Classics • Dhdosoph4

• T he T h ¡ r d Wa y
'f.W DIRECTIO 5 IN PLAT IC 51 UDlf 5

Edil e d by francisco J . Gonzolez

ABO T THE ED TOR

F.or orders and informat1on J)lease contact the J)Ublisher

ROWMAN & l! lifl.'. EF.IEL! D P.UBLISRERS INC

ISBN 0-8476-8114-9 Ed i ted by

J~~~ll~l l lilJl~lfül 11 íllllllllíllll francisco J. Go n zolez
 The Third Way
N ew Directions in Platonic Studies

t - edited by
FRANCISCO J. GONZALEZ
1

1t

ROWMAN & LITTLEFIELD PUBLISHERS, INC.

ROWMAN & LfITLEFIELD PUBLISHERS, INC.

Published in the United States ofAmerica

by Rowman & Littlefield Publishers, Inc.
4720 Boston Way, Lanham, Maryland 20706
3 Henrietta Street
London WC2E 8LU, England

Copyright © 1995 by Rowman & Littlefield Publishers, Inc.

Ali rights reserved. No part ofthis publication may be reproduced,

stored in a retrieval system, or transmitted in any form or by any Contents
means, electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of the publisher.
Preface vii
British Cataloging in Publication Jnformation Available

Library of Congress Cataloging-in-Publication Data Introduction

Toe third way: new directions in Platonic studies / [edited by] Francisco J. Gonzalez. l. A Short History of Platonic Interpretation and the "Third Way"
p.cm. Francisco J. Gonzalez 1
Includes bibliographical references and indexes.
l. Plato. I. Gonzalez, Francisco J. Part I: Philosophy and Speech: Plato's Dialogues Between Oral and Written
B395.T52 1995 184--dc20 95-30779 CIP
Discourse
ISBN 0-8476-8113-0 (cloth: alk. paper)
ISBN 0-8476-8114-9 (pbk.: alk. paper) 2. Reflections on the Orality and Literacy of Plato's Dialogues
Jackson P. Hershbell 25
Printed in the United States of America
3. Plato's Audiences, or How Plato Replies to the Fifth-Century
8"'Toe paper used in this publication meets the mínimum requirements of Intellectual Mistrust of Letters
American National Standard for Jnfonnation Sciences---Permanence of Elinor J. M. W!st 41
Paper for Printed Library Materials, ANSI Z39 .48-1984.
4. Neither Published Nor Perished: Toe Dialogues as Speech, Not
Text
Joa,nne Wiugh 61

Part II: Philosophy and Rhetoric: Between Persuasion and Proof

5. Socrates' Rhetorical Attack on Rhetoric

David Roochnik 81

V
 6. Apodeiknunai, Dialegesthai, Peithein: A Reconstruction of Plato's
Methods of Argument in the Phaedo
P. Christopher Smith 95

Part ID: Philosophy as Myth, Drama, and Enactment: Between Imagination

and Reason, Plot and Argument

7. Methodology in the Reading of the Timaeus and Politicus

T. M. Robinson 111

8. How To Read a Platonic Dialogue

James A. Arieti 119

9. Plato's Dialogues as Enactments Preface

Gerold A. Press f33
Now, had Thshtego perished in that head, it had been a very precious perishing,
Part N: Dialectic and Dialogue: Between Skepticism and Dogmatism smothered in the very whitest and daintiest of fragrant spennaceti; coffined,
hearsed, and tombed in the secret inner chamber and sanctum sanctorum of the
10. Self-Knowledge, Practical Knowledge, and Insight: Plato's whale. Only one sweeter end can readily be recalled-the delicious death of an
Dialectic and the Dialogue Form Ohio honey-hunter, who seekiµg honey in the crotch of a hollow tree, found such
Froncisco J. Gonzalez 155 exceeding store of it, that leaning too far over, it sucked him in, so that he died
embalmed. How many, think ye, have likewise fallen into Plato's honey head, and
sweetly perished there?
11. Dogma, Skepticism, and Dialogue
189 Herman Melville, Moúy Dick, ch. 77
Vtálter Vtátson
Those observing from the outside the busy industry of contemporary
12. Toe Dialogical Composition of Plato's Rlrmenides Platonic studies must wonder how anything new can remain to be said about
Victorino Tejero 211
Plato after more than two thousand years. Yet, to judge from the issues that
are currently being debated, there is little exaggeration in saying that the study
Part V: Between Unwritten Doctrines and Written Dialogue of Plato is still in its injancy. Scholars not only sharply disagree on the details
of Plato's positions; they disagree on how we are to go about discovering these
13. Toe Choice Between the D1"alogues and the "Unwnºtten positions, indeed, they disagree on whether Plato even had positions!
Teachings": A Scylla and Charybdis for the Jnterpreter? Furthermore, the purposes of the dialogue furm in which Plato chose to write
Mitchell Miller 225
are as much in debate today as they were almost two thousand years ago in the
anonymous Prolegomena to Plato 's Philosophy.' One should not ignore the
Bibliography 245
positive gains that have been made recently in our understanding of specific
dialogues. But can the lack of consensus on fundamental issues fail to be a
Index 259
source of profuund embarrassment to us?
In the introduction I attempt to explain, if not excuse, this state of affairs by
About the Contributors 267
showing that the history of Platonic scholarship has been a succession of

1. Anonymous Prolegomena to Plato's Philosophy, ed. & trans. L. G. Westerink

(Amsterdam: North-Holland Publishing Company, 1962).

vi vii
viii Toe Third Way
competing paradigms and that, as the contributors to this volume in general
agree, a completely adequate paradigm has yet to be discovered. Toe
Neoplatonic paradigm ruled for approximately one thousand years, mu~h
longer than the competing skeptical paradigm. When the. Neo~latomc
interpretation ca.me to be rejected by at least the 1700s, Platomc ~tudies had
in essence to begin ali over again. Plato's philosophy was now mterpreted
according to the paradigm of "systematic" philosophy of the kind found in
Descartes, Kant, or even Hegel. When this paradigm could not account for
contradictions in the dialogues, it was emended with the notion of "evolution,"
so popular in the nineteenth century. Though this characterization of .Pla~'s
philosophy as a system evolving from one dialogue to another 1s stlll
"orthodox" in sorne quarters, it has been challenged by the advocates_ of
Plato's "unwritten teachings" who characterize their view as a "new
paradigm. "2 Both interpretations are in turn being challenged by a resurgence
of the skeptical paradigm. Toe history of Platonic scholarship has therefore
alternated between radically different ways of characterizing Plato's philosophy
and thus of interpreting bis philosophical texts. What is distinctive of our own
time is that such diverse paradigms are existing simultaneously and that
therefore fundamentally different presuppositions are guiding the work of
different scholars. Toe only thing that provides the appearance of consensus
and stability in this country is the insularity of much English-language
scholarship.
Is there really any way of deciding on an adequate paradigm? Is it not
inevitable that the dialogues, given their elusive character, will always be
2. In / tre paradigmi storici nell'interpretazione di P/atone e i fondamenti del nuovo
paradigma (Napoli: Istituto Suor Orsola Benincasa, 1991), Giovanni Reale distinguish:s
between three paradigms in the interpretation of Plato's philosophy: 1) the Neoplatoruc
paradigm that !asted fifteen hundred years and interpreted Plato allegorically; 2) the
paradigm, beginning in the 1800's and lasting one hundred and fifty years, that sought
to interpret Plato solely on the basis of bis writings and that ignored bis "unwritten
teachings" as handed down through the indirect tradition; 3) the "esotericist" paradigm
that first became prominent in the 1950's with the Tübingen School and seeks to make
the "unwritten teachings" central to the interpretation of Plato's thought. This
characterization of the bistory of Platonic interpretation is inadequate for a number of
reasons: i) it ignores the New Academy's interpretation of Plato; ii) it lumps together in
2) very divergent positions; and ili) it exaggerates the extent to wbich 3) is a radical new
departure. As I try to show in the introduction to this collection, the "unitarian," the
"developmentalist" (both of wbich Reale includes in the second paradigm) and the
"esotericist" interpretations can ali be seen as attempts to reconcile a shared conception
of Plato's philosophy as "systematic" with the clearly unsystematic, fragmentary and
open-ended forro of bis writings. An alternative to this "dogmatic" paradigm is the
"skeptical" paradigm. If this latter paradigm is itself defective, as I argue it is, then a
third paradigm is needed distinct from any of those Reale recognizes.
Preface ix
mirrors reflecting the interpreter? Will not Plato always be a skeptic to the
skeptics and a dogmatist to the dogmatists? I do not think this is necessary as
long as we attempt to meet one essential requirement, which scholars have so
far appeared unwilling or unable to meet: that in attempting to understand the
dialogues, we be willing to put aside as much as possible ali of our
assumptions concerning the nature of philosophy. This is the most difficult
requirement for an interpreter to meet, and perhaps it can never be fully met. 3
Yet we can certainly avoid seeing Plato as a would-be analytical philosopher
or a would-be deconstructionist. What is required, for example, is that the
analytical philosopher allow that perhaps Plato was neither a good nor a bad
logician but something totally different, and that the deconstructionist allow
that pe~baps Plato ~as neither a "modernist" nora "postmodernist" but again,
something totally d1fferent. More generally, wbat is required is that we keep
3. In a very important article, Hans Kriimer uncovers sorne of the fundamental
philosopbical presuppositions of both the "developmentalist" and "skeptical" or
"antisystematic" interpretations . of Plato current today ("Pichte, Schlegel und der
Infinitis~us in de~ Platondeutung," Deutsche Vierteljahrschrift für Literaturwissenschaft
und Ge1stesgesch1chte 4 [1988): 583-621). More of this kind of work needs to be done
in order to check the natural tendency of Plato's interpreters to see bim in their own
image. It is odd that K.rii.mer appears to believe that an acceptance of the evidence of
Plato' s "unwritten teachings" renders an interpretation proof against reading into Plato· s
thought alíen philosophical assumptions. This is clearly false. Not only are these
teachings avail~ble to us only through the second-hand reports of other philosophers
whose assumpt1ons, we have much reason to believe, differed from Plato's own, but a
coherent system can be extracted from these reports only through a reconstruction. Here
there is obviously much opportunity for missing the spirit and context of Plato's
philosophizing. In fact, K.rii.mer's own view that Plato's philosophy was an axiomatic and
deductive system more geometrico is not derived from the evidence of the "unwritten
teachings," but is guided by a conception of "system" that is distinctly Post-Cartesian
(as, for example, Heidegger has shown in bis lecture course on Schelling: Schelling: Vom
Wesen der Menschlichen Freiheit [1809), in Gesamtausgabe 42 [Prankfurt am Main:
Vittorio Klostermann, 1988], 47-52). Portunately, K.rii.mer's assumption that an
acceptance of the evidence of the "unwritten teachings" commits one to bis
characterization of Plato's pbilosophy has been challenged by Rafael Perber's book, Die
Unwissenheit des Philosophen oder Warum hat Plato die 'ungeschriebene Lehre • nicht
geschrieben? (Sankt Augustin: Academia Verlag, 1991). Perber has shown that an
acceptance of this evidence is compatible with a characterization of Plato's philosophy
compl~tely ~ppo~ed to that of K.rii.mer, i.e., a characterization of it as essentially
aporellc. K.ramer s assumptions concerning the nature ofphilosophy, and those of other
"esotericist" interpreters, may therefore be no less aliento Plato's thought than those of
the "developmentalist" and "sceptical" interpretations K.rii.mer criticizes. Por sorne
interesting observations on the philosopbical agenda behind the "esotericist"
interpretation see Enrico Berti, "Strategie di interpretazione dei filosofi anticbi: Platoni
e Aristotele," Elenchos 10 (1989): 289-300.
X Toe Third Way
open the possibility that Plato was neither a dogmatist seeking to establish
doctrines nor a skeptic seeking to undermine them but, once again, something
totally different.
But these examples show why this is such a hard requirement to meet. Toe
analyst is being asked to grant that maybe for Plato formal logic was not as
central to the philosophical enterprise as analytical philosopby assumes it to be;
the deconstructionist is being asked to grant that perhaps Plato does not fit into
any of the categories recognized by the critic of "modernism;" philosophers
in general are being asked to consider the possibility of a philosophy whose
main aim has nothing to do with refuting or establishing doctrines. Difficult
or not, however, this requirement cannot be ignored. Indeed, if there is on~
thing we must learn from Plato's dialogues, it is that we should always be
willing to expose to questioning not simply sorne of our specific philosophical
positions, but even our most fundamental assumptions about the nature, point,
and value of philosopby. It is likely that this willingness will enable us to do
much more justice to the dialogues and allow them in turn to offer us much
more. Though I cannot defend this view here, I would even suggest that the
reason why none of the traditional paradigms have proveo adequate is that they
depend on one-sided and impoverished conceptions of philosophy that could
never do justice to the much richer notion of philosophy in the dialogues. In
this case we would indeed have much to gain from abandoning our
preconceptions and actually leaming from , which for Plato meaos being
transformed lJy, the dialogues.
Toe attempt to meet the above requirement is perhaps the most imponant
common characteristic of the contributors to this volume. They ali pursue a
"third way" by bringing into question traditional oppositions between
philosophy and oral discourse, philosophy and rhetoric, philosophy and drama,
philosophy and imagination, skepticism and dogmatism, and written and
unwritten teachings. Philosophy has always defined itself in opposition to
something else; therefore, one way in which we can examine our assumptions
about the nature of philosophy in the attempt to understand Plato is by bringing
sorne of these oppositions into question. Toe "third way" represented by the
contributors is therefore not a new school or a new interpretation: it is instead
a way of exploring relatively new territory beyond the boundaries of the
skeptical and "doctrinal" conceptions of philosophy that have so far ruled
Platonic studies. In the introduction I provide detailed summaries of ali of the
papers in order to guide the reader along the different paths they follow,
showing where they diverge and where they intersect. Toe bibliography at the
end of the volume (more than a list of works cited) is meant to be an aid to the
reader in exploring further the issues they address. I have also sought to make
the volume accessible to the beginning student of Plato (perhaps the best
reader because still innocent of the preconceptions from which the collection
seeks to break away). Foreign language quotations and Greek expressions have
Preface xi
been translated, while also being retained in the original for the scholar.
I wish to thank two people without whom this book could not have been.
Toe first is Gerald Press, who originally suggested that I edit this book and
who set a welcome precedent with his own collection, Plato's Dialogues: New
Studies and Interpretations (Lanham, Maryland: Rowman & Littlefield, 1993).
Jerry p~ovided indispensable help at every stage of the process, from
substan~1ve comments on the introduction to tecbnical advice on editing and
~rmatt1ng: He also deserves credit for organizing panels on "nondoctrinal"
mte1:Pretat1ons of _Plato in conjunction with the meetings of the Society for
An~1ent Greek ~hilosophy. These panels have provided a stimulating forum in
which the contnbutors to this volume and others have been able to share their
work. and ideas. The second person to thank is Joanne Waugh, who was my
co-:~tor before family obligations prevented her from continuing in this
pos1non. I doubt that I would have undertaken this project if I had not had her
help at its inception. Our numerous initial conversations were indispensable to
me. !º~e specifically assisted me in the process of evaluating potential
contnbut1ons and deciding which ones to include. Unfortunately for me of
course, I can blame neither Jerry nor Joanne for the results! '
Editing a collection of papers today requires more expertise in computer use
than I had at the outset. I therefore also need to thank Bill Duffy of the Center
~r. Inform~tion Te_chnology Services at Skidmore College for his eager and
dihgent ass1stance m converting files and scanning documents.
My ho~e is that this collection and others like it will encourage a kind of
sch?larship_~at seeks neither to "vindicate" Plato as "up to date" (i.e., as
ha~u~ ant_1c1pated our own wisdom) nor to condemn him as hopelessly
pn~nve (1.e., as not poss:ssi~ our enlightened logical tools and concepts),
but mstead to learn from his dialogues new and hitherto unimagined ways of
philosophizing.
F. J. G.
Saratoga Springs, NY
May 1995
 1

