the antihedonist
stance of the Gorgias is in line with the position in the Phaedo and the
Republic. But since I take it that the position of the Protagoras is merely
the exploration of a hypothesis anyway, which Plato did not adopt for
himself (as his contempt for such "vulgar" calculations in Phd. 69a
shows), it is immaterial at what point of time he made it. For differing
possible interpretations see G. Vlastos, "Introduction," in Plato: Prota-
goras (Indianapolis, 1956); T. Irwin, Plato: Gorgias (Oxford, 1979), 121//;
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 459
D. Zeyl, "Socrates and Hedonism - Protagoras 35ib-358d," Phronesis
25 (1980)1250-69; and, for the most recent discussion of the question, R.
Weiss, "Hedonism in the Protagoras and the Sophist's Guarantee/' An-
cient Philosophy 10 (1990)117-39.
12 The reason for calling this a "medical" point of view will be discussed
more extensively later. By refraining from mixing the question of moral
standards with that of the ontology of pleasure, Socrates can avoid the
battle at two fronts he had to fight against Callicles, by forcing him to
grant that "leaks" are never desirable while at the same time conceding
that some are especially bad.
13 In Rep. VI 5O5b-c pleasure and knowledge are both rejected as defini-
tions of the good; pressed for specification the champions of knowledge
will have to beg the question and resort to "knowledge of the good,"
while the hedonist will have to admit the existence of bad pleasures. No
other differentiations of pleasures, such as the possibility of pleasures of
knowledge, are mentioned here.
14 The term eskiagraphemene (583b) means literally "a thing painted with
its shadow" in order to produce the illusion of a three-dimensional physi-
cal object. For a survey of "unsatisfactory points in the Republic ac-
count," see also Gosling and Taylor, Greeks on Pleasure, 126-8.
15 That it is the need for orderliness that dictates the setup in the discus-
sion of the Philebus is indicated by this: Plato mentions the postpone-
ment of the discussion of certain shortcomings of pleasures because he
did not want to run different points of criticism together (see 47C-d:"I
did not mention earlier . . .").
16 The same objection has been raised by Donald Davidson, "Plato's Phi-
losopher," The London Review of Books 7, no. 14 (1985): 15-17. He
draws the conclusion that the Philebus represents a renunciation of
Plato's preoccupation with dialectic as the philosopher's concern. It is
difficult to say whether this is really a final renunciation on Plato's part
rather than an ad hoc solution for the comparison of pleasure and knowl-
edge. In what follows it will be argued that it is the nature of pleasure as
the filling of a lack itself that makes a complete division undesirable, if
not outright impossible.
17 Cf. Tht. 156//. Whether Plato ever agreed with such a Heraclitean stance
cannot be discussed here; his earlier theory of the Forms seems to have
provided for a definite nature of changeable objects only insofar as they
participated in the Forms. On Plato and the flux see T. Irwin, "Plato's
Heracliteanism," Philosophical Quarterly 27 (i977):i-i3.
18 That the restoration of being (ousia) can at best be the means (as genera-
tion) to the good is indeed the conclusion of the long discussion of
pleasures in 5 3C-5 5a: "But if pleasure really is a kind of generation, then
Cambridge Companions Online © Cambridge University Press, 2006
460 THE CAMBRIDGE COMPANION TO PLATO
we will rightly place it in a class different from that of the good." This
only rules out that pleasure is good by definition, it does not preclude
that some pleasures are good; the fact that true and pure pleasures are on
the final list of goods even though they presuppose an unfelt lack shows
that processes are not eo ipso excluded from the realm of the good.
19 Later (43c) a further specification will be added: only significant changes
are pleasures and pains, while moderate ones have no such effect.
20 Such a suspension between a present pain and a future pleasure is the
new "in-between" situation (35c) that replaces the Republic's "rest" as a
state between true pleasure and liberation from pain.
21 Among the skeptics are J. C. B. Gosling, "False Pleasures: Philebus 35c-
41b," Phronesis 4 (19591:44-54; to which there is a reply by A. Kenny,
"False Pleasures in the Philebus," Phronesis 5 (19601:45-92; and brief
rejoinder by Gosling, "Father Kenny on False Pleasures," Phronesis 5
(19601:41-5. See further J. C. Dybikowski, "False Pleasures in the
Philebus/' Phronesis 15 (19701:147-65; Gosling, Plato: Philebus, 214-
20; Gosling and Taylor, Greeks on Pleasure, Appendix A, 429-53.
22 Plato: Philebus, 212.
23 The immense range of possible disturbances and their restorations in
soul and body may be one of the main reasons for Socrates' reluctance to
engage in dihairesis proper.
24 Most of the literature (see note 21) focuses on the "propositional" sense
of "false."
25 Cf. Bernard Williams, "Pleasure and Belief," Proceedings of the Aristote-
lian Society (19591:57-72. Williams's articles started the long debate on
the justifiability of Plato's theory in the Philebus. For a defense of Plato's
conception of false pleasures, see T. Penner, "False Anticipatory, Plea-
sures," Phronesis 15 (19701:166-78; and my "Rumpelstiltskin's Plea-
sures," 165-79. Hampton's ("Pleasure, Truth and Being in Plato's
Philebus," Phronesis 32 [19871:25 3-62) objecting to my defense of "propo-
sitional truth" — that this does not do justice to Plato's conception of
truth as a whole - misses my (explicitly limited) point; no more than an
explanation of the possibility of such "false pleasures" was intended.
26 "If we notice that a pain or pleasure is mistaken in what it is pleased or
pained about" (37e). Protarchus finds it as hard to accept the view that
pleasures are not merely accompanying our judgments as do some of the
modern critics of "propositional pleasures." Socrates for the moment
retreats to accept the point that pleasures and pains "accompany" judg-
ments. But the subsequent elaboration makes it clear that the whole
enjoyment provided by the soul's inner dialogue or the painter's work
consists in nothing but the logoi or pictures in the soul. So Socrates'
momentary retreat is not an admission of defeat.
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 461
27 ". . . it follows that there are false pleasures in the human soul that are
quite ridiculous imitations of true ones, and also such pains" (40c).
28 Plato's seeming reserve, when he lets Protarchus continue with his lin-
gering doubts about false pleasures does not weaken the case. For Socra-
tes is quite in the affirmative: "There certainly are, I at least am con-
vinced" (41b). That Plato often refrains from driving a point home with
vigor must be seen against the background of Phdr. 273-6: He does not
hand out truths that can be learned by rote, but leaves it to the intelli-
gent reader to draw his own conclusions.
29 This (in Plato's eyes false) conception of pleasure as peace of mind was
later developed by Epicurus, who ranked the stationary (katastematic)
pleasures above the dynamic (kinetic) pleasures. Cf. Diogenes Laertius,
X, 136-138. Plato would, of course, agree with Epicurus that imperturb-
ability is best; he would disagree about calling it pleasure. On Epicurus
see Gosling and Taylor, Greeks on Pleasure, 365-74; Philip Mitsis, Epi-
curus' Ethical Theory (Ithaca, N.Y., 1988), chap. 1.
30 For this question see M. Schofield, "Who Were 01 bvoxegeic, in Plato,
Philebus 4.4a ff.?" Museum Helveticum 28 (1971)12-20. He argues for
Plato's nephew Speusippus and assumes that the "difficulties" are logical
ones. My own suspicion is that Plato here humorously characterizes his
own attitude of an earlier stage (in his "scientific" phase). Not only is the
"unsoundness" of pleasures discussed in Rep. IX, but the same phrase is
used in the passage in the Phaedo, where Plato has nothing good to say
about pleasure at all (ouden hygies oud'alethes echei). On Speusippus in
the Philebus see also Gosling and Taylor, Greeks on Pleasure, 231-40.
31 This does not rule out that moral evaluation of the propositional con-
tent of the relevant pleasures is important for Plato. If he does not here
appeal to our moral intuitions to rule out certain pleasures as wrong, it is
because his purpose is to show first on purely ontological, i.e., non-
moral, grounds that and why pleasures can be false. So the Philebus
contains, in Kantian terms, the groundwork of any future morality of
the emotions.
32 On pleasure in tragedy and comedy cf. also Rep. X 6o6b-c.
33 This is true even for the "unquenchable laughter" of the immortal gods
in their amusement at limping Hephaistos's peace-making efforts, Iliad
I, 599-600.
34 It is difficult to say, ex silentio, whether Plato has given up other points
of criticism against fiction, or whether he refrains from it because its
distortion of reality is less relevant for the critique of pleasure than the
unsoundness of the emotions themselves.
35 Poetics 6, I449b28. The basic idea would be that our emotions can be
freed of unjust "edges," as well of all unhealthy inflation.
Cambridge Companions Online © Cambridge University Press, 2006
462 THE CAMBRIDGE COMPANION TO PLATO
36 This need not constitute any revision in Plato's psychology; it may
rather be connected with Plato's concern to maintain the generic unity
of all pleasures.
37 Plato seems to have conceived of thought as "motionless/' so that
knowledge itself, like its objects, is always in the same state (cf. Phil.
55^, 59c). For Aristotle's theory of pleasure (see N. E. X) see Gosling and
Taylor, Greeks on Pleasure, 301-14; cf. also J. Urmson, "Aristotle on
Pleasure," in Aristotle, ed. Julius Moravcsik (Garden City, N.Y., 1967),
323-33; J. Annas, "Aristotle on Pleasure and Goodness," in Essays on
Aristotle's Ethics, ed. A. O. Rorty (Berkeley, 1980), 285-99.
38 This question has been further pursued in my paper "The Impossibility
of Perfection," Review of Metaphysics 39 (19861:729-53.
39 For a further discussion of the dimension between being and becoming,
see M. Frede, "Being and Becoming in Plato," Oxford Studies in Ancient
Philosophy, supp. vol. (19881:37-52.
40 "Revisionists" go on the assumption that the difficulties for the theory
of the Forms raised by Plato in the first part of the Parmenides led to a
revision of the theory in the late dialogues. The "Unitarians" deny any
significant change. Among the revisionists for the Philebus are Gosling,
Plato: Philebus; Shiner, Knowledge and Reality; Shiner, "Must Phile-
bus 59a-c Refer to Transcendent Forms?" Journal of the History of
Philosophy 17 (1979): 71-7; Henry Teloh, The Development of Plato's
Metaphysics (University Park, Pa., 1981). A Unitarian interpretation is
argued for by R. Hackforth, Plato's Examination of Pleasure (The Phile-
bus) (Cambridge, 1945); R. Mohr, "Philebus 55c-62a and Revisionism,"
in New Essays on Plato, ed. F. J. Pelletier and J. King-Farlow (Guelph,
1983), 165-70, with a response by Shiner at 171-80; Waterfield, "Place
of the Philebus"; and Hampton, Pleasure, Knowledge, and Being. For a
careful review of the history of the controversy see W. Prior, Unity and
Development in Plato's Metaphysics (London, 1985).
41 The insistence on the continuity in doctrine of the Philebus with the
earlier dialogues is a major virtue of Hampton's book. It is to blame also
for its main shortcomings, however. In battling against all "revisionist"
tendencies by interpreters who want to see major changes in the theory
of Forms of Plato's late dialogues she overlooks important changes, espe-
cially changes concerning the conception of pleasure itself between Rep.
IX and the Philebus.
42 The allusion is to the Socratic notion that human beings need tending
that ought to be in the hands of experts, analogous to the doctor's solici-
tude for the body and the trainer's for the "virtues" of his horses or dogs.
See Ap. 25b, Grg. 464a//; Callicles even loses patience with Socrates'
analogies (Grg. 491a).
Cambridge Companions Online © Cambridge University Press, 2006
Pleasure and pain in the Philebus 463
43 For a constructive criticism of the shortcomings of Plato's conception of
love, see G. Vlastos, The Philosophy of Socrates (Garden City, N.Y.,
I97 I) 3~42- The issue is further discussed in M. Nussbaum, The Fragil-
ity of Goodness (Cambridge, 1986), chaps. 6 and 7.
Cambridge Companions Online © Cambridge University Press, 2006
TREVOR J. SAUNDERS
15 Plato's later political thought
Does Plato have a "later" political theory, as distinct from an earlier?
It is certainly easy to suppose so. Read the Republic and you con-
front the most radical political system ever devised. Plato's key argu-
ment is that ruling a state is or should be a skill, based on precise
knowledge of certain suprasensible, eternal, and unchanging reali-
ties called Forms [ideal], notably those of the social and political
virtues, but also those of the rest of reality. On the strength of such
insight, the rulers of Callipolis ("Splendid City"), a highly trained
cadre of Guardians, or Philosopher-Kings, exercise direct and total
political control; and lawmaking is accordingly treated by Plato in a
decidedly offhand manner, as a humdrum business on which the
Guardians will spend little time.1
If you are by now gasping with astonishment or even indignation
at the audacity of these proposals, you have every justification; for at
first sight Plato's thought seems almost unintelligible. For what are
Forms, and why ought they to have such a drastic effect on practical
politics? The explanation, though full of philosophical problems
that cannot be explored here, can be stated quite briefly. A Form, say
of virtue, is the essence of virtue, what it really is, as distinct from
the individual things or actions or people that instantiate it in this
world; for they only share in the Form of Virtue in varying degrees;
they are not Virtue Itself. Now the Greek word usually translated
"virtue" is in fact better rendered by "excellence" - excellence for
something. The kind of excellence that interests Plato is human
excellence, that set of qualities thanks to which we are excellently
equipped to perform human functions excellently, and so achieve
human eudaimonia, "happiness," "success," "fulfillment." The
Philosopher-Kings, equipped as they are with precise knowledge of
464
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 465
the Form of arete,1 will naturally embrace it and be supremely virtu-
ous themselves; and they will also be able to discern infallibly
whether and how far a given action or institution is a virtuous one.
Further, they will take care, by various means, to transmit their
knowledge, or rather an approximate version of it called "right opin-
ion," to the rest of the state,* which will in proportion to its limited
understanding practice "ordinary" or "demotic" virtue, and so enjoy
a limited measure of eudaimonia.
The central proposition of Plato's political thought in the Repub-
lic is therefore that human eudaimonia depends on possessing arete;
that fully to possess arete requires an understanding of its Form,- and
understanding the Form is a philosophical activity. The dependence
of morality and politics on philosophy is the distinctive mark of the
Republic.
Turn now to the Laws, the last and longest of Plato's dialogues, and
prepare for a shock. With the possible exception of a few final pages, it
is wholly devoid of the theory of Forms. After three preliminary
"books" on moral and educational theory, the entire work is devoted
to a discursive and painstaking account of the formation and adminis-
tration of a "practical Utopia," Magnesia. The Philosopher-Kings
have vanished without trace; the only vestige of rule based on meta-
physical or theoretical insight of a more than ordinary kind is the
supremacy of the Nocturnal Council, which is composed of gentle-
men-farmers - intelligent and educated indeed, but hardly philoso-
phers. The state is administered under an extensive and detailed
corpus of constitutional, civil, and criminal law, some of it of vertigi-
nous complexity. It is very hard not to feel that one has entered a
different world, in which the cutting edge of Plato's political thought,
metaphysics, has been lost.
The simplest explanations of this startling state of affairs are meth-
odological. One could straightforwardly infer that Plato has now
abandoned the theory of Forms. Or one could argue that the mere
absence of its metaphysical underpinning from a description of a
political structure does not in itself show that that underpinning
does not exist. The political structures both of the Republic and of
the Laws are, in the most general terms, hierarchical and authoritar-
ian; that of the Republic purports to depend on the theory of Forms;
so it is natural to suppose that the same is true of that of the Laws.
The Laws would then take the shape it does because Plato chooses
Cambridge Companions Online © Cambridge University Press, 2006
466 THE CAMBRIDGE COMPANION TO PLATO
to present the results of that dependency, in a practical work in-
tended for consumption by nonphilosophers/ without dwelling on
the metaphysics, which he presupposes but either suppresses or at
any rate keeps in the background. Therefore, one might conclude,
Plato does not have a "later" political philosophy; the Republic and
the Laws are simply different sides of the same cloth.
Fortunately, however, we are in a position to do rather more than
conjecture. There is a limited amount of evidence to suggest that in
his later years Plato was still trying to root politics in the terra firma
of philosophy. 5
The Politicus ("Statesman/7 i.e., one who engages in politika, the
affairs of the state, polis) is ostensibly a search for the definition of the
Statesman, that is to say the person who possesses politike episteme,
political knowledge. Such a person, it is argued, would be entitled to
rule even without laws, and even without the consent of his subjects;
for after all, by definition, he knows what is best for them.6 So far, this
view is very much in the spirit of the Republic, which distinguished
sharply between the sure "knowledge" of the Guardians and the mere
"right opinion" of the rest of the state. So too is the firm distinction
between the overarching and all-controlling function of the States-
man on the one hand, and the essentially ancillary functions of all
other occupations, both of peace and of war.? But it is the method,
apparently a difficult one,8 by which the Statesman is separated off
from all these, that is important: dihairesis, "division," of things
according to their real kinds; for we are explicitly told that the pur-
pose of the search for the Statesman is not for its own sake, but in
order to make us dialektikoteroi, "more dialectical," "better at philo-
sophical discussion," on all subjects.9 Evidently an ability to perform
division, to establish an accurate taxonomy of things, is a philosophi-
cal ability; and it is by implication part of the Statesman's intellectual
equipment, of his political knowledge and skill.10 And the "kinds"
into which he "divides" things have apparently suprasensible coun-
terparts, which it is his ultimate aspiration to understand. Though
Plato's language in describing these counterparts is ambiguous and
allusive, it strongly suggests Forms.11
The ideal ruler would in that case be one with politike episteme
somehow based on a knowledge of relevant Forms. He would be
qualified to prescribe what ought to be done in every detailed situa-
tion and by every person without being constrained by any law, even
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 467
his own. Such a paragon is, at best, rare.12 The second-best form of
government, therefore, is a good code of laws, "imitations of truth"
in that they incorporate as far as possible the insights of a Statesman
with genuine politike episteme.1* And the most important task of
such political skill is to weave together by a variety of social contriv-
ances the active and aggressive members of the polis with the inac-
tive and compliant, so that the social structure achieves a firm tex-
ture, neither too hard nor too soft.1* With this aim of a harmony of
belief and action in the polis the dialogue closes.
So it looks as if Plato still maintains, as a theoretical ideal, the
ruler with insight into Forms. Politics is, or should be, still anchored
in metaphysics. Plato may have believed that proposition fervently;
but for the purposes of the Politicus he keeps it in a shadowy back-
ground. The ideal, perfectly knowledgeable, godlike Statesman re-
treats to the wings; and into the limelight comes the wise, benevo-
lent Statesman with practical experience, a human being legislating
for human beings.^ The emphasis shifts from the rule of philoso-
phers to the reign of good law, tempered in its flexibility by the
prudent practical judgment with which it is applied.16
The Laws exhibits a similar but a more interesting pattern of
evidence. In all its 327 Stephanus pages there is not a single passage
that indisputably refers to the Forms,- but there are several that give
hints. The most explicit passage is the one that describes the pro-
gram of higher study to be carried out by the Nocturnal Council, a
body that Plato intends to be the supreme intellectual force in Mag-
nesia, its "anchor. "^ Its membership will be the state's wisest and
most distinguished officials.18 Its curriculum includes mathematics,
astronomy, theology, law, and moral theory.^ Its members are ex-
pected to study the problem of the one and the many; they must
attempt to look beyond the many dissimilar instances of something
pros mian idean, "to a single shape/form"; and they must concen-
trate on the way in which the virtues are individually distinct, and
yet "one."20
This and other features of the Council's curriculum are strongly
reminiscent of Plato's discussions of Forms in earlier dialogues. To
anyone disposed to believe that Plato never abandoned the Forms,
such passages will seem to bring them very near indeed. But his
words are both brief and vague. Attenuated interpretations of them
are available; and in particular it is impossible to know whether the
Cambridge Companions Online © Cambridge University Press, 2006
468 THE CAMBRIDGE COMPANION TO PLATO
full hierarchy of Forms elaborated in the Republic, culminating in
the Form of the Good, is implied. Certainly much of the rest of the
Council's curriculum seems more physical than metaphysical: for
example, the rational movements of the heavenly bodies as evidence
of design in the cosmos; the priority of soul over matter; and the
nature and functions of the gods.21 However, none of these topics
can be shown to preclude a continued belief in Forms; indeed, such a
belief may be implied by a number of passages that betray a keen
interest in division.22 The most reasonable conclusion is surely that
Plato did still hold to at least some version of the theory of Forms,
and for the purposes of the Laws was careful, right at the end of that
work, where philosophy is presented as the savior of the state, to
direct the studies of the members of the Nocturnal Council in that
direction, so as still to do what he could to ground practical politics
in philosophy.2*
But there lies the rub. We may speak airily of "grounding" or
"rooting" politics in philosophy, or of "basing" it on metaphysics, or
of "linking" to the Forms; but what does such language mean? Sup-
pose you are a Guardian or a member of the Nocturnal Council, and
after much strenuous study and contemplation you arrive at the
vision or knowledge of the Form of a particular virtue (e.g., courage)
or of Virtue in general. Just how are you better off? What sort of tool
have you for the conduct of politics, the framing of a constitution, or
the drafting of moral rules? If the Form of Virtue can be expressed in
the form of a definition of Virtue, could one simply describe some
action that is a candidate for the title "virtuous," put it against the
yardstick of the definition, and see if it matches? These matters are
highly uncertain: Plato is very clear that there is a connection, but
never tell us what it is.2* He seems simply to assume that knowledge
of the higher reality constitutes some sort of control on that of the
lower, such that the knower always knows what is virtuous in this
world, and acts accordingly. As for division, even in his relatively
realistic mood in the Politicus, his keen awareness of its difficulty
and complexity never leads him to conclude that divisions of things
might be arbitrary or conventional; the divisions always have to be
genuine, reflecting the real suprasensible structure that somehow
lies behind them.2*
It is therefore worth underscoring what a remarkable phenome-
non the Laws is.26 The longest work of the philosopher who is com-
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 469
monly thought of as the arch-idealist among idealists is stuffed with
a mass of constitutional, administrative, legal, religious, and social
detail; and nowhere does it tell us, except in the foggiest terms, how
that detail is supposed to be related to metaphysical realities. We
ought therefore to wonder about the status of what we read. Plato
may have felt he had attained full metaphysical understanding, and
may therefore have framed the institutions of the Laws in the per-
fect confidence that they reflected metaphysical realities at a level of
approximation appropriate to the circumstances he envisaged in
Magnesia. But this seems improbable, on simple a priori grounds. It
is much more likely that in the Laws Plato formulates a set of
proposals in such a manner as he feels sure, from a combination of
experience and philosophical reflection, he would have to formulate
them, if only he did in fact have a full understanding of the Forms.
The Laws, on this view, is a work written on the basis of an incom-
plete understanding. Hence, though Plato is emphatic on some mat-
ters (on the tight connection between arete and eudaimonia, for
instance),2? on others he can be quite tentative.28 The Laws is not a
work that suggests its author is confident about everything.2?
That means, if we wish to understand what Plato is doing in the
Laws, to analyze his proposals as he himself would, and to judge
why he prescribed institutions, a, b, c, and not institutions x, y, z,
we have to do some work of our own; and that, of course, is what
Plato's dialogues always do require of us. Where may we start?
Observe first the very strongly marked hierarchical structure of
the state. The fiction of the Laws is that an Athenian Stranger
(clearly Plato himself), Cleinias (a Cretan), and Megillus (a Spartan)
fall to discussing plans for a colony (Magnesia) to be founded in the
South of Crete. In the course of Book IV*0 the Stranger invites his
companions to suppose that the colonists have arrived and are wait-
ing to be addressed. What they hear is an edifying protreptic on their
forthcoming religious duties. Here is another speech, which the
Athenian could have delivered but did not:
"Future members of Magnesia: This new state of yours is intended
to be a very good one. It will be administered not in the spirit of any
current ideology, but in accordance with the truth about the world.
After all, you would not wish to conduct your lives in any manner
that is not based on the truth, would you? Now you may not be
aware that philosophy has shown us that the truth is expressed by
Cambridge Companions Online © Cambridge University Press, 2006
47O THE CAMBRIDGE COMPANION TO PLATO
Forms - abstract ideas of things, very difficult to understand except
by intelligent people who have put much effort into the job. How-
ever, that effort has been made,- and you will find that tKe state you
are about to join is run largely by people who have attained some
insight into these things, and to a lesser degree by those who have
lesser insight, but who have accepted the doctrines taught them by
the others. It is all a question of reason: The more rational you are,
the more you can understand the truth, and the more you will have
authority in the state you are about to found."
Hence as one works one's way through the detail of Magnesia's
structure and administration, one becomes aware of an elaborate
hierarchy of knowledge, control, and influence. To speak very com-
pendiously, gods rule (or in some sense oversee) human beings, offi-
cials rule nonofficials, free human beings rule slaves, citizens rule
foreigners, the old rule the young; and the title of the rulers to rule is
some form of superior knowledge.*1
Some peculiarly Platonic features apart (e.g., the Nocturnal Coun-
cil and the ability of women to hold office), this is very much the
sort of pecking order that might appeal to a wide range of political
and social allegiances. Plato is in this instance working with the
grain: He finds it possible to use common Greek assumptions and
common Greek institutions for his own purposes, namely, to ensure
that reason and knowledge of the Forms, however dimly grasped by
the Nocturnal Council, and however imperfectly transmitted by
them to the rest of the state, have the maximum possible diffusion
in Magnesia.32 In this sense, what dictates the structure and adminis-
tration of Magnesia is, however remotely, the Forms.
But Plato clearly feels that his success in constructing a philo-
sophically based state will vary. Like the divine Demiurge ("crafts-
man") of the Timaeus, who in constructing the world has to make
the best of the material that lies to hand, Plato has to do what he can
with the whole array of contemporary institutions and laws, and
with his new colonists, who have a wide variety of assumptions,
expectations, and political, social, and religious beliefs." His job,
therefore, as a craftsman-lawgiver, becomes complex: It calls for
selection, judgment, and indeed ingenuity.
His first job is to select his human "material." Only those persons
who seem suitable will be admitted to the new state.34 Even then,
Plato wants to know as much about their habits and opinions as he
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 471
can: He has great mistrust of the secrecy inseparable from private
family life, and fears that all sorts of undesirable practices may go on
undetected. There is some evidence that he would have extended
the research, presumably legal and philosophical, for which he
pleaded (in terms that remind us of Socrates) in the Politicus,^ to
sociological investigation of the Magnesians' characters, practices,
and opinions. Before the craftsman can fashion, he must know his
materials.*6
As a result of such investigation, Plato will be able, in all realism,
to achieve some things but not others. He will aim as high as he can,
but will be careful to retain a fall-back position. This is the principle
of the sliding scale. It emerges in many contexts,^ but perhaps at its
most explicit in the matter of property. The ideal would be a total
community of possessions, which would discourage the evils associ-
ated with the institution of private ownership and foster the commu-
nity of sentiment that Plato wishes to achieve. But he knows it is
impractical; so the Magnesians will be permitted, within certain
limits, to own their own property. And he acknowledges that in
certain circumstances he may have to relax standards even further. &
But the obvious way to avoid having to relax standards is to ensure
that no member of the state should even desire it; and Plato accord-
ingly provides for a comprehensive program of persuasion and educa-
tion to make as sure as he can that the Magnesians' likes and dislikes
and reasoned convictions become what he wishes them to be. Mere
vocational training exists in the system, but the chief emphasis is
moral; and the Minister of Education provides a direct doctrinal/
philosophical link with the Nocturnal Council. The whole of Book
VII is devoted to education, but the topic recurs constantly through-
out the work; and the Laws itself is in effect claimed to be Magnesia's
best educational text.39
It is hardly necessary to give a resume of all Plato's educational
measures, for good accounts exist elsewhere.*0 But it is worth dwell-
ing on one central point. Magnesia's program of education is Plato's
bid for the hearts and minds of the Magnesians; and he is not pre-
pared to brook rivals. His intention is that Platonic tastes, Platonic
values, and Platonic ideology should be fully accepted by the Magne-
sians at large, not merely in conscious preference to other values,
but because other values are simply not there to compete; for the
notion that the world might be other than it is constitutes a danger.
Cambridge Companions Online © Cambridge University Press, 2006
472 THE CAMBRIDGE COMPANION TO PLATO
To be sure, he wants the Magnesians to understand their laws
gnomei, "by reason, judgment"; but only when a man has shown
himself to be beyond likely corruption is he allowed to travel abroad
and make principled comparisons with foreign institutions,- the Noc-
turnal Council has oversight of such excursions, and is encouraged
to have wider mental horizons.*1 But the general run of Magnesians
are to be totally immersed in Platonic values. Plato's pressures on
them are intense and relentless.
Let us glance at a few of his techniques. The most obvious is the
strict censorship of doctrinally undesirable artistic productions.*2
The most insidious is perhaps the use of the pleasure to be had from
the arts to recommend doctrinally correct tenets to the feelings.**
The best-known is the provision of dignified and edifying "pream-
bles" to the legal code in general and to the constituent parts of it,
designed, by a combined appeal to sensibility and reason, to per-
suade the bearer that the laws are to be obeyed.** And the most
elaborate is the full-dress refutation, occupying almost the whole of
Book X, of the three heresies that in Plato's view encourage us to
break the link between virtue and eudaimonia, a cardinal doctrines
All these and similar techniques amount to a most determined at-
tempt to ensure that the Magnesians think as he wishes. He is in a
sense cultivating the "consent of the governed" to his political struc-
tures; but it is perhaps truer to say that by exerting pressure he
forces this consent into existence.
Another excursus, less long but even more radical, concerns penol-
ogy.*6 It is addressed to the Magnesian citizens in their capacity as
jurors. Athenian courts followed a partly retributive and partly deter-
rent policy. Plato, however, is wholly utilitarian. He separates off
recompense to the injured party (which must be paid in every case)
as a nonpenal measure, designed to restore the two sides to friend-
ship. The purpose of punishment proper is not primarily to deter the
offender by the prospect of further suffering, but to cure him of his
psychic injustice, a "disease" that like medical disease he cannot
have chosen to contract. For bad moral states, like bad physical
states, are disadvantageous to their possessors; therefore, "no one is
evil voluntarily. "*7 Offenses are involuntary, and thus retributive
punishment is misdirected. The purpose of punishment ought to be
the "cure" of mistaken opinions and bad moral states. To achieve
that, any measure, whether painful or not, may be used. In order to
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 473
give effect to this policy, Plato writes a model penal code, based on
Athenian law but freely modified both in large matters and in count-
less significant details.*8
Here again there is a vigorous attempt to persuade the Magne-
sians to think Platonically.49 A modern law school tries to teach its
students not just the law but to think like lawyers; it inculcates a
cast of mind, not just a body of knowledge. Plato approaches his
task in the same way. The Magnesians jurors have to act in accor-
dance with special assumptions and policies. Instead of asking,
What does this convicted offender deserve to suffer retributively, in
the light of the damage or injury he has caused?, they must ask,
What does that damage or injury tell us about the state of his soul,
and what penalty does he deserve to undergo to cure it? "Deserve,"
axios, takes on a different meaning. So too do timoria, "ven-
geance," "punishment," and dike, justice," "punishment." Timoria
acquires a disreputable connotation: "backward-looking retributive
suffering that does the offender no good"; dike comes to mean
"punishment systematically calculated to reform him for the fu-
ture. "5° Familiar vocabulary is used, but it acquires a new content.
No wonder the penological excursus is prefaced by a long apologia
for philosophical legislation.
But it is time to turn from philosophical and educational theory to
more concrete matters. One enters Magnesia (visitors are not ex-
cluded^1,- one strolls around; what does one see?
One sees a state whose wealth comes from agriculture, not com-
merce. The land is divided into 5,040 inalienable "portions" or
"lots" (kleroi), one for each of the 5,040 adult male citizens,-*2 420
such lots constitute the territory of each of the twelve "tribes." On
the way to the central city, one notices that each tribal division has a
village. Dotted around are the dwellings of the citizen-farmers; and
there are various other buildings for administrative, legal, religious,
educational, and social purposes. One finds that the people one
meets fall into four social categories: the free citizens and their
wives and children; their slaves; resident aliens; and visiting aliens
of diverse kinds. After a while, two features of Magnesian life be-
come conspicuous: There is no undue poverty or wealth, and the
range of economic activity permitted to those of citizen rank is
strictly limited, since slaves do much of the manual work and trad-
ing and handicrafts are in the hands of resident aliens (metics).*^
Cambridge Companions Online © Cambridge University Press, 2006
474 T H E CAMBRIDGE COMPANION TO PLATO
All these provisions are the practical expressions of well-known
Platonic preoccupations. Wealth, poverty, trade, and handicrafts
tend to diminish a man's virtue and to unfit him for a civilized life;
they lead him to value the interests of the body over those of the
soul.54 The relationship of Plato's economic regulations to historical
practice is a fascinating topic, but too extensive for consideration
here.55 Suffice it to note that Plato manipulates: He grants with one
hand and claws back with the other. His central concern is to control
private life.*6 He grants private property - but wealth over a certain
level is subject to a tax of ioo percent.*? He allows private families
(they are indeed the bedrock institution of the entire state)*8 - but he
wants all citizens and their wives to attend communal meals orga-
nized on the Spartan/Cretan model.59
How is the state governed and administered? The Stranger de-
scribes its constitution as "midway" between monarchy and democ-
racy.60 Summarized and slightly expanded, his explanation is as
follows. Ideally, a state should be run by a single person of supreme
virtue and political wisdom. Such a person rarely or never exists,-
and in any case we shall always be faced with the democratic de-
mand for purely arithmetical61 equality of political power (one man,
one vote, in the modern expression). But the many are always less
capable of political wisdom;62 so the best way of framing a constitu-
tion is to combine two modes of distributing power: election,
which (if we are lucky) will enable the state at large to appoint to
office the (moderately numerous) persons best qualified by char-
acter and intelligence to hold it, and the lot, which guarantees some
political influence to the common man, however rudimentary his
political virtue. Political power will thus be diffused throughout the
state; but the use of the lot should be confined to the smallest
possible compass.
Hence the Stranger's task is to inject into the constitutional struc-
ture various means of maximizing the influence of those Magne-
sians who are more rational and virtuous than others. But does not
Magnesia have the same array of constitutional features as is com-
monly found in Greek states at large, and conspicuously in demo-
cratic Athens - an Assembly of adult male citizens, a Council, and
an Apparat of officials, not to speak of courts with large juries ensur-
ing popular participation in the administration of the law? Can the
Stranger manipulate such institutions in a Platonic direction? In-
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 475
deed he can; and his methods are full of interest. I instance four
only; close inspection of the fine print could well reveal others.
1. The Nocturnal Council, which is obviously intended to me-
diate philosophy to the state in the course of its strong influ-
ence on the state's day-to-day administration, is composed
of persons who have attained membership on the basis of
personal qualities only. As far as I can see, the lot plays no
part whatever in the sequence of events leading to their ap-
pointment; they are members either by virtue of some con-
spicuous attainment or ability or by virtue of some office to
which they have at some time been elected, not chosen by
lot to any degree at all.6* Here then, in such supreme gover-
nance as Magnesia has, election holds the field and the lot is
nowhere. The Nocturnal Council is in effect an intellectual
and moral aristocracy. As far as constitutional machinery
goes, it has only a slender connection with the rest of the
state; it does not have to seek approval for its decisions; and
much of its activity and influence is no doubt intended to be
informal, and perhaps covert.6*
2. In the election of the most powerful officials, the Guardians
of the Laws/5 the ten most senior of whom are members of
the Nocturnal Council, the number of candidates is whittled
down first to 300, then to 100, and finally to 37. Electors
who wish to vote in the final st#ge are obliged to confer
solemnity on their vote by walking between the victims of a
sacrifice. The effect will be to exclude from the final and
decisive vote a number of the poorer citizens, who cannot
afford the expense.66
3. The Assembly, to which all adult male citizens belong sim-
ply by having that status, and the Council, a select executive
body of 360 members, seem to have on the whole a more
restricted range of powers than in Athens.6? Two details are
to be noticed:
a. In the elections to the Council,68 all members of all four
property classes vote for ninety members from each
class. Voting is compulsory, on pain of a fine. But mem-
bers of the fourth (poorest) class need not take part in the
election of Councillors from the third class, and mem-
Cambridge Companions Online © Cambridge University Press, 2006
476 THE CAMBRIDGE COMPANION TO PLATO
bers of the fourth and third classes need not take part in
the election of the Councillors from the fourth class. In
both these elections there will thus be a preponderance
of "wealthy votes" over poor ones; for the well-off will
vote because of the fine, and the poorly off will tend not
to (no doubt they will be glad to get on with their work).
On the standard assumptions6* that there are fewer rich
than poor, or even if they are equal in numbers, there
will in this method of voting be a pronounced oligarchic
influence in the election of Councillors from among the
poor.
b. Similarly, members of the two wealthier classes are fined
if they fail to attend meetings of the Assembly, whereas
no such obligation is normally imposed on members of
the two poorest classes.7°
The upshot of (2) and (3) is a slightly curious one. To wisdom and
virtue as qualifications for political power we are to add wealth.
Plato perhaps feels that membership in one of the two higher prop-
erty classes is a sign of industry, thrift, and modest desires. These are
no small virtues; and since there is a strict upper limit on wealth in
Magnesia, they are unlikely to degenerate into greed. At any rate,
possession of wealth in Magnesia confers some modest political
clout. The concession to demotic virtue is a shrewd one, and goes
some way to meeting the oligarchic sentiment that some Magne-
sians may have, namely that wealth deserves power.?1
Although Plato talks in rather static terms of the Magnesian con-
stitution as "midway" between monarchy and democracy, what he
in effect does is to allow contrary pressures or tensions to conflict
dynamically. A certain shifting balance is struck not only between
the two conceptions of equality represented by the use of election
and the lot, and so between rich and poor/2 but (i) between an offi-
cial's freedom and discretion on the one hand, and on the other the
risk of prosecution for misconduct during his tenure^ and the neces-
sity to undergo "scrutiny" at the end of it,™ and (ii) between the
authority of boards of officials to take binding decisions on their
own, and their obligation to act in concert with other officials or
pass cases on to a superior authority.?* But we must not suppose that
this attempt to hold opposing tensions in equilibrium is either (a)
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 477
the modern doctrine of the separation of powers, legislature, execu-
tive, and judiciary each checking the other, or (b) a way of running
harmoniously a "mixed society," in which various ideologies or eth-
nic groups are accorded esteem in their own right, but have to coex-
ist indefinitely. Plato's playing off of one pressure against another
takes place within, and should serve to reinforce, what is intended in
time to be a single ideology, in which the ends of society will not be
in dispute. If that is the mark of a totalitarian society, then Plato is a
totalitarian.?6
Another balance he strikes is between fidelity to the letter of the
law and discretion in its application. Magnesia is a "law state," and
laws as instruments of government are imperfect: They cannot meet
the peculiar circumstances of every case, and need interpretation
and application by persons acting according to their spirit rather
than their bare wording. ~n Hence although Plato is emphatic that the
Magnesians should give the laws unconditional obedience, ?8 he is
well aware that they will have to exercise enlightened flexibility in
their day-to-day enforcement.^ In fact, one of the reasons why cer-
tain selected Magnesians are encouraged to travel abroad is that no
state that obeys its laws merely by habit without grasping them by
"understanding," "judgment," "reason" [gnomei) will in the end be
able to preserve them intact.80 The "observers," "ranging over land
and sea," are therefore to consult foreign experts and inquire into the
laws and jurisprudence of other countries, with a view to strengthen-
ing the legal system of Magnesia. Insofar as this is a philosophical
activity (questions of intent, responsibility, penology, and psychol-
ogy could presumably arise), Plato is trying to ensure that Magnesia
is founded on philosophical insight: The observers report to a full
meeting of the Nocturnal Council,81 which no doubt attempts to
feed into the legal code any new ideas and practices of which they
approve.82
Plato adopts a good deal of contemporary Athenian law and legal
practice, but in a decidedly critical spirit, and the list of modifica-
tions he makes is long and various. Apart from his radical new
penology, which we have already described, perhaps his most signifi-
cant innovations are (i) the institution of a less confrontational and
more interventionist and inquisitorial legal procedure than pre-
vailed in Athens,83 and (ii) appeal, not allowed in Athens, from the
verdicts of the "tribal" courts (the Magnesian counterparts of the
Cambridge Companions Online © Cambridge University Press, 2006
478 THE CAMBRIDGE COMPANION TO PLATO
regular Athenian courts, dikasteha) to a higher tribunal.8* As for his
other measures, I venture to quote again Morrow's splendid tribute:
The pattern he lays down is in the main the procedure of Athenian law, with
its freedom of prosecution and its rich variety of actions and remedies; but it
is Athenian law modified at many points in directions, we may say, that are
suggested by that law itself. In giving to the presiding magistrates power to
control the pleading and prevent the introduction of irrelevant and mislead-
ing matter, in introducing something like inquisition of witnesses and prin-
cipals, in excluding the opportunities for rhetorical jousting afforded by the
archaic challenge to the oath and the challenge to the torture, in enlarging
the range of competent witnesses and enforcing a litigant's right to compel
their assistance, in eliminating the oath of witnesses and principals, in
relying at all stages upon written documents, and in invoking the power of
the state to assist a litigant in enforcing a judgment obtained in court - in all
these provisions Plato's law, while still essentially Attic in character, em-
bodies a conception of the judicial process broader and more enlightened
than ever characterized Athenian practice at its best.8*
If we ask what specifically philosophical principles, or in particular
what connections with the Forms, lie behind this impressive collec-
tion of measures, it is once again hard to give ̂ n answer. For they are
reforms such as could perfectly well be arrived at empirically by any-
one concerned with the most efficient ways to ensure justice in the
courts. But perhaps it is this very notion of justice that gives us a clue.
As we have seen, Plato redefines it. The purpose of his legal code is not
to make a criminal suffer "justly/7 that is, reciprocally, but to treat
him in such a way that he will become "more just." Now the various
unjust states of offenders are hard to diagnose with precision, and
there is a constant risk that misdiagnosis will result in prescription of
the wrong cure. Hence Plato's close attention to calm, prolonged,
impartial, and scrupulously accurate legal procedure.86 For only if the
criminal is made "just," that is, virtuous, by "cure," will he have
correct moral belief or disposition and thus attain some measure of
human eudaimonia. Legal provisions are in the service of a penology
that depends on a certain philosophical position. So if we wish to
describe them as "philosophically grounded," it will be in some end-
related sense: they are the most efficient means by which a man
whose moral beliefs are (involuntarily) erroneous or whose desires are
(involuntarily) wrongly directed may have his psychic state diag-
nosed and then cured. Bad court procedure would be concerned with
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 479
something other than that diagnosis, and hence ultimately with
something other than human eudaimonia; hence the very severe
view Plato takes of the kind of pleading that perverts the work of the
courts8? - for it impedes human "happiness."
The same end-related view should be taken of the entire constitu-
tional and economic structure of Magnesia. Insofar as the Nocturnal
Council attains knowledge of the Forms, or the "genuine" divisions
of reality, and mediates it by a variety of methods to its fellow
citizens at large, its task can only be hindered by excessive poverty
and excessive wealth,88 or by indulgence in retail trade, or by the
temptations put in a man's way by (say) unlimited tenure of office or
the lack of scrutiny at its expiration. Economics and constitutional
law dance to the tune of moral theory. 89
So too do theology and religion, though at a higher intellectual
level. Most of Book X is an elaborate theodicy designed to persuade
Magnesians of the following propositions:
1. That the world is in some sense a rational construction, in
which thought, design, and calculation are "prior" to nature,
matter, and chance.
2. That the heavenly bodies, which move in regular and ra-
tional patterns, are gods.
3. That the gods are not indifferent to mankind; they have a
warm regard for the virtuous and a detestation of the wicked.
They reward the former and punish the latter; they are virtu-
ous themselves, and cannot be begged off by sacrifices and
prayers.
Strict laws are laid down for the repression of both impious actions
and heretical opinions.9° The Magnesians are to take part in a regular
series of religious ceremonies and the accompanying social celebra-
tions; worship at private shrines is forbidden,^ for it can all too easily
reinforce the belief that the gods can be bribed to overlook misdeeds.
The reason why Plato pays extraordinary attention to religious belief
and practice is his firm conviction that no one persuaded of proposi-
tions (1)-(3) above, and in particular (3), will go to the bad morally*2 in
the belief that eudaimonia is independent of aretes Hence he presses
theology and religion into the service of the moral, legal, and constitu-
tional orthodoxies of Magnesia. Book X is in fact his most sustained
attempt to mold the minds of the Magnesians into the shape he re-
Cambridge Companions Online © Cambridge University Press, 2006
48O THE CAMBRIDGE COMPANION TO PLATO
quires. It does of course contain much passionate protreptic; but its
ratiocination is more than elementary: There is an analysis of the
various kinds of motion, some discussion of how the heavenly bodies
propel themselves, and an important attempt to present eschatology
in terms of cosmic physics.94 Plato is only too aware that the Magne-
sians, and particularly potential heretics among them, have brains as
well as feelings.95
Plato's radicalism in confronting and manipulating historical insti-
tutions is by now, I hope, apparent. But his most radical step has yet to
emerge: It is to require that the legislator and statesman, in exercising
politike episteme, concern himself with the arete and hence eudaimo-
nia not merely of the adult male citizen, but of all other parts of the
population, namely, women, slaves, and foreigners. Even in these
days of women's liberation, it is still not generally known that in the
Laws Plato allows and indeed encourages women to take a large share
in public life. Not only are they to join in the program of athletic and
military training; not only are they to dine communally, like their
husbands, and be educated precisely as the men are in political vir-
tue;*6 they are actually t6 hold public office. True, they have to wait
until the age of forty to do so, presumably in consideration of their
careers as childbearers, whereas a male may hold office from the age
of thirty.97 But Plato is simply acting consistently with the principles
of the Republic: In the crucial respects, a woman's potential ability is
not inferior to a man's, and to refuse to use them is to waste half the
state's human resources.98 In the Republic there were accordingly
Philosopher-Queens as well as Philosopher-Kings.99 In the Laws, the
principle has been universalized: All citizen women, however unlike
Philosopher-Queens they may be, are to take part in public affairs.
One may wonder how often they would in fact be elected, given the
likelihood of male prejudice; perhaps Plato was covertly relying on
the results of theTuse of the lot to familiarize the practice. At all
events, it is important to realize that Plato is not actuated by any
special admiration for the female sex as such,100 nor by any sympathy
with the idea that women are or should be equal to men in every
social and legal particular, or with any such thing as "women's libera-
tion"; he regards the matter purely as one of efficient use of available
abilities.
Plato treats the handling of slaves as something of a problem.101
He admits that some slaves are sterling characters who do their
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 481
owners great good; but he is sharply conscious of the possibility of
insubordination or even revolt. He recommends that masters should
refrain from arrogantly ill-treating their slaves, but that they should
be decidedly firm with them, chastising them justly; one should
never be familiar with them, and everything said to them should be
an order. In the legal code, slaves are accorded less protection against
harm than free men, and some of the penalties visited on them for
serious offenses are of horrible savagery.102 In general, Plato thinks of
a slave as a creature with a low level of reason and moral virtue, and
hence of little capacity to live that rational and civilized life which
constitutes human eudaimonia. But he does not believe they have
no capacity for eudaimonia at all: One almost incidental remark
seems to indicate that although bringing up slaves properly is primar-
ily in their masters' interests, it is also in the interests of the slaves
themselves, that is, their eudaimonia.10* In a hard-nosed and calcu-
lating way, Plato brings the happiness of slaves within the benevo-
lent concern of political science.10* And to an even more attenuated
degree he is concerned with the virtue of foreigners too, both resi-
dent and itinerant: The former should possess sophrosune, "modera-
tion," "restraint"; and both, like the slaves, are capable of moral
improvement as a result of punishment for offenses.10*
This care for the admittedly limited arete and eudaimonia of the
subordinate sections of the population raises interesting questions
of theory. Is it simply instrumental, and intended merely to subserve
the eudaimonia of adult male citizens? Or is there the germ of the
notion that the legislator ought to have some universal concern for
all human beings, regardless of political status? Whatever its motiva-
tion and purpose, Plato's concern for the eudaimonia of the entire
state at least exists; and it is pregnant with possibilities for future
philosophies, notably Stoicism and Christianity.
After these heady speculations, it is time to tug ourselves back to
the central problem: Is the Laws a work of philosophy or not? We
may consider general principles first, then specific constitutional
and social detail; and essentially the same problem arises in both
cases.
(1) Obviously Magnesia is constructed on certain principles - for
example, moderation, compromise, the rule of law. But that does not
make the Laws a work of philosophy in any strong sense. For these
principles are pragmatic: One does not have to be a philosopher to
Cambridge Companions Online © Cambridge University Press, 2006
482 THE CAMBRIDGE COMPANION TO PLATO
excogitate them. They are of course open to philosophical analysis;
but they do not have to be based on any metaphysical belief in
suprasensible realities such as Forms. Admittedly, one possible way
of connecting moderation, compromise, and so on, to such realities
would be to treat them as examples of "due" or "appropriate" mea-
sure, which in the Politicus106 is "divided" from "relative" measure;
for as we have seen, Plato seems to think that "divisions" reflect
some higher structure of reality. But the nature of this relationship
remains obscure: The difficulty is that achieving "due" measure
looks like an entirely empirical exercise.
(2) So what relationship could there be between the detail of
Plato's concrete practical proposals and his metaphysics? Consider
the concept of the sliding scale. There is in the Laws a clear tension
between what Plato would like to prescribe and what he feels he can
achieve in practical terms - in a word, a tension between ideal and
real. Now our immediate temptation is to suppose that the ideal
must be Forms: Plato brings each institution of Magnesia as near as
possible to its perfect exemplar. That is, he has his eye constantly on
the Forms of (say) Court, Office, or Property Class, and seeks to
perform accurate "divisions" of such things for deployment in Mag-
nesia. IO? But if that is what he thinks he is doing, he keeps his modus
operandi veiled.
(1) and (2) sketch possibilities only. What is possible to say with
safety is that, if his concern is with Forms at all, it appears to be with
the Forms of moral virtues. In that case, the sliding scale of preferred
and less preferred institutions in Magnesia simply reflects a prag-
matic judgment about conditions facilitating an intellectual grasp of
those virtues, and hence the invariable practice of them.108
If we were to tax Plato on the point and ask him why this or that
institution of the Laws subserves this purpose, while some other does
not, he would make, I think, a twofold reply. "First, you are quite right
in supposing I am working on a purely pragmatic level, in this sense:
No man who is excessively poor or rich or money-loving or mean or
ambitious or belligerent can acquire virtue, even on the humdrum
level of 'demotic' virtue; for that he needs security, order, rational
cooperation with his fellow men, and leisure (not idleness on the one
hand nor luxury on the other). These are the sort of criteria on which I
have constructed Magnesia. But second, if my intentions are fulfilled,
and the Magnesians succeed under the guidance of the Nocturnal
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 483
Council in obtainiong some intellectual grasp of Virtue, it is surely
improbable that the kind of life they will then lead will be wholly
different from the one they have led so far - rather, their lives will
gradually improve in moral quality, as a result of their improved un-
derstanding. After all, there must be some relationship between 'ordi-
nary7 or customary virtue based on imperfect understanding induced
by myth and so forth, on the level of 'right opinion,' and a perfect grasp
of what Virtue Itself is, based on reason. The life you led when you
were in the process of improving your understanding of virtue is thus
an excellent preparation and foundation for the life you will lead
when you have obtained it.10? In that sense, the Laws is a work of
philosophy - in spite of some people who fondly think I have aban-
doned metaphysical idealism. This work is my final attempt to mark
out for mankind the path to eudaimonia; for as I say in it clearly and
often enough,110 no man without virtue can be happy. That is the
philosophy on which my Laws is based."
We are now in a position to face the question, What is the relation-
ship between the state Magnesia and the state Callipolis?111 Ex-
pressed in its sharpest form, my answer would be: There is no rela-
tionship. They are the same Platonic state - but placed at two points
on a single sliding scale of political maturity. Now a politically ma-
ture Platonic state is, essentially, one governed by persons with meta-
physical insights; and the hypothesis of Callipolis is that that kind of
rule is achievable. The hypothesis of Magnesia is that it has not yet
been achieved, and may indeed never be; nevertheless, Magnesia
contains as an integral part of itself machinery embodying a continu-
ing aspiration to it. It is crucial to this aspiration that the demotic
virtues possessed by the second and third classes of Callipolis be
fostered and stabilized in Magnesia as far as circumstances and the
character of the inhabitants permit: hence the elaborate social and
educational provisions that occupy so many pages of the Laws. In
this carefully prepared soil, the rule of philosophy may take root and
gradually grow. But even if the philosophical aspiration were to come
to nothing, at least a superior "law state" will have been constructed.
Plato's political theory, so often thought of as "top-down," is in
this important sense "bottom-up." Even in the Republic, it was a
worry that the conditions for the building of Callipolis would never
be ripe;112 and it seems that when Plato came to select states to
which he would be prepared to send political advisers from the Acad-
Cambridge Companions Online © Cambridge University Press, 2006
484 THE CAMBRIDGE COMPANION TO PLATO
emy, he rejected those where he felt conditions were unpropi-
tious.11* The whole purpose of the Laws is to produce a state in
which conditions would be very propitious indeed.
It is, I believe, in this positive, forward-looking, and aspirational
way that the Laws is interpreted best; and like Plato himself at the
end of many of his works, I will now tell a likely story. When he
wrote the Laws, he was well into his seventies. He knew he could
not live much longer. The search for the nature of virtue was in his
view still unfinished business. He therefore set out, in fine, vigorous
detail, a blueprint for a second-best ideal state, which would foster
the demotic virtues to the greatest extent he deemed practical; and
he directed the members of the Nocturnal Council toward further
philosophical inquiry, both by themselves and in consultation with
the Academy (there is a strong hint of help from that quarter)."4 Of
course, Plato could hardly have expected a body such as the Noctur-
nal Council to succeed on its own in a philosophical inquiry in
which the Academy had so far failed. But the final paragraph of the
Laws keeps the lines of communication open from the Academy to
the Council,"5 and so ultimately from the Academy to the world of
even plainer men. There is no dimming of his zeal for the arete and
eudaimonia of mankind; for that, after all, is what philosophy is for.
If this admittedly unprovable reconstruction is correct, then the
Laws has a good claim to be the most ambitious of all Plato's writ-
ings: It supplies a program of combined practical work and theoreti-
cal inquiry for the years after his death.
NOTES
1 Rep. 425a-427a; but laws of some kind there will certainly be: 458c,
5O2b-c.
2 In the Republic Plato is especially concerned with justice, that virtue
specifically relevant to social and political relationships,- but that does
not affect this very compendious account of his fundamental position.
3 I.e., Assistant Guardians (less highly trained philosophically), and the
so-called third class (without philosophical education).
4 The "popular" character of the Laws is the subject of Herwig Gorge-
manns, Beitrdge zur Interpretation von Platons Nomoi (Munich, i960).
5 Dates: Republic, probably 380s and/or 370s; Politicus, probably middle
or late 360s; the Laws, a jumbo-dialogue, presumably occupied Plato at
least intermittently for some years before his death in 347.
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 485
6 Pol. 292b-293C; cf. 296a//.
7 Pol. 287bff. W. K. C. Guthrie, A History of Greek Philosophy, vol. 5
(Cambridge, 1978), 188-9, provides a handy list of them. Plato would
have agreed with Talleyrand: "War is much too serious a thing to be left
to military men."
8 Phil. i6b-c.
9 Pol. 28 5d.
10 I cannot find that this point is made in so many words,- but it is hard to
see how the Statesman can carry out his task of "weaving" [Pol. 302b)
without performing many "divisions," notably of characters of men and
the offices they are to fill (311a). Cf. the final sentence of Julius M. E.
Moravcsik, "The Anatomy of Plato's Divisions," in Exegesis and Argu-
ment, ed. E. N. Lee, Alexander P. D. Mourelatos, and R. M. Rorty (Assen,
1973)/ 324-48: "Within the context of Plato's thought, the Method of
Division should be viewed as yet one more attempt to link the theoreti-
cal and the practical, and to insist that the latter can be solved success-
fully only if such solutions are based on an adequate conception of the
former."
n Guthrie, History of Greek Philosophy, 5:168, 172-3, 175-80. Con-
straints of space forbid a rehearsal of the crucial passages, which are Pol.
262a-b, 285d—286b (on which see G. E. L. Owen, "Plato on the Undepict-
able," in Exegesis and Argument, ed. Lee, Mourelatos, and Rorty [Assen,
1973]: and J. B. Skemp, Plato's Statesman, id ed. [Bristol, 1987], 241//),
286d8-9, 300C-C The problem is bound up with the wider and contro-
versial issue of what role, if any, the Forms play in Plato's later thought
in general, not merely in political theory; see G. E. L. Owen, "The Place
of the Timaeus in Plato's Later Dialogues," Classical Quarterly 3(1953):
79-95; and the huge later literature, summarized by Guthrie, History of
Greek Philosophy, 5: 243. On the ontological status of the Forms as-
sumed in "division," see Moravcsik, "Anatomy of Plato's Divisions."
12 Pol. 293a, 297b-c; cf. Laws 875C-d.
13 Pol. 300c. Six constitutions are then distinguished, apart from rule by a
statesman with politike episteme: (i-iii) with law, democracy (least
good), aristocracy (middlingly good), monarchy (best); (iv-vi) without
law, tyranny (worst), oligarchy (middlingly bad), democracy (least bad).
The key to the gradings is that democracy is least powerful for good or ill
(303a), because of the fragmentation of power among many people (pre-
sumably agreement and concerted action are hard to achieve); contrari-
wise, single rule is most powerful (7o8e-7i2a, in the course of a differ-
ent grading of constitutions, makes explicit the point about the relation-
ship between numbers and power).
14 Cf. Laws 734e-735a.
Cambridge Companions Online © Cambridge University Press, 2006
486 THE CAMBRIDGE COMPANION TO PLATO
15 The point of the myth of the Politicus (268d-274d) is that in the current
cosmic era mankind is on its own: Men, not gods, administer men's
lives (ij4<lff; cf. Laws 732e, 853c//). The value of practical experience is
recognized at Pol. 300b.
16 Pol. 294a//, 3ooc-d.
17 Laws 96ic.
18 The various categories of members are set out, with slight discrepancies,
at 95id//and 96ia-b.
19 96ic-968e describes the Council's functions and course of study. There
are adumbrations of it at 632c and 8i7e-8i8a.
20 965C2//. Presumably a preliminary answer is that the virtues are all
"one" in that they are all knowledge.
21 966c—968b; cf. 817e—818a, and indeed the whole of Book X.
22 See, e.g., the passages cited in note 28, and cf. R. F. Stalley, An Introduc-
tion to Plato's Laws (Oxford, 1983) 136, on 6306-631 a.
23 Useful discussions of the Forms in the Laws are V. Brochard, "Les Lois
de Platon et la theorie des Idees," in Etudes de philosophie ancienne et
de philosophie moderne, ed. V. Delbos (Paris, 1926), 151-68; Guthrie,
History of Greek Philosophy, 5:378-81 (both asserting their presence);
and Stalley, Introduction to Plato's Laws, 133-6 (skeptical). 967c seems
to refer to the use to be made of philosophy (if that is what "the Muse"
means: see Saunders, "Notes on the Laws of Plato," Bulletin of the
Institute of Classical Studies, supp. 28 [London, 1972], n. 10) to "arrive
at practices of moral character and at rules consistently." He who can do
this has more than "the demotic virtues" (968a2).
24 Phil. 6i2iff even toys with the notion that a man could know the Forms
but have an inadequate grasp of their instantiations in this world. Nor
does Laws 967c tell us how philosophy is used to arrive at moral rules
(cf. note 23).
25 Pol. 262a//, 285a—b; cf. Phdr. 265c and note 28.
26 I cannot go into its literary merits and demerits. All human life is in it,
and one can perhaps catch something of Plato himself: the benevolent
testiness of 76iC5-d3 (did he enjoy a hot bath, and had he suffered from
incompetent doctors?) and the almost comic outrage of 9i8e-9i9b (had
he been stung by innkeepers?).
27 Laws 66od-663d, 732c—734c, 743c; cf. the unyielding assertiveness of
86od//, on "no one is evil voluntarily" (on which more later in this
chapter).
28 For instance, there is hesitation reminiscent of the Politicus in the diffi-
culties over division into precise categories at S66djff. The division he
eventually adopts is "closest to the truth" (not just "most convenient"
or the like, 867b3). Contrast 86ib, where Plato is certain that existing
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 487
legislators have divided wrongly. On the other hand, he is not at all sure
how to distinguish magic from medicine, 932c//.
29 859b: "We are turning into legislators, but are not so y e t " - a remark
that is (perhaps) a disclaimer of the status of a "true" legislator.
30 7i5e#
31 I forbear to list the dozens of passages in the Laws necessary for setting
out the hierarchy in its full complexity: They would amount to a large
proportion of the text. As to knowledge, see esp. 690b and 875C-d, in
addition to the description of the Nocturnal Council, 95 icff, 960//.
32 964a//.
33 858a-c, yoye—yoSG) see Glenn R. Morrow, "The Demiurge in Politics:
the Timaeus and the Laws," Proceedings of the American Philosophical
Association 27 (1953-4): 5-23; Andre Laks, "Raison et plaisir: pour une
characterisation des Lois de Platon," in La naissance de la raison en
Grece, ed. J. F. Mattei (Paris, 1990), 291-303; and Laks, "Legislation and
Demiurgy: On the Relationship between Plato's Republic and Laws,"
Classical Antiquity 9 (1990): 209-29.
34 736a-c.
35 Pol. 299b-e ; cf. Phaedrus on the orator's need to "know souls" (269c//)
and Laws 738c See Saunders, " T h e RAND Corporation of Antiquity'?
Plato's Academy and Greek Politics," in Studies in Honour of T. B. L.
Webster, ed. J. H. Betts, J. T Hooker, and J. R. Green (Bristol, 1986), 1:207-
8 and refs. Reporting on opinion and practice in the state is presumably
one of the functions of the promising young men who are the "eyes" of
the Nocturnal Council (9646-965a). We have to realize, of course, that
information gathering is, for Plato, not simply an exercise of academic
sociology, much less undertaken in order to respond to whatever "public
opinion" wants; his clipboard is in the service of political control.
36 Cf. the Stranger's topographical inquiries about Magnesia at the start of
Book IV.
37 See, e.g., 84ia-842a, on high and less high sexual standards, and cf.
858a.
38 739a-e; but the whole sequence from 736c to 746d should be read.
39 8nc-8i2a.
40 Glenn R. Morrow, Plato's Cretan City: A Historical Interpretation of
the Laws (Princeton, 1900), 297-398; Stalley, Introduction to Plato's
Laws, 123-36.
41 951 a//; cf. later in this chapter.
42 802a//, 8i7a-e, 858C-C
43 65 3b, and Book II in general.
44 7i8a-723d. The language of the preambles is somewhat heightened, and
there is frequent use of colorful myth.
Cambridge Companions Online © Cambridge University Press, 2006
488 THE CAMBRIDGE COMPANION TO PLATO
45 See further discussion near the end of this chapter.
46 856e-864C, a difficult text, elucidated in Saunders, Plato: The Laws
(Harmondsworth, 1970), 361, 367-9; cf. 933e//.
47 86od; cf. 734b. The formulation is a version of the "Socratic" paradox,
"no one does wrong voluntarily." For early formulations of this paradox,
see Prt. 345d-e and Grg. 509c
48 This is the subject of my Plato's Penal Code (Oxford, 1991). Plato's
modifications are rarely evident from a mere reading of the Laws in
isolation from the sources for Greek law.