Introduction: A Short History of Platonic

Interpretation and the "Third Way"

Francisco J. Gonzalez

Toe papers brought together in this collection display a wide range of

interpretative strategies and aims. What unites them, however, is the general
goal of finding a "third way" between oppositions or dichotomies that have
traditionally divided Platonic scholarship. While each paper characterizes this
"third way" and, therefore, the other two ways, somewhat differently, they ali
seek to avoid, in varying degrees, two general altematives that tend to be seen
as exclusive and exhaustive. These altematives are clearly stated by Diogenes
Laertius when he tells us that an interpretation of a Platonic dialogue must
decide, among other things, whether the statements in it are meant to establish
Plato's own doctrines (eíc; óo-yµchwv KCJ.TCXCTKEIJTJV) orto refute the interlocutor
(ill.65). On this view, to the extent that a dialogue is not simply refutative, we
can assume that its primary or sole aim is to expound and argue for certain
philosophical theses. Toe tasks of the interpreter are therefore to identify these
theses, to analyze the arguments and, finally, to determine the truth of the
theses by evaluating the arguments. But are these the only altematives? Must
a dialogue be interpreted either as "aporetic" in the sense of containing only
refutations, questions, and problems or as "constructive" in the sense of
defending doctrines? While the contributors ro this collection may lean more
one way or another (e.g., doctrines play a more important role for sorne than
for others), they all seek sorne kind of third altemative here. Because they give
2 Toe Third Way
more importance than usual to the literary, dramatic, and rhetorical aspects of
the dialogues, they try to avoid identifying their aims with either the
establishment or the refutation of doctrines. Instead, they characterize the
dialogues as "inspiring" us, as providing us with a "vision" of the world, as
"exhorting" us to action, as expanding our imaginations, as "orienting" us in
our own inquiry, as communicating a forro of reflexive, practical, and
nonpropositional knowledge, or as inviting us to a conversation in which we
must actively participate in order to arrive at the truth. These are different
suggestions, sorne compatible and sorne not, but they ali share the general goal
of finding a "third way."
Yet why is a "third way" needed? Why should we avoid interpreting the
dialogues as aiming either to establish or refute philosophical doctrines? To
answer these questions, we need to look at the two paradigms between which
the interpretation of Plato's philosophy has traditionally oscillated. 1 Toen we
will be able to see how both paradigms, along with their historical offshoots,
have failed to do justice either to the content or to the form of the dialogues.
Toe first paradigm, beginning with the Old Academy, continuing through the
Neoplatonists, and still prevailing in our own day, regards the airo and final
product of Plato's philosophizing to have been a systematic body of
philosophical doctrines.2 Another tradition, mainly associated with the New
1. In his book The Decline and Fall of the Neoplatonic lnterpretation of Plato
(Helsinki: Societas Scientarum Pennica, 1974) E . N. Tigerstedt argues that the
Neoplatonic and the skeptical interpretations of Plato's philosophy were the only two
major ones available until the former carne to be rejected in favor of a "third
alternative," an alternative that Tigerstedt sees as first emerging with the work of Ioannes
Serranus in the sixteenth century (42). However, Tigerstedt also recognizes that this new
alternative that defines modero Platonic scholarship is in a different sense not really a
third alternative, since it shares with the Neoplatonic interpretation the basic assumption
that Plato had a philosophical system. Tigerstedt therefore observes that the scholar
Wilhelm Gottlieb Tennemann bequeathed to modero Platonic scholars "the belief that
Plato bad a philosophic system. This assumption was, of course, no invention of
Tennemann's, for it was more or less shared by ali earlier Platonists, save the New
Academy, and can be traced back to Plato's immediate successors in the Old Academy,
Speusippus and Xenocrates, though it culminated in the Neoplatonists" (68). See also
Tigerstedt's other important book, lnterpreting Plato (Stockholm: Almquist & Wiksell,
1977), for a discussion of modero representatives of the traditional "skeptical" and
"dogmatic" interpretations. Toe view that inspires the present volume is that what is
presently needed in Platonic scholarship is finally a genuine third alternative to these
interpretations.
2. Toe attempt to systematize Plato's thought began with his immediate successors in
the Academy, Speusippus and Xenocrates (heads ofthe Academy 347-339 B.C. and 339-
314 B.C., respectively). They seem to have had little interest in the dialogues, being
more conceroed with developing the theory of ultimate principies that Plato apparently
Introduction 3
Academy under Arcesilaus and finding an occasional exponent in the
subsequent his~ry of P~atonic sch_olarship, portrays Plato as a skeptic refusing
to advanc: philos?phical doctnnes of his own and instead using his
argumentahve, poet1c, and rhetorical skills to undermine conceit and promote
3
open-ended inquiry. Thus, as a member of this latter tradition Cícero writes
c?mm~nicated only orally. Their systematizing in fact led them to positions greatly
d1vergmg from what is found in the dialogues (see Eduard Zeller, Plato and the O/der
Academy, trans. Sarah Prances A1leyne & Alfred Goodwin [London: Longmans, Green,
and Co:, 18761: 55~-622; W . K. C. Guthrie, A History of Greek Philosophy, vol. 5
[C~bndge Urnvers1ty Press, 1978], 446-492; Giovanni Reale, A History of Ancient
Ph1losophy. III. The Systems of the Hellenistic Age, trans. John R. Catan [Albany:
SUN~, 1985], 67-83). This enterprise soon met resistance within the Academy itself (see
foll~wmg note). Much later (third century A.D.) the Neoplatonists, beginning with
Plotmus, s_ystematized Plato's philosophy (making it in fact essentially identical to the
Neo~lat~rnc sy~t~m) through a selective reading of bis works that ignored their aporetic
and rrornc qualit1es. Por a good, succinct account of this reading see Giovanni Reale A
History ofAncient Philosophy. IV. The Schools ofthe Imperial Age, trans. John R. ca~n
(Albany: SUNY, 1990), 307. Aécording to Reale, "the dialogues which are valuable to
him [Plotinus] are the Phaedo, Phaedrus, Symposium, Timaeus, the central books ofthe
Re;:ublic, and, in a lesser way, sorne aspects of the Sophist, the Parmenides, the
Ph1lebus, and the Second Letler" (307). Here we already see a general characteristic of
the ''.doctrinal" _inte1:J>retation: that it either ignores Plato's aporetic dialogues altogether
or g1ves them linte unportance. Yet even the dialogues that were read could be shown
to contain the Neoplatonic system only by means of allegorical interpretation. on which
se~ Reale's s~ccinct discussion in/ tre paradigmi storici nell'interpretazione di Platone
e 1fondament1 del nuovo paradigma (Napoli: Instituto Suor Orsola Benincasa, 1991), 22-
25. Por a general account of the Neoplatonists' principies of interpretation see J. A.
Coulter, The Literary Microcosm: Theories of lnterpretation of the Later Neoplatonists
(Leiden: Brill, 1979).
. 3. Toe s~te of Platonic interpretation during Arcesilaus' lifetime (c. 315-240 B.C.)
IS well ~ecnbed by A. A . Long and D. N. Sedley, The Hellenistic Philosophers, vol. 1
~Cambnd~e University Press, 1987), 445: "In fact there had been no consensus
mterpretauo~ of ~lato's_philosophy at any time since bis death , sorne seventy-five years
before Arcesilaus elect1on. If, as seems likely, Polemo [head of Academy c. 314 to 276
B.C.] and his contemporaries had already begun to react against the efforts of their
p~edecessors: Spe_usippus and Xenocrates, to create a hard-and-fast system out of Plato's
d1al~gues, this will have encouraged Arcesilaus to challenge the wbole enterprise of
readmg Pla~o as a doctrinaire philosopher." Por a thorough though generally
unsympathetJc account of Arcesilaus' interpretation see Julia Annas' article: "Platon le
sce~~ique," Revue de Metaphysique et Morale 95 (1990): 267-291. Por a somewhat more
~os1t1ve and ~~mpathetic,account, however, see Carlos Lévy's reply to Julia Annas:
Platon, Arcesilas, Carneade. Reponse a J. Annas," ibid.: 293-306. See also Annas'
mo:e recent version of this article, "Plato the Sceptic," Oxford Studies in Ancient
Philosophy, supp. vol. (1992), ed. James C. Klagge and Nicholas D. Smith, 43-72. Sorne
of the arguments with which the ancient skeptics tried to show that Plato was not a
4 Toe Third Way
that in Plato 's books "nothing is affirmed, many arguments are presented on
both sides of a question, and ali is inquired into without anything certain being
said. " 4
Toe first paradigm constitutes the orthodoxy bebind the dissension and
divergence in what can be called "mainstream" Platonic studies. Many
scholars today agree in assumi~ that Plato's thought takes the forro of a
roughly gystematic body of doctrines, and that the main point of the dialogues
is to present arguments in defense of these doctrines. Divergences are to be
found, however, in how this assumption has been reconciled with the actual
forro of the dialogues. Toe death of the Neoplatonic interpretation of Plato
simply gave way to other attempts to reconstruct Plato's philosopbic syste~.
One such attempt was made by Wilhelm Gottlieb Tenneman who, in bis book,
System der P/atonischen Philosophie (Leipzig, 1792-95), could get a system
out of the ungystematic and fragmentary dialogues only by ignoring their forro
and using Kant's critical pbilosophy to fill in gaps. 5 Friedricp
Schleiermacher's famous introduction to bis translations of the dialogues
(1804) was in large part directed against the kind of violent treatment of the
dialogues characteristic of Tenneman's approach. However, Schleiermacher
did not abandon the assumption that Plato, like any other pbilosopher, must be
seen as expounding a system. 6 What distinguished bim from Tenneman and
dogmatic philosopher can be found in the Anonymous Prolegomena to P/,ato 's Philosophy,
ed. L. G. Westerink (Amsterdam: North-Holland Publishing Company, 1962), 20-24.
Por a more comprehensive history of this alternative tradition see Tigerstedt's The
Decline ami Fall ofthe Neop/,atonic lnterpretation of P/,ato, 7-10, 12-13, 31-38, and 49.
4. "cuius in Jibris nihil adfinnatur et in utramque partem multa disseruntur, de
omnibus quaeritur, nihil certi dicitur" (Academica 1.46).
5. Por an account of Tennemann's interpretation see Tigerstedt, The Decline ami Fall,
64-68.
6. See Schleiennacher's claim that Plato was the first systematic philosopher, in
Geschichte der Philosophie, aus Schleiermachers hamischriftlichen Nach/,ass, ed. H.
Ritter in Siimmtliche Werke III 4, 1 (Berlin: G. Reimer, 1839), 98 . Schleiennacher's
systematic approach is evident even in the famous "lntroduction," where he divides the
dialogues into the following three groups: 1) the earliest group, which focuses on the
preliminaries of a science, i.e., Jogic, epistemology, and philosophical method in general
(this group includes the Phaedrus, Protagoras, and Parmenides as well as the "aporetic"
dialogues); 2) a middle group, which further develops the general principies of
philosophy and begins to apply them in a tentative way to the concrete subject matter of
ethics and physics (here are included the Theaetetus, Sophist, Statesman, Phaedo, and
Philebus, in that order); 3) a final group, which finally provides a constructive and
scientific exposition of the main areas of philosophy (here are included the Republic,
Timaeus, Critias, and Laws) . What is described by this grouping is a gradual introduction
to a system of doctrines. Since most of the dialogues are far from having even the
appearance of being systematic or fully conclusive, Schleiermacher must characterize
Introduction 5
probably constituted bis most important bequest to modero Platonic studies was
bis insistence that the pbilosopbical content of the dialogues was inseparable
7
from their forro. His own way of reconciling forro with content was to see
the dialogues as a gradual and pedagogically ordered exposition of Plato's
system. This interpretation committed Schleiermacher to the view that Plato's
doc~nes ~emain fundamentally the same throughout the dialogues: only the
imy 10 wbich they are presented changes. Schleiermacher therefore represents
a "unitarian" view of Plato's doctrines that is primarily associated in this
century with Paul Shorey. 8
This view, however, has always been threatened by the existence of
apparent contradictions between what is said in different dialogues. 9 It has
therefore been superseded by the "developmentalist" interpretation dominant
in Platonic scholarsbip today, wbich was inaugurated by Karl Friedrich
Hermano, the nineteenth-century scholar who combined the notions of
"system" and "evolution"-a synthesis pronounced by the very title of bis
?ook: (!eschichte und System der Platonischen Philosophie. 10 Toe
10troduct1on of the notion of "evolution" allowed Hermano unlike
Schleiermacher, to explain contradictions among the dialogues" (the; are due
to the fact that Plato's doctrines developed throughout bis writing career) and
thus to vindicate the systematic character of Plato's thought. 12 This is in fact
most of them as preliminary or tentative.
7. "So_also will those spectators of the analysis fail altogether to attain to a knowledge