49 Note how they are encouraged to attend trials: 85sd. Good law edu-
cates: it is an expression of reason (957c; cf. 714a 1-2, 857c-e).
50 728c, as elucidated by Saunders, "Notes on the Laws, n. 23; and
Saunders, "On Plato, Laws 728bc," Liverpool Classical Monthly 9
(1984): 23—4. Cf. Mary Margaret Mackenzie, Plato on Punishment
(Berkeley, 1981), 196.
51 95 2d//.
52 A number convenient for administrative purposes: 7376-738a.
53 This paragraph extracts salient points from 736c-738a, 7 396-741 a,
744a-745e, 778a-779d, 794a, 8420-850^ 9i9d-92oc, which should be
read in extenso. As to trade, though Magnesia is not a commercial state,
it recognizes the necessity of a certain minimum of trade internal to
itself; see the stringent regulations at 9i5C-922a. The lot is not only
inalienable, but must always retain a certain minimum of property
(744d-e); poverty below this level is not permitted.
54 63ib-d, 66ia-e, 726e//, 741c
55 On family law, see Morrow, Plato's Cretan City; Josef Bisinger, Der
Agrarstaatin Platons Gesetzen (Wiesbaden, 1925); E. Klingenberg, Pla-
tons VOJXOI yscogyixoi und das positive griechische Recht (Berlin, 1976);
and Walter G. Becker, Platons Gesetze und das griechische Familien-
recht (Munich, 1932).
56 Cf. 779d-78oa, 788a-c, 9O9d-9io.
57 744e-745b. There are also strict rules of bequest and inheritance, 923a//.
58 See, e.g., 7170-718^ 729c, 77ia-776b, 783b-78sb, 84ic-842a.
59 779d-78id. See Morrow, Plato's Cretan City, 393-8 for the interpreta-
tion of this section. The inclusion of women is, as the Stranger indicates,
a sensational innovation,- see in general E. David. "The Spartan Sussitia
and Plato's Laws, " American Journal of Philology 99 (1978): 486-95,
60 756e//; cf. Pol. 3ooe// and note 13; he means in effect an aristocracy. On
7o8e//, where for the founding of the state Plato prefers a good single
ruler, see Stalley, Introduction to Plato's Laws, 90-2; cf. also 691b//on
"middle" or "mixed" constitutions in history.
61 On "arithmetical" as opposed to "geometrical" equality (inequality
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 489
granted to imequals) see also Aristotle, Politics III ix, xii; V i, and the
illuminating article by F. David Harvey, "Two Kinds of Equality/' Clas-
sica et Mediaevalia 16 (1965): 101-46.
62 Pol. 292e-293a, 297b; cf. Rep. 494a.
63 Admittedly, the lot may enter into the appointment of priests (759b-c)
and certain priests are indeed members of the Nocturnal Council
(95id8). But priesthood is not in itself a qualification for membership:
awards of merit are necessary. This requirement may point to the high
officials known as Scrutineers (of other officials at the end of their term,
946b), who may in exceptional circumstances be elected by lot (946a//);
but it is not clear that Scrutineers are priests, nor whether, if so, they
automatically count as having won merit awards. Cf. Morrow, Plato's
Cretan City, 503.
64 964e-965a; cf. 758cs-d2. I therefore find it hard to accept Charles L.
Griswold, Jr.'s recent conclusion, which in any case conflicts with 756e,
that "politike episteme and the virtues will best flourish in the context
of a democracy [italics added] ruled by law.;/ See Griswold, "politike
episteme in Plato's Statesman," in Essays in Ancient Greek Philosophy,
vol. 3, ed. John P. Anton and Anthony Preus (Albany, 1989), 162.
65 753^//. No doubt this title is intended to be reminiscent of the Guardians
of the Republic.
66 Cf. Rep. 378a.
67 Morrow, Plato's Cretan City, 157-78.
68 756b-e.
69 Aristotle, Politics i 2 8 i a n - 2 8 , I3i8a26. Even if all members of all
classes elect from all classes compulsorily, a small first (rich) class still
has ninety Councillors, just as many as a large fourth (poor) class.
70 764a.
71 744a//; cf. Aristotle, Politics I283ai6//, 130^25//; cf. i28oa22-5.
72 Compare the mixing of social classes in the arrangement of marriages,
773a-e.
73 76ie/7, 92.8b, 941a, 946d-e, 947e-948b, 955C-d.
74 Office holding in Magnesia, as in Athens, is generally limited to a set
number of years. On the scrutiny, euthuna, an Athenian practice that
Plato adopts enthusiastically, see 945b//.
75 E.g., 76id-e, 847b.
76 The world's experience of Nazism, fascism, and Communism prompted
from the 1930s onward a series of attacks on Plato's political thought on
grounds of totalitarianism. The most vigorous assault was by K. R. Pop-
per, The Open Society and its Enemies, vol. 1 (New York, 1963); the
fullest defense was by Ronald B. Levinson, In Defense of Plato (Cam-
bridge, Mass., 1953).
Cambridge Companions Online © Cambridge University Press, 2006
49O THE CAMBRIDGE COMPANION TO PLATO
77 875c!; cf. earlier in this chapter.
78 700a, 7i5b-d, "servants of the laws"; cf. 957b. Law is after all "the
distribution of reason" (y 14.^2).
79 92 5df, and cf. their discretion in the fixing of punishments, discussed
earlier in the chapter. The discretion is, of course, not unfettered, but
subject to firm guidelines.
80 95ia-c, and note the description of virtue as the harmony of emotion
and reason (653a-C; cf. 688b).
81 952b.
82 Just how they are to do so is, however, not clear. Plato is reluctant to
countenance frequent changes to Magnesian law, since it is already very
good (and probably also because such changes suggest a state unsure of its
moral aims). But he allows refinement of them, especially in the early
years of the state, at least in some areas (769b//, 722a-d, 957a-b). No
doubt principled adjustments suggested by the Nocturnal Council would
be more acceptable to him than ad hoc tinkering. See Morrow's discus-
sion, Plato's Cretan City, 500-3; cf. Pol. 299b//on excessive fidelity to
existing laws, and the need for zetein, "inquiry," "research."
83 766d, 768a, 875e-876b, 855c-856a; cf. Auguste Dies and Louis Gernet,
"Introduction," Platon: Oeuvres completes, vol. 2, part 1, Les Lois
(Paris, 1951), cxl-cxliv.
84 768b-c, 956c.
85 See Saunders, Plato: The Laws, 31-2; Morrow, Plato's Cretan City,
295-6.
86 766d-768a, 855C-856a, 875e-876e, 936e-938c, 949a-b, 956d-958a.
87 937<l-938c.
88 Cf. 679D-C. Note how the discussion of these states at 742c// is related
to the "starting-point" and wish" of the legislator (on which see also
770c//). There is no asceticism here, no belief that poverty ennobles the
soul. Cf. Benjamin Franklin: "[I inserted in my Almanac] proverbial
sentences, chiefly such as inculcated industry and frugality as the means
of procuring wealth, and thereby securing virtue; it being more difficult
for a man in want to act always honestly, as, to use here one of those
proverbs, it is hard for an empty sack to stand upright" [The Autobiogra-
phy of Benjamin Franklin, chap. 7; his italics).
89 See in particular 742c-744a and yjob-yjia, and cf. 802C2-4 on the
control of dancing, etc.
90 9O7b-9iod.
91 ji6dff, 8o3ei-2, 8i6c-d, 828a-d,
92 885b.
93 905b.
94 903b//.
Cambridge Companions Online © Cambridge University Press, 2006
Plato's later political thought 491
95 See 885c//, esp. 886d sophoi, "clever," and 908c His description of the
euetheia, "naivete," of early men, who believed what they were told, is
a real cri de coeur (679c).
96 77oc-d, 780a//, 796c, 8o4d//, 833C-d.
97 785b. It is slightly curious that we hear directly of the provision only
this once. No song-and-dance is made about it (contrast the cases of the
communal meals, education, and the army). But David Cohen's review
of the indirect evidence ("The Legal Status and Political Role of Women
in Plato's Laws," Revue Internationale des Droits de l'Antiquite 34
[1987]: 27-40) convinces me that this passage is to be taken seriously.
Yet what voting rights they have is rather less clear. I add one more
indication that Plato regarded female virtue as not essentially inferior
to male: In his penal code, penalties are frequently graded in severity on
the basis of the social class (free, slave; citizen, alien) of the offenders,
and therefore the presumed level of their virtue, but never by reference
to their sex (except the one special case at 932b-c, on the basis of the
same age distinction as at 805e). For a full review of the evidence for
Plato's treatment of women in the Laws, and a less favorable assess-
ment than mine, see Susan Moller Okin, Women in Western Political
Thought (Princeton, 1979).
98 Laws 805a, 806c.
99 Rep. 45 ib// may well imply that in Callipolis women of the second and
third classes will be doctors, soldiers, etc., on an equality with men.
100 See Laws 78ia-b, Rep. 455d-e.
101 The main discussion is at Laws 776b-778a.
102 On slaves in the Laws, the classic investigation is Glenn R. Morrow,
Plato's Law of Slavery in its Relation to Greek Law (Urbana, 111., 1939).
103 777^: "[Masters should] train them properly, according them respect not
only for their (i.e., the slaves') sakes but even more for their own." Good
training of slaves conduces to masters' eudaimonia, which is dimin-
ished if their slaves do not enjoy such eudaimonia as they are capable of;
for then they become rebellious. Refraining from ill-treatment of slaves
will "sow the seeds of virtue" (77761) - in slaves? or in their masters,
who thereby avoid sullying their own souls?
104 Cf. Morrow, Plato's Law of Slavery, 43-4.
105 85oa-d, 941c
106 Pol. 283b-287b.
107 See note 10.
108 This position is in many ways Aristotle's too: The good polis provides
the material environment, and social and political structure, in which
human arete, and hence human eudaimonia, can flourish.
109 At 967c—968a the members of the Nocturnal Council are required to
Cambridge Companions Online © Cambridge University Press, 2006
THE CAMBRIDGE COMPANION TO PLATO
have certain attainments "in addition to the demotic virtues": Some
sort of progress from a lower level of virtue to a higher seems to be
envisaged.
n o He means, e.g., 66od—663d.
i n For further discussion of this important question, see Laks, "Raison et
plaisir," and Laks, "Legislation and Demiurgy."
112 Rep. 471C-474I); cf. 54od—541b.
113 E. g., Diogenes Laertius, III 23, and in general Saunders, "The RAND
Corporation of Antiquity?/' 1:202-3 and refs. The evidence for the
Academy's interventions in Greek politics is, however, of uncertain
reliability. Some of it comes from the Letters that pass under Plato's
name; but some of them are, and all of them may be, spurious. Even so,
one should read at least the Seventh and Eighth Letters; there are sev-
eral points of contact with the Laws (see Glenn R. Morrow, Plato's
Epistles, 2nd ed. [Indianapolis, 1962]). On the relevance to his political
theory of Plato's own attempts at political reform in Sicily, see my
translation, Plato: The Laws, 27-8, 545-7.
114 Laws 968b.
115
Cambridge Companions Online © Cambridge University Press, 2006
BIBLIOGRAPHY
All works cited in this volume, together with some additional material, are
listed below under one or more headings. Editions of the Greek texts of
individual works of Plato are classified with the secondary literature on
those texts, as are translations of individual works with commentary. The
following headings are used:
I. Plato
A. Comprehensive Greek texts
B. Comprehensive translations
C. Background to Plato: History and culture
D. Background to Plato: Philosophy and science
E. General discussions of Plato's thought
F. Methods of interpretation
G. Chronological and linguistic studies
H. Socrates and the early dialogues
I. Metaphysics and epistemology
J. Ethics, political philosophy, and moral psychology
K. Art and poetry
L. Eros
M. Mathematics
N. Studies of single works
i. Meno
i. Phaedo
3. Symposium
4. Republic
5. Phaedrus
6. Parmenides
7. Theaetetus
8. Timaeus
9. Sophist
493
Cambridge Companions Online © Cambridge University Press, 2006
494 Bibliography
10. Statesman
11. Philebus
12. Laws
13. Letters
O. Oral doctrines
II. Aristotle and later classical thought
III. Platonism after Plato
IV. Miscellaneous
A. Editions, transmission, and study of ancient texts
B. Modern and contemporary works
I. PLATO
A. COMPREHENSIVE GREEK TEXTS
A commonly used edition of the entire corpus is:
Burnet, John. Platonis Opera. 5 vols. Oxford: Clarendon Press, 1900-7. (Of-
ten called the Oxford Classical Text [O.C.T.].)
There is also a twelve volume edition of Plato's works (called the Loeb
Classical Library) with Greek text and English translation (by various
hands) on facing pages published in Cambridge, Mass., by Harvard
University Press and in London by William Heinemann (reprinted
1961-84). Similarly, there is a fourteen-volume Greek-French edition
by various hands (published "sous le patronage de VAssociation Guil-
laume Bude" and hence referred to as the Bude edition) published in
Paris by Societe d'Edition, Les Belles Lettres, 1951-64.
B. COMPREHENSIVE TRANSLATIONS
New translations into English, many of them accompanied by com-
mentary, emerge nearly every year. In addition, one can generally rely
on (1) translations of Plato's works by various hands published by
Penguin Books (Harmondsworth, England) and by Hackett Publishing
Company (Indianapolis). The translations in the Loeb Classical Li-
brary and the Bude edition mentioned above are also worth consulting
in many cases. Careful study of particular passages is often aided by
consulting different translations. A widely used and convenient one-
volume translation by various hands of all of the dialogues (except
those of questionable authenticity) and letters can be found in:
Hamilton, Edith, and Cairns, Huntington, editors. The Collected Dialogues
of Plato. New York: Pantheon, 1961. Princeton: Princeton University
Press, 1971.
Cambridge Companions Online © Cambridge University Press, 2006
Background to Plato: History, culture 495
A multivolume translation, widely used for nearly a century and still
worth consulting, is:
Jowett, Benjamin. The Dialogues of Plato. 3d ed. 5 vols. Oxford: Clarendon
Press, 1892. 4th ed. 4 vols. Oxford: Clarendon Press, 1953.
An excellent translation of the entire corpus by a single hand is under
way, beginning with:
Allen, R. E. The Dialogues of Plato. Vol. 1. New Haven: Yale University
Press, 1984. (Contains Euthyphro, Apology, Crito, Meno, Gorgias, and
Menexenus.)
C. BACKGROUND TO PLATO! HISTORY AND CULTURE
Adkins, Arthur W. H. Merit and Responsibility: A Study in Greek Values.
Oxford: Clarendon Press, i960.
"Homeric Values and Homeric Society." Journal of Hellenic Studies 91
(1971): 1-14.
"Merit, Responsibility, and Thucydides.;/ Classical Quarterly 25 (1975):
209-20.
"Problems in Greek Popular Morality/7 Classical Philology 73 (1978):
143-58.
Borgeaud, Phillipe. The Cult of Pan in Ancient Greece. Chicago: University
of Chicago Press, 1988.
Bremmer, fan. The Early Greek Concept of the Soul. Princeton: Princeton
University Press, 1983.
Burkert, Walter. Orphism and Bacchic Mysteries: New Evidence and Old
Problems of Interpretation. Berkeley: Center for Hermeneutical Studies
in Hellenistic and Modern Culture, 1977.
Homo Necans: The Anthropology of Ancient Greek Sacrificial Ritual and
Myth. Berkeley: University of California Press, 1983.
Greek Religion. Translated by John Raff an. Cambridge, Mass.: Harvard
University Press, 1985.
Ancient Mystery Cults. Cambridge, Mass.: Harvard University Press,
1987.
Cohen, David. "Law, Society, and Homosexuality in Classical Athens/7 Past
and Present 117 (1987): 3-21.
Cole, Susan Guettel. "New Evidence for the Mysteries of Dionysos.77 Greek,
Roman and Byzantine Studies 21 (1980): 223-38.
Creed, J. L. "Moral Values in the Age of Thucydides.77 Classical Quarterly
23(1973): 213-31.
de Romilly, Jacqueline. Magic and Rhetoric in Ancient Greece. Cambridge,
Mass.: Harvard University Press, 1975.
Cambridge Companions Online © Cambridge University Press, 2006
496 Bibliography
de Ste. Croix, G. E. M. The Class Struggle in the Ancient Greek World from
the Archaic Age to the Arab Conquests. Ithaca, N.Y.: Cornell Univer-
sity Press, 1981.
Detienne, Marcel. Dionysos Slain. Baltimore: Johns Hopkins University
Press, 1979.
Deubner, L. Attische Feste. Berlin: H. Keller, 1932.
Dodds, E. R. The Greeks and the Irrational. Berkeley: University of Califor-
nia Press, 1951.
"The Religion of the Ordinary Man in Classical Greece/' In The Ancient
Concept of Progress and Other Essays on Greek Literature and Belief,
edited by E. R. Dodds, 140-55. Oxford: Clarendon Press, 1973.
Dover, K. J. Greek Popular Morality in the Time of Plato and Aristotle.
Oxford: Basil Blackwell, 1974.
Greek Homosexuality. Cambridge, Mass.: Harvard University Press, 1978.
Easterling, P. E., and Muir, J. V., editors. Greek Religion and Society. Cam-
bridge: Cambridge University Press, 1985.
Ehrenberg, Victor. From Solon to Socrates: Greek History and Civilization
during the Sixth and Fifth Century B.C. id ed. London: Methuen, 1973.
Finley, M. I., editor. The Legacy of Greece: A New Appraisal. Oxford: Claren-
don Press, 1981.
Fontenrose, Joseph. The Delphic Oracle. Berkeley: University of California
Press, 1978.
Foucault, Michel. Uusage des plaisirs. Paris: Gallimard, 1984.
Graf, Fritz. Eleusis und die orphische Dichtung Athens in vorhellenist-
ischer Zeit. Berlin: Walter de Gruyter, 1974.
Grote, George. A History of Greece. 6th ed. 10 vols. London: Dent, 1888.
Guthrie, W. K. C. The Greeks and their Gods. London: Methuen, 1950.
Havelock, Eric. Preface to Plato. Cambridge, Mass.: Harvard University
Press, 1963.
Heinrichs, Albert. "Changing Dionysiac Identities/' In Jewish and Chris-
tian Self Definition, edited by Ben F. Meyer and E. P. Sanders. Vol. 3.
London: SCM Press, 1982.
Hornblower, Simon. The Greek World 479-322 BC. London: Methuen,
1983.
Hussey, E. L. "Thucydidean History and Democritean Theory." In Crux
(Essays Presented to G.E.M. de Ste Croix), edited by Paul Cartledge and
F. D. Harvey, 118-38. London: Duckworth, 1985.
Jones, A. H. M. Athenian Democracy. Oxford: Basil Blackwell, 1957.
Kidd, Ian. "The Case of Homicide in Plato's Euthyphro." In Owls to Athens:
Essays on Classical Subjects presented to Sir Kenneth Dover, edited by
E. M. Craik, 213-22. Oxford: Clarendon Press, 1990.
Kitzinger, Rachel. "Alphabets and Writing." In Civilization of the Ancient
Cambridge Companions Online © Cambridge University Press, 2006
Background to Plato: History, culture 497
Mediterranean, edited by Michael Grant and Rachel Kitzinger, 1: 397-
420. New York: Scribner's, 1988.
Lasserre, Francois. La figure d'Eros dans la poesie Grecque. Lausanne:
Imprimeries Reunies, 1946.
Lewis, David. "The Political Background of Democritus." In Owls to Ath-
ens: Essays on Classical Subjects presented to Sir Kenneth Dover, ed-
ited by E. M. Craik, 151-4. Oxford: Clarendon Press, 1990.
Lloyd-Jones, Hugh. The Justice of Zeus, id ed. Berkeley: University of Cali-
fornia Press, 1983.
Long, A. A. "Morals and Values in Homer." Journal of Hellenic Studies 90
(1970): 121-39.
MacDowell, Douglas M. Andokides: On the Mysteries. Oxford: Oxford Uni-
versity Press, 1962.
The Law in Classical Athens. London: Thames and Hudson, 1978.
Mikalson, Jon D., Athenian Popular Religion. Chapel Hill: University of
North Carolina Press, 1983.
Mosse, Claude. Athens in Decline 404-86 B.C. Translated from the French
by Jean Stewart. London: Routledge & Kegan Paul, 1973.
Mylonas, G. Eleusis and the Eleusinian Mysteries. Princeton: Princeton
University Press, 1961.
Nilsson, Martin. Greek Folk Religion. New York: Columbia University
Press, 1940.
Greek Piety. Oxford: Oxford University Press, 1948.
A History of Greek Religion. Oxford: Oxford University Press, 1952.
Ober, Josiah. Mass and Elite in Democratic Athens: Rhetoric, Ideology, and
the Power of the People. Princeton: Princeton University Press, 1989.
Parke, H. W. Greek Oracles. London: Hutchinson, 1967.
The Oracles of Zeus. Cambridge: Harvard University Press, 1967.
Parke, H. W. Festivals of the Athenians. Ithaca, N.Y.: Cornell University
Press, 1977.
Parker, Robert. "Greek Religion." In The Oxford History of the Classical
World, edited by John Boardman, Jaspar Griffin, and Oswyn Murray,
254-74. Oxford: Oxford University Press, 1986.
Miasma: Pollution and Purification in Early Greek Religion. Oxford: Cla-
rendon Press, 1983.
Powell, C. "Religion and the Sicilian Expedition." Historia 28 (1979): 15-31.
Richardson, N. J. "Early Greek Views about Life after Death." In Greek
Religion and Society, edited by P. E. Easterling and J. V. Muir. Cam-
bridge: Cambridge University Press, 1985.
Roberts, J. W. City of Sokrates: An Introduction to Classical Athens. Lon-
don: Routledge &. Kegan Paul, 1984.
Samuel, Alan E. "Calendars and Time-Telling." In Civilization of the An-
Cambridge Companions Online © Cambridge University Press, 2006
498 Bibliography
cient Mediterranean, edited by Michael Grant and Rachel Kitzinger,
1:389-95. New York: Scribner's, 1988.
Scully, Vincent. The Earth, the Temple, and the Gods: Greek Sacred Archi-
tecture. New Haven: Yale University Press, 1979.
Snell, Bruno. The Discovery of the Mind in Greek Philosophy and Litera-
ture. New York: Dover, 1982.
Solmsen, Friedrich. Intellectual Experiments of the Greek Enlightenment.
Princeton: Princeton University Press, 1975.
Stephens, Susan A. "Book Production/7 In Civilization of the Ancient Medi-
terranean, edited by Michael Grant and Rachel Kitzinger, 1:421-36.
New York: Scribner's, 1988.
Taylor, C. C. W. "Popular Morality and Unpopular Philosophy." In Owls to
Athens: Essays on Classical Subjects presented to Sir Kenneth Dover,
edited by E. M. Craik, 233-43. Oxford: Clarendon Press, 1990.
Walbank, F. W. "The Problem of Greek Nationality." In Selected Papers:
Studies in Greek and Roman History and Historiography, chap. 1. New
York: Cambridge University Press, 1985.
D. BACKGROUND TO PLATO! PHILOSOPHY AND
SCIENCE
Barnes, Jonathan. The Presocratic Philosophers. 2 vols. London: Routledge
& Kegan Paul, 1979.
Cope, E. M. "The Sophists." Journal of Philology 1 (1854): 145-88.
Cornford, Francis MacDonald. Principium Sapientiae. Cambridge: Cam-
bridge University Press, 1952.
Dicks, D. R. Early Greek Astronomy to Aristotle. Ithaca, N.Y.: Cornell
University Press, 1970.
Diels, Hermann, and Kranz, Walther, editors. Die Fragmente der Vorsokrat-
iker, 6th ed. Berlin: Weidmann, 1952.
Furley, David J. "Anaxagoras in Response to Parmenides." In New Essays on
Plato and the Pre-Socratics, edited by R. A. Shiner and J. King-Farlow.
Canadian Journal of Philosophy, supplementary vol. 2 (1976): 61-85.
The Greek Cosmologists. Vol. 1, The Formation of the Atomic Theory
and its Earliest Critics. Cambridge: Cambridge University Press, 1987.
Furth, Montgomery. "Elements of Eleatic Ontology," Journal of the History
of Philosophy 6 (1968): 111—32. Reprinted in The Pre-Socratics: A Col-
lection of Critical Essays, edited by Alexander P. D. Mourelatos, 241-
70. Garden City, N.Y.: Anchor Books, Doubleday, 1974.
Gerson, Lloyd. God and Greek Philosophy. London: Routledge, 1991.
Gillispie, Charles Coulston, editor. Dictionary of Scientific Biography. 16
vols. New York: Scribner's, 1970-80.
Cambridge Companions Online © Cambridge University Press, 2006
General discussions of Plato's thought 499
Heath, Thomas. A History of Greek Mathematics. Vol. 1. Oxford: Claren-
don Press, 1921.
Irwin, Terence. Classical Thought. Oxford: Oxford University Press, 1989.
Jaeger, Werner. The Theology of the Early Greek Philosophers. Oxford: Ox-
ford University Press, 1947.
Kahn, Charles H. "Pythagorean Philosophy Before Plato/7 In The Pre-
Socratics: A Collection of Critical Essays, edited by Alexander P. D.
Mourelatos, 161-85. Garden City, N.Y.: Anchor Books, Doubleday,
1974-
The Art and Thought of Heraclitus. Cambridge: Cambridge University
Press, 1979.
Kerferd, G. B. The Sophistic Movement. London: Cambridge University
Press, 1981.
Kirk, G. S., Raven, J. E., and Schofield, M. The Presocratic Philosophers, id
ed. Cambridge: Cambridge University Press, 1983.
Lloyd, G. E. R. The Revolutions of Wisdom: Studies in the Claims and
Practice of Ancient Greek Science. Berkeley: University of California
Press, 1987.
Morrison, J. S. "Antiphon." In The Older Sophists: A Complete Translation
by Several Hands of the Fragments in Die Fragmente der Vorsokratiker,
edited by Diels-Kranz, edited by Rosamund Kent Sprague. Columbia:
University of South Carolina Press, 1972.
Owen, G. E. L. "Eleatic Questions." Classical Quarterly 10 (i960): 84-102.
Reprinted in G. E. L. Owen, Logic, Science, and Dialectic: Collected
Papers in Greek Philosophy, edited by Martha Nussbaum, 3-26. Ithaca:
N.Y.: Cornell University Press, 1986.
Sidgwick, Henry. "The Sophists." In Lectures on the Philosophy of Kant and
Other Philosophical Lectures and Essays. London: Macmillan, 1905.
Taylor, C. C. W. "Pleasure, Knowledge, and Sensation in Democritus."
Phronesis 12 (1967): 6-27.
van der Waerden, B. L. Science Awakening. New York: Oxford University
Press, 1963.
Vlastos, Gregory. "On Heracleitus." American Journal of Philology 76
(i955): 337-68.
"Ethics and Physics in Democritus." In Studies in Presocratic Philosophy,
edited by David J. Furley and R. E. Allen, 2:381-408. London: Routledge
&KeganPaul, 1975.
E. GENERAL DISCUSSIONS OF PLATO'S THOUGHT
Cherniss, Harold. "The Philosophical Economy of the Theory of Ideas."
American Journal of Philology 57 (1936): 445-56. Reprinted in Studies
Cambridge Companions Online © Cambridge University Press, 2006
500 Bibliography
in Plato's Metaphysics, edited by R. E. Allen, 1-12. London: Routledge
& Kegan Paul, 1965. Also reprinted in Harold Chemiss, Selected Papers,
edited by Leonardo Tar an, 121-32. Leiden: E.J. Brill, 1977.
"Some War-Time Publications Concerning Plato/' American Journal of
Philology 68 (1947): 113-46, 225-65. Reprinted in Harold Cherniss,
Selected Papers, edited by Leonardo Taran, 142-216. Leiden: E.J. Brill,
1977-
"Review of G. C. Field, Plato and his Contemporaries." American Journal
of Philology 54 (1978): 79-83. Reprinted in Harold Cherniss, Selected
Papers, edited by Leonardo Taran, 133-7. Leiden: E.J. Brill, 1977.
Crombie, I. M. An Examination of Plato's Doctrines. 2 vols. London: Rout-
ledge & Kegan Paul. New York: Humanities Press, 1962, 1963.
Field, G. C. Plato and his Contemporaries: A Study in Fourth Century Life
and Thought. London: Methuen, 1930. 3d ed., 1967.
Findlay, J. N. Plato: The Written and Unwritten Doctrines. New York: Hu-
manities Press, 1974.
Friedlander, Paul. Plato. Translated by Hans Meyerhoff. 3 vols. New York:
Pantheon, 1958-69.
Gosling, J. C. B. Plato. London: Routledge & Kegan Paul, 1973.
Grote, George. Plato and the Other Companions of Sokrates. 2d ed. 3 vols.
London: J. Murray, 1867.
Grube, G. M. A. Plato's Thought, id ed. Indianapolis: Hackett, 1980.
Guthrie, W. K. C. A History of Greek Philosophy. Vols. 1-5. Cambridge:
Cambridge University Press, 1962, 1965, 1969, 1975, 1978.
Reale, Giovanni. A History of Ancient Philosophy. Vol. 2, Plato and Aris-
totle, edited and translated from the 5 th Italian edition by John R.
Caton. Albany: State University of New York Press, 1990.
Ritter, Constantin. Platon: Sein Lebens, seine Schriften, seine Lehre. 2 vols.
Munich: C.H. Beck, 1910.
Robinson, R., and Denniston, J. D. "Plato/7 In Oxford Classical Dictionary,
id ed., edited by N. G. L. Hammond and H. H. Scullard, 839-42. Ox-
ford: Clarendon Press, 1970.
Ryle, Gilbert. "Plato/' In The Encyclopedia of Philosophy, edited by Paul
Edwards, 6:314-33. New York: Macmillan and Free Press, 1967.
Shorey, Paul. What Plato Said. Chicago: University of Chicago Press, 1933.
The Unity of Plato's Thought. Chicago: University of Chicago Press,
i960.
Taylor, A. E. Plato the Man and his Work. 5th ed. London: Methuen, 1948.
Vlastos, Gregory. Platonic Studies, id ed. Princeton: Princeton University
Press, 1981.
Wilamowitz-Moellendorff, Ulrich von. Platon. id ed. 2 vols. Berlin: Weid-
mann, 1920.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Methods of interpretation 501
F. METHODS OF INTERPRETATION
Burnyeat, Myles. "Sphinx Without a Secret/' New York Review of Books 32
(May 30, 1985): 30-6.
Coventry, Lucinda. "The Role of the Interlocutor in Plato's Dialogues." In
Characterization and Individuality in Greek Literature, edited by
Christopher Pelling, 174-96. Oxford: Clarendon Press, 1990.
Derrida, Jacques. "Plato's Pharmacy." In Dissemination, translated, with an
introduction and additional notes by Barbara Johnson, 61-171. Chi-
cago: University of Chicago Press, 1981.
Griswold, Charles L., Jr., editor. Platonic Writings, Platonic Readings. New
York: Routledge, 1988.
Shorey, Paul. The Unity of Plato's Thought. Chicago: University of Chicago
Press, i960.
Stokes, Michael C. Plato's Socratic Conversations: Drama and Dialectic in
Three Dialogues. Baltimore: Johns Hopkins University Press, 1986.
Strauss, Leo. Persecution and the Art of Writing. Glencoe, 111.: Free Press,
1952.
Tigerstedt, E. N. Interpreting Plato. Uppsala: Almquist & Wiksell Interna-
tional, 1977.
Weingartner, Rudolf H. The Unity of the Platonic Dialogue. Indianapolis:
Bobbs-Merrill, 1973.
G. CHRONOLOGICAL AND LINGUISTIC STUDIES
Arnim, H. von. "Sprachliche Forschungen zur Chronologie der platonischen
Dialoge." Sitzungsberichte der Kaiserlichen Akademie der Wissen-
schaften in Wien: Philos. Hist. Klasse 169.1 (1912): 1-210.
Baron, C. "Contributions a la chronologie des dialogues de Platon." Revue
des Etudes grecques 10 (1897): 264-78.
Billig. L. "Clausulae and Platonic Chronology." Journal of Philology 35
(1920): 225-56.
Blass, Friedrich W. Die attische Beredsamkeit. Leipzig: Teubner, 1874.
Brandwood, Leonard. "The Dating of Plato's Works by the Stylistic Method:
A Historical and Critical Survey." Ph.D. thesis, University of London,
1958.
A Word Index to Plato. Leeds: Maney & Son, 1976.
The Chronology of Plato's Dialogues. Cambridge: Cambridge University
Press, 1990.
Diaz, Tejera A. "Ensayo de un metodo linguistico para cronologia de
Platon." Emerita 29 (1961): 241-86.
Cambridge Companions Online © Cambridge University Press, 2006
502 Bibliography
Dittenberger, W. "Sprachliche Kriterien fur die Chronologie der platon-
ischen Dialoge." Hermes 16 (1881): 321-45.
Janell, G. "Quaestiones Platonicae." Jahrbucher fur classische Philologie,
Supp. 26 (1901): 263-336.
Kaluscha, W. "Zur Chronologie der platonischen Dialoge." Wiener Studien
26 (1904): 190-204.
Ledger, Gerard R. Re-counting Plato: A Computer Analysis of Plato's Style.
Oxford: Clarendon Press, 1989.
Lutoslawski, Wincenty. The Origin and Growth of Plato's Logic With an
Account of Plato's Style and of the Chronology of his Writings. London:
Longmans, Green, 1897.
Mueller, Ian. "Joan Kung's Reading of Plato's Timaeus." In Nature, Knowl-
edge and Virtue: Essays in Memory of foan Rung, edited by Terry
Penner and Richard Kraut. Edmonton, Alberta: Academic Printing and
Publishing. Apeiron 22 (1989): 1-27.
Ritter, Constantin. Untersuchungen uber Platon: Die Echtheit and Chronol-
ogie der Platonischer Schriften. Stuttgart: Kohlhammer, 1888.
"Unterabteilungen innerhalb der zeitlich ersten Gruppe platonischer
Schriften." Hermes 70 (1935): 1-30.
Ross, W. D. Plato's Theory of Ideas. Oxford: Clarendon Press, 1951. 2d ed.
1953-
Ryle, Gilbert. Plato's Progress. Cambridge: Cambridge University Press,
1966.
Schanz, M. "Zur Entwicklung des platonischen Stils." Hermes 21(1886):
439-59-
Siebeck, Hermann. Untersuchungen zur Philosophie der Griechen. Halle:
J.C.B. Mohr, 1888.
Thesleff, Holger. Studies in Platonic Chronology. In Commentationes Hu-
manarum Litterarum 70 (1982). Helsinki: Societas Scientarum Fen-
nica, 1982.
Wishart, D., and Leach, S. V. "A Multivariate Analysis of Platonic Prose
Rhythm." Computer Studies in the Humanities and Verbal Behavior 3
(1970): 90-9.
H. SOCRATES AND THE EARLY DIALOGUES
Allen, R. E. Plato's "Euthyphro" and the Earlier Theory of Forms. London:
Routledge & Kegan Paul, 1970.
Beversluis, John. "Socratic Definition." American Philosophical Quarterly
11 (1974): 331-6.
Bolotin, David. Plato's Dialogue on Friendship. Ithaca, N.Y.: Cornell Univer-
sity Press, 1979.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Socrates and the early dialogues 503
Brickhouse, Thomas C, and Smith, Nicholas D. Socrates on Trial. Prince-
ton: Princeton University Press, 1989.
Dodds, E. R. Plato: Gorgias. A Revised text with introduction and commen-
tary. Oxford: Clarendon Press, 1959.
Frede, Dorothea. "The Impossibility of Perfection: Socrates7 Criticism of
Simonides' Poem in the Protagoras." Review of Metaphysics 39 (1986):
729-53.
Geach, P. T. "Plato's Euthyphro: An Analysis and Commentary." Monist 50
(1966): 369-82.
Gentzler, Jyl. "Knowledge and Method in Plato's Early through Middle Dia-
logues." Ph.D. thesis, Cornell University, 1991.
Gilford, Edwin Hamilton. The Euthydemus of Plato. With revised text,
introduction, notes and indices. Oxford: Clarendon Press, 1905. Re-
print. New York: Arno Press, 1973.
Glidden, David K. "The Lysis on Loving One's Own." Classical Quarterly
31(1981): 39-59-
Heidel, W. A. Plato's Euthyphro. With introduction and notes. Greek Series
for Colleges and Schools. New York: American Book Company, 1902.
Irwin, Terence. Plato's Moral Theory: The Early and Middle Dialogues.
Oxford: Clarendon Press, 1977.
Plato: Gorgias. Translated with notes. Oxford: Clarendon Press, 1979.
"Coercion and Objectivity in Plato's Dialectic." Revue Internationale de
Philosophie 40 (1986): 49-74.
"Socrates and the Tragic Hero." In Language and the Tragic Hero, edited
by Pietro Pucci, 55-83. Atlanta: Scholars Press, 1988.
"Socrates and Athenian Democracy." Philosophy and Public Affairs 18
(1989): 184-205.
Kahn, Charles H. "Did Plato Write Socratic Dialogues?" Classical Quar-
terly 31 (1981): 305—20.
"The Beautiful and the Genuine: A discussion of Paul Woodruff, Plato,
Hippias Major." Oxford Studies in Ancient Philosophy 3 (1985): 261-88.
Kidd, Ian. "The Case of Homicide in Plato's Euthyphro." In Owls to Athens:
Essays on Classical Subjects Presented to Sir Kenneth Dover, edited by
E. M. Craik, 213-22. Oxford: Clarendon Press, 1990.
Kraut, Richard. "Comments on Gregory Vlastos, 'The Socratic Elenchus.' "
Oxford Studies in Ancient Philosophy 1 (1983): 59-70.
Socrates and the State. Princeton: Princeton University Press, 1984.
Nehamas, Alexander. "Socratic Intellectualism." In Proceedings of the Bos-
ton Area Colloquium in Ancient Philosophy, edited by John J. Cleary.
Vol. 2. Lanham, Md.: University Press of America, 1987.
Patzer, Andreas, editor. Der Historische Sokrates. Darmstadt: Wissenschaft-
liche Buchgesellschaft, 1987.
Cambridge Companions Online © Cambridge University Press, 2006
504 Bibliography
Penner, Terry. "Socrates on Virtue and Motivation." In Exegesis and Argu-
ment: Studies in Greek Philosophy Presented to Gregory Vlastos, ed-
ited by E. N. Lee, Alexander P. D. Mourelatos, and R. M. Rorty, 133-51.
Assen, Netherlands: van Gorcum & Comp., 1973.
"The Unity of Virtue." Philosophical Review 82 (1973): 35-68.
"Socrates on the Impossibility of Belief-Relative Sciences." In Proceedings
of the Boston Area Colloquium in Ancient Philosophy, edited by John J.
Cleary, 3:263-325. Lanham, Md.: University Press of America, 1988.
"Power and Desire in Socrates: The Argument of Gorgias 466a-468e that
Orators and Tyrants have no Power in the City." Apeiron 24 (1991):
147-202.
Reeve, C. D. C. Socrates in the Apology. Indianapolis: Hackett, 1989.
Robinson, David B. "Plato's Lysis: The Structural Problem." Illinois Classi-
cal Studies 11 (1986): 63-83.
Robinson, Richard. Plato's Earlier Dialectic. Oxford: Clarendon Press, 1953.
Ross, W. D. "The Problem of Socrates." Proceedings of the Classical Asso-
ciation 30 (1933): 7-24. Reprinted in Der historische Sokrates, edited
by Andreas Patzer, 225—39. Darmstadt: Wissenschaftliche Buchgesell-
schaft, 1987.
Santas, Gerasimos Xenophon. Socrates: Philosophy in Plato's Early Dia-
logues. London: Rout ledge & Kegan Paul, 1979.
Stokes, Michael C. Plato's Socratic Conversations: Drama and Dialectic in
Three Dialogues. Baltimore: Johns Hopkins University Press, 1986.
Stone, I. F. The Trial of Socrates. Boston: Little, Brown, 1988.
Taylor, A. E. Varia Socratica: First Series. Oxford: James Parker, 1911. Re-
print. New York: Garland Press, 1987.
Taylor, C. C. W. Plato's Protagoras. Translated with notes. Oxford: Claren-
don Press, 1976.
Vlastos, Gregory. Plato: Protagoras. Edited with an introduction. Indianapo-
lis: Bobbs-Merrill, 1956.
"Introduction: The Paradox of Socrates." In The Philosophy of Socrates: A
Collection of Critical Essays, ed. Gregory Vlastos. Garden City, N.Y.:
Anchor Books, Doubleday, 1971.
Platonic Studies. 2d ed. Princeton: Princeton University Press, 1981.
"Socrates on the Tarts of Virtue/ " In Platonic Studies, 418-23. 2d ed.
Princeton: Princeton University Press, 1981.
"The Unity of the Virtues in the Protagoras." In Platonic Studies, 221-
69. 2d ed. Princeton: Princeton University Press, 1981.
"What did Socrates Understand by his 'What is F Question?" In Platonic
Studies, 410-17. 2d ed. Princeton: Princeton University Press, 1981.
"The Socratic Elenchus." Oxford Studies in Ancient Philosophy 1 (1983):
27-58.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Metaphysics and epistemology 505
"Socrates' Disavowal of Knowledge." Philosophical Quarterly 35 (1985):
i-3i.
"Elenchus and Mathematics." American Journal of Philology 109 (1988):
362-96.
"Socrates." Proceedings of the British Academy 74 (1988): 89-111.
Socrates: Ironist and Moral Philosopher. Cambridge: Cambridge Univer-
sity Press, 1991.
Vlastos, Gregory, editor. The Philosophy of Socrates: A Collection of Criti-
cal Essays. Garden City, N.Y.: Anchor Books, Doubleday 1971.
Weiss, Roslyn. "Hedonism in the Protagoras and the Sophist's Guarantee."
Ancient Philosophy 10 (1990): 17-39.
Woodruff, Paul. Plato: Hippias Major. Translated, with commentary and
essay. Indianapolis: Hackett, 1982.
"Plato's Early Theory of Knowledge." In Companions to Ancient Thought
1: Epistemology, edited by Stephen Everson, 60-84. Cambridge: Cam-
bridge University Press, 1990.
Zeyl, Donald. "Socrates and Hedonism - Protagoras 35 ib-358d." Phronesis
25 (1980): 250-69.
I. METAPHYSICS AND EPISTEMOLOGY
Ackrill, John. "In Defense of Platonic Division." In Ryle: A Collection of
Critical Essays, edited by Oscar P. Wood and George Pitcher, 373-92.
Garden City, N.Y.: Anchor Books, Doubleday, 1970.
"Gwilym Ellis Lane Owen." Proceedings of the British Academy 70
(1984): 481-99.
Allen, R. E. "Participation and Predication in Plato's Middle Dialogues."
Philosophical Review 69 (i960): 147-64. Reprinted in Plato: A Collec-
tion of Critical Essays. Vol. 1, Metaphysics and Epistemology, edited by
Gregory Vlastos, 167-83. Garden City, N.Y.: Anchor Books, Doubleday,
1970.
Allen, R. E., editor. Studies in Plato's Metaphysics. London: Routledge &
Kegan Paul, 1965.
Burnyeat, Myles. "Conflicting Appearances." Proceedings of the British
Academy 65 (1979): 69-111.
Cherniss, Harold. Aristotle's Criticism of Plato and the Academy. Balti-
more: Johns Hopkins University Press, 1944.
Fine, Gail. "The One Over Many." Philosophical Review 89 (1980): 197-240.
"Separation." Oxford Studies in Ancient Philosophy 2 (1984): 31-87.
"The Object of Thought Argument." Apeiron 21 (1988): 137-42.
"Plato on Perception." Oxford Studies in Ancient Philosophy, supplemen-
tary vol. (1988): 15-28.
Cambridge Companions Online © Cambridge University Press, 2006
506 Bibliography
Frede, Michael. ''Being and Becoming in Plato/7 Oxford Studies in Ancient
Philosophy, supplementary vol. (1988): 37-52.
Gentzler, Jyl. "Knowledge and Method in Plato's Early through Middle Dia-
logues." Ph.D. thesis, Cornell University, 1991.
Irwin, Terence. "Plato's Heracliteanism." Philosophical Quarterly 27 (1977):
i-i3-
Kramer, Hans Joachim. Plato and the Foundations of Metaphysics. Edited
and translated by John R. Caton. Albany: State University of New York
Press, 1990.
Lloyd, A. C. "Non-Discursive Thought - An Enigma of Greek Philosophy."
Proceedings of the Aristotelian Society 70 (1969-70): 261-74.
Mohr, Richard D. "Forms as Individuals: Unity, Being and Cognition in
Plato's Ideal Theory." Illinois Classical Studies 11 (1986): 113-28.
Moravcsik, Julius M. E. "The Anatomy of Plato's Divisions." In Exegesis
and Argument: Studies in Greek Philosophy Presented to Gregory
Vlastos, edited by E. N. Lee, Alexander P. D. Mourelatos, and R. M.
Rorty, 324-48. Assem, Netherlands: van Gorcum & Comp., 1973.
"Understanding and Knowledge in Plato's Philosophy." Neue Hefte fur
Philosophie 15/16 (1979): 53-69.
Morgan, Michael L. Platonic Piety: Philosophy and Ritual in Fourth-
Century Athens. New Haven: Yale University Press, 1990.
Owen, G. E. L. "A Proof in the Peri Ideon." Journal of Hellenic Studies 77
(1857): 103-11. Reprinted in Studies in Plato's Metaphysics, edited by
R. E. Allen, 293-312. London: Routledge & Kegan Paul, 1965. Also
reprinted in G. E. L. Owen, Logic, Science, and Dialectic: Collected
Papers in Greek Philosophy, edited by Martha Nussbaum, 165-79. Ith-
aca, N.Y.: Cornell University Press, 1986.
Patterson, Richard. "The Eternality of Platonic Forms." Archiv fur Ge-
schichte der Philosophie 67 (1985): 27-46.
Image and Reality in Plato's Metaphysics. Indianapolis: Hackett, 1985.
Penner, Terry. The Ascent from Nominalism: Some Existence Arguments in
Plato's Middle Dialogues. Dordrecht: D. Reidel, 1987.
Peterson, Sandra. "The Greatest Difficulty for Plato's Theory of Forms: the
Unknowability Argument of Parmenides 133c-134c" Archiv fur Ge-
schicte der Philosophie 63 (1981): 1-16.
Prior, William J. Unity and Development in Plato's Metaphysics. London:
Croom Helm, 1985.
Robinson, Richard. Plato's Earlier Dialectic. Oxford: Clarendon Press, 1953.
Ross, W. D. Plato's Theory of Ideas. Oxford: Clarendon Press, 1951. 2d ed.
1953-
Sayre, Kenneth M. Plato's Late Ontology: A Riddle Resolved. Princeton:
Princeton University Press, 1983.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Ethics, politics, and psychology 507
Schofield, Malcolm. "The Denouement of the Cratylus." In Language and
Logos: Studies in Ancient Greek Philosophy Presented to G.E.L. Owen,
edited by Malcolm Schofield and Martha Nussbaum, 61-81. Cam-
bridge: Cambridge University Press, 1982.
"Editor's Notes/' Phronesis 35 (1990): 327-34.
Scott, Dominic. "Platonic Anamnesis Revisited." Classical Quarterly 37
(1987): 346-66.
Teloh, Henry. The Development of Plato's Metaphysics. University Park:
Pennsylvania State University Press, 1981.
Vlastos, Gregory. "Degrees of Reality in Plato." In New Essays on Plato and
Aristotle, edited by Renford Bambrough. London: Routledge & Kegan
Paul, 1965. Reprinted in Gregory Vlastos, Platonic Studies, 58-75. 2d
ed. Princeton: Princeton University Press, 1981.
Vlastos, Gregory, editor. Plato: A Collection of Critical Essays. Vol. 1, Meta-
physics and Epistemology. Garden City, N.Y.: Anchor Books, Double-
day, 1970.
Wedberg, A. "The Theory of Ideas." In Plato: A Collection of Critical Es-
says. Vol. 1, Metaphysics and Epistemology, edited by Gregory Vlastos,
28-52. Garden City, N.Y.: Anchor Books, Doubleday, 1970.
White, Nicholas P. Plato on Knowledge and Reality. Indianapolis: Hackett,
1976.
"Perceptual and Objective Properties in Plato." In Nature, Knowledge and
Virtue: Essays in Memory of Joan Kung, edited by Terry Penner and
Richard Kraut. Edmonton, Alberta: Academic Printing and Publishing.
Apeiron 22 (1989): 45-66.
Williams, Bernard. "Cratylus7 Theory of Names and its Refutation." In Lan-
guage and Logos: Studies in Ancient Greek Philosophy Presented to
G.E.L. Owen, edited by Malcolm Schofield and Martha Nussbaum, 83-
94. Cambridge: Cambridge University Press, 1982.
Woodruff, Paul. "Plato's Early Theory of Knowledge." In Companions to
Ancient Thought 1: Epistemology, edited by Stephen Everson, 60-84.
Cambridge: Cambridge University Press, 1990.
J. ETHICS, POLITICAL PHILOSOPHY, AND MORAL
PSYCHOLOGY
Bambrough, Renford, editor. Plato, Popper, and Politics. Cambridge: Cam-
bridge University Press, 1967.
Barker, Ernest. Greek Political Theory: Plato and His Predecessors. London:
Methuen, 1918.
Gosling, J. C. B., and Taylor, C. C. W. The Greeks on Pleasure. Oxford:
Clarendon Press, 1982.
Cambridge Companions Online © Cambridge University Press, 2006
508 Bibliography
Griswold, Charles L., Jr. "Politike episteme in Plato's Statesman." In Essays
in Ancient Greek Philosophy, vol. 3, edited by John P. Anton and An-
thony Preus. Albany: State University of New York Press, 1989.
Harvey, F. David. "Two Kinds of Equality/' Classica et Mediaevalia 26
(1965): 101-46. Emended in "Corrigenda." Classica et Mediaevalia 27
(1966): 99-100.
Irwin, Terence. Plato's Moral Theory: The Early and Middle Dialogues.
Oxford: Clarendon Press, 1977.
"Socrates and Athenian Democracy." Philosophy and Public Affairs 18
(1989): 184-205.
Kahn, Charles H. "Plato's Theory of Desire." Review of Metaphysics 41
(1987): 77-103.
Klosko, George. The Development of Plato's Political Theory New York:
Methuen, 1986.
Kraut, Richard. "Review of Plato's Moral Theory, by Terence Irwin." Philo-
sophical Review 88 (1979): 633-9.
Levinson, Ronald B. In Defense of Plato. Cambridge, Mass.: Harvard Univer-
sity Press, 1953.
Mackenzie, Mary Margaret. Plato on Punishment. Berkeley: University of
California Press, 19 81.
Nussbaum, Martha C. The Fragility of Goodness: Luck and Ethics in
Greek Tragedy and Philosophy. Cambridge: Cambridge University
Press, 1986.
Penner, Terry. "Thought and Desire in Plato." In Plato. Vol. 2, Ethics, Poli-
tics, and Philosophy of Art and Religion, edited by Gregory Vlastos,
96-118. Garden City, N.Y.: Anchor Books, Doubleday, 1971.
"Plato and Davidson: Parts of the Soul and Weakness of Will." In Cana-
dian Philosophers. Canadian Journal of Philosophy, supplementary vol.
16(1990): 35-74-
Popper, K. R. The Open Society and its Enemies. Vol. 1, The Spell of Plato.
4th ed. New York: Harper & Row, 1963. 5th ed. 1966.
Price, Anthony W. Love and Friendship in Plato and Aristotle. Oxford:
Clarendon Press, 1989.
"Plato and Freud." In The Person and the Human Mind: Issues in Ancient
and Modern Philosophy, edited by Christopher Hill, 247-70. Oxford:
Oxford University Press, 1990.
Sabine, George Holland. A History of Political Theory. 4th ed. Hinsdale, 111.:
Dryden Press, 1973.
Saunders, T. J., " 'The RAND Corporation of Antiquity'? Plato's Academy
and Greek Politics." In Studies in Honour of T.B.L. Webster, edited by J.
H. Betts, J. T Hooker, and J. R. Green, 1:200-10. Bristol: Classical Press,
1986.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Art and poetry 509
Smith, Nicholas D. "Plato and Aristotle on the Nature of Women." Journal
of the History of Philosophy 21 (1983): 467-78.
Strauss, Leo. The City and Man. Chicago: University of Chicago Press,
1964.
Vlastos, Gregory. "Socratic Knowledge and Platonic 'Pessimism/ " Review
of The Development of Plato's Ethics, by John Gould. Philosophical
Review 66 (1957): 226—38. Reprinted in Gregory Vlastos, Platonic Stud-
ies, 204-17. 2d ed. Princeton: Princeton University Press, 1981.
Wender, Dorothea. "Plato: Misogynist, Paedophile and Feminist." Arethusa
6(i973): 75-8o.
K. ART AND POETRY
Annas, Julia. "Plato on the Triviality of Literature." In Plato on Beauty,
Wisdom, and the Arts, edited by Julius Moravcsik and Philip Temko,
1-28. Totowa, N.J.: Rowman &. Littlefield, 1982.
Battin, M. Pabst. "Plato on True and False Poetry." Journal of Aesthetics and
Art Criticism 36 (1977): 163-74.
Belfiore, Elizabeth. "A Theory of Imitation in Plato's Republic/' Transac-
tions of the American Philological Association 114 (1984): 121-46.
Bosanquet, Bernard. A History of Aesthetic. London: Swan Sonnenschein,
1892.
Brock, Roger. "Plato and Comedy." In Owls to Athens: Essays on Classical
Subjects Presented to Sir Kenneth Dover, edited by E. M. Craik, 39-51.
Oxford: Clarendon Press, 1990.
Clay, Diskin. "The Tragic and Comic Poet of the Symposium." In Essays in
Ancient Greek Philosophy, edited by John P. Auton and Anthony Press,
2: 186-202. Albany: State University of New York Press, 1983.
Collingwood, R. G. "Plato's Philosophy of Art." Mind 34 (1925): 154-72.
Ferrari, G. R. F. "Plato and Poetry." In The Cambridge History of Literary
Criticism, edited by George A. Kennedy, 92-148. Cambridge: Cam-
bridge University Press, 1989.
Greene, W. C. "The Spirit of Comedy in Plato." Harvard Studies in Classi-
cal Philology 31 (1920): 63-123.
Menza, V. "Poetry and the Techne Theory." Ph.D. dissertation, Johns Hop-
kins University, 1972.
Moravcsik, Julius, and Temko, Philip, editors. Plato on Beauty, Wisdom,
and the Arts. Totowa, N.J.: Rowman & Littlefield, 1982.
Murdoch, Iris. The Fire and the Sun: Why Plato Banished the Artists. Ox-
ford: Clarendon Press, 1977.
Woodruff, Paul. "What Could Go Wrong with Inspiration? Why Plato's Poets
Fail." In Plato on Beauty, Wisdom, and the Arts, edited by Julius
Cambridge Companions Online © Cambridge University Press, 2006
51 o Bibliography
Moravcsik and Philip Temko, 137-50. Totowa, N.J.: Rowman & Little-
field, 1982.
L. EROS
Armstrong, A. H. "Platonic Eros and Christian Agape." Downside Review
79(1961): 105-21.
"Platonic Love: A Reply to Professor Verdenius.;/ Downside Review 82
(1964): 199-208.
Carson, Anne. Eros the Bittersweet. Princeton: Princeton University Press,
1986.
Cornford, Francis MacDonald. "The Doctrine of Eros in Plato's Sympo-
sium." In Plato: A Collection of Critical Essays. Vol. 2, Ethics, Politics,
and Philosophy of Art and Religion, edited by Gregory Vlastos. Garden
City, N.Y.: Anchor Books, Doubleday, 1971.
Gagarin, Michael. "Socrates' Hybris and Alcibiades' Failure." Phoenix 31
(1977): 22-37.
Gould, Thomas. Platonic Love. New York: Free Press of Glencoe, 1963.
Halperin, David. "Platonic Eros and What Men Call Love." Ancient Philoso-
phy 5 (1985): 161-204.
"Plato and Erotic Reciprocity." Classical Antiquity 5 (1986): 60-80.
Kosman, L. A. "Platonic Love." In Facets of Plato's Philosophy, edited by W.
H. Werkmeister, 53-69. Assen, Netherlands: van Gorcum &. Comp.,
1976.
Price, Anthony W. Love and Friendship in Plato and Aristotle. Oxford:
Clarendon Press, 1989.
Robin, Leon. Theorieplatonicienne del'amour. Paris: Alcan, 1908.
Santas, Gerasimos Xenophon. Plato and Freud: Two Theories of Love. Ox-
ford: Basil Blackwell, 1988.
Vlastos, Gregory. "The Individual as an Object of Love in Plato." In Gregory
Vlastos, Platonic Studies, 3-37. 2d ed. Princeton: Princeton University
Press, 1981.
M. MATHEMATICS
Burnyeat, Myles. "Platonism and Mathematics: A Prelude to Discussion."
In Mathematics and Metaphysics in Aristotle, edited by Andreas
Graeser, 213-40. Bern and Stuttgart: P. Haupt, 1987.
Cherniss, Harold. "Plato as Mathematician." Review of Metaphysics 4
(1951): 395-425. Reprinted in Harold Cherniss, Selected Papers, edited
by Leonardo Taran, 222-52. Leiden: E. J. Brill, 1977.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Meno 511
Cornford, Francis MacDonald. "Mathematics and Dialectic in the Republic
VI-VII." Mind 41 (1932): 37-52, 173-90. Reprinted in Studies in
Plato's Metaphysics, edited by R. E. Allen, 61-96. London: Routledge &
Kegan Paul, 1965.
Frajese, Attilio. Platone e la matematica nel mondo antico. Rome: Editrice
Studium, 1963.
Gillispie, Charles Coulston, editor. Dictionary of Scientific Biography. 16
vols. New York: Scribner's, 1970-80.
Gulley, Norman. "Greek Geometrical Analysis/7 Phronesis 3 (1958): 1-14.
Heath, Thomas. A History of Greek Mathematics. Vol. 1. Oxford: Claren-
don Press, 1921.
Lasserre, Francois. Die Fragmente des Eudoxos von Knidos. Edited and trans-
lated, with commentary. Berlin: de Gruyter, 1966.
De Leodamas de Thasos a Philippe d'Oponte. Vol. 2, La scuola di
Platone. Naples: Bibliopolis, 1987.
Morrow, Glenn R., translator. Proclus: A Commentary on the First Book of
Euclid's Elements. Princeton: Princeton University Press, 1970.
Morrow, Glenn R., and Dillon, John M., translators. Proclus' Commentary
on Plato's Parmenides. With introduction and notes by John M. Dillon.
Princeton: Princeton University Press, 1987.
Mueller, Ian. "On the Notion of a Mathematical Starting Point in Plato,
Aristotle, and Euclid." In Science and Philosophy in Classical Greece,
edited by Alan Bowen, 59-97. London and New York: Garland, 1991.
N. STUDIES OF SINGLE WORKS
1. Meno
Bluck, R. S. Plato's Meno. Edited with an introduction and commentary.
Cambridge: Cambridge University Press, 1961.
Heitsch, Ernst. "Platons hypothetisches Verfahren im Menon." Hermes 105
(i977): 257-68.
Moravcsik, Julius M. E. "Learning as Recollection/7 In Plato: A Collection
of Critical Essays. Vol. 1, Metaphysics and Epistemology, edited by
Gregory Vlastos, 53-69. Garden City, N.Y.: Anchor Books, Doubleday,
1970.
Nehamas, Alexander. "Meno's Paradox and Socrates as a Teacher." Oxford
Studies in Ancient Philosophy 3 (1985): 1-30.
Sharpies, R. W. Plato, Meno. Edited with translation and notes. Warminster,
Wiltshire: Aris & Phillips, 1985. Chicago: Bolchazy-Carducci, 1985.
Thompson, E. Seymer. The Meno of Plato. With introduction, notes and
excurses. London: Macmillan, 1901.
Cambridge Companions Online © Cambridge University Press, 2006
512 Bibliography
Vlastos, Gregory. "Anamnesis in the Meno." Dialogue 4 (1965): 143-67.
White, Nicholas P. "Inquiry." Review of Metaphysics 28 (1974): 289-310.
2. Phaedo
Bostock, David. Plato's Phaedo. Oxford: Clarendon Press, 1986.
Brentlinger, J. "Incomplete Predicates and the Two-World Theory of the
Phaedo. Phronesis 17 (1972): 61-79.
Castaneda, Hector-Neri. "Plato's Phaedo Theory of Relations." Journal of
Philosophical Logic 1 (1972): 467-80.
Gallop, David. Plato, Phaedo. Translated with notes. Oxford: Clarendon
Press, 1975.
Hackforth, R. Plato's Phaedo. Translated, with introduction and commen-
tary. Indianapolis: Bobbs-Merrill, 1955.
Kung, Joan. "Plato's Criticism of Anaxagoras on Mind and Morality in Na-
ture." Proceedings of the Tenth University of Dayton Philosophy Collo-
quium, University of Dayton Review 16 (1982): 65-76.
Matthews, Gareth B., and Blackson, Thomas A. "Causes in the Phaedo."
Synthese 79 (1989): 581-91.
Nehamas, Alexander. "Plato on the Imperfection of the Sensible World."
American Philosophical Quarterly 12 (1975): 105-17.
Plass, Paul. "Socrates' Method of Hypothesis in the Phaedo." Phronesis 5
(i960): 103-15.
Vlastos, Gregory. "Reasons and Causes in the Phaedo." Philosophical Re-
view 78 (1969): 291-325. Reprinted in Plato: A Collection of Critical
Essays. Vol. 1, Metaphysics and Epistemology, edited by Gregory
Vlastos, 132-66. Garden City, N.Y.: Anchor Books, Doubleday, 1970.
White, Nicholas P. "Forms and Sensibles: Phaedo 74B-C." Philosophical
Topics 15 (1987): 197-214.
3. Symposium
Allen R. E. "A Note on the Elenchus of Agathon: Symposium 199C-201C."
Monist 50 (1966): 460-3.
Anton, John P. "The Secret of Plato's Symposium." Diotima 2 (1974): 27-
47-
Bacon, Helen. "Socrates Crowned." The Virginia Quarterly Review 35
(19591:415-30.
Brenkman, John. "The Other and the One: Psychoanalysis, Reading, the
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Symposium 513
Symposium." In Literature and Psychoanalysis, edited by Shoshana
Felman, 396-456. Baltimore: Johns Hopkins University Press, 1982.
Bury, Robert Gregg. The Symposium of Plato, id ed. Cambridge: Cambridge
University Press, 1932.
Dover, K. J. "The Date of Plato's Symposium." Phronesis 10 (1965): 2-20.
Plato's Symposium. Cambridge: Cambridge University Press, 1980.
Hackforth, R. "Immortality in Plato's Symposium." Classical Review 64
(1950): 43-5.
Halperin, David. "Why Is Diotima a Woman?" In One Hundred Years of
Homosexuality, 113-51. New York: Routledge, 1990.
Isenberg, Meyer W. The Order of the Discourses in Plato's Symposium.
Private Edition. Chicago: University of Chicago Libraries, 1940.
Markus, R. A. "The Dialectic of Eros in Plato's Symposium." In Plato: A
Collection of Critical Essays. Vol. 2, Ethics, Politics, and Philosophy of
Art and Religion, edited by Gregory Vlastos. Garden City, N.Y.: Anchor
Books, Doubleday, 1971.