of the Philosop~! of_Plato, for in that, if in any thing, form and subject are inseparable,
and ~o p~opos1t100 1s to be rightly understood, except in its own place, and with the
combinattons and limitations which Plato has assigned to it" (lntroductions to the
Dialogues of P/,ato, trans. William Dobson, reprinted [New York: Arno Press 1973]
14). ' '
8. Paul Shorey, The Unity of Plato's Thought (University of Chicago Press, 1960).
9. One way in which scholars in the nineteenth century sought to eliminate these
contradictions was to reject from a few to most of the dialogues as spurious. Por an
excellent account and critique of the excesses of this "solution" see Luigi Stefanini,
P/,atone, 2d. ed. (Padova: CEDAM, 1949; repr. 1992), xviii-xx.
10. Heidelberg, 1839; reprinted New York: Arno Press, 1976.
11 . Por Hennann's argument that Schleiermacher's view is incompatible with
contradictions existing between the dialogues see ibid., 352 and 369 (examples of these
contradictions can be found on p. 371 ). An excellent survey of sorne of the contradictions
can also be found in Stefanini, xiv-xvü.
12. Hennann_ is no less_~onvinced than Schleiermacher that Plato's philosophy is a
system of doctnnes (see 1bid., 347). This is why, apart from the fact that Hennann
places th~ Phaedrus_ in the /,ate period of Plato's career, his division of the dialogues into
three p~nods essenl!ally agrees with Schleiennacher's division (see note above; the only
other differences are that Hermann places the Parmenides in the middle period and the
Phaedo in the late period). Por both scholars the final group of dialogues is
6 Toe Third Way
the virtue of the "developmentalist" hypothesis that makes it today the most
13
widespread form of the "doctrinal" interpretation. • •
Yet even this hypothesis is starting to face criticism. One problem 1s that 1t
depends on establishing the order in which Plato actually wrote the dialogues,
and the meaos by which this has been attempted are beginning to betray flaws
and serious limitations. 14 Another problern is that this interpretation requires
"constructive" because in them Plato finally offers an exposition of bis whole system.
Of course, geneticists in the twentieth century depart from the division of both scholars
by inverting (in large part due to stylometric studies) their m.iddle and late periods (so
that the Republic is placed in the m.iddle period and the Sophist and Statesman in the late
period). Despite, however, these disagreements about the order of the dia~og~es (and
therefore about the character of Plato's final "system"), the fundamental pnnc1ples and
presuppositions of the geneticist interpretation have remained constant since the
publication of Herroann's monumental work.
13. Por a few out of the numerous examples of this approach see: I. M. Crombie, An
F.xamination of Plato's Doctrines (London: Routledge, 1962); R. C. Cross and A. D.
Woozley, Plato's Republic: A Philosophical Commentary (London: Macmillan, 1964);
Plato, a Collection of Critical Essays /: Metaphysics and Epistemology, ed. Gregory
Vlastos (Garden City, New York: Anchor Books, Doubleday, 1970); Plato, A Collection
of Critical Essays, II: Ethics, Politics, and Philosophy of Art and Religion, ed. Greg~ry
Vlastos (Garden City, New York: Anchor Books, Doubleday, 1970); Terence Irwm,
Plato 's Moral Theory: The Early and Middle Dialogues (Oxford: Clarendon Press, 1977);
Nicholas P. White, Plato on Knowledge and Reality (lndianapolis: Hackett Publishing
Company, 1976); Gregory Vlastos, Socrates: Ironist and Moral Philosopher (Ithaca:
Comell University Press, 1991); The Cambridge Companion to Plato, ed. Richard Kraut
(Cambridge: Cambridge University Press, 1992). Though the latter volume claims to
presenta "conspectus of recent developments in the interpretation of Plato," it essenti~lly
ignores alternatives to the developmental, doctrinal interpretation, and the editor
categorically states that "the dialogue forro of [Plato's] 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 philosopbical treatises, many of them share these
purposes with philospbical treatises" (26). As Mary Margaret McCabe rightly observes,
"the editorial tendency of this col!ection seems to be more towards the setting out of
Platonic doctrine rather than the assessment ofbis arguments in context" ("Porro, Porros,
and Reforro," Oxford Studies in Ancient Philosophy 12 [1994]: 220).
14. Por serious challenges to attempts to establish a chronology see Holger Thesleff,
Studies in Platonic Chronology (Helsinki, 1982) and "Platonic Chronology," Phronesis
34 (1989): 1-26; Jacob Howland, "Re-reading Plato: Toe Problem of Platonic
Chronology Reconsidered," Phoenix 45 (1991): 189-214; Debra Nails, "Platonic
Chronology Reconsidered," Bryn Mawr Classical Review 3 (1992): 314-27, "Problems
with Vlastos' Platonic Developmentalism," Ancient Philosophy 13 (1993): 273-91,
"Plato's Middle Cluster," Phoenix 48 (1994): 62-7. Ledger's recent stylometric study in
Re-counting Plato: A Computer Analysis of Plato 's Style (Oxford: Clarendon Press, 1989)
is important because it draws our attention to the circularity of previous stylometric
Introduction 7
us to read the dialogues as treatises in whicb Plato defended wbatever
doctrines he was committed to at the time of composition and thereby requires
us in principie to deemphasize their dramatic and literary aspects. 15 Finally,
there are questions central to Plato's system (e.g., "What is the Good?") that
apparently never receive answers in the dialogues. 16
One way out of these problems is to abandon the "developmentalist"
hypothesis in favor of another form of "doctrinal" interpretation that is
becoming increasingly influential: the "esotericist" hypothesis. 17 According
studies. However, Ledger's analysis itself does not fully avoid circularity, as is pointed
out by Nails in "Plato's Middle Cluster" and by T. M. Robinson in "Plato and the
Computer," Ancient Philosophy 12 (1992): 378-9. Por other problems with Ledger's
ana~ysis and with stylometry in general see Charles M . Young, "Plato and Computer
Datmg," Oxford Studies in Ancient Philosophy 12 (1994): 227-250.
15. Toe major defect ofthe "developmental" interpretation is that it is not compatible
with reading the dialogues as dialogues. It is therefore not surprising that Karl Priedrich
Herroann who, as noted above, was the principie originator of this interpretation,
dismissed the dialogue forro as "eine beliebte und hergebrachte Einkleidungsweise" (354-
5). Toe incompatibility is well-expressed by Jacob Howland: "Toe key point here is that
one cannot consistently appeal both to what Plato is alleged to have thought at a certain
time a_nd to features interna! to a given dialogue (including its dramatic time and setting,
narrat1ve structure, the character of its interlocutors, literary and bistorical allusions. and
the like) in order to explain the kinds of styles of argument one fmds in it. Toe
Traditionalist may claim to be sensitive to the fact that the dialogues, as written records
of living conversations, must be understood in terros of their rhetorical and dialectical
dimensions, but he undercuts this claim precisely at the point where he appeals (as
~evitably he must, gi:en bis fundamental interpretative presupposition) to putatively
mdependent chronolog1cal considerations" ("Philosophy as Dialogue," Reason Papers 17
[Pall 1992]: 118-19). See also the fol!owing critiques of Terence Irwin: David Roochnik,
"Terence Irwin's Reading of Plato," in Platonic Writings, Platonic Readings, ed. Charles
L. Griswold, Jr. (New York: Routledge, Chapman & Hall, 1988), 183-193; Victorino
Tejera, "Methodology of a Misreading: A Critica! Note on T . Irwin's 'Plato's Moral
Theory,"' lntemational Studies in Philosophy 10 (1978): 131-6.
16. Por a thorough defense ofthe claim that what is said about dialectic and the Good
in ~e m.iddle book~ of the Republic cannot be unders~?od and explained without going
ou~1~e. of the d1al~gue see: H. J. Krlimer. "Uber den Zusammenhang von
Pnnz1p1enlehre und D1alektik bei Plato; zur Definition des Dialektikers Politeia 534B-C "
Philologus 100 (1966): 35-70; and Giovanni Reale, Per una nuova interpretazione di
Platone: Relettura della metajisica dei grandi dialoghi alle luce delle 'Dottrine non
scritte,' 6th ed (Milan: Universita Cattolica del Sacro Cuore. 1989), 293-333.
17. Toe classic presentation of the "esotericist" interpretation is to be found in
Krlimer's Arele bei Platon und Aristoteles: zum Wesen und zur Geschichte der
Platonischen Ontologie (Heidelberg: C. Winter, 1959) and Gaiser's Platons
Ungeschriebene Lehre, 2d ed. (Stuttgart: E. Klett, 1968). G. Reale leads the Italian
branch of the Tübingen position: his exposition and development of it can be found in
8 Toe Third Way
to tbis view, Plato's philosopbical system neither develops nor re~ains the
same throughout the dialogues because, it is not to be found in the dialogues.
Instead, Plato communicated bis system only orally witbin the Academy and
therefore its most fundamental doctrines are available to us only through the
reports of Aristotle and other members. Toe problem with tbis interpr7tati?n
is that it requires us to look for Plato's pbilosophy not in the extraordinanly
rich and suggestive body of writings he left us, but instead in a bighly abstract
system of principles reconstructed from . incomplete, biased, and bighl_y
ambiguous reports in secondary sources. 18 What needs to be note~ here. 1s
that all three of the above interpretations found in modero Platomc studies
assume that Plato's pbilosophy is a system of pbilosopbical doctrines and ~en
pursue different ways of reconciling tbis assumpti_on with _the unsys~matic,
nondoctrinal form of the dialogues, namely, by seemg the dialogues e1ther as
a gmdual exposition of the system ("unitarianism"), or as records of the
system's evolution ("developmentalism"), or as propaedeutic to a system they
do not contain ("esotericism"). 19
Partly as a result of dissatisfaction with all of these "doctrinal"
interpretations, the skeptical interpretation is finding an increasing number of
exponents today. 20 Sorne scholars have emphasized, like Cicero, the absence
Per une nuova interpretazione di Platone, cited above. Now we also hav~ a very
thorough explanation and defense of this position in English, though unfortunately very
bad English: Krlimer's Plato and the Foundations ofMetaphysics, ed. and trans. John R.
Catan (Albany: SUNY, 1990).
18. Por an excellent recent critique of the "esotericist" interpretation see Kenneth
Sayre, "Review of Krlimer's Plato and tbe Foundations of Metapbysics," Ancient
Philosophy 13 (1993): 167-184. . . . .
19. "Esotericism" sees itself as a radical departure from the rnterpretat1ve pnnc1ples
of Schleiermacher. This is true to the extent that it rejects the autonomy that
Schleiermacher grants the dialogues, but it does so only because it refuses to abandon
Schleiermacher 's other presupposition: that Plato 's philosophy is a system of doctrines.
Por example, Reale's objection to seeking Plato's thought in the dialogues alone is that
the latter provide us with neither an adequate "systematic framework" nor the "doctrinal
nuclei of Plato's thought" ("i nuclei dottrinali del pensiero platonico;" I tre paradigmi
storici, 28 & 30).
20. It is significant that in addition to the old question ofwhich views in the dialogues
are Plato's and which are Socrates' , one now finds with increasing regularity the question
of whether any of the views in the dialogues are Plato's. Por example, in his prologue
to the recent collection on "Methods of Interpreting Plato and his Dialogues," James C.
Klagge asks: "Does Plato expound his own views, as opposed to not expounding views
at ali?" (Oxford Studies in Ancient Philosophy, suppl. vol. [1992), ed. James C. Klagge
and Nicholas D. Smith, 4). Another paper in the same collection begins with the
question: "Does Plato have a philosophy?" (S. Marc Cohen and David Keyt, "Analy~ing
Plato's Arguments: Plato and Platonism," ibid., 173). A few years ago these questions
Introduction 9
in the dialogues of final answers to the questions Plato considered most
important. 21 Other scholars have been led to deny the existence of doctrines
in the dialogues by a desire to place greater emphasis on their dramatic and
literary character. 22 Leo Strauss and bis followers have been the main
representatives of tbis approach in North America. Their view that every
dramatic detail has significance and must be given weight in an interpretation
generally leads them to minimize the existence or importance of "doctrines."
Thus, Strauss claims that Plato's "dialogues supply us not so much with an
answer to the riddle of being as with a most articulate 'imitation' of that
riddle. His teacbing can never become the subject of indoctrination. In the last
analysis bis writings can not be used for any other purpose than for
pbilosophizing. "23 Similarly, one of bis students, Stanley Rosen, writes that
"Pbilosophy as portrayed in the Sophist, and in the entire Platonic corpus, is
would have provoked only laughter from most scholars. There are two distinct issues
here, however, which are not kept distinct by Klagge when he responds to his question
by concluding that "the evidence establishes a slight presumption in favor of the view
that sorne of the main philosophical and political positions articulated by the leading
characters in the dialogues do indeed represent the views of Plato himself" (Klagge, 11).
I frnd it plausible that the dialogues contain sorne of Plato's own views, justas I frnd it