Moravcsik, Julius M. E. "Reason and Eros in the 'Ascent'-passage of the
Symposium." In Essays in Ancient Greek Philosophy, edited by John P.
Anton and G. L. Kustas, 1:285-302. Albany: State University of New
York Press, 1972.
Nehamas, Alexander, and Woodruff, Paul. Plato: Symposium. Translated
with introduction and notes. Indianapolis: Hackett, 1989.
Neumann, Harry. "Diotima's Concept of Love." American Journal of Philol-
ogy 86(1965): 33-59-
O'Brien, Michael J. " 'Becoming Immortal' in Plato's Symposium." In Greek
Poetry and Philosophy, edited by Douglas E. Gerber, 185-206. Chico,
Cal.: Scholars Press, 1984.
Plochmann, G. K. "Hiccups and Hangovers in the Symposium." Bucknell
Review 11 (1963): 1-18.
Robin, Leon. Platon: Le Banquet. In Platon: Oeuvres completes, edited and
translated. Vol. 4, Part 2. Paris: Societe d'Edition, Les Belles Lettres,
1966.
Rosen, Stanley. Plato's Symposium. 2d ed. New Haven: Yale University
Press, 1987.
Stokes, Michael C. Plato's Socratic Conversations: Drama and Dialectic in
Three Dialogues. Baltimore: Johns Hopkins University Press, 1986.
Szlezak, Thomas A. Platon und die Schriftlichkeit der Philosophie. Berlin:
de Gruyter, 1985.
White, F. C. "Love and Beauty in Plato's Symposium." Journal of Hellenic
Studies 109 (1989): 149-57.
Wolz, H. G. "Philosophy as Drama: An Approach to Plato's Symposium."
Philosophy and Phenomenological Research 30 (1969-70): 323-53.
Cambridge Companions Online © Cambridge University Press, 2006
514 Bibliography
4. Republic
Adam, James. The Republic of Plato. Edited with critical notes and an intro-
duction. 2 vols. Cambridge: Cambridge University Press, 1902.
Annas, Julia. "Plato's Republic and Feminism." Philosophy 51 (1976): 307-
21.
An Introduction to Plato's Republic. Oxford: Clarendon Press, 1981.
Bloom, Allan. The Republic of Plato. Translated with notes and an interpre-
tive essay. New York: Basic Books, 1968.
Cherniss, Harold. "On Plato's Republic X 597B." American Journal of Phi-
lology 53 (1932): 233-42. Reprinted in Harold Cherniss, Selected Pa-
pers, edited by Leonardo Taran, 271-80. Leiden: E.J. Brill, 1977.
Cooper, John M. "The Psychology of Justice in Plato." American Philosophi-
cal Quarterly 14 (1977): 151-7.
"Plato's Theory of Human Motivation." History of Philosophy Quarterly
1 (1984): 3-21.
Cross, R. C, and Woozley, A. D. Plato's Republic: A Philosophical Commen-
tary. London: Macmillan, 1964.
Demos, Raphael. "A Fallacy in Plato's Republic2." The Philosophical Re-
view 73 (1964): 395-8. Reprinted in Plato: A Collection of Critical
Essays. Vol. 2, Ethics, Politics, and Philosophy of Art and Religion,
edited by Gregory Vlastos, 52-6. Garden City, N.Y.: Anchor Books,
Doubleday, 1971.
Fine, Gail. "Knowledge and Belief in Republic V." Archivfur Geschichte der
Philosophie 60 (1978): 121-39.
"Knowledge and Belief in Republic V-VII." In Companions to Ancient
Thought: Epistemology, edited by Stephen Everson. Cambridge: Cam-
bridge University Press, 1990.
Foster, M. B. "A Mistake of Plato's in the Republic." Mind 46 (1937): 386-93.
Gosling, J. C. B. "Republic V: ta polla kala." Phronesis 5 (i960): 116-28.
Joseph, Horace W. B. Essays in Ancient and Modern Philosophy. Freeport,
N.Y.: Books for Libraries Press, 1971.
Kirwan, C. A. "Glaucon's Challenge." Phronesis 10 (1965): 162-73.
Kraut, Richard. "Egoism, Love and Political Office in Plato." Philosophical
Review 82 (1973): 330—44.
"Reason and Justice in Plato's Republic. In Exegesis and Argument: Stud-
ies in Greek Philosophy Presented to Gregory Vlastos, edited by E. N.
Lee, Alexander P. D. Mourelatos, and R. M. Rorty, 207-24. Assen, Neth-
erlands: van Gorcum & Comp., 1973.
Review of Philosopher-Kings, by C. D. C. Reeve. Political Theory 18
(1990): 492-6.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Republic 515
"Return to the Cave: Republic 519-521." In Proceedings of the Boston
Area Colloquium in Ancient Philosophy, edited by John J. Cleary. Vol.
7. Forthcoming.
Laks, Andre. "Legislation and Demiurgy: On the Relationship between
Plato's Republic and Laws." Classical Antiquity 9 (1990): 209-29.
Mabbott, J. D. "Is Plato's Republic Utilitarian?" Mind 46 (1937): 468-74.
Reprinted in Plato. Vol. 2, Ethics, Politics, and Philosophy of Art and
Religion, edited by Gregory Vlastos, 57-66. Garden City, N.Y.: Anchor
Books, Doubleday, 1971.
Mourelatos, Alexander P. D. "Plato's 'Real Astronomy': Republic 527d-
53id." In Science and the Sciences in Plato, edited by John P. Anton,
33-73. Delmar, N.Y.: Caravan Books, 1980.
Mueller, Ian. "Ascending to Problems: Astronomy and Harmonics in Repub-
lic VII." In Science and the Sciences in Plato, edited by John P. Anton,
103-21. Delmar, N.Y.: Caravan Books, 1980.
Murphy, N. R. The Interpretation of Plato's Republic. Oxford: Clarendon
Press, 1951.
Nehamas, Alexander. "Plato on Imitation and Poetry in Republic 10." In
Plato on Beauty, Wisdom, and the Arts, edited by Julius Moravcsik and
Philip Temko, 47-78. Totowa, N.J.: Rowman & Littlefield, 1982.
Nettleship, Richard. Lectures on the Republic of Plato, id ed. London: Mac-
millan, 1962.
Okin, Susan Moller. "Philosopher Queens and Private Wives: Plato on
Women and the Family." Philosophy et) Public Affairs 6 (1977): 345-69.
Women in Western Political Thought. Princeton: Princeton University
Press, 1979.
Reeve, C. D. C. Philosopher-Kings: The Argument of Plato's Republic.
Princeton: Princeton University Press, 1988.
Sachs, David. "A Fallacy in Plato's Republic." Philosophical Review 72
(1963): 141-58. Reprinted in Plato: A Collection of Critical Essays. Vol.
2, Ethics, Politics, and Philosophy of Art and Religion, edited by Greg-
ory Vlastos, 35-51. Garden City, N.Y.: Anchor Books, Doubleday, 1971.
Santas, Gerasimos Xenophon. "The Form of the Good in Plato's Republic.
In Essays in Ancient Greek Philosophy, edited by John P. Anton and
Anthony Preus, 2: 232-63. Albany: State University of New York Press,
I983-
"Aristotle's Criticism of Plato's Form of the Good: Ethics without Meta-
physics?" Philosophical Papers 18 (1989): 137-60.
Smith, J. A. "General Relative Clauses in Greek." Classical Review 31
(1917): 69-71.
Sorabji, Richard. "Myths about Non-Propositional Thought." In Language
Cambridge Companions Online © Cambridge University Press, 2006
516 Bibliography
and Logos: Studies in Ancient Greek Philosophy Presented to G.E.L
Owen, edited by Malcolm Schofield and Martha Nussbaum, 295-314.
Cambridge: Cambridge University Press, 1982.
Stokes, Michael C. "Adeirhantus in the Republic." In Law, Justice and
Method in Plato and Aristotle, edited by Spiro Panagiotou, 67-96. Ed-
monton: Academic Printing & Publishing, 1985.
Vlastos, Gregory. "Justice and Happiness in the Republic." In Plato. Vol. 2,
Ethics, Politics, and Philosophy of Art and Religion, edited by Gregory
Vlastos, 66-95. Garden City, N.Y.: Anchor Books, Doubleday, 1971.
Reprinted in Gregory Vlastos, Platonic Studies, 111-39. 2d ed. Prince-
ton: Princeton University Press, 1981.
"Was Plato a Feminist?" Times Literary Supplement, March 17-23, 1989:
276, 288-9.
White, Nicholas P. A Companion to Plato's Republic. Indianapolis: Hack-
ett, 1979.
"The Classification of Goods in Plato's Republic." Journal of the History
of Philosophy 22 (1984): 393-421.
"The Ruler's Choice." Archiv fur Geschichte der Philosophie 68 (1986):
22-46.
"Happiness and External Contingencies in Plato's Republic." In Moral
Philosophy, edited by William C. Starr and Richard C. Taylor, 1-21.
Milwaukee: Marquette University Press, 1989.
Williams, Bernard. "The Analogy of City and Soul in Plato's Republic." In
Exegesis and Argument: Studies in Greek Philosophy Presented to
Gregory Vlastos, edited by E. N. Lee, Alexander P. D. Mourelatos, and
R. M. Rorty, 196-206. Assen, Netherlands: van Gorcum & Comp.,
1973-
5. Phaedrus
Asmis, Elizabeth. "Psychagogia in Plato's Phaedrus." Illinois Classical
Studies 11 (1986): 113-28, 153-72.
Burger, Ronna. Plato's Phaedrus: A Defense of a Philosophic Art of Writing.
University, Ala.: University of Alabama Press, 1980.
Burnyeat, Myles F. "The Passion of Reason in Plato's Phaedrus." Unpub-
lished manuscript.
Derrida, Jacques. "Plato's Pharmacy." In Dissemination, translated with an
introduction and additional notes by Barbara Johnson, 61-171. Chi-
cago: University of Chicago Press, 1981.
de Vries, G. J. A Commentary on the Phaedrus of Plato. Amsterdam:
Hakkert, 1969.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Parmenides 517
Dorter, Kenneth. "Imagery and Philosophy in Plato's Phaedrus." Journal of
the History of Philosophy 9 (1971): 279-88.
Ferrari, G. R. F. Listening to the Cicadas: A Study of Plato's Phaedrus.
Cambridge: Cambridge University Press, 1987.
Griswold, Charles L., Jr. Se1' Knowledge in Plato's Phaedrus. New Haven:
Yale University Press, 1986.
Hackforth, R. Plato's Phaedrus. Cambridge: Cambridge University Press,
1952.
Heath, Malcolm. "The Unity of Plato's Phaedrus." Oxford Studies in An-
cient Philosophy 7 (1989): 151-74.
Howland, R. L. "The Attack on Isocrates in the Phaedrus." Classical Quar-
terly 31 (1937): I5I-9-
Lebeck, Anne. "The Central Myth of Plato's Phaedrus." Greek, Roman and
Byzantine Studies 13 (1972): 267-90.
Philip, A. "Recurrences thematiques et topologie dans le Phedre de Platon."
Revue de Metaphysique et de Morale 86 (1981): 452-76.
Rowe, C. J. "The Argument and Structure of Plato's Phaedrus." Proceedings
of the Cambridge Philological Society 212 (1986): 106-25.
Plato: Phaedrus. With translation and commentary. Warminster: Aris &
Phillips, 1986.
Thompson, W. H. The Phaedrus of Plato. With English notes and disserta-
tions. London: Bell & Whittaker, 1868.
6. Parmenides
Allen, R. E. "Participation and Predication in Plato's Middle Dialogues."
Philosophical Review 69 (i960): 147-64. Reprinted in Plato: A Collec-
tion of Critical Essays. Vol. 1, Metaphysics and Epistemology, edited by
Gregory Vlastos, 167-83. Garden City, N.Y.: Anchor Books, Doubleday,
1970.
Plato's Parmenides. Minneapolis: University of Minnesota Press, 1983.
Code, Alan. "On the Origins of Some Aristotelian Theses about Predica-
tion." In How Things Are: Studies in Predication and the History of
Philosophy, edited by J. Bogen and J. E. McGuire, 101—31, 323-6.
Dordrecht: D. Reidel, 1985.
Cohen, S. Marc. "The Logic of the Third Man." Philosophical Review 80
(i97i):448-75-
Cornford, Francis MacDonald. Plato and Parmenides. Translated with an
introduction and running commentary. London: Routledge & Kegan
Paul, 1939. Reprint. Indianapolis: Bobbs-Merrill, no date.
Dies, Auguste, editor. Platon: Oeuvres completes. Vol. 8. Paris: Societe
d'Edition, Les Belles Lettres, 1923.
Cambridge Companions Online © Cambridge University Press, 2006
Si8 Bibliography
Dodds, E. R. "The Parmenides of Plato and the Origin of the Neoplatonic
One." Classical Quarterly 22 (1928): 129-42.
Geach, P. T. "The Third Man Again." Philosophical Review 65 (1956): 72-82.
Meinwald, Constance C. Plato's Parmenides. New York: Oxford University
Press, 1991.
Miller, Mitchell, Jr. Plato's Parmenides. Princeton: Princeton University
Press, 1986.
Morrow, Glenn R., and Dillon, John M., translators. Proclus' Commentary
on Plato's Parmenides. With introduction and notes by John M. Dillon.
Princeton: Princeton University Press, 1987.
Owen, G. E. L. "Notes on Ryle's Plato." In Ryle: A Collection of Critical
Essays, edited by Oscar P. Wood and George Pitcher, 341-72. Garden
City, N.Y.: Anchor Books, Doubleday, 1970. Reprinted in G. E. L. Owen,
Logic, Science, and Dialectic: Collected Papers in Greek Philosophy,
edited by Martha Nussbaum, 85-103. Ithaca, N.Y.: Cornell University
Press, 1986.
Peterson, Sandra. "A Reasonable Self-Predication Premise for the Third Man
Argument." Philosophical Review 82 (1973): 451-70. Emended in "A
Correction." Philosophical Review 84 (1975): 96.
Proclus. Commentarium in Platonis Parmenidem. In Procli Philosophi Pla-
tonici Opera Inedita, edited by Victor Cousin. Paris: Minerva, 1864.
Ryle, Gilbert. "Plato's Parmenides." Mind 48 (1939): 129-51, 302-25. Re-
printed in Studies in Plato's Metaphysics, edited by R. E. Allen, 97-
147. London: Routledge & Kegan Paul, 1965.
Sayre, Kenneth M. Plato's Late Ontology: A Riddle Resolved. Princeton:
Princeton University Press, 1983.
Sellars, Wilfred. "Vlastos and the Third Man." Philosophical Review 64
(19551:405-37.
Strang, Colin. "Plato and the Third Man." Proceedings of the Aristotelian
Society, supplementary vol. 37 (1963): 147-64. Reprinted in Plato: A
Collection of Critical Essays. Vol. 1, Metaphysics and Epistemology,
edited by Gregory Vlastos, 184-200. Garden City, N.Y.: Anchor Books,
Doubleday, 1970.
Vlastos, Gregory. "The Third Man Argument in Plato's Parmenides." Philo-
sophical Review 63 (1954): 319-49. Reprinted in Studies in Plato's
Metaphysics, edited by R. E. Allen, 231-63. London: Routledge & Ke-
gan Paul, 1965.
"Plato's 'Third Man' Argument (Parm. 131A1-B2): Text and Logic." Philo-
sophical Quarterly 19 (1969): 289-301. Reprinted with appendices in
Gregory Vlastos, Platonic Studies, 342-65. 2d ed. Princeton: Princeton
University Press, 1981.
Wundt, Max. Platons Parmenides. Stuttgart: Verlag W. Kohlhammer, 1935.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Theaetetus 519
7. Theaetetus
Annas, Julia. "Knowledge and language: the Theaetetus and the Cratylus."
In Language and Logos: Studies in Ancient Greek Philosophy Pre-
sented to G.E.L. Owen, edited by Malcolm Schofield and Martha
Nussbaum, 95-114. Cambridge: Cambridge University Press, 1982.
Bostock, David. Plato's Theaetetus. Oxford: Clarendon Press, 1988.
Burnyeat, Myles. "Socrates and the Jury: Paradoxes in Plato's Distinction
between Knowledge and True Belief." Proceedings of the Aristotelian
Society, supplementary vol. 54 (1980): 173-92.
The Theaetetus of Plato. With a translation by M. J. Levett, revised by
Myles Burnyeat. Indianapolis: Hackett, 1990.
Fine, Gail. "False Belief in the Theaetetus." Phronesis 24 (1979): 70-80.
McDowell, John. Plato: Theaetetus. Translated with notes. Oxford: Claren-
don Press, 1973.
8. Timaeus
Cherniss, Harold. "The Sources of Evil According to Plato." Proceedings of
the American Philosophical Society 98 (1954): 23-30. Reprinted in
Plato. Vol. 2, Ethics, Politics, and Philosophy of Art and Religion, ed-
ited by Gregory Vlastos, 244-58. Garden City, N.Y.: Anchor Books,
Doubleday, 1971. Also reprinted in Harold Cherniss, Selected Papers,
edited by Leonardo Taran, 253-60. Leiden: E.J. Brill, 1977.
"The Relation of the Timaeus to Plato's Later Dialogues." American
Journal of Philology 78 (1957): 225-66. Reprinted in Studies in
Plato's Metaphysics, edited by R. E. Allen, 339-78. London: Rout-
ledge & Kegan Paul, 1965. Also reprinted in Harold Cherniss, Se-
lected Papers, edited by Leonardo Taran, 298-339. Leiden: E.J. Brill,
1977-
Cornford, Francis MacDonald. Plato's Cosmology: The Timaeus of Plato.
Translated with a running commentary. London: Kegan Paul, 1937.
Kung, Joan. "Tetrahedra, Motion, and Virtue." Nous 19 (1985): 17-27.
"Why the receptacle is not a mirror." Archiv fur Geschichte der Philoso-
phie 70 (1988): 167-78.
"Mathematics and Virtue in Plato's Timaeus." In Essays in Ancient
Greek Philosophy, edited by John Anton and Anthony Preus, 3: 309-39.
Albany: State University of New York Press, 1989.
Morrow, Glenn R. "Necessity and Persuasion in Plato's Timaeus." Philo-
sophical Review 59 (1950): 147-64. Reprinted in Studies in Plato's
Metaphysics, edited by R. E. Allen, 421-37. London: Routledge &. Ke-
gan Paul, 1965.
Cambridge Companions Online © Cambridge University Press, 2006
520 Bibliography
"The Demiurge in Politics: The Timaeus and the Laws." Proceedings of
the American Philosophical Association 27 (1953-4): 5-23.
Mueller, Ian. "Joan Kung's Reading of Plato's Timaeus." In Nature, Knowl-
edge and Virtue: Essays in Memory of Joan Rung, edited by Terry
Penner and Richard Kraut. Edmonton, Alberta: Academic Printing and
Publishing. Apeiron 22 (1989): 1-27.
Owen, G. E. L. "The Place of the Timaeus in Plato's Dialogues." Classical
Quarterly 3 (1953): 79-95. Reprinted in Studies in Plato's Metaphysics,
edited by R. E. Allen, 313-38. London: Routledge & Kegan Paul, 1965.
Also reprinted in G. E. L. Owen, Logic, Science, and Dialectic: Col-
lected Papers in Greek Philosophy, edited by Martha Nussbaum, 65-84.
Ithaca, N.Y.: Cornell University Press, 1986.
Sorabji, Richard. Time, Creation and the Continuum. Ithaca, N.Y.: Cornell
University Press, 1983.
Taylor, A. E. A Commentary on Plato's Timaeus. Oxford: Clarendon Press,
1928.
Vlastos, Gregory. "The Disorderly Motion in the Timaeus." Classical Quar-
terly 33 (1939): 71-83. Reprinted in Studies in Plato's Metaphysics,
edited by R. E. Allen, 379-420. London: Routledge & Kegan Paul, 1965.
Plato's Universe. Seattle: University of Washington Prei>s, 1975.
"The Role of Observation in Plato's Conception of Astronomy." In Sci-
ence and the Sciences in Plato, edited by John Anton, 1-31. Albany:
Eidos Press, 1980.
9. Sophist
Ackrill, J. L. "ZYMFIAOKH EIAQN." Bulletin of the Institute of Classical
Studies 2 (1955): 31-5. Reprinted in Plato: A Collection of Critical
Essays. Vol. 1, Metaphysics and Epistemology, edited by Gregory
Vlastos, 201-9. Garden City, N.Y.: Anchor Books, Doubleday, 1970.
"Plato and the Copula: Sophist 251-59." Journal of Hellenic Studies 77
(1957): 1-6. Reprinted in Plato: A Collection of Critical Essays. Vol. 1,
Metaphysics and Epistemology, edited by Gregory Vlastos, 210-22. Gar-
den City, N.Y.: Anchor Books, Doubleday, 1971.
Bostock, David. "Plato on 'Is Not.' " Oxford Studies in Ancient Philosophy
2 (1984): 89-119.
Cambell, Lewis. The Sophistes and Politicus of Plato. With a revised text
and English notes. Oxford: Clarendon Press, 1867.
Frede, Michael. Prddikation und Existenzaussage. Hypomnemata 18. Got-
tingen: Vandenhoeck & Ruprect, 1967.
McDowell, John. "Falsehood and Not-being in Plato's Sophist." In Language
and Logos: Studies in Ancient Greek Philosophy Presented to G.E.L.
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Statesman 521
Owen, edited by Malcolm Schofield and Martha Nussbaum, 115-34.
Cambridge: Cambridge University Press, 1982.
Moravcsik, Julius M. E. "Being and Meaning in the Sophist." Acta Philo-
sophica Fennica, Fasc. 14. Helsinki, 1962.
Owen, G. E. L. "Plato on Not-Being/7 In Plato: A Collection of Critical
Essays. Vol. i, Metaphysics and Epistemology, edited by Gregory
Vlastos, 223-67. Garden City, N.Y.: Anchor Books, Doubleday, 1970.
Reprinted in G. E. L. Owen, Logic, Science and Dialectic: Collected
Papers in Greek Philosophy, edited by Martha Nussbaum, 104-37. Ith-
aca, N.Y.: Cornell University Press, 1986.
Roberts, Jean. "The Problem About Being in the Sophist." History of Philoso-
phy Quarterly 3 (1986): 229-43.
Vlastos, Gregory. "An Ambiguity in the Sophist." In Platonic Studies, 270-
322. 2d ed. Princeton: Princeton University Press, 1981.
Wiggins, David. "Sentence Meaning, Negation, and Plato's Problem of Non-
Being." In Plato: A Collection of Critical Essays. Vol. 1, Metaphysics
and Epistemology, edited by Gregory Vlastos, 268-303. Garden City,
N.Y.: Anchor Books, Doubleday, 1970.
10. Statesman
Ackrill, John. "In Defense of Platonic Division." In Ryle: A Collection of
Critical Essays, edited by Oscar P. Wood and George Pitcher, 373-92.
Garden City, N.Y.: Anchor Books, Doubleday, 1970.
Cambell, Lewis. The Sophistes and Politicus of Plato. With a revised text
and English notes. Oxford: Clarendon Press, 1867.
Dorter, Kenneth. "Justice and Method in the Statesman." In Law, Justice
and Method in Plato and Aristotle, edited by Spiro Panagiotou, 105-22.
Edmonton: Academic Printing & Publishing, 1985.
Griswold, Charles L., Jr. "Politike episteme in Plato's Statesman." In Essays
in Ancient Greek Philosophy, edited by John P. Anton and Anthony
Preus. Vol. 3. Albany: State University of New York Press, 1989.
Miller, Mitchell H., Jr. The Philosopher in Plato's Statesman. The Hague:
Martinus Nijhoff, 1980.
Moravcsik, Julius M. E. "The Anatomy of Plato's Divisions." In Exegesis
and Argument: Studies in Greek Philosophy Presented to Gregory
Vlastos, edited by E. N. Lee, Alexander P. D. Mourelatos, and R. M.
Rorty, 324-48. Assen, Netherlands: van Gorcum & Comp., 1973.
Owen, G. E. L. "Plato on the Undepictable." In Exegesis and Argument-
Studies in Greek Philosophy Presented to Gregory Vlastos, edited by E.
N. Lee, Alexander P. D. Mourelatos, and R. M. Rorty, 349-61. Assen,
Netherlands: van Gorcum &. Comp., 1973. Reprinted in G.E.L. Owen,
Cambridge Companions Online © Cambridge University Press, 2006
522 Bibliography
Logic, Science, and Dialectic: Collected Papers in Greek Philosophy,
edited by Martha Nussbaum, 138-47. Ithaca, N.Y.: Cornell University
Press, 1986.
Skemp, J. B. Plato's Statesman. A translation of the Politicus of Plato, with
introductory essays and footnotes. London: Routledge & Kegan Paul,
1952. id ed. Bristol: Bristol Classical Press, 1987.
11. Philebus
Bury, Robert Gregg. The Philebus of Plato. Edited with introduction, notes
and appendices. Cambridge: Cambridge University Press, 1897. Re-
print. New York: Arno Press, 1973.
Dancy, R. M. "The One, The Many, and the Forms: Philebus i5bi-8."
Ancient Philosophy 4 (1984): 160-93.
Davidson, Donald. "Plato's Philosopher." The London Review of Books 7,
no. 14(1985): 15-17-
Dybikowski, J. "False Pleasure and the Philebus." Phronesis 15 (1970): 147-
65.
Frede, Dorothea. "Rumpelstiltskin's Pleasures: True and False Pleasures in
Plato's Philebus." Phronesis 30 (1985): 151-80.
Gadamer, Hans-Georg. PlatosDialektischeEthik - Pha'nomenologishe Inter-
pretation zum Philebus. Leipzig: Felix Meiner, 1931. Reprinted in Platos
Dialektische Ethik und andere Studien zur platonishen Philosophie.
Hamburg: Felix Meiner, 1968.
Gosling, J. C. B. "False Pleasures: Philebus 35c~4ib." Phronesis 4 (1959):
44-54-
"Father Kenny on False Pleasures." Phronesis 5 (i960): 41-5.
Plato: Philebus. Translated with notes and commentary. Oxford: Claren-
don Press, 1975.
Hackforth, R. Plato's Examination of Pleasure (The Philebus). Cambridge:
Cambridge University Press, 1945. Reprint. Indianapolis: Bobbs-Merrill,
no date.
Hampton, Cynthia. "Pleasure, Truth and Being in Plato's Philebus: A Reply
to Professor Frede." Phronesis 32 (1987): 253-62.
Pleasure, Knowledge, and Being: An Analysis of Plato's Philebus. Al-
bany: State University of New York Press, 1990.
Jackson, H. "Plato's Later Theory of Ideas: The Philebus and Aristotle's
Metaphysics 1.6." Journal of Philosophy 10 (1882): 253-98.
Kenny, A. "False Pleasures in the Philebus: A Reply to Mr. Gosling."
Phronesis 5 (i960): 45—92.
Letwin, Oliver. "Interpreting the Philebus." Phronesis 26 (1981): 187-206.
Mohr, Richard D. "Philebus 55c-62a and Revisionism." In New Essays on
Cambridge Companions Online © Cambridge University Press, 2006
Plato: Laws 523
Plato, edited by F. J. Pelletier and J. King-Farlow, 165-70. Guelph: Cana-
dian Association for Publishing in Philosophy, 1983. Canadian Journal
of Philosophy, supplementary vol.
Penner, Terry. "False Anticipatory Pleasures: Philebus 36a3-4ia6." Phrone-
sis 15 (1970): 166-78.
Ross, W. D. Plato's Theory of Ideas. Oxford: Clarendon Press, 1951. id ed.
1953
Schofield, Malcolm. "Who Were 01 SuaxeoeTg in Plato, Philebus 44a ff.?"
Museum Helveticum 28 (1971): 2-20.
Shiner, Roger. Knowledge and Reality in Plato's Philebus. Assen, Nether-
lands: van Gorcum & Comp., 1974.
"Must Philebus 59a-c Refer to Transcendent Forms?" Journal of the His-
tory of Philosophy 17 (1979): 71-7.
Striker, Gisela. Peras undApeiron: Das Problem derFormen in Platons Phile-
bus. Hypomnemata 30. Gottingen: Vandenhoeck & Ruprecht, 1970.
Waterfield, Robin A. H. "The Place of the Philebus in Plato's Dialogues."
Phronesis 25 (1980): 270-305.
Plato: Philebus. Translation with introduction and notes. Harmonds-
worth: Penguin Books, 1982.
12. Laws
Becker, Walter G. Platons Gesetze und das griechische Familienrecht. Mu-
nich: C.H. Beck, 1932.
Bisinger, Josef. Der Agrarstaat in Platons Gesetzen. Wiesbaden: Dieterich,
1925.
Brochard, V. "Les Lois de Platon et la theorie des Idees." Etudes de
philosophie ancienne et de philosophie moderne, edited by V. Delbos,
151-68. New ed. Paris: J. Vrin, 1926.
Cohen, David. "The Legal Status and Political Role of Women in Plato's
Laws." Revue Internationale des Droits deVAntiquite 34 (1987): 27-40.
David, E. "The Spartan Sussitia and Plato's Laws." American Journal of
Philology 99 (1978): 486-95.
Dies, Auguste, and Gernet, Louis. "Introduction." In Platon: Oeuvr'es com-
pletes. Vol. 2, Part 1, Les Lois, edited and translated by Edouard des
Places. Paris: Societe d'Edition, Les Belles Lettres, 1951.
Gorgemanns, Herwig. Beitrdge zur Interpretation von Platons Nomoi. Mu-
nich: C.H. Beck, i960.
Klingenberg, E. Platons vo\iol yewgyiKoi und das positive griechische Recht.
Berlin: Schweitzer, 1976.
Laks, Andre. "Raison et plaisir: pour une caracterisation des Lois de
Platon." In La naissance de la raison en Grece. Actes du Congres de
Cambridge Companions Online © Cambridge University Press, 2006
524 Bibliography
Nice, mai 1987, edited by J. F. Mattei, 291-303. Paris: Presses Universi-
taires de France, 1990.
Morrow, Glenn R. Plato's Law of Slavery in its Relation to Greek Law.
Urbana, 111. University of Illnois Press, 1939.
"The Demiurge in Politics: the Timaeus and the Laws." Proceedings of
the American Philosophical Association 27 (1953-4): 5-23.
Plato's Cretan City: A Historical Interpretation of the Laws. Princeton:
Princeton University Press, i960.
Pangle, Thomas L. The Laws of Plato. Translated, with notes and an interpre-
tive essay. New York: Basic Books, 1980.
Saunders, T. J. Plato: The Laws. Translated with an introduction. Har-
mondsworth: Penguin Books, 1970.
"Notes on the Laws of Plato." Bulletin of the Institute of Classical Stud-
ies, Supplement 28. London: University of London, 1972.
"Penology and Eschatology in Plato's Timaeus and Laws." Classical Quar-
terly 23 (1973)- 2,32-44.
Bibliography on Plato's Laws. 1920-76, with additional citations through
March 1979. 2d ed. New York: Arno Press, 1979.
"On Plato, Laws 728bc." Liverpool Classical Monthly 9 (1984): 23-4.
Plato's Penal Code. Oxford: Clarendon Press, 1991.
Stalley, R. F. An Introduction to Plato's Laws. Oxford: Basil Blackwell, 1983.
13. Letters
Edelstein, Ludwig. Plato's Seventh Letter. Leiden: E.J. Brill, 1966.
Gulley, Norman. "The Authenticity of Plato's Epistles." In Pseudepigrapha
I, chap. 5 Fondation Hardt: Geneva, 1972.
Levison, M., Morton, A. Q., and Winspear, A. D. "The Seventh Letter of
Plato." Mind 77 (1968): 309-25.
Morrow, Glenn R. Plato's Epistles. A translation with critical essays and
notes. 2d ed. Indianapolis: Bobbs-Merrill, 1962.
von Fritz, Kurt. "The Philosophical Passage in the Seventh Platonic Letter."