plausible that King Lear and Hamlet contain sorne of Shakespeare' s own views. Yet this
leaves open a different and more important question: is the aim of Plato's dialogues , and
of bis philosophizing generally, to expound and defend such views? It is a positive
answer to this question, and not simply the belief that Plato's views can be found in the
dialogues, that defrnes what I call the "doctrinal" approach. Toe papers in the present
collection, on the other hand, seek to ascribe to Plato different aims, aims more
consonant with the characteristics of the dialogue fonn.
21. A skeptical interpretation of the "Socratic dialogues" has been defended by
Richard Kraut, Socrates and the State (Princeton: Princeton University Press, 1984), a
skeptical interpretation ofthe dialogues as a whole has been defended by Peter Stemmer,
Platons Dialektik: Die Früheren und Mittleren Dialoge (Berlin: Walter de Gruyter,
1992), especially 222-3 and 272-3, and a skeptical interpretation of both the dialogues
as a whole and the "unwritten teachings" has been defended by Rafael Ferber, Die
Unwissenheit des Philosophen oder Warum hat Plato die "ungeschriebene Lehre" nicht
geschrieben (Sankt Augustin: Akademia Verlag, 1991). Ali three authors, in opposing
the mainstream "dogmatic" interpretation, work entirely within the dichotomy that the
present volume seeks to transcend.
22. Annas comments on the parallel between Arcesilaus' interpretation of Plato and
that of modern commentators "who emphasize the 'literary' aspects of the Socratic
dialogues" ("Plato the Sceptic," n. 28, p. 60; "Platon le sceptique," n. 18, p. 281).
23. "On a New Interpretation of Plato's Political Philosophy," Social Research 13
(1946): 351.
10 Toe Third Way
not a doctrine but a problem. " 24 However, the nondoctrinal approach is by
25
no means confined to Strauss and bis followers, even in North America. In
Europe existentialism and phenomenology have encouraged this kind of
interpretation. Sorne of the major representatives in Germany are H. G.
Gadamer, Hermann Gundert, and Wolfgang Wieland.26 In France, the
literary dimension of the dialogues has always been given much attention by
interpreters: a classic example is R. Schaerer;27 a more recent one is Jacques
Derrida. 28 Despite the great diversity to be found in the different strands of
24. Plato 's Sophist (New Haven: Yale University Press, 1983), 331. See also Rosen's
"Platonic Hermeneutics: On the Interpretation of a Platonic Dialogue," in Proceedings
of the Boston A rea Colloquium in Ancient Philosophy, vol. 1, ed. John Cleary (New
York: University Press of America, 1986), 271-288. In tum, a student ofStanley Rosen
has defended a similar position: Drew Hyland, "Why Plato Wrote Dialogues,"
Philosophy and Rhetoric 1 (1968): 38-50; see especially p. 41.
25. Two other North American representatives of this approach are Frederick J. E.
Woodbridge, The Son of Apollo (Boston: Houghton Mifflin, 1929; repr. New York:
Columbia University Press, 1965) and J. Randall, Plato: Dramatist of the Lije of Reason
(New York: Columbia University Press, 1970).
26. See especially H. G. Gadamer, Dialogue and Dialectic, trans. P. Christopher
Smith (New Haven: Yale University Press, 1980) and Plato's Dialectical Ethics:
Phenomenological Interpretations Relating to the Philebus, trans. R. M. Wallace (New
Haven: Yale University Press, 1991); Hermann Gundert, Der Platonische Dialog
(Heidelberg: C. Winter, 1968) and Dialog und Dialektik (Amsterdam: B. R. Grüner,
1971); and Wolfgang Wieland, Platon und die Formen des Wissens (Goningen:
Vandenhoeck & Ruprecht, 1982). Wieland claims that dialectic, unlike mathematics, does
not have the goal of winning and establishing true propositions about certain facts. He
can therefore assert that while there are "mathematical propositions," there are no
"dialectical propositions" ("Platon und der Nutzen der Idee. Zur Funktion der Idee des
Guten," Allgemeine Zeitschrift jür Philosophie 1 [1976]: 32).
27. R. Schaerer, La question platonicienne: Étude surles rapports de la pensée et de
l'expression dans les dialogues, 2nd ed. (Paris: J. Vrin, 1969). (Note the very
Schleiermacherian subtitle).
28. See his very influential essay: "La Pharmacie de Platon," in La dissémination
(Paris: Éditions du Seuil, 1972), 69-198. It is important to note that Derrida himself
tends to see Plato as a metaphysical dogmatist. Sorne of those influenced by his reading
of the dialogues, however, deny this characterization of Plato and thus make Plato more
like Derrida than Derrida himself does. See, for example, David M. Halperin's
statement: " ... I read Plato in opposition to Derrida not as a metaphysical dogmatist but
as a kind of deconstructionist avant la lettre, a cunning writer fully alive to the
doubleness of his rhetoric who embraces différance and who actively courts in his writing
an effect of undecidability" ("Plato and the Erotics of Narrativity," Oxford Studies in
Ancient Philosophy, suppl. vol. [1992), ed. James C. Klagge and Nicholas D. Smith,
118). Por another, more thorough critique of Derrida for not recognizing in Plato a
fellow "deconstructionist," see Stanley Rosen, "Platonic Hermeneutics," 271-288.
Introduction 11
this "nondoctrinal" approach, what unifies it is the view it shares with the
skeptical paradigm described above: that the form of the dialogues is
incompatible with the characterization of Plato as a pbilosopher whose aim was
to prove and defend specific pbilosopbical doctrines. 29
This approach does in fact appear best able to explain the dialogue form. If
Plato was a skeptic or, in more modero terms, a "deconstructionist," then he
understandably would have wanted to distance bimself from what is said in the
dialogues by biding bebind their dramatis personae and never stating anything
in bis own person. This distancing, on the other hand, appears pointless from
the perspective of the other tradition that depicts Plato as a pbilosopher seeking
to defend bis own doctrines. Toe sceptical interpretation can also account for
another important characteristic of the dialogues: ad hominem argumentation.
If Plato's aim was to present universally valid arguments for the establishment
ofpbilosophical doctrines (something that could be done in a treatise), then the
dialogue, which at least in form relativizes argumentation by referring it to a
specific context and a specific interlocutor, appears not only unnecessary but
even obstructive. If, on the other hand, Plato's aim was to expose and refute
the presuppositions of specific types of interlocutors, then the necessity of ad
hominem argumentation is evident. Its importance would in turn explain the
literary and dramatic aspects of the dialogues, since in order for it to be
effective, the interlocutor would have to be given a concrete character and
placed in situations where bis behaviour and the presuppositions upon which
this behaviour rests are made clear. 30 Toe literary and dramatic details of the
29. Por examples of the nondoctrinal approach in ali of its diversity, see the following
two collections of essays and discussions: Platonic Writings, Platonic Readings, ed.
Charles L. Griswold Jr. (New York: Routledge, Chapman & Hall, Inc., 1988); Plato's
Dialogues: New Studies and lnterpretations, ed. Gerald A. Press (Rowman & Linlefield
Publishers, Inc. , 1993). Many of those who could be classified under this approach are
offering a genuinely new "third way" of interpreting Plato's dialogues, but too often they
do not clearly distinguish their approach from the "skeptical" one.
30. Julia Annas shows how the New Academy, by finding skeptical ad hominem
argumentation in the dialogues, coutd do justice to their literary and dramatic aspects:
"lf an argument is ad hominem then fully to understand it we must know about what the
int~rlocutor thinks, and focus on the actual moves and why they are made, rather than
seemg the argument as part of a built-up 'Socratic ethics.' Much recent focus on the
'dialogue forro' and treatment of Socratic arguments piecemeal with stress on the
particular context implicitly revives Archelaus' Socrates. Toe Academy may even have
stressed th~se literary aspects themselves, since we know that an interest in oratory
developed m the Academy, and Socratic arguments provide much natural material for
studying how to (and how not to) convince various types of people" ("Plato the Sceptic,"
P- 60, n. 28). It is worth noting that Arcesilaus apparently owned Plato's books, perhaps
even the original manuscripts (" KOlL rcx {3i(J>.fo eKÉKTTJTO Dlvroü," Diogenes Laertius,
IV.32-3; A. A. Long and D. N. Sedley interpret this as probably meamng that
12 Toe Tbird Way
dialogues appear as so much pointless and even misleading omamentation, on
the other hand, if Plato's objective was to transmit and defend pbilosopbical
doctrines, unless one is willing to give these details allegorical significance in
the manner of the N eoplatonists.
Toe irreconcilable opposition between the "doctrinal" interpretation and the
dialogue forro is most evident if one considers the comerstone of this
interpretation: its attribution to Plato of a "theory of forros." To the
unprejudiced reader, it is evident that there is no "theory of forros " in the
dialogues. What one instead finds is what one scholar has called
"impressionistic language"31 from wbich readers, starting with Aristotle, have
attempted to construct a "theory of forms." This is to say that the the9ry,
whether rightly or wrongly attributed to Plato, is a construction by the reader
of what is not to be found as such in the dialogues. But then the simple
question is unavoidable: why did Plato, rather than writing a treatise in wbich
this comerstone of bis metaphysics is fully worked out, instead write dramatic
dialogues in wbich references to the forms are not only scattered and ad hoc,
but are also generally couched in poetic images and metaphors? Toe skeptical
interpretation would have no difficulty in accounting for Plato's choice of
dialogue over treatise, since it does not see him as committed to any theory.
Yet despite this virtue, the skeptical interpretation is flawed. While the
dialogues may be unsystematic, it is clearly not the case that they contain only
problems, refutations, and questions. Tbey suggest sorne sort of positive
pbilosophy. If one wants to claim that this positive pbilosophy is not a
"system" nor a set of "doctrines," then the burden is on one to explain what
exactly it is. If Plato did not see pbilosophy as a matter of advancing and
defending pbilosophical theses, then how did he understand it? And what is the
connection between pbilosophy as thus understood and the dramatic, literary
form of the dialogues? If it fails to answer these questions, the "nondoctrinal"
interpretation being advocated today runs two risks: either it will "tum
Platonism into empty 'pbilosopbizing"' 32 and the dialogues into "mere
"A.rcesilaus was in possession of Plato's own library and manuscripts" [The Hellenistic
Philosophers, vol. 2 (Cambridge: Cambridge University Press, 1987), 435)).
31. Gail Fine, "Aristotle's Criticisms of Plato," Oxford Studies in Ancient Philosophy,
suppl. vol. (1992), ed. James C. Klagge and Nicholas D. Smith, 13-41. Toe most
thorough and convincing defence of the claim that there is no "theory" of forms in the
dialogues is to be found in Wieland's book, Platon und die Formen des Wissens, 95-150.
Wieland does not deny that forms are recognized in the dialogues, but claims that they
are not and cannot be made objects of any "theory." Toe phrase, "Ideen ohne
ldeenlehre," succinctly expresses his positiofü
32. This is the phrase used by E. N. Tigerstedt to describe what Plato's thought is
transformed into by certain "existentialist" interpreters (lnterpreting Plato, 103). lt is
also worthwhile citing here Findlay 's sharp criticism of the "nondoctrinal" interpretation:
Introduction 13
literature" or it will surreptitiously reintroduce a "doctrinal" conception of
pbilosophy into the dialogues by, for example, seeing their literary details as
a kind of secret code for bidden doctrines. 33 Many of those classified above
under the "nondoctrinal" approach have already at least begun to answer these
questions, and the contributors to this volume are indebted to sorne of their
suggestions. To this extent, however, they are moving away from the purely
skeptical interpretation. wbich is itself inadequate because it cannot explain the
positive pbilosopbical content of the dialogues and therefore either ignores it
altogether or lapses back into the doctrinal interpretation.
What these observations leave us with is a paradox: the skeptical
interpretation can account for the form of Plato's writings only by minimizing
their positive pbilosopbical content, wbile the "doctrinal" interpretation can
uncover their content only at the cost of considering their form little more than
a curiosity and even an embarrassment. Tbis state of affairs cannot but leave
us wisbing for sorne third alternative: an interpretation that, unlike the
sceptical one, grants positive content to Plato 's pbilosophy, and that, unlike the
"doctrinal" one, is able to show sorne necessary connection between this
pbilosophy and the dialogue forro. Is there such an alternative?
It is precisely such an alternative that the papers in this collection seek to
define. Toe different ways in wbich this alternative is sought are reflected by
the division of the collection into five parts representing five sets of
oppositions. Toe papers in the first part observe that Plato's critique of the