In Essays on Ancient Greek Philosophy, edited by John P. Anton and
George L. Kustas, 1:408-47. Albany: State University of New York
Press, 1971.
White, Nicholas P. Plato on Knowledge and Reality. Indianapolis: Hackett,
1976.
O. ORAL DOCTRINES
Cherniss, Harold. The Riddle of the Early Academy. Berkeley: University of
California Press, 1945.
Cambridge Companions Online © Cambridge University Press, 2006
Aristotle and later classical thought 525
Findlay, J. N. Plato: The Written and Unwritten Doctrines. New York: Hu-
manities Press, 1974.
Gaiser, Konrad. "Plato's Enigmatic Lecture On the Good." Phronesis 25
(1980): 5-37.
Kramer, Hans Joachim. Plato and the Foundations of Metaphysics. Edited
and translated by John R. Caton. Albany: State University of New York
Press, 1990.
Sayre, Kenneth M. Plato's Late Ontology: A Riddle Resolved, Princeton:
Princeton University Press, 1983.
Vlastos, Gregory. Review of Arete bei Platon und Aristoteles, by H. J.
Kramer. Gnomon 41 (1963): 641-55. Reprinted as "On Plato's Oral
Doctrine." In Gregory Vlastos, Platonic Studies, 379-403. 26. ed. Prince-
ton: Princeton University Press, 1981.
II. ARISTOTLE AND LATER CLASSICAL THOUGHT
Ackrill, J. L. "Aristotle on Eudaimonia." Proceedings of the British Acad-
emy 60 (1974): 339-59. Reprinted in Essays on Aristotle's Ethics, edited
by Amelie Oksenberg Rorty, 15-33. Berkeley: University of California
Press, 1980.
Annas, Julia. "Aristotle on Pleasure and Goodness." In Essays on Aristotle's
Ethics, edited by Amelie Oksenberg Rorty, 285-99. Berkeley: Univer-
sity of California Press, 1980.
Barnes, Jonathan, editor. The Complete Works of Aristotle. 2 vols. Prince-
ton: Princeton University Press, 1984.
Irwin, Terence. Aristotle's First Principles. Oxford: Clarendon Press, 1988.
Kung, Joan. "Aristotle on Thises, Suches, and the Third Man Argument."
Phronesis 26 (1981): 207-47.
Lynch, John Patrick. Aristotle's School: A Study of a Greek Educational
Institution. Berkeley: University of California Press, 1972.
Mitsis, Philip. Epicurus' Ethical Theory: The Pleasures of Invulnerability.
Ithaca, N.Y.: Cornell University Press, 1988.
Owen, G. E. L. "Aristotle on the Snares of Ontology." In New Essays on
Plato and Aristotle, edited by Renford Bambrough, 69-95. London:
Routledge & Kegan Paul, 1965. Reprinted in G.E.L. Owen, Logic, Sci-
ence, and Dialectic: Collected Papers in Greek Philosophy, edited by
Martha Nussbaum, 259-78. Ithaca, N.Y.: Cornell University Press,
1986.
Urmson, J. "Aristotle on Pleasure." In Aristotle: A Collection of Critical
Essays, edited by Julius M. E. Moravcsik, 323-33. Garden City, N.Y.:
Anchor Books, Doubleday, 1967.
Cambridge Companions Online © Cambridge University Press, 2006
526 Bibliography
von Fritz, Kurt. The Theory of the Mixed Constitution in Antiquity: A
Critical Analysis of Polybius' Political Ideas. New York: Columbia
University Press, 1954.
Walsh, James J. Aristotle's Conception of Moral Weakness. New York: Co-
lumbia University Press, 1963.
III. PLATONISM AFTER PLATO
Armstrong, A. H., editor. The Cambridge History of Later Greek and Early
Medieval Philosophy. Cambridge: Cambridge University Press, 1967.
Blumenthal, H. J., and Markus, R. A., editors. Neoplatonism and Early Chris-
tian Thought: Essays in Honour ofA.H. Armstrong. London: Variorum,
1981.
Cassirer, Ernst. The Platonic Renaissance in England. Edinburgh: Nelson,
1953.
Chadwick, Henry. Early Christian Thought and the Classical Tradition.
Oxford: Clarendon Press, 1966.
Dillon, John M. The Middle Platonists, 80 B.C. to A.D. 220. Ithaca, N.Y.:
Cornell University Press, 1977.
Dodds, E. R. "The Parmenides of Plato and the Origin of the Neoplatonic
One." Classical Quarterly 22 (1928): 129-42.
Findlay, J. N. "Appraisal of Platonism and its Influence." In Plato: The
Written and Unwritten Doctrines, 350-412. New York: Humanities
Press, 1974-
Gersh, Stephen. Middle Platonism and Neoplatonism: the Latin Tradition.
2 vols. Notre Dame, Ind.: University of Notre Dame Press, 1986.
Hankins, James. Plato in the Italian Renaissance. 2 vols. Leiden: E.J. Brill,
1991.
Klibansky, Raymond. The Continuity of the Platonic Tradition During the
Middle Ages: Outlines of a Corpus Platonicum Medii Aevi. London:
Warburg Institute, 1939.
Lloyd, A. C. The Anatomy of Neoplatonism. Oxford: Clarendon Press, 1990.
Merlan, Philip. From Platonism to Neoplatonism. 3d ed. The Hague: Marti-
nusNijhoff, 1969.
Pater, Walter. Plato and Platonism: A Series of Lectures. Oxford: Basil Black-
well, 1973.
Patrides, C. A., editor. The Cambridge Platonists. Cambridge: Cambridge
University Press, 1969.
Reale, Giovanni. A History of Ancient Philosophy. Vol. 4, The Schools of the
Imperial Age, edited and translated from the 5 th Italian edition by John
R. Catan, 165-449. Albany: State University of New York Press, 1990.
Cambridge Companions Online © Cambridge University Press, 2006
Editions and study of ancient texts 527
Rees, D. A. "Platonism and the Platonic Tradition." In The Encyclopedia of
Philosophy, edited by Paul Edwards, 6: 333-41. New York: Macmillan
and Free Press, 1967.
Rist, J. M. Plotinus: The Road to Reality. Cambridge: Cambridge University
Press, 1967.
Schmitt, Charles B., editor. The Cambridge History of Renaissance Philoso-
phy. Cambridge: Cambridge University Press, 1988.
Sorabji, Richard. "Myths about Non-Propositional Thought/7 In Language
and Logos: Studies in Ancient Greek Philosophy Presented to G.E.L.
Owen, edited by Malcolm Schofield and Martha Nussbaum, 295-314.
Cambridge: Cambridge University Press, 1982.
Turner, Frank M. The Greek Heritage in Victorian Britain. New Haven:
Yale University Press, 1980.
Wallis, R. T. Neoplatonism London: Duckworth, 1972.
IV. MISCELLANEOUS
A. EDITIONS, TRANSMISSION, AND STUDY OF
ANCIENT TEXTS
Badawi, Abdurrahman. La transmission de la philosophie grecque au
monde arabe. 2d ed. Paris: Libraire Philosophique, J. Vrin, 1978.
Bond, Godfrey W. Euripides: Heracles. Oxford: Clarendon Press, 1981.
Cousin, Victor. Proclus: Commentarium in Platonis Parmenidem. In Procli
Philosophi Platonici: Opera Inedita. Paris: Minerva, 1864.
Couvreur, Paul, editor. Hermiae Alexandrini in Platonis Phaedrum Scholia.
Hildesheim: Olms, 1971.
Creutzer, F. Olympiodorus: In Platonis Alcibiadem Commentarii. Frank-
furt: Officina Broenneriana, 1821.
Diels, Hermann, and Kranz, Walther, editors. Die Fragmente der Vorsokrati-
ker. 6th ed. Berlin: Weidmann, 1952.
Dover, K. J. Aristophanes' Clouds. Edited with an introduction and commen-
tary. Oxford: Clarendon Press, 1968.
Lysias and the Corpus Lysiacum. Berkeley: University of California
Press, 1968.
Friedlein, G. Proclus: In Primum Euclidis Elementorum Librum Comen-
taria. Leipzig: Teubner, 1873. (English translation by Glenn Morrow.
Princeton: Princeton University Press, 1970.)
Gaiser, Konrad. Philodemus: Academica. Supplementum Platonicum, I.
Stuttgart-Bad Cannstatt: Fromann-Holzboog, 1988.
Griffith, Mark. Aeschylus: Prometheus Bound. Cambridge: Cambridge Uni-
versity Press, 1983.
Cambridge Companions Online © Cambridge University Press, 2006
528 Bibliography
Heyduck, Michael. Commentaria in Aristotelem Graeca. Vol. i: Alexander
of Aphrodisias. In Ahstotelis Metaphysica Commentaria. Berlin: Geor-
gius Reimerus, 18 91.
febb, R. C. The Attic Orators from Antiphon to Isaeos. 2 vols. London:
Macmillan, 1893.
Kock, Theodorus, editor. Comicorum Atticorum Fregmenta. 3 vols. Leipzig:
Teubner, 1880-8.
Lasserre, Franc,ois. Die Fragmente des Eudoxos von Knidos. Edited and trans-
lated, with commentary. Berlin: de Gruyter, 1966.
De Leodamas de Thasos a Philippe d'Oponte. Naples: Bibliopolis, 1987.
Lobel, Edgar, and Page, Denys, editors. Poetarum Lesbiorum fragmenta.
Oxford: Clarendon Press, 1955.
Long, H. S. Diogenes Laertius: Vitae Philosophicorum. 2 vols. Oxford: Cla-
rendon Press, 1964. (Translated into English by R. D. Hicks. Diogenes
Laertius: Lives of Eminent Philosophers. 2 vols. Cambridge, Mass., and
London: Harvard University Press and William Heinemann, 1925.)
Page, Denys L. editor. Poetae Melici Graeci. Oxford: Clarendon Press,
1962.
Ross, W. D. Aristotle's Metaphysics. A revised text with introduction and
commentary. 2 vols. Oxford: Clarendon Press, 1924.
Aristotelis Fragmenta Selecta. Oxford: Clarendon Press, 1955.
Aristotle's Prior and Posterior Analytics. With introduction and commen-
tary. Oxford: Clarendon Press, 1957.
Reynolds, L. D., and Wilson, N. G. Scribes and Scholars: A Guide to the
Transmission of Greek and Latin Literature. 3d ed. Oxford: Clarendon
press, 1991.
Sandys, John Edwin. A History of Classical Scholarship. 3 vols. New York:
Hafner, 1958.
Westerink, L. G. Olympiodorus: Commentary on the First Alcibiades of
Plato. Amsterdam: North Holland, 1956.
Anonymous Prolegomena to the Philosophy of Plato. Amsterdam: North
Holland, 1962.
B. MODERN AND CONTEMPORARY WORKS
Abrams, M. H. The Mirror and the Lamp: Romantic Theory and the Critical
Tradition. New York: Oxford University Press, 1953.
Burnyeat, Myles. "Wittgenstein and De Magistro." Proceedings of the Aris-
totelian Society, supplementary vol. 61 (1988): 1-24.
Frege, Gottlob. "On Sense and Reference/7 In Translations from the Philo-
sophical Writings of Gottlob Frege, edited by Peter Geach and Max
Black. 2d ed. Oxford: Basil Blackwell, i960.
Cambridge Companions Online © Cambridge University Press, 2006
Modern and contemporary works 529
Geach, P. T. "Good and Evil." Analysis 17 (1956): 33-42. Reprinted in Theo-
ries of Ethics, edited by Phillipa Foot, 64-73. London: Oxford Univer-
sity Press, 1967.
Goldman, Alvin. Epistemology and Cognition. Cambridge, Mass.: Harvard
University Press, 1987.
Hampshire, Stuart. Thought and Action. London: Chatto & Windus, 1959.
Harman, Gilbert. Change in View. Cambridge, Mass.: Bradford Books, 1986.
Kitcher, Philip. "Positive Understatement: The Logic of Attributive Adjec-
tives." Journal of Philosophical Logic 7 (1978): 1-17.
Kripke, Saul. Naming and Necessity. Cambridge, Mass.: Harvard University
Press, 1980.
Larmore, Charles. Patterns of Moral Complexity. Cambridge: Cambridge
University Press, 1987.
Leibniz, G. W. New Essays on Human Understanding. Translated and ed-
ited by Peter Remnant and Jonathan Bennett. Cambridge: Cambridge
University Press, 1981.
Philosophical Essays. Translated by Roger Ariew and Daniel Garber. Indi-
anapolis: Hackett, 1989.
Locke, John. An Essay Concerning Human Understanding. Edited with an
introduction by Peter H. Nidditch. Oxford: Clarendon Press, 1975.
Putnam, Hilary. "The Meaning of 'Meaning/ " In Philosophical Papers, 2:
215-71. Cambridge: Cambridge University Press, 1975.
Russell, Bertrand. The Problems of Philosophy. London: H. Holt, 1912.
Smyth, Herbert Weir. Greek Grammar. Cambridge, Mass.: Harvard Univer-
sity Press, 1963.
Stich, Stephen, editor. Innate Ideas. Berkeley: University of California
Press, 1975.
Wallace, John. "Positive, Comparative, Superlative." Journal of Philosophy
69(1972): 773-82.
Wheeler, Samuel. "Attributives and Their Modifiers." Nous 6 (1972): 310-
34-
Whitehead, Alfred North. Process and Reality: An Essay in Cosmology.
Edited by D. R. Griffin and D. W. Sherburne. Corrected edition. New
York: Free Press, 1978.
Williams, Bernard. "Pleasure and Belief." Proceedings of the Aristotelian
Society (1959): 57-72.
Ziff, Paul. Semantic Analysis. Ithaca, N.Y.: Cornell University Press, i960.
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (1 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
INDEX OF NAMES AND SUBJECTS
Abrams, M.H., 362n21
Academy, xiii, 20, 28, 31n3, 38n25,
50nn, 170-5, 195n8, 304,
492nll 3
Achilles, 58, 82n35
Ackrill, J.1., 150nl4, 367
Adam, James, 87n78
Adkins, A.W.H., 82n34
Aeschylus, 53, 82n36, 140; see also
drama
agape, 272n30
akrasia, 129-30
Alcibiades I and II, 35m8, 112
Alcibiades, 58, 82n36, 234, 250, 261-2,
263, 347
Alexander the Great, 58, 61
Alexandria, 3In3
Allen, R.E., 40n32, 271n22, 390, 392nlO,
nl5
Anacreon, 249
Anaxagoras, 51, 86-7n75, 240, 375
Anaximander, 52
Anonymus Iamblichi, 83n41 '
Annas, Julia, 44D49, 197n30, Hln2,
332n2, n7, 336n29, n30, 337n32,
364D35,462n37
Antiochus of Ascalon, 31n3
Antiphon, 61, 65
Antisthenes, 400, 414-16
Anton, John P., 274D54
Anytus, 64-5, 165n63
Apology, 3-5, 33n7, n8, 36n20, 37n22,
94, 96, 104, lOS, 106, 112, 114, r I 5,
118n41, 124, 125, 134, 136, 138,
Arcesilaus, 3In3
Archilochus, 340
Archytas, 37n24, 173, 176
aristocracy, 48 5n I 3
Aristophanes, 54,61,74-5, 80n20,
87n78 n79, 362n18; see also drama
Aristotle, xiii, 4, 5,7, 18, 20, 22, 23, 28,
30n2, 31n2, 34D13, nl5, 35n 17,
36n2o, 38n26, 39n27,41n34,45n51,
48n70, 50n77, 5I, 52, 55, 58,73,
77-8, 79n S, 86n 73,87-8n81,
89n85, 90, 107, 122-3, 148n7,
155n24, 159n42, 162n53, 170, 171,
217n13, 224D40, n41, 240, 303,
310n49, 311, 320, 333nlo, 357, 361,
365,414-15, 452,453,457n 5,
489n61,49Inl08
Aristoxenus, 4sns I
arithmetic, 3, 126-7, 191
Armstrong, A.H., 272n30
Arnim, H. von, 107-9, 114, 120n69
Asmis, Elizabeth, 275n67, 276n71,
364D38
astronomy, 3, 171, 174, 192-4,479-80
Athens, 2, 31n3, 38n25, 58-9, 61-2, 64,
67-8,227-30,233
Bacon, Helen, 274D54
Bambraugh, Renford, 44D48
Barker, Ernest, 44D48
Barnes, Jonathan, 39n27, 82n32
Baron, C., 100, 120n69
Battin, M. Pabst, 364D36
Beauty, Form of, 43n41, 256, 259-60,
'\~R '\nT
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
INDEX OF NAMES AND SUBJECTS
Abrams, M.H., 3621221
Academy, xiii, 20, 28, 31223, 381225,
502277, 170-5, 195^8, 304,
49222113
Achilles, 58, 822235
Ackrill, J.L., 1502214, 367
Adam, James, 872278
Adkins, A.W.H., 822234
Aeschylus, 53, 822236, 140; see also
drama
agape, 2722230
akrasia, 129-30
Alcibiades I and //, 352218, 112
Alcibiades, 58, 822236, 234, 250, 261-2,
263, 347
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (2 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Alexander the Great, 58, 61
Alexandria, 31223
Allen, R.E., 402232, 2712222, 390,
3922210,
2215
Anacreon, 249
Anaxagoras, 51, 86-72275, 240, 375
Anaximander, 52
Anonymus Iamblichi, 832241
Annas, Julia, 442249, 1972230, 331222,
332222, 227, 3362229, 2230, 337*332,
3642235, 4622237
Antiochus of Ascalon, 31223
Antiphon, 61, 65
Antisthenes, 400, 414-16
Anton, John P., 2742254
Anytus, 64-5 , 1652263
Apology, 3-5 , 33237/ *38, 362220, 372222,
94, 96, 104, 105, 106, 112, 114, 115,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (3 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
1182241, 124, 125, 134, 136, 138,
34^, 355
Aquinas, Thomas, 2472232
Arcesilaus, 31223
Archilochus, 340
Archytas, 372224, 173, 176
aristocracy, 4852213
Aristophanes, 54, 61, 74-5, 802220,
872278 2279, 3622218; see also drama
Aristotle, xiii, 4, 5, 7, 18, 20, 22, 23, 28,
30222, 31222, 34*313, *3I5, 35«i7,
362220, 382226, 392227, 412234, 452251,
482270, 502277, 51, 52, 55, 58, 73,
77-8, 79225, 862273, 87-82281,
892285, 90, 107, 122-3, 148227,
1552224, 159*342, 1622253, 170, 171,
2172213, 2242240, 2241, 240, 303,
3102249, 311, 320, 333*3io, 357, 361,
365, 414-15, 452, 453, 457*35/
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (4 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
4892261, 49122108
Aristoxenus, 452251
I
9
I
arithmetic, 3, 126-7,
Armstrong, A.H., 2722230
Arnim, H. von, 107-9, "4 / 1202269
Asmis, Elizabeth, 2752267, 2762271,
364*338
astronomy, 3, 171, 174, 192-4, 479~8o
Athens, 2, 31223, 382225, 58-9, 61-2, 64,
67-8, 227-30, 233
Bacon, Helen, 2742254
Bambraugh, Renford, 442248
Barker, Ernest, 442248
Barnes, Jonathan, 392227, 822232
Baron, C , 100, 1202269
Battin, M. Pabst, 3642236
Beauty, Form of, 432241, 256, 259-60,
268, 291
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (5 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Becker, Walter G., 4882255
531
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (6 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
5 32 Index of names and subjects
being, 3101247, 399~4O2, 439, 454; see
also not-being
Bekker, Immanuel, 3 1222
Belfiore, Elizabeth, 3632227
Beversluis, John, 216227
Billig, L., 106-8, n o
Bisinger, Josef, 4882255
Blass, Friedrich W., 100, 102
Bluck, R.S., 372225, 1542221, 1552223,
2252242
Bolotin, David, 269224
Borgeaud, Phillipe, 2462220
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (7 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Bosanquet, Bernard, 3632225
Bostock, David, 217229, 367, 401-2, 422,
423221, 424228
Brandwood, Leonard, 342216, 362221,
422239,2240, 472261, 116228,2216,
1172220, 2223, 2231, 1182236, 2249,2252,
1192255, 1202268,2269,2271, 148227,
228, 215221
Bremmer, Jan, 2472225
Brenkman, John, 2742254
Brentlinger, J., 3942222
Brickhouse, Thomas C , 32226
Brochard, V., 4862223
Brock, Roger, 872279
Burger, Ronna, 269223
Burkert, Walter, 230, 237, 241, 242,
245221, 226,228, 229, 245-62214,
2462217, 2219,2222, 2223,2224, 2472228,
2234, 2238, 3622224
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (8 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Burnet, John, 31222, 1652263
Burnyeat, Myles, 472260, 862269, 1992243,
2182219, 225-62242, 2752264, 305229
Bury, Robert Gregg, 269222, 2702210,
2722231, 2732247, 2253, 457122
Callicles, 58, 61, 65, 434, 449, 458229,
4592212
Campbell, L., 91, 94, 109, 113
Carson, Anne, 270228, 2752265
Cassirer, Ernst, 31223
Castaneda, Hector-Neri, 3062215
cause, 53, 54,428
cave, allegory of 10-12, 299, 327-9,
335^25
censorship, 472
Chaerephon, 832239
change, 71
Charmides, 3, 5, 362220, 96, 109, 112,
124, 1662269
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (9 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Charmides, 60-1
Cherniss, Harold, 382226, 462258, 482269,
2270, 78221, 194223, 1962218, 2236,
2252242, 3092242, 3632230, 366-7
chronology of dialogues, xiii, 5, 9, 15-
20, 462257, 862272, 90-120, 215221,
458228, 484225
Clay, Diskin, 872279, 2702213
Cleitophon, 352218, 112
Cleon, 842250, 852261
Code, Alan, 3942220
Cohen, David, 2702216, 4912297
Cohen, S. Marc, 3932213
Cole, Susan G., 2462219
Collingwood, R.G., 353
comedy, see drama
common meals, 474
convention, 54-5, 61
Cooper, John Mv 3332212, 3342214,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (10 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
3362227, 3371232
Cope, E.M., 842253
Cornford, Francis MacDonald, 791211,
2212, 194223, 1982242, 2722233,
2752268, 3922210
cosmological argument, 2472232
Coventry, Lucinda, 872280
craft[techne], 125-7, 342-3, 353, 355, 358
Crates, 31123
Cratylus, 362221, 422238, 2239, 862272,
94,
95, 96, 102, 103, 109, no , 112, 115,
124, 3092241
Cratylus, 2, 391*27, 51, 69, 77, 79125, 123
Creed, J.L., 822234
Creophylus, 354
Crete, 474
Critias, 9, 15, 16, 18, 33228, 90, 91, 96,
97, 100, 102, 103, 104, 105, 106,
107, 108, no , in , 112, 113, 114,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (11 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
1162210, 1182235, 1182237, 1182241,
1192259, 1192263
Critias, 60-1
Crito, 5, 7, 96, 104, 105, 106, 109, 112,
115, 124, 125
Crombie, I.M., 78223, 79225, 457222
Cross, R.C., 331222, 332223
Dancy, Russell M., 457222
David, E., 4882259
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (12 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Index of names and subjects
Davidson, Donald, 4592216
definition, 3-4, 5, 7-8, io, 472260, 70,
123, 127, 1532221, 163-42260, 200-4,
277-80
Definitions, 352218
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (13 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
democracy, 20, 28, 58-9, 61-2, 64, 67,
474, 476, 485*113/ 489*164
Democritus, 51, 52, 54-5, 56-7, 61, 69,
79226, 79-802212, 802217, 86-72275
Demos, Raphael, 3362229
Demosthenes, 842251
de Romilly, Jacqueline, 361229
Derrida, Jacques, 3632224
Descartes, Rene, 277
Detienne, Marcel, 2462219
Deubner, L., 245226
de Vries, G.J., 269223
dialectic, 13, 20-1, 30222, 171, 183-94,
359, 427, 437, 456; see also dialogue
form; Socratic method
dialogue form, 3, 25-30, 73-8, 122,
131-4
dianoia, 184
Diaz Tejera, A., 109-1 0
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (14 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Dicks, D.R., 1962216
Dies, Auguste, 3922210, 424222, 4902283
Dillon, John, 31223
Diogenes Laertius, 15, 33227, 352216,2218,
78223
Dion, 382225
Dionysius I, xiii, 21-2 , 382225
Dionysius II, 382225
diorismoi, 175-7
Diotima, 872279,233-4,250, 252-62,
263, 267-8, 2732251, 344-7, 354, 356
Dittenberger, W., 91-4, 113
divided line, analogy of, 183-94,
2222234, 2262243
division, method of, 422239, 359, 427,
438-9 , 466, 4852210
Dodds, E.R., 362220, 372225, 802216, 124,
1492210, 2462219
Dorter, Kenneth, 2742255
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (15 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Dover, K.J., 422239, 822234, 832242,
872279,
1162211, 1492210, 269222, 2702216,
2712219,2227, 2742256, 2752261,
3622220
drama, 25, 67-8, 74-5, 338, 341, 343~4,
348,451-2
Dybikowski, J., 4602221
Edelstein, Ludwig, 78224
education, 58, 64, 68, 126, 130, 134,
136-8, 1502216, 170-5, 255-62, 265,
323, 339-4O, 344-9, 353-5, 358-
6i,47i-3
egoism, 134-7, 1552223, 1622256
Ehrenberg, Victor, 832238
elenchus; see Socratic method
emotions, 341, 356-7, 450-2, 456
Empedocles, 51, 73, 230, 240
Epicrates, 171-2
Epicurus, 4612229
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (16 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Epinomis, 352218, 96, 112, 113
Epistles, see Letters; Letter VII
equality, 474-6
eristic, 66, 68, 76, 852257, 872280
error, involuntariness of; see involuntari
ness of error
eternity, 392228
Euclid, 175-7, 178
Eudemus, 1962223
Eudoxus, 173, 175, 1952212, 1962217
Euenus, 138
Euripides, 68, 74, 802221, 822236, 872277,
235, 344, 3622218; see also drama
Euthydemus, 5, 362221, 66, 862272, 94,
96, 109, 112, 115, 124, 135, 138,
414-16
Euthyphro, 3, 5, 7, 362220, 53-4, 96, 109,
112, 124, 277-8
falsity, 397-8, 412-23, 442-3; see also
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (17 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
truth
Ferrari, G.R.F., 269223, 2752266,2267,2269,
2762271, 363*227
Field, G.C., 78221, 862274, 194226
Fine, Gail, 412234, 422237, 2238, 862269,
862271, 217229, 2202224, 2222234,
2252241, 3052211, 3082223
flux, 19, 39*227, 55-6 , 70, 123, 124
Fontenrose, Joseph, 2452210
Forms, theory of, 4, 7-14, 16-19, 23-4,
28, 72, 123-4, 127, 180-94, 234,
240, 259, 265-6, 279-310, 316-23,
365-8, 372-3, 374-8, 389-91,
3952227, 454-5, 464-70
Foster, M.B., 332223
Frajese, Attilo, 1962222
Franklin, Benjamin, 4902288
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (18 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (19 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
534 Index of names and subjects
Frede, Dorothea, 4582210, 4602225,
4622238
Frede, Michael, 381, 3962231, 424225,
4602225, 4622239
Frege, Gottlob, 147
Freud, Sigmund, 2752268
Fritz, Kurt von, 79224
Foucault, Michel, 2702216
Furley, David J., 79229, 822232, 394-
52222
Furth, Montgomery, 8122 31
Gadamer, Hans-Georg, 4572*0..
Gagarin, Michael, 2742254
Gaiser, Konrad, 194225
Gallop, David, 382226, 1972226
Gandhi, Mohandas, 136
Geach, P.T. 1602249, 1682278, 216225
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (20 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Gentzler, fyl, 2212228
geometry, 3, 6, 7, 126-7, 171, 172-3,
184-6, 188, 191-3, 452
Gerson, Lloyd, 2472230
Gifford, E.H., 852263
Gillispie, Charles C., 1952212
Glidden, David K., 269224
gods, 52-4, 227-30, 241; see also inspi-
ration,- theodicy
Goldman, Alvin, 2192220
Good, Form of, 3, 13-14, 342214, 432241,
452251, 88n82, 992282, 127-8, 171,
184, 190, 193, 194, 260, 298, 303,
311, 320, 321-2, 323, 335*222,2223,
349, 468
Gorgemanns, Herwig, 484224
Gorgias, 5, 27, 362220, 372224,2225,
382225, 432242, 862266, 96, 109, 112,
124-5, 136, 1562231, 343-4, 346,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (21 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
356, 359,434,449,456
Gorgias, 138, 340-1 , 344, 348, 349,
458229
Gosling, J.C.B., 1532221, 2252242,
3052211,
332225, 395*226, 442, 456221, 457*22,
223, 224, 458226,2210, 4592214, 4602221,
4612229, 2230, 4622237, 2240
Gould, Thomas, 269227
Greene, W.C., 872279
Griffith, Mark, 842252
Griswold, Charles L., Jr., 492273, 502278,
269223, 2752266, 2267, 4892264
Grote, George, 33227, 842253, 91
guardians, 12-13 , 19-20, 130, 298-9 ,
327-9 , 347, 464-5 , 480, 4892265
Gulley, Norman, 78224, 19 62219
Guthrie, W.K.C., 30221, 33227, 352217,
2218, 422239, 482269, 502277, 802217,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (22 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
812228, 842253, 852263, 882282,
1492210, 1542221, 216224, 237,
2462216, 2702213, 485227, 2211,
4862223
Hackforth, R., 382226, 802218, 269223,
2732247, 2752266, 2267, 457222, 458226,
4622240
Halperin, David, 269221, 2702216,
2732251, 2752269,2270
Hampshire, Stuart, 1602249
Hampton, Cynthia, 457222,224, 225,
4602225, 4622240, 2241
happiness, 5, 9-10, 372222, 69, 852265,
135, 139, 1502214, 1602246, 2248, 254,
2712227, 313, 319, 331*23, 332224,
3352220, 427, 431, 464-5, 469, 472,
478-9, 481
Harman, Gilbert, 2192220
harmonics, 3, 171
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (23 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Harvey, F. David, 4892261
Havelock, Eric, 3 61221
Heath, Malcolm, 2762271
Heath, Thomas, 19 62215
Hector, 822235
Heidel, W.A., 812223
Heitsch, Ernst, 1972225
Heraclitus, 2, 392227, 51, 52, 55-6, 69-
70, 73-4, 240, 245224, 340
Herodotus, 58-9, 74, 230
Hesiod, 228, 340, 345, 346, 354
hiatus, 100-3, 113-14, 119-202263,
1202271
Hipparchus, 352218
Hippias, 138
Hippias Major, 3, 5, 362221, 96, 109, 112,
115, 124, 304222
Hippias Minor, 5, 109, 112, 124, 132-4,
145-6
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (24 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Hippocrates of Chios, 1952213, 195-
62215
Homer, 52, 53, 58, 59, 60, 138, 139, 228,
236, 245224, 338, 339, 342, 344, 345,
346, 351, 353-4, 36o, 363*224,
3642233, 2234, 450, 4612233
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (25 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Index of names and subjects
homosexuality, 251-2 , 260-1 , 265, 270-
12216, 344-5
Hornblower, Simon, 832243
Howland, R.L., 851263
hubris, 231, 242
Hume, David, 821233
Hussey, E.L., 832241
hypothesis, 37*324, i77~9 4
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (26 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Iamblichus, 31223, 2762271
imitation, see mimesis
induction, 123, 147222
innate ideas, 213, 2242240, 224-52241
inspiration, 138, 230-5 , 339, 342-3 , 359
involuntariness of error, 129-30 , 132-4 ,
156-72232,472
/022, 5, 96, 109, ii2 , 124, 138, 342, 354,
355
Irwin, T.H., 30222, 33226, 362220, 372224,
78221, 802214, 812227, 2229, 822234,
2237, 832240, 852257,2263, 2264, 862265,
2269, 882281, 147221, 1492214, 1542221,
1612249, 1662272, 2172210, 2202224,
2212226, 2227, 2222230, 2722235,
2732242, 2752259, 3052211, 3082232,
331222,223, 332223, 333*312, 334*314,
2218, 3362229, 337^32, 394*317, 458-
92211,4592217
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (27 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Isenberg, Meyer W., 2702213, 2712223,
2732250
Isocrates, 66-7 , 842252, 852263, 103,
1202271
Jackson, Henry, 457225
Jaeger, Werner, 2472230
Janell, G., 100-3
Jebb, R.C., 852261
Jones, A.H.M., 832245
Joseph, Horace W.B., 331222
Jowett, Benjamin, 3962237
justice, 13, 53, 61, 69, 70, 3H-37 , 478
Justinian, 31223
Kahn, Charles, 362218,2220, 392227,
79227,
274-52259
Kaluscha, W., 103-6, 108, n o
Kantianism, 133-7, 1622253,2255,
4612231
Kenny, A., 4602221
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (28 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Kerferd, G.B., 842252, 2253, 852257
Kidd, I.G., 812223
kinds, 3952227
Kirk, G.S., 392227, 812228, 872275
Kirwan, C.A., 332223
Kitcher, Philip, 3062213
Kitzinger, Rachel, 402229
Klibansky, Raymond, 31223
Klingenberg, E., 4882255
Klosko, George, 442248
knowledge, 3, 6, 16, 19, 472260, 71, 129-
30, 200-26, 277-80, 284, 300, 377,
387-9 , 427, 430, 454, 458227; politi-
cal, 466-7 , 470, 4892264; see also
craft; self-knowledge
Kosman, L.A., 2752269
Kramer, Hans Joachim, 452251, 482269
Kraut, Richard, 362219, 147221, 1492212,
1502214,2216, 1582240, 1592241,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (29 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
1622251,2254, 1642260, 1662268,2272,
1682279, 2182216,2217, 332225,
3362229, 337*332,2235, 394*317
Kripke, Saul, 2232238
Kung, Joan, 472262
Laches, 3, 5, 362220, 96, 109, 112,
1172226, 124, 141-2, 146
Laks, Andre, 4872233, 492221 n
Larmore, Charles, 2472239
Lasserre, Francois, 194-5227, 1952212,
269226, 270228
law, 477-8
Laws, 9, 10, 15, 16, 18, 19-20, 352217,
37-82225, 832244, 842249,2254, 88 -
92283, 90, 91, 94, 96, 97, 100, 102,
103, 104, 105, 106, 107, 108, 109,
no , in , 112, 113, 114, 1182241,
1192263, 1202265, 240-4 , 338, 361,
464-9 2
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (30 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Lebeck, Anne, 2742255
Ledger, G.R., 112, 114
Leibniz, G.W., 2242240, 2241
lemmas, 175-7
Leodamas, 176
Lesbianism, 252, 2712218
Letter VZ/xiii, 21-4 , 28, 33228, 382225,
51, 79224, 832240, 842248, in , 112,
113, 130,49222113
Letters, 33228, 352218, 382225, 78-9224,
112, 49222113; see also Letter VII
Letwin, Oliver, 457224
Leucippus, 52
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (31 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (32 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
536 Index of names and subjects
Levinson, Ronald B., 442248, 4892276 Moravcsik, J.M.E., 2222229,
2242241,
Lewis, David, 832241 2732242, 4852210,2211
limit [peras), 427-30, 438—9 Morgan, Michael L., 2462215, 2472227,
Lloyd, A.C., 1982240 2231,2237
Lloyd-Jones, Hugh, 822234, 2462218 Morrison, J.S., 852255
Locke, John, 2172210 Morrow, Glenn R., 78-9224, 802219,
logos (language, statement), 341, 822237, 842247, 3092242, 478, 487*333,
342, 344, 413-18, 445-6; see also 2240, 4882255, 2259, 4892263, 2267,
definition 4902282, 49122102,22104, 49222113
Long, A.A., 822234 Mosse, Claude, 832243
lot, 59 Mourelatos, Alexander P.D., 1992245,
love, 144, 248-76, 344-7, 456; see also 3O9JH3
sex Mueller, Ian, 342216, 472258, 1982234,
Lutoslawski, W., 97—100 1992245, 304222
Lycurgus, 345, 354 Murdoch, Iris, 442248, 3622218
Lynch, John Patrick, 194221 Murphy, N.R., 1562227, 331222
Lysis, 5, 33227, 362221, 91, 96, 109, 112, Mylonas, G., 2452212
115, 124
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (33 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
names, 422238, 413
nature, 54-5, 61
Nehamas, Alexander, 402230, 209,
Mabbott, J.D., 332223,227
216223,224, 2182216, 2219, 2212226,
MacDowell, Douglas M., 245225
2222232, 2232237, 2732244, 305228,
Mackenzie, Mary Margaret, 4882250
madness, 129, 235, 239, 358-9 363^31
Markus, R.A., 2712223, 2722230 Neoclides, 176
Matthews, Gareth B., 304221 Neo-platonism, 31223, 189-90
McDowell, John, 2202224, 422, 423221, Nettleship, Richard, 331222
Neumann, Harry, 2732243,2247, 2251
424228
Nilsson, Martin, 245221
Meinwald, Constance C , 462255, 36522,
Nocturnal Council, 243, 465, 467-8,
Menaechmus, 173 470, 471, 475, 477, 479, 484,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (34 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Menedemus, 171, 194227 4872231, 2235, 489*363, 4902282, 491
222109
Menexenus, 5, 33228, 372225, 96, 102,
not-being, 56, 71, 397~9, 403-12; see
112, 114, 115, 1172226, 1182241, 124
also being
Me22O, 6, 7, 9, 14, 362221, 372224, 382225,
Nussbaum, Martha C , 2712222, 2742254,
96, 109, ii2 , 138, 1562231, 1652263,
1662269, 177-80, 187, 200—26, 236-2752269, 334*313, 335*323, 364*339,
463*343
8,342
Menza, V., 3632227 Ober, Josiah, 842251
method, see dialectic,- division, method objectivity, 291-2
of; hypothesis; Socratic method O'Brien, Michael J., 2712222, 2722229
Middle Platonism, 31223 Okin, Susan Moller, 442249, 4912297
Mikalson, Jon D., 245221 oligarchy, 60, 62, 842246, 4852213
Mill, John Stuart, 1632256 oral opinions of Plato; see unwritten
Miller, Mitchell, Jr., 392-32210, 3932215 opinions
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (35 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
mimesis, 346, 347-59 Owen, G.E.L., 15, 462252, 2254, 472259,
Minos, 352218, 112 812231, 305226,227,228, 3062212,2214,
Mitsis, Philip, 4612229 30722221, 3082226, 3102249, 367,
Mohr, Richard D., 2732246, 4622240 3922210, 395-62229, 407-8 , 423*31,
monarchy, 474, 476, 4852213 424225,4852211
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (36 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Index of names and subjects
paradigms, 19, 462254, 293-4, 3°°/ 328-
9,366
Parke, H.W., 245226,227,2210
Parker, Robert, 80-12223, 245221
Parmenides, 9, 10, 13, 15, 16-19, 39*228,
412236, 452252, 51, 90, 94, 96, 100,
109, no , 112, 114, 115, 1162217,
1182247, 1192260, 125, 293, 301,
365-96
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (37 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Parmenides, 2, 472261, 55, 56, 70-2, 73,
240, 340
participation, 295, 301
particles, Plato's use of Greek, 91-4
Patterson, Richard, 392228, 402237, 2238,
3082228, 3342216
Pelopponesian War, xiii, 59-60, 75, 229
Penner, Terry, 382226, 147221, 150-
12217,
1512218,2219, 1522220,2221, 153,
1552224, 1562225,2227,2229, 1572232,
*239,
I
59^4
1
/ 1602242,2248, 1622253,
2254, 1632258, 1652263,2264, 1662266,
2272, 2172210, 304225, 3082225,
3332212, 334*215/ nij, 4602225
Pericles, 59, 75, 852261, 138, 1652263
Persian Wars, 59 229
Peterson, Sandra, 3932213, 3962233
Phaedo, 7, 8, 9, 14, 18, 26, 362221,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (38 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
372224, 96, 107, 109, no , 112, 115,
124, 125, 127, 180-3,
I
87, 238-9,
279, 280-5, 435, 4612230
Phaedrus, 9, 10, 13, 18, 20-2, 27, 30222,
35ni7, 4^39 , 862266, 91, 94, 96, 97,
100, 102, 103, 109, no , in , 112,
113, 114, 115, 118-192252, 1192258,
2271, 239, 2462220, 262-8, 358-61,
425-63,455
Philebus, 9, 10, 13, 15, 16, 17, 18, 26,
30222, 452251, 482270, 94, 96, 97, ioo,
102, 103, 104, 105, 106, 107, 108,
109, no , in , 112, 113, 114,
1192263, 1202265, 3
O1
/ 335222-1,
3362227, 379, 392229
Philip, A., 2742255
Philip of Opus, 352218
Philodemus, 172-3, 177, 195229
philosopher-kings, see guardians
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (39 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
philosophy, see dialectic
piety, 53-4, 232-3, 242, 244
Pindar, 230, 236, 237
Plass, Paul, 1972228
pleasure, 16, 432245, 129, 302, 312-14,
338, 344, 351, 425-63; false plea-
sure, 442-52; true pleasure, 452-6
Plochmann, G.K., 2702213
Plotinus, 31223, 392228
Plutarch, 31223, 352217, 812229, 173
Polemon, 31223
political knowledge; see knowledge, po
litical
Politicus; see Statesman
Popper, K.R., 442248, 4892276
Porphyry, 31223
Powell, C , 245225, 2462221
predication, 367, 374, 378-84, 400-2
Price, Anthony W, 269224, 2712222,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (40 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
2722234, 2732241, 2242, 2249, 2742257,
2752263, 2268,2269, 3622220
Prior, William J., 382226, 462255, 46 -
72258, 304222, 4622240
privacy, 471, 474
private property, 63, 842249, 872278, 471,
473-4
Proclus, 31223, 176-7, 3922210
Prodicus, 138
Protagoras, 5, 27, 96, 104, 105, 106, 109,
112, 1162214, 1192258, 124-5, 141-2,
1562231, 2232, 433-4, 447-8
Protagoras, 57, 61, 65, 66, 72-3 , 138,
4i5
punishment, 136, 1612250,2251, 241,
472-3
Putnam, Hilary, 2232238
Pyrilampes, 832239
Pythagoreans, xiii, 2, 382225, 51, 79227,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (41 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
123, 124-5, 151-22219, 229, 231,
236-7, 239
Reale, Giovanni, 482269
reality, degrees of, 11-12, 442246, 2247,
291-2, 298, 322, 335-62225
recollection, 6, 207-8, 213-15, 236-8
Rees, D.A., 32223
Reeve, C.D.C., 32226, 3052211, 331223,
332223, 225,226, 333*212, 334*313,
3362227,2229, 337*232
relations, 285-7
relativism, 139
reply formulae, 96-9, 107-9, IJ4 ,
1202271
Republic, 3, 4, 5, 6, 8, 9, 10-14, 15, 16,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (42 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (43 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
538 Index of names and subjects
Republic [cont.) scapegoat, 362-31224
18, 26, 28, 30122, 341215, 371224, Schanz, M., 94, 113
421238, 43^41, **45, 44**46, 45«5i/ Schmitt, Charles B., 31-2123
491272, 70-1 , 871278, 88-91283, 90, Schofield, Malcolm, 390, 4611230
94, 96, 97, 100, 109, no , HI , 112, Scott, Dominic, 2241241, 2251241
114, 115, 1161218, 1171235, 1181241, Scully, Vincent, 2451211
1191258, 124, 126-7,
I2
9 , !3° , self-knowledge, 140—1, 231, 268
1551221, 1651263, 170-1 , 183-94, self-predication, 18, 293-4, 300-1,
365,
2261243, 228, 239, 241, 242, 256, 374-5, 383-9, 400-2, 424126
2701214, 2721233, 279, 284-5 ,
2
98-9 , Sellars, Wilfred, 3931213
3H-37, 338, 346, 347-58, 366, 431, separation of Forms, 411234, 123
435-7, 440-1, 453, 456, 461/230, Seventh Letter, see Letter VII
480, 483, 4891265 sex, 248, 249, 251, 260-1 , 265, 267,
Reynolds, L.D. 481268 269121,125, 2751268, 325-6, 434,
rhetoric, 66-8, 76, 871280, 126, 137-9, 449-50, 487«37
343-4, 359-60; see also Gorgias Sharpies, R.W., 1971225
rhythm of Plato's prose, 103-8, no-n , Shiner, Roger, 457122, 4621240
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (44 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
113-14, 1201271 Shorey, Paul, 411235, 821237, 194123
Richardson, N.J., 2471226 Sicily, xiii, 381225, 91, no , 124-5,
Ritter, Constantin, 96-7 , 100, 109, 113-49212113
14, 116124,1215, 148128 Sidgwick, Henry, 841253
Rivals, 351218 Siebeck, Hermann, 114, n 61215
Roberts, J.W. 831238 Simonides, 340
Roberts, Jean, 424124 Simplicius, 174
Robin, Leon, 2721234, 2731249 Skemp, J.B., 4851211
Robinson, David B., 269124 skepticism, 31123, 57, 72, 73, 277-8, 304
Robinson, Richard, 30121, 2171215 slavery, 480-1
Rome, 31123 Smith, J.A., 41-21237
Rosen, Stanley, 269122, 270129 Smith, Nicholas D., 451249
Ross, W.D., 30121, 34^15, 421239, 79**5, Smyth, Herbert Weir, 401229
148125, 3101246,1251, 391-2122, Snell, Bruno, 269-70128
457122 Socrates, xiii, 2-7 , 17, 26-7, 52-3, 60,
Rowe, C.J., 269123, 2761271 68-70, 74-5 , 871276, 87-81281, 121
Russell, Bertrand, 297, 3091234,1236 169, 363*124
Ryle, Gilbert, 451252, 1191262, 390, Socratic fallacy, 1681278
3921210 Socratic ignorance, 3-4, 139-147
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (45 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Socra tic intellectualism, 128-9
Sabine, George, 441248 Socratic method [elenchus), 4, 65-6, 68
Sachs, David, 332123, 3361229 9, 76, 126, 131-4, 139-47, 204-13,
Samuel, Alan E., 30121 370, 432
Sandys, John Edwin, 31122 Socratic paradox; see involuntariness of
Santas, G.X., 147121, 1571232, 269127, error
2711226,1228,1229, 2731242,1249, Solmsen, Friedrich, 871277
2741257, 2751268, 335*12,3, 3361227 Solon, 58, 345, 354, 360
Sappho, 249 Sophist, 9, 10, 13, 15, 16, 18, 431244, 72,
Saunders, Trevor, 242, 243, 4861223, 90, 91, 94, 96, 97, 100, 102, 103,
4871235, 4881250, 4901285, 49212113 104, 105, 106, 107, 108, 109, no ,
Sayre, Kenneth M., 461254, 471258, in , 112, 113, 114, 293, 3Oi, 303,
481270, 393HIO, 1215, 395^23, 457*12, 3101247, 363-41232, 379, 381-
2,
395-61229, 397-424
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (46 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (47 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Index of names and subjects
sophists, 2, 63-6, 137-9, 340-2,
3641332; see also Antiphon; Euenus,-
Gorgias; Hippias; Prodicus;
Protagoras
Sophocles, 74, 872377, 230; see also
drama
Sorabji, Richard, 392328, 3082323
soul, 6, 9, 10, 13, 39*228, 1^3, 125, 129,
134-7, 207, 2252342, 230, 232, 236,
240-1, 2462216, 262-4, 316, 441
Sparta, 2, 59-60, 62, 229, 474
Speusippus, 31233, 502377, 78213, 170,
194237, 4612330
Stalley, R.F., 4862222,2323, 4872340,
4882260
Statesman, 9, 13, 15, 16, 18, 422338, 90,
91, 94, 96, 97, 100, 102, 103, 104,
105, 106, 107, 108, 109, no , in ,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (48 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
112, 114, 127, 128, 379, 466-8, 471,
482, 4862328, 4892364
Stephanus, 30-1232
Stephens, Susan A., 482268
stereometry, 171, 1992345
Stich, Stephen, 2252341
Stokes, Michael C , 502378, 2712221,
333-42313
Stone, I.F., 32226
Strang, Colin, 3932313
Strauss, Leo, 48-92372, 492375, 49-502376
Striker, Gisela, 457222, 234
stylometry, 4-5, 15, 19, 90-120, 124
Symposium, 9, 18, 362221, 422239, 432341,
872179, 91, 96, 107, 109, no , in ,
112, 113, 115, 124, 233-4, 248-62,
265, 287-8, 291, 344-7, 352, 354,
356,358-9,360,3642234,455
synthesis, 175-7
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (49 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Syracuse, xiii, 21, 382225, n92359
Szlezak, Thomas A. lyin^i, 2732353
Talleyrand, Charles Maurice de, 485237
Taylor, A.E., 792211, 2212, 852257,
1992347
Taylor, C.C.W., 822334, 832241, 1532221,
1682378, 392222
teaching, see education
technical terms, Plato's increasing use
of, 91
Teloh, Henry, 4622340
Thales, 354
Theaetetus, 9, 10, 15, 16—19, 26, 372225,
472260, 70, 72-3, 862372, 90, 94, 96,
97, 100, 109, no , 112, 114, 115,
1192259, 125, 138, 146, 2202224, 415,
439
Theaetetus, 372225, 176
Theages, 352218, 112
Themis tocles, 1652263
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (50 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
theodicy, 479
Thesleff, Holger, 352216, 116221,234,
458228
Third Bed argument, 366
Third Man argument, 18, 365-96
Thirty Tyrants, 60, 62, 832340
Thompson, E.S., 852257
Thompson, W.H., 852363, 269223,
2752266
Thrasyllus, 352216
Thrasymachus, 61, 65
Thucydides, 60-1, 63, 74
Tigerstedt, E.N., 492273
Timaeus, 9, io, 14, 15, 16, 18, 19, 22,
33228, 39*228, 42-32240, 47^58, 2359,
90, 91, 94, 96, 97, 100, 102, 103,
104, 105, 106, 107, 108, 109, no ,
in , 112, 113, 114, 1162310, 117-
182335, 1182237, 2341, 1192263, 289,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (51 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
293, 302, 304222, 366-7
Timaeus, 372224
time, 392228, 288-9, 296-7
Tolstoy, Leo, 339
tragedy, see drama
truth, 71-2, 862270, 412-23, 452-6; see
also falsity
Turner, Frank M., 32233
tyranny, 2702214, 325-6, 485*213
unexamined life, 5, 138
unity (oneness), 13-14, 23, 452251,
3362227, 360-1
unity of virtue, 127, 141-2
unwritten opinions, 20-4, 452251,
882282, 302, 3102346, 457235
Urmson, J., 4622237
Van der Waerden, B.L.. 1992343
virtue, 3, 5-6, io, 372222, 125-8, 134-6,
1522220, 1602346, 464-5; see also jus-
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (52 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
tice; piety, unity of virtue
Vlastos, Gregory, 342212, 362221, 372222,
2324, 402332, 41*234, 422239, 44«47,
452249, 462253, 482269,2371, 49*274
79237, 802212, 2214, 2219, 812228,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (53 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
540 Index of names and subjects
Vlastos, Gregory [cont.)
832241, 881282, 147221, 148/24,
1492212,2113, 1502215, 1542221,
1562227,2229, 1662268,2270, 1672274,
*375/
J
99*345,
n
46f 2172214, 2182216,
2192219, 2212227, 2228, 2222230, 2235,
2232240, 2252242, 2732242, 2752269,
305227, 30822224, 309*242, 335-62225,
3362229, 372, 39^ni, 393*313, *3i4,
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (54 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
3942218, 3961230, 2231, 4582211,
Walbank, F.W., 812225
Wallace, John, 3062213
Walsh, fames J., 1572232
Waterfield, Robin A.H., 1192262,
458228,
4622240
Wedberg, A., 392227
Weingartner, Rudolf, 492272
Weiss, Roslyn, 45 92211
Wender, Dorothea, 452249
Wheeler, Samuel, 3062213
White, F.C., 2732249
White, Nicholas P., 2192221, 2242240,
2262242, 304223, 3062212, 2215,
3082223, 3092241,2244, 331*22, 332*23,
332-3229, 334*314, 336*227, 337*331,
2232,2233
Whitehead, Alfred North, 32224
Wiggins, David, 424228
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (55 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm
Wilamowitz-Muellendorff, Ulrich von,
2732251
Williams, Bernard, 3092241, 4602225
Wishart, D., no-1 1
Wolz, H.G., 2742254
women , 44-52249, 49*272, 470, 480,
4912297,2299
Woodruff, Paul, 362218, 2182216,
2222230,
304*22, 3622215
writing, deficiencies of, 20-3
Wundt, Max, 3922210
Xenocrates, 31*23, 502277
Xenophanes, 240, 245224, 340
Xenophon, 492272, n9225 8
Zeno, 51, 68, 1982239
Zeyl, Donald, 4592211
Ziff, Paul, 3352224
file:///C|/wip/CUP/CCOL/ForUpload/0521430186/0521430186nin_p531-540_W.htm (56 of 56)19/12/2005 09:38:50
Cambridge Companions Online © Cambridge University Press, 2006
Cambridge Companions Online © Cambridge University Press, 2006
INDEX OF PASSAGES
Aeschylus
Agamemnon
67-71 802322
1481-8 802321
Choephori
924-5 802323
Prometheus Vinctus
62 842352
309 140
944 842352
Alexander of Aphrodisias
223 Metaphysica
84.21// 3942320
Anacreon
Page 413 249
Anonymus Iamblichi
D.K., 89 #6
Antiphon
D.K., 87 B 44
Aristophanes
Acharnians
757
784
Clouds
367-81
368-411
423-4
1075-82
Frogs
1009-10
1422-32
831241
842355
1162312
1162312
812326
812326
812326
61
3622318
822336
Thesmophoriazusae
191-2 2712217
Aristotle
Posterior Analytics
392233
Topics
iO4b2o//
I46b36-i47an
414
1551224
De Sophisticis Elenchis
179*3
i8ib25~35
i83b6-8
i83b7-8
Physics
2O9bn-i7
209b14-15
De Caelo
289b25-29
O23 the Soul
4O3b2o-2i
43iai6-i7
Parts of Animals
642a25~3i
Metaphysics
985b29
987320-25
987a22
987a29-bi4
987*30
987a32-b7
987a32-bio
987b!—4
987b2—4
987bi- io
987bi4-2i
987b22-24
987b27~33
99obi7
995327//
ioooa9-2O
ioo4b25~5
472362
3102349
147232, 204
341213
882382
2 2
51
423
190
58
148235
147232
148235
147232
123, 148235
391227
51
341213
411233
341214
23
147232
147232
471262
423
52
30232
Cambridge Companions Online © Cambridge University Press, 2006
542 INDEX OF PASSAGES
Aristotle [cont.)
ioo9a38-bi2
ioo9bn-i2
iO24b32-4
iO24b32/7
iO34b26//
103932
iO78bi2-i7
iO78bi2-34
iO78bi7-i9
iO78b23
iO78b3o
io86a32-bi3
io86b2-5
io86b4-7
io9ibi3-i5
57
57
4 1 4
416
4 1 4
472262
39*227
147222
41*233
862267,
148225
41*234
39*227
34*214,
41*233
41*234
45*251
Nicomachean Ethics
io96b35
iO96b35-io97a6
Magna Moralia
1182315-28
Eudemian Ethics
I2i8ai5~28
Politics
i264b24-27
I264b26
1265310-13
I266b5
I27ibi
I274b9
I28oa22~5
i28ian-28
nS^a.i6ff
1301325//
I3i8a26
Poetics
1447328-020
I449b28
320
335*222
34*215
45*251
116222
35*217
89*283
89*283
89*283
89*283
4892271
4892269
4892271
4892271
4892269
862274
4612235
Constitution of the Athenians
28.3
De Poetis
fragment 4
O22 the Good
fragment 1
fragment 2
Sophist
fragment 1
852261
862274
882282
3102245
862267
Aristoxenes
Element a Harmonica
II.30-31
Cicero
Academica
1.10.16
Democritus
D.K., 68 A 57
D.K., 68 B 125
D.K., 68 B 141
Demosthenes
33O-I
Diogenes Laertius
Lives of Eminent
Philosophers
2.6
2-35
3-1-47
3-5
3-23
335
3-37
3-57-61
4.2
9-43
9-55
10.136-8
Epicrates
T. Kock, 2:287-8
Euclid
Elements
I.22
Euripides
Hercules Furens
1340-6
Gorgias
45*25i
37*224
792212
57
792212
842251
275*265
1 1 2
78*23
872276
49222113
33*27
35*217,
352218, 90
352216
194222
79*26
852256
4612229
171-2
176
802221
Encomium of Helen
D.K., 82 B 8
D.K., 82 B 9
D.K., 82 B 11
D.K., 82 B 13-14
D.K., 82 B 15-19
D.K., 82 B 23
Heraclitus
D.K., 22 B 32
D.K., 22 B 40
D.K., 22 B 42
D.K., 2 2 B 9 1
34O-I
341
340-1
341
3 4 i
341
812224
245*24
34O
812229
87*275
Cambridge Companions Online © Cambridge University Press, 2006
Index of passages 543
Herodotus
II 35-2
II 37
III 80-2
IV 116-7
V78
VII 102.1
VII 139
VII 144
Homer
Iliad
I.599-600
Odyssey
I 3^-43
Isocrates
Antidosis
50
84-5
235
265-6
268
270
285
Helen
1
s
58
822237
74
822237
59
58
822237
812225
4611233
802221
66
67
842252
852260
842252
66
66
67
67
Panathenaicus
26 852260
Olympiodorus
Anonymous Prolegomena to the Phi-
losophy of Plato
Westerink, 5.24 90
Westerink, 5.24-27 194122
Commentary on the First Alcibiades
Creutzer, 140.16-17 194
Parmenides
D.K., 28 B 2.7-8 812231
D.K.,28B3.i 812231
D.K., 28 B 5.7-9 56
D.K., 28 B 6.1-2 812231
D.K., 28 B 8.38-9 56
Philodemus
Gaiser, 152-3 172-3
Plato
Alcibiades I
I27e-i35e 140
Apology
19b// 139
19c 802220
64
140
1602243
139
138
1652263
138
3622211
138, 140
1592241
852258
1602243
157^35
4622242
136
136
1622251
802220
1572235
79229
1622256
1652265
1602243
1602243
1622256
1602243
1652265
1622256
1602243, 1622256
1652265
372222
78222
1622256
1602243
125
1502214
852265
1602244
802213
154
2172212
154
154
154
154
154
154
2172212
i9d-e
i9d-2oc
2oa-b
2OC-d
2od-e
2IC
»IC#
22a-c
22C3
22e-23b
23c-d
24C-25C
24c-26a
25b
25C-26a
25d-e
25e
26a4-5
26b-27a
26d6-e4
29a-3oe
29a6-b9
29d-3oa
3ib
32a-e
32d
34e-35a
35a
36c
37b5-9
38a
38b6
38e-39b
39d
40-41
41c
4ic-d
Charmides
I53d-i54e
i56d-e
i59b3/7
159C1
159C4
159C6-9
I59d5
I59dio-n
I59e3~I°
i6oai-d3
i6oe6
Cambridge Companions Online © Cambridge University Press, 2006
544 INDEX OF PASSAGES
Plato (cont.)
1620-e
i64a-c
16731-7
i69C-d
I7od5
I72b-I74b7
I72c-i74d
I74b-d
i74bio—04
I74bi2-C2
I74d-I75a8
I74d3~6
i74d5-6
I74e2
Cratylus
384C-391C
385e-386d
389d-39oe
40039
400c
4OOC7-4Oib7
402a
4O2d
40467-40505
40937
439b-d
439b-e
440c
Crito
46c-47d
49a
49a-e
52e5
53b-c
Euthydemus
2780-2826
278d-e
278d-282e
278e
28id-e
28id-282a
28ie
283b3-6
283e-286d
283e9-284ai
285d7//
286a4-7
286a7-b3
15 81241
15 81241
1592241
1582241
1591241
1572235
135, 1582241
1601248
135, 153
1581241
1582241
1582241
1582241
1581241
1582241
862272
1652262
421238
79229
238
103
55
79228
103
79229
391227
862272
79228
1602243
1622251
1622252
842247
125
1602248
1572235
135
135
1501214
1552224
15 22221
415
432244
4 1 4
415
415
415
286C2-3
287C-d
288c-292e
289C-29oa
295b—c
296a—b
301b
3O5b-3o6c
Euthyphro
4b7—^3
4c—9c
5^4
5d8-6a5
5e
6b7-C4
6dn
6e3
7b7-9
9e-nb
9ei-3
iod-iib
na-b
nb-e
ne-i2e
ne//
i2e6-8
13 a—b
13 a//
13b—c
i3eio-n
i4an
i4b-c
I4b8-ci
14c//
Gorgias
447c
448e-449a
45ia-b
453a4-5
453*6-7
453b7-455a6
454C-456C
454d-457C
46iei-3
462b
462e//
463a
463a//
463c
463d2
415
852257
1601248
138
852257
852257
245223
67
53
1642260
402232
53
245122
802215
402232
402232
53
1642260
54
54, 278
216226
130
154
1572235
1572235
1642260
1641260
1602243
1572235, 1642260
1572235, 1642260
1572235
1642260
1572235
1631260
1632260
1632260
1632260
1632260
1572235
1502217
138
832245
1631260
154
1502217
138
1631260
3621216
Cambridge Companions Online © Cambridge University Press, 2006
Index of passages 545
46364
464a//
464b-465C
464b-465d
464d-46sc
465a
466-469
466a-e
466a-468e
466b//
466bn-e2
467a8-468es
4670-4686
468b6
46803
47oe
472c-d
472c-48ib
473e6
482C-486C
482c//
482d
484c
485a-486d
487a
487e-488a
489b-c
491a
493a-b
500c
5ooe-5oia
5ooe-5Oic
501a
5Oid-5O3b
502b
5 02b 1
5O2C2-d7
502C12
5O2d2
5O7a7-b4
5O7e6-5o8a8
507 e6ff
5o8a6
509a
5O9e
5i5c-5i9d
520a//
52oa4
520C4-5
52oe
3622216
4622242
852262
3622216
137
472260
138
432242
1552223, 155*^4
137
137
137
254
1552223
1652264
372222
1502214
136
842251
434
136
852259
852260
432242
458229
1502214
85^59
4622242
434
1502214
472260
3622216
434
3622217
67
343
1582239
3622217
3622217
154
79*27
148225
148225
75
4882247
1652263
1622252
1602243
1602243
1622252
52ib-c
52id
522d-e
Hippias Major
28ia-283b
286c
289a
289a-c
289a4-5
289a9-b3
293a-b
Hippias Minor
367a8-b3
372b4-d7
372c-373a
372d7-ei
372e3-d6
374e3
375a4
375a6-7
375b5
375b7-8
375C4
375C6-7
375^-376b
375d//
376b4-6
/022
5 3ob-d
53obio
533^-534^
534C
535a
535e
535e-536d
536c
536c-d
536ei-3
54ie-542b
Laches
180c
i8oe-i8ia
i85ai
i85C-e
i87e-i88a
i87e-i88c
1896-1900
I9oc8-d8
I92ai-b4
I92b-i93d
1622252
432242, 344
1622256
65
277
79228, 287
2 9 0
287
287
862268
132
1582240
132
1582240
1582240
1602245
1602245
1602245
1602245
1602245
1602245
1602245
1522221
157^36
132
3622214
342
3622213
138
138
343
3622213
1502217
138
364«33
1502217
1602244
1602244
1602244
1552224
144
140
278
1662268
154
157*235
Cambridge Companions Online © Cambridge University Press, 2006
546 INDEX OF PASSAGES
Plato [cont.)