writ!en ~ord as ~•~xed" and "unalterable" counts against the view that the goal
of bis pbilosophiz1ng was to communicate and defend pbilosopbical doctrines.
On the other hand, despite bis apparent preference for oral discourse, Plato did
wri~e. This tension in bis thought between oral and written discourse provides
fertlle ground for discovering an alternative both to interpretations that
characterize bis thought as completely fluid and open-ended (like a runny nose,
to adapt one of Plato's own images) and to interpretations that characterize bis
thought as fixed doctrine.
Jackson P. Hershbell's paper provides a good starting point for working out
such an alternative by reviewing the different theses that have been advanced
in the scholarly literature concerning the origins, nature, and purpose of the
"Many are satisified with this tentative, suggestive thought-fragmentation which is
certainly worth a pack of low-grade systematization, but the nisus of Plato's thought is
n~ne ~e l~ss towards systematic completion, and if we lose the willingness to run along
w1th this msus, the fragments lose ali their meaning, become even trivial and ridiculous.
A Plato who merely played around with notions and arguments is a Plato corrupted, a
Plato unworthy of serious study" (Plato: The Written and Unwrirten Doctrines, 6).
33. For a critique of Straussian interpreters along these lines see Harry Berger Jr.,
"Levels of Discourse in '.lato's_ Dialogues," inLiterature and the Question of Philosophy,
ed. Anthony J. Cascard1 (Baltunore: Johns Hopkins University Press, 1987), 77-100.
14 Toe Third Way
dialogue forro. In addition to the often-noted influence of Athenian drama,
Hershbell points to the Athenian symposia or "drinking parties" and to
courtroom elenchus or cross-examination as possible influences on Plato 's
choice of the dialogue forro. Hershbell also reviews the arguments made for
seeing the dialogue forro as reflecting a transition from orality to literacy.
Most importantly for the purposes of the present volume, Hershbell discusses
the pbilosopbical motive that sorne have seen bebind Plato's choice of the
dialogue forro: the view that both the written and the spoken word are
inadequate for expressi~ the truth, and that, therefore, a forro of writing is
needed that not only approximates the fluidity of oral discourse but is also
nondogmatic and open-ended, always reminding the reader that he must go
beyond what is written or spoken in order to discover the truth. Hershbell does
not choose between these different explanations of the dialogue forro nor does
he see them as mutually exclusive. His paper, however, provides us with an
excellent account of the present state of the scholarsbip on this question.
Elinor J. M . West is one of those who sees the Platonic dialogue as existing
at the crossroads of orality and literacy. She rejects the equation of an
"unlettered" society with an "uncultured" society, showing instead how an oral
culture's attitudes towards time, composition, and remembering differ from
those of a literate culture. Agreeing with those who have seen the culture of
Ancient Greece as originally and predominantly an oral one, West argues that
one should understand Plato's dialogues within the context of such a culture,
wbich, even in its transition to literacy, still exhibited a "mistrust of letters"
(a mistrust that West thoroughly documents). In this way, West is resistí~ the
common tendency to understand the dialogues within the context of our own
culture in wbich the written word is predominant. By situating Plato within an
oral culture with its different attitudes towards time, composition, and
remembering, West can claim that Plato cannot be understood as an "author"
in our sense of the word, that he cannot be seen as "developing," and that the
categories of "fact" or "fiction" are inappropriate for describing bis portrayal
of Socrates. In commenting on Plato's anonymity or silence in the dialogues.
West makes the interesting suggestions that "Plato" is the relation between
Socrates and bis respondents, and that "Plato" is the "temporal unity of each
of bis texts" (something that is similarly, but not identically, true of
"Homer"). These suggestions oppose the view that "Plato" is to be identified
with specific doctrines stated in the dialogue (by Socrates or someone else),
as well as the view that the space at the end of an "aporetic" dialogue
represents a failure. To understand "Plato," we must rehearse and recreate the
conversations preserved by the dialogues, making objections and asking
questions, _spotting relations and discontinuities. "Plato" cannot be equated
with any written words in bis texts. To this extent, he remains faithful to the
Athenian mistrust of letters.
Joanne Waugh explores further the pbilosophical implications of the
Introduction 15
distinction between spoken and written discourse. Waugh argues that the
understanding of pbilosopbical language as logically tenseless and as
completely independent of the specific context in wbich it is spoken emei:ges
only with the forro of discourse that is capable of rendering language fixed and
unalterable: writing. Toe sentences fixed in this way become "timeless"
propositions and thus the objects of knowledge. Therefore, Waugh argues that
to look for such propositions in Plato's dialogues is to miss completely their
oral dimension, their portrayal of philosopbical language as rooted in the
specific contexts of contingent, concrete discussions. This is not to say that
pbilosophy for Plato has no universal import, but rather that the universal truth
that pbilosophy actively seeks is not to be found in a system of timeless
propositions that can be written down. Waugh shows that our conception of
pbilosophy is fundamentally transforroed if we see it as an "oral" enterprise
instead of as a matter of "publisbing or perisbing."
Toe papers in the second part seek a "third way" in the uneasy kinsbip
between pbilosophy and rhetoric. One cannot deny that the dialogues are
critical of rhetoric and the relativism that it seems to presuppose. However,
the rhetorical character of pbilosophy as depicted in the dialogues also cannot
be denied and appears to tell against the possibility of pbilosophy being a
purely rational and apodictic science. Here again, Plato's pbilosophy appears
to exist between two opposites.
David Roochnik's paper brings into question the standard view that Plato
completely opposes pbilosophy to rhetoric. In focusing on the locus classicus
for this view, the Gorgias, Roochnik shows that the criticism of rhetoric there
is highly ambiguous. In addition to leaving many fundamental questions
unanswered, Socrates assumes rhetoric to be a techne in bis refutation of
Goi:gias and denies it is a techne in bis discussion with Polus, thus not only
leaving its nature questionable but also appearing to use rhetoric for bis own
purposes. Moreover, the discussion in the Gorgias enables us to make a sharp
distinction between pbilosophy and rhetoric only if pbilosophy is itself a techne
with a fully determínate, specific subject matter. However, Roochnik provides
reasons for doubting, even within the context of the Gorgias, that pbilosophy
is a forro of technical knowledge. Socrates uses the techne analogy both to
~efute the rhetorician's claim to have a techne and to encourage bis
mterlocutors to pursue knowledge. Neither use implies that Socrates bimself
as a philosopher possesses a techne. In conclusion, Roochnik sees pbilosophy
as a tension between the indeterminacy of the concrete and contingent
situations to wbich rhetoric is restricted and the determinacy of that technical
knowledge of the good that the philosopher seeks without ever attaining. Plato
chose the dialogue forro on account of its ability to portray this tension by
rooting the pursuit of knowledge of the good in the contingency of discussions
taking place at a particular time and place with particular interlocutors. Toe
dialogue is neither a rhetorical composition nor a technical treatise and is
16 Toe Third Way
therefore perfectly suited to conveying a philosophy that is neither rhetoric nor
tecbnical knowledge, but a tension between the two.
P. Cbristopher Smith also seelcs a via media between seeing the dialogues
as mere literary dramas aiming at rhetorical exhortation and seeing them as
trearises expounding and demonstrating doctrines. In interpreting what is
actually taking place in the dialogues, Smith makes use of Aristotle's
distinction between demonstrarive aigumentation and rhetorical-dialectical
argumentation. Toe former is the kind of argumentation used by mathematics
and starts with self-evident axioms in order to deduce from them sorne new
knowledge. Toe latter starts with generally held beliefs, which are understood
to be always open to question, and aims not so much to provide knowledge as
to influence one's decisions. Given this distinction, which kind of
argumentation do we find in the dialogues? Smith turns to the Phaedo, both
for its wealth of arguments and for its reflections on the nature of
argumentation. In examining the different arguments put forward by Socrates
for the immortality of the soul, Smith shows that most of them are rhetorical-
dialectical and not at ali demonstrative. After the major objections of Simmias
and Cebes, Socrates indeed attempts to demonstrote the immortality of the soul
by means of a method of "hypothesis" like that employed by the
mathematicians. Smith argues, however, that by the end of the dialogue we are
meant to see that this attempt has failed. When it comes to matters like the
soul, the good, and beauty, no unshakeable or divine logos is available to us.
Toe failure to find such a logos does not collapse the dialogue into a mere
literary drama. Instead, in its very failure, the dialogue reveals a via media
between mere exhortation or persuasion and apodictic demonstration.
Toe papers in the third part explore the "third way" offered by Plato's use
of drama and myth. Toe dialogues appeal not only to our reason, but also to
our emotions and imagination; they not only prove points, but also enact
something and tell stories. While they are not simply dramas or myths, they
are also not abstract theoretical treatises. A proper understanding of Plato's
philosophy therefore requires an understanding of the relation in bis dialogues
between imagination and reason, between drama and philosophical aigument.
Many readers of Plato's dialogues, starting with Xenocrates and Speusippus,
have denied the importance of the myths by interpreting them as mere
allegories that clothe a systematic doctrine arrived at and defended through
rational argumentation alone. This interpretation assumes that views arrived
at solely through flights of imagination (which is what the myths are) and not
based on solid evidence or argumentation are not worthy of a philosopher. T.
M. Robinson, however, shows that imagination is an indispensable and
valuable component of Plato's speculation on the nature and origin of the
cosmos. Against the allegorists, he shows that both the generation of the
cosmos described in the myth of the Timaeus and the reversal of the world's
movement (with the bizarre consequences that attend it) described in the
Introduction 17
Statesman are to be taken literally and seriously, not as playful allegories for
sorne more "rational" truth. Robinson then shows the importance of what he
calls Plato's "cosmological imaginativeness" by comparing it to the role that
imagination has · played in the thinking of modern cosmologists such as
Einstein, Heisenbeig, and Hawking. Despite the limitations imposed on Plato's
imagination by the lack of a concept of four-dimensionality, Robinson sees
Plato's imaginativeness as anticipating sorne ideas of the modero cosmologists,
such as a beginning of time (Big Bang theory), the tensility of space, and the
contraction of the universe (Oscillation theory). Robinson's article shows that
myth and imagination are for Plato an autonomous source of insights and as
such complement and enrich, rather than contradict, the purely rational
aigumentation that is also to be found in the dialogues.
According to James Arieti, a fundamental presupposition of most Platonic
scho!arship is that Plato intended the dialogues to present systematic,
cons1stent, and straightforward positions defended with sound, clear, and
unambiguous arguments. Since the inconsistent, ambiguous, and unsystematic
appeai:ance of the ~ialogues, however, does not confirm such a presupposition,
Pl~tomc scholarship has resorted to something analogous to the theory of
ep1cycles employed by the ancient astronomers to make the apparently erratic
~ovements ~f the heavenly bodies conform to the presupposition of perfectly
circular mot10n: the theory of "cycles" in Plato's life according to which the
dialogues can be divided into "early," "middle," and "late" groups. Arieti,
however, considers this theory as untenable as the theory of epicycles and
~uggests that we simply abandon the presupposition that has made it necessary,
J~st as astr~mome~ eventually abandoned the presupposition of perfectly
crrcular motion. In 1ts place we should recognize that the dialogues are meant
to_ be re~d as _dra.mas_ and that only when read in this way can they inspire us
w1th the1r philosophical point. Through observations on a number of Plato's
dialogues, Arieti demonstrates the results that a dramatic reading can have.
For example, only when read as dramas do the Phaedo and the Crito make the
philosophical points, respectively, that an aigument should be courageously
pursued despite the imminent threat of failure, and that one must act in