I94c8-d2
i94di-9
194^3
I94dio
I94e-i95ai
I94en-i95ai
i95C-d
i95C5-dio
I96a2~3
19661-19704
I97e-i99e
19834-5
19806-7
I98di-I99a4
19906-7
i99d-e
Laws
609SL-6.
63oe-63ia
631a
63ib-d
632c
638b
653a-c
653b
66od-663d
66ia-e
679b-c
679c
688b
690b
691b//
692c
700a
7O7e-7o8e
7o8e-7i2a
7o8e//
7i4ai-2
7i4a2
7i5b-d
7i5c/5f
7i6b8-9
7i6d-ei
7i6dff
7i7b-7i8a
7i8a-723d
7i9e-72oe
722a-d
141
1572235, 1582241
141
154
153
1582241
135, 1632256
157*235
135
212
141, 153, 154,
1582241, 1662268
1582241
15 82241
153
135
153
822237
4862222
84*247
4882254
4862219
352217, 382225
4902280
487*243
4862227, 49222110
4882254
4902288
491*295
4902280
487*231
4882260
822237
4902278
487*233
485*213
4882260
4882249
4902278
4902278
4872230
244
244
4902291
4882258
487*244
802213
4902282
726e//
729c
73ic
732e
732e-734e
734b
734e-735a
735C-738a
736a-c
736c-746d
737e-738a
738b
738b-c
738d
738e
739a-e
739a-74oa
739e-74ia
74ie
742C-744a
742c//
743C
744a-745e
744a//
744d-e
744e-745b
745^
753a//
756b-e
756e
756e//
758c5-d2
759a-c
759b-c
76iC5-d3
76id-e
76ie//
764a
766d
766d-768a
768b-c
769b//
77ob-77ia
77oc-d
77oc//
77i-776b
77ic-d
773a-e
776b-778a
777d
4882254
4882258
1562230
4862215
4862227
1562230, 4882247
4852214
4882253
487*234
487*238
4882252
242
2472238
242
487*235
487*238
20
4882253
4882254
4902289
4902288
4862227
4882253
4892271
4882253
4882257
242
4892265
4892268
4892264
4882260
4892264
247*238
4892263
4862226
489*275
489*273
4892270
490*283
4902286
4902284
4902282
4902289
4912296
4902288
4882258
242
4892272
49122101
49122103
Cambridge Companions Online © Cambridge University Press, 2006
Index of passages 547
777ei
778a-779d
779d-78oa
779d-78id
78oa//
78ia-b
782c
783b-785b
785b
788a-c
794
796c
802a//
802C2-4
8o3ei-2
804d//
805a
8o5e
806c
8nc-e
8uc-8i2a
8i6c-d
8i7a-d
8i7a-e
8i7e-8i8a
822a
828a-d
833c-d
8350-8423
84ia-842a
8410-8423
842b-85od
843a
847b
848c-e
85oa-d
853c//
8550-8563
855d
8566-8640
857C-e
858a
858a-c
858c-e
859b
86od
86od-863e
86od//
861b
866d7//
49122103
4882253
4882256
4882259
4912296
49122100
4882250
4882258
4912297
4882256
4882253
4912296
4872242
4902289
4902291
4912296
822237, 4912298
4912297
491229
361
4871239
4902291
338
4872242
4862219, 4862221
192
4902291
4912296
248
487*237
4882258
4882253
245*23
4892275
2472238
49122105
4862215
4902283, 4902286
4882249
4882246
4882249
487*337
487*333
4872242
4872229
4882247
1562230
4862227
4862228
4862228
867b3
875C-d
875d
875e-876b
875e-876e
885b
885c//
885d
886d
889a-89oa
89ie-899d
897b2
897C7-9
897611-89808
89833-6
89937-01
899d-9oob
899d8-ei
9OOC-9O3b
9ooe
903b//
9O3b7-9
9O4a-b
905b
9O5d
9O5d-9O7b
9O7b-9iod
9o8e
909b
9O9d-9io
9O9d//
9O9d3-9ioe4
9150-9223
9i8e-9i9b
9i9d-92oc
923a/?
925d//
928b
932b-c
932e//
933e//
936e-938c
937d-938c
941a
94ie
945b//
946b
946d-e
947a//
947a6
4862228
4852212, 4872231
4902277
4902283
4902286
4902292
4912295
247*235
4912295
53
247*232
241
241
247*333
247*333
241
247*335
241
241
247*336
490*294
241
241
490*293
241
802222
4902290
491*395
242
4882256
4902291
243
4882253
4862226
4882253
4882256
4902279
489*373
4912297
4872228
4882246
4902286
4902287
489*373
49122105
489*374
4892263
489*373
4892263
822237
Cambridge Companions Online © Cambridge University Press, 2006
548 INDEX OF PASSAGES
Plato [com.)
947e-948b
949a-b
95ia-c
95ia//
95ic//
95 id//
95id8
952b
952d//
955C-d
956c
956d-958a
957a-b
957b
957C
960//
96ia-b
961c
96ic-968e
964a//
9646-9653
965C2//
966c
966c-968b
967c
968b
969d
976e-968a
Letter II
3i4b-c
314c
Letter VII
324a
324C-325C
326b
34ib
34ib-342a
34ic
34ib-345a
34id-e
3426-3430
35oa-b
Lysis
2O3a-2O4b
2O7d-2iod
2iia4-s
2i6d-2i7a
2i7a-2i8c
218c//
4892273
4902286
4902280
4872241
4872231
4862218
4892263
4902281
4882251
4892273
4902284
4902286
4902282
4902278
4882249
4872231
243, 4862218
4862217
4862219
487*332
4872235, 4892264
4862220
243
4862221
4862223, 4862224
49222114
49222115
49122109
2 1
882282
382225
832240
382225
2 2
882282
2 2
2 1
2 2
2 2
37*124
1602244
157^35
157*335
1582241
1582241
1582241
2I9b-22Ob
22oc i -e6
22oe6//
22ia5-ci
Me22exe22izs
234c6-235b8
24063-24237
242d8-243b4
245e
Me22O
7ia5-7
7ia7
7 i b
7ib3
7ib3-4
7ib4-6
7ib4-7
7ib5
72b-c
73d-75a
73d7-e2
75C-d
76c
77b-78b
77b//
77C8
77e5-78a8
78b
78b4-8
78c-79e
78C4-5
78c5-d3
78d3~79Cio
79*3-5
80-86
80c
8od
8od5-6
8od5-8
8od6
8oei-6
8ia5-6
8iaio-b2
8iC5-d5
81C6-7
8ic8
81C9
8id4-5
8ie3-5
82b-85d
1552224
1582241
1582241
1582241
103
103
103
1 1 2
216224
200, 2192221, 2222229
278
2192221, 2222229
200, 216224
216224
2252242, 2252242
2192221, 2222229
333**io
154
154
85**57
79*3ii
135/ 157*335,254
1552222
1552223
1552224
1552222
134
1582241
1582241
1582241
154
1582241
309*339
277
236
216224
2 0 5
2192221
2 0 5
2 1 4
2 1 4
2 0 7
2212226
2212226
2232240
2212226
2 0 8
2 0 8
Cambridge Companions Online © Cambridge University Press, 2006
Index of passages 549
82b2
82C-e
82e
84a-b
84b-c
84d-85b
85b-d
85C2-8
8506-7
85C9-di
85Cio-di
85d-86c
85di
8 5 d 3 -4
85d6
85d9
85e2-3
86a8
86b
86b-c
86bi
86bi-2
86c
86e-87b
87b-87C
87b//
87c-89a
87ei-3
88b-89a
88bi-d3
88d2
88d3
89ai
89c—d
89c//
89C2-3
89d
89d—96U
91a—92c
93b—94^
95a
95e
96d-iooc
97a-98c
97a9-b3
97d-98a
98-100
98a
98bi-5
99c
2 0 8
2 0 8
2 0 8
2 0 8
2 0 8
2 0 8
2232240
2 0 9
209-10
2 0 4
2 0 9
2 0 8
2232240
2232240
2232240
2232240
2 0 9
2232240, 2252242
215, 238
277
2232240
237
209
177-8
178
154
179
2172212
1522221
154
154
154
154
179
179
1552221
179
1592241
64
1652263
1652263
1652263
1592241
2 1 0
2262242
130
3092239
472260, 204, 2182219
215
214
99C-d
99e
99e6
1 oob
ioob-c
Parmenides
I26a-i35d
I27a-i28e
i27bi-C5
I27b2
127C4-5
I27d-i3oa
i29d-e
130b
i3ob-e
I3oe-i3ia
131-135
I3ie-i32b
i3ie8//
i32d
I32d-i33b
i33b6-9
I33e-i34a
I35a7-b2
I35b-c
I35b5-C3
I35c8-d3
135C8-137C3
I35d
I35d8-e4
I37C
137^4//
I4oe-i42a
Phaedo
59bio
64c
64d—69c
65d
65d-66a
65e-67b
66e
68c-69d
68e
69a
69C8-12
72c
72e-73a
74a-c
74b-c
3622212
138
159-41
1652263
278
16
792210
3942219
116223
116223
1982239
439
402231
412236
402231
301
301
373
482266
301
372
285
372
452252
372
372
369
17
372
472261, 115
392224
392228
78222
281
435
281
334^15
9
435
127
435
4582211
238
79229
7
862268
280, 283, 284, 285,
292,293,3342215
Cambridge Companions Online © Cambridge University Press, 2006
550 INDEX OF PASSAGES
Plato (cont.)
74b-e
74C4-5
74d-e
74d-75b
75a-d
75b-c
75C-d
76b
78c-e
78c-8ob
78d
78d-e
78d-79a
78d-79d
7 8e
79c
79d2
8oa3
80b
80b 1
8od6
82a-84b
95^7
96a—99d
9639—10
97a-b
97b-98c
99d-iO5b
iooa3-7
1 ooc
100C4-6
iood
IOIC
ioid3-ei
iO2a-b
iO2a-d
102b
IO2b-C
IO2C
io2d7
IO2C2
107b
ii7d
Phaedrus
229e-23oa
230b
23ob2-C5
23od3~5
23006-64
7, 382226
293
402230
283
298
298
412236
472260
862268
334*315
3102250
79228
288
7
392227
288
2472229
2472229
288, 3102250
2472229
2472229
127
180
53
180
2 9 0
79229
37*224
181
402231
39i*2i
259, 301
402231
181
281
298
402231, 286
286
281
286
286
183
273*252
268
2762271
268
268
262
23ia-d
23idi-3
23163-23234
232C-d
232C3-6
232e
234d
235c
237a9
237aio-bi
237d-238c
23805^3
238d
239a
239a7-b3
239e-24oa
24oe8-24ib
24ia2-4
24iei-3
24iei-5
24ie4-243e
243e
244//
244a
244*̂ 6—8
244ds—24Sai
245a
2450-24632
246a
246c
247b2-5
247c
248ai-5
248b5—C2
248d-e
249d—254c
249d—256c
249e
25OC-d
25oc-25ia
25oe-25ia
252a4~5
252C-253C
252d5-253C2
252e-253c
253d-254e
253d6
253e3
263
263
263
263
263
263
2762271
263
2762271
2762271
1562227
262
2762271
263
263
263
263
263
3642238
2 6 2
263
2762271
455
358
263
235
358
240
264
245222
264
264, 334*215
266
264
359
263
235
2752266
2 6 0
1 0 2
268
263, 265, 267
264
267
265
268
267
265
264
264
Cambridge Companions Online © Cambridge University Press, 2006
Index of passages 55 i
254a7-b3
254b
254b-c
254b3-4
254b5-C3
254b6-7
254b7
254b7-ci
254CI-2
254di
254e8-255ai
255b6
255b7-d3
255d3-6
256a-e
256a7-b7
256b
256bi
256b2-7
256d-e
256e3-257a2
257ai-6
257^3-4
257^5
257C1-2
258d
259b-e
25 9e//
261a
26ie-262C
264c
265a-d
265CI
265d-266b
265c
265e-266a
265e-267d
267//
268c-d
269c//
26963-27038
270b
27oc-e
272d-274a
273-6
273e8-274a2
274-278b
274b-278b
275a-276a
276d
265
263
235
265
265
266
266
266
266
265
265
267
268
268
267
263
267
267
2 6 4
267
2762271
262
2762271
3642238
262
2762271
1 0 2
1502217
359
1502217
3 6 i
1 0 2
3642238
30222, 42.2239
4862225
263
1 0 2
299
3 6 i
4871235
2752265
299
802213
1502217
4612228
2762271
2 0
882282
2 1
2 1
277a-b
278a5-b2
278b-e
278b7-e2
279a
Philebus
na-i2b
n d
i 2 d
i3b-d
I4c-3ib
14C1//
i5a-c
i6-i7a
i6b-c
16c
i6c-i7a
i 7 e
19b
20b
2 l d
2id-23b
22c
23b
23b-c
23c
23c-d
23C-26d
23C-27C
24a-26b
24b
24d
25a
25d5
25e
26c
26d
26e
27e-3ib
28a
28c-3oe
3 i b
3ib-55C
3ib-66d
3ib-67b
32b
32d-33c
33b
33C-d
34b
299
2762271
360
2722232
67
427
425
444
458229
425
392129
455
30222
485228
427
437
438
458229
427
4 3 2
428
454
428
429
439
428
3362227
428
451251
428
439
439
439
428
428
439
428
429
429
53
425
43O
427
426
4 4 0
448
44O
441
441
Cambridge Companions Online © Cambridge University Press, 2006
552 INDEX OF PASSAGES
Plato [cont.)
35b
35C-d
35e
35e-36c
36c
36c-4ia
36c-4ib
36C-42C
36e
37e
38c
38e-39b
39e-4ob
4oa6-7
40c
4ib-42c
4id
42C-44d
43a
43C
44a
44b-c
44c-d
44d-5oe
44e-5oe
45b
45d-47a
45e
46b
46d
47a
47b
47c
47C-d
47d-5oc
48a
49C-d
5 i b
5ib-55
5 i e
52a-b
52c
53C-55a
54e
55a
55C-59b
55d//
57C-59C
58a
441
441
4601220
4 4 2
4 4 2
442
443
445
448
4602226
445
445
446
446
4612227
443
447
443, 448
4 4 0
4602219
448
802212
448
443
449
449
435
449
4 5 0
449
449
4 5 0
4 5 0
4592215
4 5 0
4 5 1
4 5 1
45*
452
453
453
453
4592218
454
4622237
43O, 454
1992248
30222
455, 458229
59b
59c
59d
60c
62a//
63e
64d-e
64d-65a
65a
65a-67b
66a-c
66b-c
66c-d
Protagoras
3O9a-3i4e
3i5C-d
3i5d-e
316-317C
3i8d-3i9a
329c6-di
33OC2-e2
33ic
334C-335C
338e-339a
345d-e
349b2-3
349e3-5
35oc-35ib
35ib-358a
352b-d
352e7
353CI-2
354a-b
354e6-7
355a6
355e2-3
356c8-e4
356d
356d4
357a5-b6
357C
357C7-di
357e2
358d5-7
359e4-7
36oa8
36oc6-d5
36oe6-36ia3
36oe6-36ic
365d5-7
43O
3342215,4622237
454
454
4862224
430, 458226
3362227
432241
45*251
427
43O
458227
458226
64
65
2712217
34O
64
1522221, 1632259
374, 39ifli
66
66
34O
4882247
1522221, 1632259
2172212
65
433
1 2 9
1632260
1632260
279
1632260
75
1632260
153
433
862272
153
1 2 9
1632260
1632260
1 4 1
2172212
153
141, 153
1682278
278
1 3 0
Cambridge Companions Online © Cambridge University Press, 2006
Index of passages 553
Republic
329di
33od-e
3 3 i
33ic-332a
335b-e
335C
338d
339b
35ob-d
35od
35ia-c
352d
352e-354a
353a9-n
353<l-354a
353d3-en
353e
353e-354b
354a
354a-b
357bi
358a3
36oa-c
360C8
3606-3620
36id3
36ie-362a
363a-366e
366c-4i2b
366e9
368C6
372C-d
373b
376e//
377a-b
377b-392a
378a
378d-e
379a-c
380C-383C
3 80c 1
382c-d
387b4
389b-c
392d-394C
394<*
395<*
2752265
802222
2 1 2
862268
1622252
1522221, 1572236
85*359
157*335
1522221, 1572236
157*337
1522221, 1572236,
157*337
1502214
1602245
1602245
135
1602245
1572237
1522221, 1572236
157*335
278, 279
313
313
326
313
331*31
313
331*33
247*337
126
334*319
313
313
75
348
126
2622222
53
126, 4892266
3622222
247*336
2 4 0
3622222
126
3622222
126
347
348
348
39605—e2
396e
396e2
398a
4oia-b
401b
402c-e
4O2d-4O3c
4O2d8
4i4b-4i5d
4i6e-4i7a
4i9a-42ia
42id-422a
422e-423b
4226-4230
4253-4273
427b-c
427e-428a
433a-d
433a-435a
435C-d
436-440
44id-442b
4416-442
443e-444a
444c-e
444C-445C
44467-44534
444e7-445b7
445b5-7
45ib//
455d-e
457d-466b
458c
459-461
462a-b
468b
47ic-474b
472b7
474C-476C
474d-475a
475-479
475-480
475b
475C
475C-48oa
475e-48oa
476b
476c-d
476d
3622223
348
3622223
349
349
348
2722237
248
2722237
126
126
13
63
63
3362227
484221
247*338
333*313
279
333*313
127
129
279
316
279
315
312
332*37
333*39
333*39
491*399
49122100
43*341
484221
248
452251, 336*227
2722237
49222112
332228
317
2722237
862271
300
298
472260
284
3052211
334*319
317
402231
Cambridge Companions Online © Cambridge University Press, 2006
554 INDEX OF PASSAGES
Plato (com.)
477a
477a-48oe
478d
479a
479a-b
479a-c
479a-e
479b
479b-c
479b6-7
4-79b9-io
479C3-5
479C7
479d
479d3-4
480a
484a7-8
484b
484C-d
484d2
485b
49oa-b
493a
493a-c
493C
494a
495a-b
495a7
497C
498a-c
498d
499e-sooe
500c
50002
50004
5ooc9-di
Sood-soia
5ooe
5Oid
5O2b-c
5O4a-b
5053-5090
5O5b-c
5O5b-5o6e
5osd-e
5O5e-5o6a
5o6d-e
507b
335*325
334*315
335*3^5
298
285, 286, 3052211
862268, 284
391227
3102149
296
296
2 8 4
296
2 8 4
335*3^5
298
334*319
33*218
3052211, 334*319
292, 3052211
298
79228, 862268, 3052211
3342219
65
1502217
472260
4892262
862268
331^3
472260
262
334*313
323
13, 412234, 292, 328,
334*319
322
322
323
324
482266
334*319
484221
127
13
459*313
3^2
1582239,264
1552224
3^3
3052211
5O9d-5i id
510b
5 ioc
5ioc3-d3
5iod6-5iiai
511b
5iibi-2
5iib5-c2
51107
5i4a-5i7C
5i4a-5i9a
5i4a-5i9d
5i5a-e
5i5a5-6
5i6c-d
5i6cio-di
5i6d2
5i6e-5i7a
517a
5i7b-c
518c
5i9C-d
5i9d-52ib
52oe-52ia
52oei
52ic-d
522C-525C
5233-525
523a//
5^3b-c
523cn-d3
523e-524a
5^3e4-5
5^3e7
524d-525a
525d8-526a7
526e
527a6-bi
527b
529C7-53ob4
529di-5
53ob6-ci
53id-534e
53ie
532a//
533b
533b-c
533C
533C8
534a
30222
185, 303
472260
186
184
303
184
187
1982238
335*325
1 0
317
1 1
65
299
299
299
2752265
1 1
303
11
303
327
337*334
172
79*37
305226
189
282, 285, 287
288
285, 286, 287, 288
288
288
285
191
288, 303
185
288
323
1 9 2
1 9 2
30222
472260
190
288
188
472260
188
288
Cambridge Companions Online © Cambridge University Press, 2006
Index of passages 555
534*-b
534b
534b-d
537^-539^
538c
540a
54oa-b
540b
54od-54ib
543C7-544b3
544c
545^
547b-548d
547d-e
548d-e
555b-565e
557b-c
566c
57ib-572b
573d
576a
579a-c
579d-e
5 8oa-c
58ob-c
5800-5833
58od-583b
58od-588a
58137
58ie
5810-5833
58107-58232
583b
583b-5863
583b-587b
583b-5883
583b6-7
5830-5843
583c
5843-b
584b-e
5853-586b
585b
585b-e
586d
587c
587d-e
587c
58837-10
58901-4
298
471260, 279
30122, 188, 190
851258
2721238
481266, 293
327
298
49212112
315
841247
841247
62
841247
2721237
28
831345
821237
325
325
325
325
325
312
279
312
2 6 0
435
2 6 4
453
297
332*25
4591214
297
302
312
312
436
435
4 4 1
436
436
435
436
264
436
437
314,435
332*35
332135
595a
595^5
595b9-C3
5963
5963-597b
596di
5960-5973
597b6
597C7-598d6
597e-598d
597e6
598b-d
598d-e
598d3-4
598d7-6o2bn
5993-6ooe
599c-d
6oob6-9
60004-5
6ooe4-6
60134
6010-6023
602b
6o2C-6o3b
6o2C-6o6d
6o2Ci-6o8b2
6020-6033
6o33i2-b4
603c
6o4C-d
6046—6053
60539—b2
605b
6o5C-6o6d
6o5C-6o8b
6o6b-c
6o6b-d
6073-e
60732
60734
60705
612b//
Sophist
2163
2173
217c
2193—2323
2270-2300
23108-07
2323-236d
35O
3631226
2721232
8, 333*3io
335*325
3 5 i
1 1
363*330
363*329
352
351
1501217
353
351
363*329
353
354
354
354
3631228
3631228
355
355
355
67
363*329
283
3641237
356
299
356
364*337
357
357
364*335
4611232
357
338
363*328
351
3631228
331123
90
90
791210, 90
438
1561230
397
65
Cambridge Companions Online © Cambridge University Press, 2006
556 INDEX OF PASSAGES
ato \cont.)
232ai//
235b-236c
235d2
23609//
236d9//
236e-237a
236e//
23561-2
236e4-237ai
23909-24006
23909//
24133
24107-25904
242b6//
24205-6
242d-e
242e
24262-3
24302-5
247d-e
248
249Cio-d4
249d6//
25oci-d5
25005//
25ia5-6
251*5//
25ib6
254C-d
255-6
255012-13
255012//
255d4-5
255d5
255e8
25568-257312
255e8//
256cio-dio
256dn-i2
256di2-e4
256e6
256e6-7
257al//
257a4-5
257bi-257C3
257bi//
257b3
257b3~4
257b3//
397
3642232
397
397
397
72
4 1 9
398
398
398
397
397
398
399
399
792211
79228
55
399
309^37
2472232
399
399
4 0 0
399
4 0 0
399, 400,415
802212
439
3962231
381-2, 401
4 0 0
4 0 0
4 0 2
4O3
403, 406, 411
4 0 4
4O3
4O3
403
418
403, 412, 421, 422
4 0 4
4 0 4
403, 405
4 2 2
406, 407
405
4O7
257b5
257b6
257b6-7
257bio-c2
25704-25805
25704//
258an//
258b
258b9-c3
258bn-C4
25806-25904
25807//
258ei
259*5//
259a6-bi
259bi-5
259b5-6
25905//
26004-5
262a3-7
26204//
262d4
262d5
262d5-6
26204//
26205
262e6
26209
262011-263310
262014
263a2
263a4
26335
26339
263a9-io
263an
263an-d5
263b
263b3-4
263b4-5
263b7
263b9
263b!1
263bn-i2
26302
26305-12
26307
26309-11
263d!
263d!—2
406
407
405
410
403
407, 408, 409
408
422
391221
402
403
410
410
410
410
410
410
398
417
413
413
413
413
413
414
417
416
417
1 414
416
417
416
416
416
416
417
417
421, 422
420
417, 419, 420
419, 420, 421
418, 420
402, 418, 421
418, 421, 422
417
414
416
417
418
421
Cambridge Companions Online © Cambridge University Press, 2006
Index of passages 557
263d2
26362
264b-268d
268a
Statesman
257a
258b
258b-268d
262a-b
262a//
262c-e
266d
268d-274d
274d//
274e-end
283b-287b
284b
285a-b
285d
285d-286b
286b
286d8-9
287b//
292b-293C
292e-293a
293a
294a//
296a//
297b
297b-c
299b-e
299b//
300b
300c
3ooc-d
3ooc-e
3ooe//
302b
303a
3o6a-3iic
311a
Symposium
174^9
I76a4-b8
17603-5
I76c5-d4
I76d4
17664-10
I77a-c
I77d8-ei
418
417
438
3647232
462256, 90
461256, 90
438
4852211
4861225
8
462256
4862215
4862215
438
49122106
462256
4862225
485229
4852211
462256
4852211
485227
485226
4892262
4852212
4862216
485226
4892262
4852212
487*235
4902282
4862215
4852213
4862216
4852211
4882260
4852210
4852213
1 2 7
4852210
2712217
249
2 6 2
249
2 5 0
249
249
2712217
178b
I78b-c
17803-5
I79c-i8ia
i8oc4-d3
i8ib3-4
i8ib5-6
i82d7
i83a-b
i83d8-ei
i83d8-e2
i83e5
18404-7
i85d-e
I85e6-i86a3
i86b-e
i86b2-3
i87a-b
i87e6-8
18902-3
I9ic-d
I9ie-i92a
i92ai
i92eio
I93a3-d5
i93b6-7
I94e4-i95a5
I96b4~5
196C4-7
I96d6-ei
I97a6-b5
i97b3-9
i97b8-9
197CI-3
i97e7-8
i98d5-6
199b
i99d-2ooa
200a
2Oia-c
2oie
2O2d-2O4C
2O2e//
2O3d8
2O4d-2O5a
205a
2O5a-2o6a
2O5d-2o6a
2O5di-8
2O5d2-3
249, 2702215
792210
2 5 1
2 5 1
2 5 1
2 5 1
2 5 1
2 5 1
2752265
2 5 1
2 5 1
2 5 1
2 5 1
2 5 0
2 5 1
802213
2 6 1
79228
2 6 1
2 5 1
872279
2 5 2
2712219
2 5 1
2 6 1
2712217
2 5 2
2712220
270229
2 6 1
2712220
2 5 2
2 6 1
2 5 2
2 5 2
2 6 1
2 6 1
285
2 5 2
2 5 2
253
253
455
3632224
254
253
254
872279
254
255
Cambridge Companions Online © Cambridge University Press, 2006
558 INDEX OF PASSAGES
256
253, 254, 264
250
255
250
2712228
254
254
233
255
3622219
255
255
2722230
255
79228
256
256
256
255
256
3622220
3622220
354
255
256
3622220
3622220
255
842247
255,256
2722231, 345
254
432241
233
256
257
258
256
256
257
258
2722240
257,258
258
257
258
256,257
345
258
2722239
257
258, 272224O
259, 2732244
273JH5
288
255
258
259
256
259
29O
334*315
26O
259
392227, 862268, 259
3IO225O
259
4O223I
259
257
26O
273*243
26O
254
258
259, 2732243
26O
259/ 345
260
2722236
273*243
259
261
2712217
261
261
234
347
263
261
262
262
261
262
261
262
250
347
Plato (cont.)
2O5d5
2O5dio—2o6ai
2O5e7-2o6ai
206a
2o6an-i2
206b
2o6bi-4
2o6b3
2o6b7-8
206CI-5
2o6e-2O7a
2o6e5
2o6e7-2O7a4
2o6e8
2O7a7-d3
2O7d
20803
2o8d7-8
2o8d8
2o8ei-5
2o8e2-4
2O9ai-e4
20934
20935-7
20935-07
20938-02
2093-9
209b 1
20906-7
2O9d
20965—2ioa2
210a
2ioa-2i2a
2ioa-2i2b
2ioai
2ioa2
2ioa4
2ioa7
2ioa8
2ioa8-b2
2ioa8-b6
2iob5
2iob6
2iob6-7
2iob6-d6
2iob8
2iob8-ci
2iob8-C3
2 ioc
2IOCI
2IOC3
2IOC3-5
210C5-6
2IOC7
2iod
2iod-2iia
2I0dl-2
2iod4
2iod5
2iod6
2iod6-b5
2ioe-2iia
2ioe-2iie
2ioe3-2iib5
2ioe4-6
2iia-b
2iia5-bi
2iia7
211b
2iib3-5
2iib7-ci
2IIC
2IIC8
2Ildl-2I2a2
2iid3-8
2iid8-e3
2Ild8-2I2a2
2iie4-2i2a2
212a
2I2a2-7
2i2a3~4
2i2a4-5
2I2bl-4
2I2b2-6
2i2e8
21367-21431
21406-8
2i5b-2i6e
2i5d-2i6c
2i5e-2i6c
21561-21603
2i6a8-b2
2i6b5
2i6b5-C3
2i8d6-b2
2I9e-22IC
22oai-5
22oa4~s
222a
Cambridge Companions Online © Cambridge University Press, 2006
Index of passages 559
222CI-3
Theaetetus
i42a-b
I43C
I48e-i5id
152a
i52a-b
i52b-c
I52d2~3
i52e
i54C-d
i65e-i68c
i66d4-8
167c
i7oa-c
i7ia-c
I7id-i72b
i72ai-5
i72b2-6
I73c-i76a
i76bi
178-180
I78a-i79a
i79d-e
183c
184-186
i86a-b
187C-200C
i88a-c
i88a-i89b
2OIC-2IOa
2O2a6-8
2o6c-2o8b
Timaeus
17b-19b
25c
27a
27d-28a
27e-28a
28e
29a-3ob
29b
29b4-5
29c
36b-d
37b-38c
37c-38b
37d
37d-e
2 6 1
37*225
472261, 90
17
57
1642262
57
862269
79228, 792211
3072220
459*217
65
1642262
65
297
3082231
299
1642262
1642262
275*265
2 3 0
3072219
299
79*28
472261, 792210, 90
3082230
299
43*244
2202224
72
472260
415
303
9 0
822237
90
334*215
392227, 482264
293
14
482265
2722236
53
1 9 2
3072217
288
392228, 482264
289
37e-38b
38a
46d-e
47e-48e
48e
48e-49a
49a
49d-5od
5oa-b
5od
5id-e
5id-52a
5ie-52a
5ie-52b
52a
52a-b
53C
61//
6id-62b
62C-63e
73d7
77C6
7 8e
86c-e
Plotinus
Enneads
3.7.2
Plutarch
De Exilio
392b-c
39*227
334*215
14
53
3 0 0
482265, 288
293
392227, 482264
288
482265
472260
300
288
482264
392227, 482265
334*215
482265
309*238
3102248
3102248
79*2H
79*2ii
79*2H
1562230
392228
812229
De Iside et Osiride
37off. 35*217
Ouaestiones Convivales
7i8e-f
Proclus
173
Commentary on the First Book of Eu
clid's Elements
66.7-8
66.8-67.1
211.1-23
Protagoras
D.K., 80 A 1
D.K., 80 B 1
D.K., 80 B 6
195*213
176-7
176
852256
852256
852256
Sextus Empiricus
Adversus Mathematicos
VII. 135
Simplicius
55
Commentary on Aristotle's "On the
Heavens "
Cambridge Companions Online © Cambridge University Press, 2006
560 INDEX OF PASSAGES
Simplicius [cont.)
488.7-24 174
Thucydides
III 38.7 842250
III 82-5 60
III 82.2 60
V84-113 74
V89 60
V 105.2
VII 68.1
VII 69.2
Xenophanes
D.K., 21 B 11
Xenophon
Memorabilia
I.2.39-46
60
852255
83JH5
802221, 245224, 340
852258