accord~ce with the best argu01ent available to one at the time, despite the
uncertamty and the momentous consequences. Such points do not constitute
any "theory" or "doctrine," but they do have the positive content of what
Arieti prefers to call inspirotion.
Like_Arieti, Gerald A. Press focuses on the dramatic aspect of the dialogues
by c~hng them "enactments." Press, however, is primarily concemed with
show1ng how understanding the dialogues as philosophical enactments allows
us to overcome many of the dichotomies that have divided traditional Platonic
scholars~p. Toe ~iew that the primary purpose of the dialogues is to
~omm~m_cate doctnnes, by either asserti~ them as disguised treatises or
1nstant1ating them as "mimes," gives rise to a number of unresolved either/or
18 Toe Third Way
questions. For example, which doctrines are Socrates' and which are Plato's?
Are Plato's doctrines to be found in the dialogues or outside them in
"unwritten teachings?" Did Plato's doctrines develop or remain the same
throughout the dialogues? Was Plato a dogmatist who held certain doctrines
to be indubitable or was he a skeptic who sought only to refute them and show
their limitations, as the New Academy believed? Instead of choosing one side
of these dichotomies, Press advocates a "Copernican revolution" that would
allow us to transcend them: we should see the dialogues not as saying
something, but as enactments that do something. But what exactly do the
dialogues do? According to Press, they create a worúi that the reader is
expected to enter into and experience intellectually, imaginatively, and
emotionally. What one gains through entering the world of a dialogue is not
a "theory" in the sense of a "doctrine," but a "theory" in the sense of a
"vision" : a certain way of seeing, experiencing, and doing things. Toe world
that the dialogues embody and enable us to envision is characterized by Press
as a simultaneous opposition and interpenetration of the real and the ideal, the
everchanging and the stable, the concrete and the universal. Press concludes
by indicating how sorne of the traditional dichotomies of Platonic interpretation
can be overcome by understanding the dialogues in this way: e.g., the point
of the dialogues is not to present doctrines, whether Socrates' or Plato's;
Plato's philosophy consists neither of doctrines found in the dialogues nor
"unwritten doctrines," but is instead that experience or vision with which the
dialogues provide us; the dialogues offer neither developing doctrines nor
unchanging doctrines, but instead a constant vision; Plato is neither a skeptic
nor a dogmatist, since philosophy for him is not a matter of either asserting
doctrines or refuting them. An important hermeneutical principie running
through Press's paper is that we should make our concept of Plato's
philosophy conform to tbe dialogues rather than make the dialogues conform
to it. Press's suggestion is that if we do so. we will see that the dialogues
evade the oppositions in terms of which many scholars have tried to
understand them.
Toe papers in the fourth part seek to avoid characterizing Plato's philosophy
as either "dogmatic" or "skeptical" by placing the emphasis on its dialectical
or dialogical character. Dialectic can be seen as a process that will never
termínate in final and universal conclusions or definitions, but that nevertheless
has positive universal content. lt is neither a dogmatic monologue nor purely
ad hominem refutation.
In my own paper, I develop an alternative to the skeptical and "doctrinal"
interpretations by arguing that neither can do justice to the unity of form and
content in Plato's dialogues. An explanation of this unity requires a third
interpretation that characterizes Plato's philosophy as a dialectic with the
following three characteristics: 1) it is reflexive in the sense that its content is
not objectifiable as a result separable from its method. My suggestion is that
Introduction 19
the dialectical method itself instantiates the truth it seeks and that to this extent
the method is itself the content. Furthermore, because the method of
philosop~ has this char~cter, philosophical knowledge is not of sorne purely
external, impersonal obJect and therefore is not "objective" in the way the
sciences are, but is instead inseparable from self-knowledge. Toe treatise
presents formal ai:guments in the defense of certain conclusions while in Iarge
~art. abstracti~ from the philosopher by whom, and the actual process of
mqwry by which these conclusions were reached; it is precisely the Iatter, on
the other hand, that the dialogue form strives to capture. This is why the
"reflexive" character of philosophy demands the dialogue form. 2) Dialectic
for Plato therefore also is and provides a form of practica/ knowledge
("knowledge how"). Accordingly, Plato's philosophy cannot be expressed in
the "_knowledge that" provided by the treatise, but reveals itself only in the
practtce so succesfully portrayed by the dialogues. 3) Finally, the insight with
which dialetic also provides us is nonpropositional. Such insight cannot be
commu~cated by a treatise, but it can be sparked by the suggestive silences
of the dialogues. These three mutually dependent characteristics, which this
paper can only briefly illustrate, map out a new direction in which Platonic
sc~olarship can head_in the attempt to understand both Plato's conception of
philosophy and the dialogue form that this conception demands.
Walter Watson asks why the skeptical and dogmatic interpretations have
been _so preval:nt in the history of Platonic scholarship. His answer is that
both mterpre~t1ons d~w our attention to opposed tendencies that are actually
to be found m the dialogues, but fail to see that these tendencies do not
necessaril~ contradi_ct each other. For example, the skeptic rightly points out
the ~llo~1ng: ali v1ews found in the dialogues are presented as the views of
spec1fic mterlocutors and notas the views of Plato oras "divine truth·" what
is s_aid i~ the _di~ogues is rooted in the concrete existential situatio~ they
d~p1ct; d1ale~ttc 1s ~ed to refute those who claim to have knowledge; each
d1alogu~ has 1ts own mtemal unity and is thus independent of the others. Toe
dogm_at1st, on the other hand, rightly asserts the following: the dialogues
co~tam truth that is "divine" in the sense of being independent of the human
beings who express it; the dialogues acknowledge universal "forms" that
transcend any specific_ concrete situation; dialectic aims to provide knowledge
through formally vahd arguments; the dialogues are to be understood in
relation to each other and thus as constituting sorne kind of "system. " Watson
argu~s that once these apparently contradictory positions are rid of the
?n:s1dedness and exaggeration that historically has characterized them, their
ms1gh~ can be reconciled in a third interpretation (the "dialogic") that
recogmzes the paradoxical but intelligible unity of opposites characteristic of
Pl~to's dialogues_: the dialogues are human and divine, particular and
umvers~, refutat1ve and affirmative, rhapsodic and systematic. Through a
companson of the Republic, the Statesman, and the Laws, Watson shows
20 Toe Third Way
concretely how the skeptical and dogmatic approaches, when properly
restricted, can complement both each other and the dialogical approach.
Victorino Tejera defends and explains his own version of a dialogical
approach through a discussion of the fbrmenides. Tejera insists that the
dialogues are always dialogical, even when one interlocutor (e.g., Socrates or
the Stranger) dominates. It is therefore wrong to see any of the dialogues as
a disguised monologue in which Platonic doctrines are being expounded. Toe
difference that reading the dialogues "dialogically" (and this means with
sensitivity to the dramatic context) can make is indicated by Tejera when he
points out that if the fbrmenides is read in terms of dramatic chronology as
the first of Plato's dialogues (since it is the dialogue in which the character
Socrates is youngest), then we must see Socrates in the PluJedo and the
Republic as already aware of the criticisms to which the Theory of Forms is
subjected there. As for these criticisms, Tejera sees them as logically refuting
the theory of forms but without denying the fonns their role as necessary
assumptions of ali intelligible discourse. What the young Socrates learns is to
become what he will become in other dialogues: a practitioner of elenchus
rather than an advocate of any "theory." Similarly, Tejera sees the difficult
second half as intended to exhibit for the young Socrates the limitations of
refutationism and deductivism. There is much matter in the hypotheses of this
second half for serious reflection, but this should not blind us to their satirical
and ironical character. As interpreted by Tejera, then, the fbrmenides portrays
the training of a philosopher who will become neither a dogmatist nor a
skeptic.
Toe paper that comprises the fifth and final part of the book attempts to find
a third way by relating the dialogues to Plato's so-called "unwritten
teachings." These teachings reported by Aristotle and others are normally
avoided like the plague by those who seek to remain faithful to the form of the
dialogues. As suggested by the brief history above, the unwritten teachings
have allowed the esotericists to preserve the systematic, doctrinal character of
Plato's philosopby by taking this philosopby out of the dialogues altogether,
safe from the contaminating influence of their irony, ambiguity, and skeptical
bent. However, it is possible that if the unwritten teachings are put back into
the dialogues, they may simply constitute a philosophical orientation that steers
clear of both skepticism and "doctrinalism."
In showing how the unwritten teachings are put to work in the third
hypothesis of the fbrmenides and the diaireses of the Statesman. Mitchell
Miller can reconcile the apparently doctrinal character of these teachings with
the ironic and indirect character of the dialogues. When we understand the
unwritten teachings in the context of their use in the dialogues, we recognize
that they do not constitute an apodictic system in which the fonns are deduced
from higher causes and their natures fully defined. Instead, because their role
is to guide us in the effort to understand the causal pawer of the forms in
Introduction 21
relation to sensible things, their purpose is to orient inquiry rather than resolve
it. This downward path from the forms to sensible objects that the unwritten
teachings orient cannot be substituted for the upward path to the forms
themselves: the absolute transcendence of the forms affirmed in the dialogues
cannot be denied. Once the role of the unwritten teachings is limited and
contextualized in this way, they can be reconciled with both the existential and
the ironic/indirect character of the dialogues. By indicating the order and
arrangement by which being is brought into becoming, they provide us with
a model to imitate in the fonnation of our character and actions, the kind of
formation at which the dialogues as "psychagogical rhetoric" aim. Because
their limitations make them subject to misunderstanding and misuse. the
unwritten teachings need the irony of the dialogues to awaken in the reader
self-understanding and awareness of his or her own limitations (among those
lacking such awareness and thus potentially misusi~ the teachings must be
included members of the Academy). In this way, Miller manages to avoid both
the Scylla of dogmatism that threatens the interpreter who abandons the
dialogues in favor of the unwritten teachings and the Charybdis of skepticism
that threatens the interpreter who abandons the unwritten teachings in favor of
the dialogues. ·
These different ways of inaugurating a new approach to Plato display the
wealth of possibilities available to those who are willing to break away from
the two general interpretations that to this day tend to be considered exclusive
alternatives. Since ali of the contributors acknowledge intellectual forebears
this approach is "new" only in the sense that it is still far from mainstream'.
Fu~e:°11or~, i~ is not the ambition of this collection finally to provide a
defimtlve third mterpretation to replace the traditional ones. Its aims are more
modest: to point out that there is a third way, to encourage its pursuit, and to
provide sorne assistance in the form of preliminary and diverse maps of where
it might lie.
This collection therefore raises more questions than it answers. Yet one can
say without exaggeration that what is most lacking in Platonic studies today are
good questions. Toe dominance of the "doctrinal" interpretation has prevented
the most important questions from even being asked. Because it has been
assumed that philosopby for Plato is completely "other" than rhetoric is like
modera professional philosopby, primarily written discourse, aim~ a~ the
establishment and defense of philosophical doctrines, is only accidentally
related to drama and myth-because these assumptions have not been doubted,
the urgency of the crucial and terribly difficult questions that the articles
collected here raise and address has for the most part not been felt. If this
collection succeeds in conveying that urgency, it has accomplished much. And
this is not simply a matter of understanding Plato. lf our assumptions about
what is and what is not "philosophical" prevent us from seeing the uniqueness
of Plato's thought, they may also blind us to a whole dimension of doing
22 Toe Third Way

philosophy. Our inability to understand Plato 's dialogues may stem from our
own philosophical impoverishment. In this case, it is ali the more important
that we fullow Plato on his path, however strange, rugged, and labyrinthine
it may seem, rather than dragging him along our own familiar, well-paved but
perhaps also too predictable thruways.

Philosophy and Speech:

Plato' s Dialogues Between

Oral and Written Discourse
 2

Reflections on the Orality and Literacy

of Plato's Dialogues

Jackson P. Hershbell

According to Diogenes Laertius (ill. 5), Plato "wrote poems, first

dithyrambs, afterwards lyric poems and tragedies," 1 and in 1895 R. Hirrel
commented on the "poetic" character of Plato's dialogues in his monumental
Der Dialog.2 Hirrel wrote of the "poetic talent" ("poetische Anlage") of
Plato's mind3 and claimed that sorne "anachronisms" in his dialogues, the

l. Diogenes' report is preceded by reference to Dicaearchus' On Lives (IlEp, {3iw11).

Since Diogenes drew not only from this but from several other works for his life of
Plato, it cannot be concluded that Dicaearchus was his source for this report.
Immediately after mentioning Plato's poetic activity, Diogenes cites On Lives by
Timotheus of Athens and Successions of Philosophers by Alexander (Polyhistor), and
recounts how Plato burned a tragedy he wrote after listening to Socrates talle before the
theatre of Dionysus. Por recent comment on Diogenes' life of Plato, see J. Mejer,
Diogenes Laertius and his Hellenistic Background, Hermes Einzelschriften 40
(Wiesbaden: Steiner, 1978), 36-38. Several charming epigrams attributed to Plato are
generally considered genuine. See, for example, W. K. C. Guthrie, A History of Greek
Philosophy, IV (Cambridge: Cambridge University Press, 1975), 13, n. 1 for
bibliography.
2. Rudolf Hi.rzel, Der Dialog, Ein literarhistorischer Versuch , Pt. I (Leipzig: S.
Hirzel, 1895), 179ff.
3. /bid., 176.
26 Toe Third Way
Symposium, for example, are reminiscent of the high spirits, even audaci~ of
Attic comedy. 4 Taking note of Diderot's attempt to transform the Phaedo mto
a tragic drama (on Socrates' death scene, the philosophe wrote, "What a
canvas for a poet" ["Quel canevas pour un poete"]), Hirzel commented at
length on the differences and similarities between Platonic dialogues and
drama. 5
In Plato's Progress (1966), G. Ryle reported how sorne of Plato's earlier
dialogues (which are more "dramatic" than his later ones) were performed "by
6
schoolbc,ys and students and, as we should expect, they act extremely well. "
Yet few scholars doubt that despite their similarities, ancient dialogue and
drama are different genres: a dramatic poet's concem with characters and
chorus and emphasis on plot and traditional tales (mythoi) are features, for
example, that distinguish drama from dialogue. Although both oft:n portray
conversation between participants, only dramas involve costunung, stage
7
settings and, in the case of ancient Athens, religious festivals.
But whatever the differences between dialogue and drama, Plato grew up
4. [bid., 182-83. See also R. G. Bury, The Symposium of Plato (Cambridge:
Cambridge University Press, 1909, repr. 1969), 66, who found the reference to
Mantinea's "division" (&oua.uµóc;) into several villages "audacious" (see 193a). This did
not occur until 385, well after Socrates' death.
5. Hirzel, Dialog, 202-203. See also a quite recent attempt to view Plato's dialogues
as individual plays by J. A. Arieti, lnterpreting Plato: The Dialogues as Drama (Savage,
Maryland: Rowman & Littlefield, 1991). Arieti does not mention Hirzel's work, ~ut his
study is interesting even though it draws little attention to drama and theatre m the
fifth-fourth centuries B.C.E. A.E. Taylor, Plato. The Man and His Work (New York:
Meridian Books, 1956), 2.1, objected to viewing Plato's works as dramas. Por him, it
was absurd to think of Plato as "'dramatizing' the sayings and doings ofthe living man
(Socrates) whom he revered above all others .. . . " C. Kahn, "Did Plato Write Socratic
Dialogues?" in Essayson the Philosophy ofSocrates, ed. H.H. Benson (Oxford: Oxford
University Press, 1992), 35-52, claims that the dialogues are works of dramatic art, and
although he does not doubt the historicity of Socrates presented in the dialogues, he
questions the historicity of the dialogues as reports of fifth-century philosophical
conversations. Kahn's essay appeared earlier in Classical Quarterly 31 (1981): 305-20.
6. GilbertRyle, Plato's Progress(Cambridge: Cambridge University Press, 1966), 23.
7. As Hirzel remarked, Dialog, 203-204, drama and dialogue in the ancient world are
perhaps related somewhat as Shakespeare's works, for example, may have in~irectly
influenced the dialogues of Berkeley. Hume. and Shaftesbury, "but drama and dialogue
are still essentially different" ("aber Drama und Dialog sind doch auch wesentilich
verschieden"). Much has been written on Greek drama and theater production, and a
recent work of B. Zimmermann, Greek Tragedy, trans. T. Marier (Baltimore: Johns
Hopkins University Press, 1991) provides a good bibliography and overview of the
genre. K. J. Dover's Aristophanic Comedy (Berkeley: University of California Press,
1972) is a good survey of ancient or "political" comedy.
Orality and Literacy of Plato's Dialogues 27
whe.n .drama (tragedy and comedy) flourished in Athens and, not surprisi~ly,
rellllmscences of plays by Euripides and Sophocles can be found in his
dialogues. For example, the argument between Zethus and Amphion in
Euripides' now-lost Antiope was perhaps dramatic inspiration for the debate
between Callicles and Socrates in the Gorgias on philosophy's value. 8 Further
according ~ Hirzel, the use of two persons in several Platonic dialogues, fo;
example, m the Grito, Euthyphro, and Phaedrus, shows drama's effect
("Wirkung") on Plato, for until Aeschylus introduced a third actor in the
Oresteia, only two actors assumed various personae in his plays. 9 In Plato 's
~gress, Ryle . also suggested drama's influence on Plato, referring to
D1ogenes Laertms who at m. 56 mentions Thrasylus' report that Plato
"published his tetralogies like those of the tragic poets. Thus, they contended
with the four plays at the Dionysia, the Lenaea, the Panathenaea, and the
festival of the Chytroi. " 1º Connection of these latter two festivals with
dramatic performances is possible but, as Ryle observed, it remains unclear
how 1:h,rasylus ~eant to assimilate Plato's publication of dialogues to that of
tragedies. Desp1te the apparent absunlity of Thrasylus' groupings of Platonic
dialogues into four, he did, according to Ryle, "have something to go on. " 11
For example, the Sophist, Statesman, and unwritten Philosopher were planned
12
in sequence. But contrary to Diogenes Laertius' report of Thrasylus-or
Ryle's comments on it-a genuine tetralogy of Plato's dialogues is nowhere to
8. See, for example, Hirzel, Dialog, 206, n. l. The evidence is tenuous.
9. /bid., 206-207. J. Andrieu, Le dialogue antique, Structure et présentation (Paris:
Les "Belles Lettres, " 1954) also discusses drama's influence on Plato, and considers a
transition "de la narration au drame;" 316-319. His examination of the dialogue form is
somewhat narrow, focusing on papyrological research, and contributes little to
understanding Plato's choice ofthis genre. "Platon et la perspective temporelle" receives
special attention in his study.
" 10. ~ee Ryle, P~t~•s Progress, 34, where he remarked that Thrasylus' groupings into
four, are often nd1culous. On the festivals themselves, see A . Picard-Cambridge The
Dramatic Festivals of Athens, 2nd ed. rev. by J. Gould and D. M. Lewis (Oxford:
Clarendon Press, 1973). Por Thrasylus' report, see Picard-Cambridge's assessment on
p. 56. ·
. 1 ~. R~le, Plato 's Progress, 34. Diogenes Laertius (III. 56-62) attributes an arrangement
m trilog1es to "sorne, including Aristophanes the grammarian" (61-62). Publication in
tetralogies he assigns to Thrasylus, whose identity remains somewhat of a puzzle. See
Guthrie, A History, lV, 39, n.2.
12. '!he question at_the opening ofthe Sophist is whether sophist, statesman (politicus)
and philosopher are different designations for the same person. The "Eleatic " considers
~e three distinct, and debate has arisen about whether Plato intended to devote a further
dialogue to the "philosopher." Guthrie, A History, V, 123, n. 1, believes that the
beginning of the Statesman, and Sophist 253E leave "no room for reasonable doubt that
P. planned the Philosopher."
28 Toe Third Way
be found. It is teinpting, of course, to regard the Apology, Crito, and Phaedo
as a trilogy and to add the Euthyphro to form a tetralogy (as Thrasylus does;
see D.L. ni. 58), but it is unlikely that they were composed or "published"
at the same time. More intriguing is Ryle's suggestion that sorne dialogues,
especially the Theaetetus, Sophist, and Statesma,n, forma trilogy, and that the
Olympic games, according to Plato's Hippias Minar (363C-364a and 368b ff.)
had competitions for prose and verse compositions. 13 From Hippias' rem~ks,
however, it is not wholly clear that formal literary contests were held or pnzes
given, and Ryle's suggestion that the dialogues were composed for Games
audiences suffers, as he himself notes, in regard to the "mammoth dialogues,"
the Republic and Laws. 14 In brief, there is not enough evidence to reach firm
conclusions about Plato's original audience or readers.
Toe suggestion that the Platonic dialogue had its origins in drama may also
overlook another important social phenomenon of the ancient Athenian world,
the symposium or drinking party, which, in addition to conversation or
dialogue, involved riddles, fables, and drinking songs or skolia. A famous
skolion often alluded to in Plato's dialogues involves the "four best things:"
beauty, health, strength, and wealth. 15 Moreover, besides Plato's Symposium
and that of Xenophon, which are early masterpieces of symposium literature,
dialogues with the same name were composed by later Ac~demics and
Peripatetics as frameworks for philosophical discussion. Even Ep1curus wrote
a Symposium, criticized by Athenaeus (186e) for lack of artistry. Further
evidence for the symposium's influence on Plato are the frequent references
to it in his works, and also to the sympotic dialogue or discourse (lagos
13. Ryle, Plato 's Progress, 33f. According to Ryle, not ali the dialogues were so
written. Toe Timaeus, for example, was written for members of the Academy, and the
"mammoth dialogues" were composed for private clubs in Athens. As for the evidence
in the Hippias Minor, it is perhaps worth noting that what Hippias read were epideictic
speeches, not dialogues. Given that the Hippias Minor is one of the earliest of Plato's
works and contains "manifest absurdities" (see Guthrie, A History, IV, 195f.), its
evidential value for determining Plato's own audience seems minimal.
14. Ryle, Pl.ato's Progress, 44ff. who speculates that "if there did exist in Athens the
postulated literary or literary-cum-social-cum-political clubs, one of th~m could_ in the
course, say, of a winter have gone through the whole of a mammoth dialogue like the
Republic or the Laws," 48. Despite such speculation, Ryle raises an important questi~n
often ignored by scholars of Plato: for whom were bis works composed? Who read bis
dialogues in the fourth century, and why?
15. M. Tecusan, Lagos Sympotikos: Patterns of the Irrational in Philosopbical
Drinking: Plato Outside the Symposium," in Sympotica. A Symposium on the Symposion,
ed. o. Murray (Oxford: Clarendon Press, 1990), 229-260, has a thorough discussion of
the sympotic institution in Plato's pbilosophy. On p. 240, for example, she discusses this
skolion.
Orality and Literacy of Plato's Dialogues 29
sympotikos). Tecusan's recent study of the lagos sympotikos makes clear how
important the symposium was in Plato's whole philosophy. 16 According to
Tecusan, elements from the symposium can be found in Plato 's
epistemological thought, and the Timaeus begins as a symposium in which
Critias, Timaeus, and Hennocrates pay back Socrates' hospitality. 17 Their
task is to feast Socrates in retum, and Plato sometimes described Socratic
conversation in terms of a "word-feast." At Lysis 2llcl0-dl, for example,
Ctesippus expresses a jealous wish to hear the debate in terms of a neglected
banqueter asking for bis due meal. Toe Laws opens with discussion on the
symposium and drunkenness, and this extends over two books of this work.
There seems to be little doubt that the symposia described by Plato involved
the kaloikagathoi, the Athenian gentlemen or nobility, and that the aims of the
symposium were connected with "education" (paideia), and at Protagoms
347c-e may be found an account of the "good" symposium as Plato envisioned
it. Unlike ordinary or simple men (the phauloi), Plato's friends, the
kaloikagathoi, knew how,to converse and debate topics on their own. They did
not need entertainers or musical perfonners; the entertainment and education
involved in the symposia of Athen's kaloikagathoi were largely intellectual and
pbilosopbical. 18 Finally, Plato's own Symposium may show how an "ideal"
drinking party was conducted by Athenian intelligentsia. Given the
symposium's importance in Athenian life, and its conversational nature, it
seems likely that this social institution, as well as drama, influenced Plato's
choice of the dialogue fonn.
In any case, it seems impossible to ascertain for whom Plato's dialogues
were composed, "games-audiences," Athenian literary-political clubs, and
well-to-do participants in symposia, notwithstanding. Speculation that Plato
wrote for members of the Academy or as a means of recruiting new members
is also unsupported by the evidence. Although it is not known when the
Academy was founded or when Plato composed bis works, sorne Socratic
16. See preceding note. Interestingly enough, the later Platonist, Plutarch of
Chaironeia (45-125 C.E.) wrote a number of sympotic dialogues, the Quaestiones
con vi vales.
17. Tecusan, "Logos Sympotikos," 243.
18. /bid., 257-258. Contrary to Tecusan, it is not clear that the first two books of the
~ws wbich describe sympotic customs, represent different patterns ofpaideia from those
m the P7:o~agoras (ibid., 259ff.). Wine may play a decisive part in Plato's thinking about
~ympos1a m the Laws, but Plato's awareness of human irrationality seems justas strong
m the Protagoras as in the Laws. In any case, Tecusan's views of these dialogues do
show that the symposia were often gatherings of intellectuals, and there is sorne
likelihood that Plato's written dialogues appealed to the same audiences. On the
kaloikagathoi, see the brief discussion in V. Ehrenberg, The People ofAristophanes: A
Sociology of O/d Attic Comedy (London: Methuen, 1974), 97-99.
30 The Third Way
dialogues and parts of the Republic may have been written before the Academy
was organized. 19 Moreover, if Plato's dialogues were written for the
Academy, what was their purpose? To teach its members to think
philosophically? To remember Socrates? To provide texts for the members
that, much like the quodlibeta of medieval thinkers, would then be discussed
and debated? To draw attention to the Academy and advertise its existence?
Perhaps ali of the above, but once the Academy was founded, there is sorne
evidence that Plato's instruction in it was oral: according to Aristotle, Plato
Iectured; the dialogues may have supplemented his oral teaching and provided
a basis for discussion and further inquiry. 20
Turning from these reflections to studies by the late E. A. Havelock, sorne
of which remain unpublished,21 it seems appropriate to reveal a somewhat
19. Much has been written about the founding of Plato's Academy and the dating of
his works. Suffice it to say that the Academy was established sometime after Plato's
retum from his first Sicilian joumey (about 388-7). See C. W. Müller, "Die
hellenistische Akademie" in Kleines Worterbuchdes Hellenismus. ed. H. H. Schmitt and
E. Vogt (Wiesbaden: O. Harrassowitz, 1988), 23ff. On the chrooology of the dialogues,
see, for example, Guthrie, A History , IV, 41-54. Oo the whole. attempts to coordioate
the compositioo of the dialogues with the fouoding of the Academy are a waste of time.
20. See, for example, A. E. Taylor, Plato: The Man and His Work, 6f. who refers to
Plato's lectures in the Academy as beiog delivered "without a manuscript." Taylor
perhaps thought of Aristoxeous' report that Aristotle was fond of telling the story of
Plato's lecture "On the Good," at which maoy left in disgust since his talk was ali about
astrooomy and mathematics. Por the full text, see l. Düring, Aristotle in the Ancient
Biographical Tradition (Goteborg aod Stockholm: Almquist & Wiksell, 1957), aod K.
Gaiser, "Plato's Enigmatic Lecture 'Oo the Good'. " Phronesis 25 (1980): 5-37.
21. See, for example, E. A. Havelock's Preface to Plato (Cambridge, Mass: The
Belkoap Press of Harvard University Presses, 1963), The Literate Revolution in Greece
and lts Cultural Consequences (Princeton: Princeton University Press, 1982), aod The
Muse Leams to Write (New Haven: Yale University Press, 1986). lo his unfinished aod
unpublished work, The Preplatonic Thinkers of Greece: A Revisionist History, Havelock
discusses the "mimes" of Plato. Toe term "mime" is reminisceot. of course, of the
mimes of Sophroo of Syracuse, who was admired by Plato (see Douris of Samos in P.
Jacoby, Fragmente der griechischen Historiker(Berlin and Leiden: Weidmann aod E. J.
Brill, 1923- ) 76, F 72. An attempt has been made to recoostruct one of Sophron's
mimes from Plato's Gorgias 493d-494aff. , but this was rightly rejected by Hirzel,
Dialog, I, 24, o.2. The fullest study of Sophron's influence on Plato is J. M. S.
McDonald, Character-Portraiture in Epicharmus, Sophron, and Plato (Sewaoee,
Tennessee: Toe University Press, 1931). McDonald's doctoral thesis cootaios a good
assessment of the ancient evidence about Sophron's possible influence on Plato. The
"mime" itself was coosidered a form of drama (see McDonald, 134), aod preseoted
scenes from everyday life "with situatioos aod characters of a sort that the reader or
spectator of that day recognized. " Besides Douris, other ancient authors reported oo
Plato's admiration for Sophron's mimes (see 129ff.) and, Iike Sophron aod Epicharmus,
Orality and Literacy of Plato's Dialogues 31
pers~~al bias: wh~tever. the genesis of Plato's dialogues, they reflect a
trans1t10n from oral1ty to literacy. His dialogues and even those of Iater authors
such _as Plutarch are perhaps best understood as having their origin in what
r:mamed an oral-aural culture until book printing in the fifteenth century. Ryle
himself argued that Plato and others composed dialogues for audiences Iater
making them available to readers, and at Phaedrus 274b-277a, So,crates
deplores the use of books except as aide-mémoires. 22 Making do with straws
Ryle went on to argue against the hypothesis that Plato wrote fo;
"book-rea~ers." _Ryl_e took no account of Havelock's 1963 Preface to Plato
(the work 1s not m his acknowledgements); since then, other pioneering works
of ~avelo~k have appeared. Given these works of Havelock, W. V. Harris'
¿nc~ent Lztero.cy23 and K. Robb's Litero.cy and Riideia in Ancient Greece, it
1s dtfficult to understand why sorne scholars persist in denying that ancient
H~Ilas :as Iargely an "oral culture." O. Murray's recent essay on "Sympotic
Htsto~ also acknowledges Havelock's work and its importance for drawing
~ttent1on ~ v~rbal perform:mce ~d understanding the orality of early Greek
hterature. D1sputed questions wdl probably remain: for example, when or
~ow the alphabet was introduced into HeIIas; estimates of the extent of
ltte~cy; what lit~racy meaos (or meant in the ancient world); or the influence
of Itteracy on anctent thought and behavior. These are ali difficult matters. But
f?
ther~ see~ be no reason to doubt that Plato, searching for a suitable meaos
of d1sse~natu~ his own "ideas" or beliefs, saw purpose in using the dialogue,
an. ostenstbly o~al phenomenon, for what was to be a reader's corpus of
wntten prose. Gtven the plausible tradition reponed by Diogenes Laertius that
Plato started his literary career as a composer of poems and plays, it is not
Plato to~k great int~res: in cha~acter portrayal in his dialogues. McDonald also rejects
Schuster s attempt m Heraklit uod Sophroo in platonischeo Citateo," Rheinisches
Mus~um ~9 (1874): 590-632, to regard Plato's Gorgias 493dff. as ooe of Sophroo's
P~SS!ble influences on Plato as not very useful for understaoding the origios of the
dialogue form.
22. Ryle, Plato's Progress, 22ff.
23. W. V. Harris, Ancient Literacy (Cambridge, Mass: Harvard University Press
1989). On p. 86, Harris_ remarks that the works of Aoaxagoras, Socrates, aod Protagora~
we_~ read al_oud. Hams also has sorne important observatioos oo Plato's mistrust of
wn_tmg, a IOlstrust ~hared by his cootemporary rival, Isocrates. Both seemed to have
believed that_the wntte~ word cannot clarify or defend itself. See pp. 91-92 aod 30-31.
See also Kevm Robb, Llleracy ~nd Padeia in A.ncient Greece (Oxford: Oxford University
Press, 1~4), 159-2~1, ~oran unportant assessmeot of literacy's influence on Plato. His
speculattoos oo orality, literacy, and the dialogue form (233ft) are especially worthwhile
24IM
·. n un:ay, Sympollca, · ·
8-9. Murray specifically mentioos Havelock's role in·
focus_mg attenlloo on performance as a crucial factor in explaining the nature aod
functtoo of early Greek poetry.
32 Toe Third Way
surprisi~ that he carne to write a "prose drama of ideas," ~ quote
Havelock. 25 But Plato's doubts about the written word as a vehicle for
philosophical expression seem clear not only from the ~fte,n-m~ntion:d
Pha,edrus passages,26 but also from his Seventh L~tter ~ D1on s frie~ds m
Sicily, in which he suggests that he never committed his actual behefs to
writing (see 34lcff.).27 At least one problem arises from Plato's ~emarks, and
that concerns whether the spoken word is really more effect1ve than the
written. If Aristotle's report about Plato's lecture "On the Good" is
trustworthy, his audience tumed away in disgust or disappointment, an
28
indication that Plato was a better writer than speaker. In any case, the
effectiveness of both oral and written communications depends on a
well-informed and educated audience, but almost nothing is known about
Plato's audience or their preparation for understanding his dialogues.
Plato lived at a time when he could expect circulation of his works among
well-to-do readers and, as Havelock noted, it was an ingenuous method to
construct "characters engaged in dialogue, as in a stage play, and to give them
names of well-known thinkers designated to serve his own philosophical
purposes. " 29 Thus, for example, Plato used Parmenides, whom he revered,
25. Havelock, The Preplatonic Thinkers, 24.
26. Por exarnple, at 275e Socrates says: "and every discourse (t-ó,yo~) once written,
is bandied about, both arnong those who understand it, and those who take no interest
in it. And it knows neither to whorn to speak nor not to speak. When ill-treated ... it
has no power to defend or help itself." See also the critique of writing at 275 A ff. as
producing "forgetfulness," and at 277eff. similar doubts are again expressed about the
written word.
27. Whether the Seventh Letter is genuine or not and its authenticity, like those of the
other twelve, have been endlessly discussed. On the authenticity of the Seventh Letter,
see Guthrie, A History V, 402, and T. A. Szlezák, Platon und die Schriftlichkeit (Berlin:
Walter der Gruyter,, 1985), 388-390, who, with sorne resignation, rernarks that "the
burden of proof lies with the opponents of its genuineness" ("die Beweislast liegt bei den
Gegnem der Echtheit"). In a later study, "Gespriiche unter Ungleichen. Zur Struktur und
Zielsetzung der platonischen Dialoge," Antike und Abendland 34 (1988): 99-116, Szlezák
rightly notes that what is said in the Seventh Letter agrees with Plato's "Schriftkritik" in
the dialogues. See p. 102.
28. See n. 20 above. According to Diogenes Laertius (111. 37) citing Pavorinus, Plato
also read a dialogue "On the Soul" at which only Aristotle stayed to the end. Toe rest
of the audience left.
29. In his Preplatonic Thinkers, 3lff. As Havelock noted, "this free rnanipulation
within limits" is found also arnong the tragic poets who presented personalities of past
history, e.g., Xerxes in Aeschylus' Persians, and the comic poets who "went so far as
to allow the adaptation of Athenian historical personalities, living and dead, to the
purposes of drarnatic invention." B. Gentili, Poetry and lts Public in Ancient Greece.
From Homer to the Fifth Century, trans. A. Thornas Cole (Baltimore: Johns Hopkins
Orality and Literacy of Plato's Dialogues 33
as. a persona in the dialogue named after him. Sometimes Plato's purpose in
us1ng_a c~ara_cter is polemical , and in the dialogue named after Protagoras, the
Sophist 1s g1ven a defensive role in an argument supporting Plato's own
philosophical interests. Similarly, Gorgias is a given brief identity as he
speaks, but is then connected to an argument by a possibly fictional character
(Callicles) in order that the argument, Plato's own literary creation can be
refuted in Platonic terms. 30 '
According to Havelock, the "management of characters" in Plato's works
presents a forro of composition that is neither fact nor fiction, but a mode of
memory . s~ited to the interests of the remembrancer, a mode "highly
chara tenst1c of the w_ay recollected language behaves in an oral culture, put
7
to a hterate use at a time when orality was yielding to Iiteracy. " 31 Ryle also
noted _that Plato's dialogues, although obtainable by readers, were composed
"for hsteners," and that Plato carne to abandon the eristic dialogue with its
practice of the Socratic method, because he had no more "moot records or
memories to dramatize. " 32 When Socrates, Plato's inspiration for elenctic
arguments, was executed and his personal participation in dialectical debates
stopped, the eristic dialogue disappeared. Ryle's views seem to be not very
different from those of Havelock.
~e hypotheses of Havelock and Ryle may help in understanding Plato's use
of dialogue, but they do not exhaust comprehension of what, in essence, was
Pl~to's literary creation or the superiority of bis dialogues over those of other
wnters. For students of Socrates- e.g., Aeschines, Antisthenes and
Arist~ppus-did compose dialogues, and according to Diogenes Laerti~s (II.
62), 1t was at Megara that Aeschines read aloud one of the first Socratic
University Press, 1988), 155-176, has sorne interesting remarks on "intellectual activity
and ~~cioecono~c situation," especially in regard to the book trade. According to
Gentili, Thucyd1des, Plato, Socrates, and Aristotle are the "vanguard of the new book
culture in ancient Greece" (see p. 169). But how expensive were books, and who could
afford them? Here Plato may provide sorne evidence. At Apology 26D, Socrates refers
to the purchase of Anaxagoras • "book" or papyrus rolls (ft,{3>úcx.) for a drachma.
C~Iculations of daily wages in early fourth century Athens are düficult, but in G. s.
Kirk, J. E. Raven, and M. Schofield, The Presocratic Philosophers, 2nd ed. (New York:
Cambridge University Press, 1983), 357, there is a calculation by A. H. M. Jones ofthe
arnount of wages paid to a slave copyist, and the time required for copying Anaxagoras •
work. A bo~k_sold for a drachma would amount almost toan ordinary man's daily wage.
In _the Eleus!Il1an accounts, according to Jones, skilled Iaborers, e.g. , stone masons, are
pa1d two to two anda half drachmae a day. In sum, a "book" such as Anaxagoras' was
not cheap, and Plato's longer dialogues would have been expensive.
30. !bid., 31-32. Taylor's assessment, Plato, 105, seems more cautious: "Callicles of
Acharnae, of whom we only know what Plato has thought fit to tell us."
31. Havelock, Preplatonic Thinkers, 38-39.
32. Ryle, Plato's Progress, 204.
34 Toe Third Way
dialogues. Ironically, it was Socrates, who himself apparently wrote nothing
and believed that truth can be attained only by verbal cross-examination (the
elenchus), who inspired the beginnings of written dialogue. Probably part of
the purpose of these early dialogues was to preserve a memory of Socrates.
Mention of the Socratic elenchus is a reminder that this word suggests the
legal terminology of ancient Athens, 33 and if characters in Aristophanes'
plays, such as The Birds, can be trusted, it was a society plagued by lawsuits.
Plato's own Apology claims to be Socrates' defense against the charges of
Anytus and Meletus before an Athenian jury and, given the oral exchange
between prosecutor and defendant (speechwriters or Logogrophoi aside),
another likely ímpetus for Plato's use of the dialogue form was the elenchus
of the Athenian law-courts. As D. M. MacDowell observes, the magistrate's
preliminary inquiry, the anachrisis, for example, involved interrogation in
which the magistrate put questions to the claimants or disputants. They could
also question one another, thus giving each a "clearer idea of what the other
was alleging and what were the exact points in dispute. " 34 In the dialogues,
Socrates often interrogates his respondents, and Plato was probably familiar
with the practices of the Athenian lawcourts. Toe production of dialogues
must, of course, be distinguished from that of dialectical arguments. Toe two
are sometimes confused, and dialectic in turn needs to be distinguished from
eristic. 35 Certainly the latter term has been used in conjunction with the
Socratic method, the elenchus or cross-examination, and it too was an
essentially oral phenomenon. But Diogenes Laertius reports at m 24 that Plato
coined for "eristic" the noun "dialectic," derived from the word óto:AÉ-yeCT0m,
33 . Toe term e>-.e-yxoc; appears in the works of sorne of the Attic orators, e.g., lsaeus,
8.42. Demosthenes, 22.22ff. distinguishes between aiTía and e>-.e-yxoc; as,
respectively, "accusation" and producing evidence of the matter asserted. "E>-.e-yxoc;
seems to be characterized as a quasilegal concept involving a certain procedural
sufficiency (the production of evidence transforms what would otherwise be mere insult
and abuse into a charge requiring the defendant to respond on the merits). My thanks to
my colleague George Sheets for calling attention to the Demosthenes passage. A very
good study of the Socratic elenchu.s remains that of R. Robinson, Plato 's F.arlier
Dialectic, 2nd ed. (Oxford: Clarendon Press, 1953) 7-19. See also Ryle's discussion of
elenchu.s and the eristic dialogue in Plato's Progress, 193-215. His frequent references
to "moots" (imaginary cases argued by law students asan exercise) are a reminder of the
legal connotations of the elenchu.s. Although a cross-examination in an Athenian
dikasterion or a court of justice is not a written philosophical dialogue, it is not amiss to
see sorne connection.
34. D. M. MacDowell, The Law in ClassicalAthens (lthaca: Comell University Press,
1978), 241.
35. They are confused by Diogenes Laertius, for example, in his life of Plato (111. 48
and 79). On the contrast between eristic and dialectic, see Ryle, Plato 's Progress, 126-
129. His whole discussion of dialectic on pp. 102-145 is worth reading.
Orality and Literacy of Plato's Dialogues 35
meaning. "to discuss, debate," specifically "to debate by question and answer,"
once aga.in, an oral phenomenon. So at Cra.tylus 390c Socrates asks, "and the
perso_n who knows h?~ to ~sk and answer questions you call the dialectician?"
Desp1te the ~ofty pos1t1on g1ven to the results of dialectical thinking in Republic
Vil, the Phzlebus, or Sophist, dialectic remains connected with spoken or
verbal exch~nges, and _as Plato wrote in the Seventh Letter (341C-45): "there
does_ not exist,_ nor wll_l there ever exist, any treatise (CTÚ"f'ypo:µµo:) of mine
deal!ng therew1th. For 1t does not at ali admit of verbal expression like other
stud1es, _but as a ~esu!t _of continued application to the subject itself and
commu~on _therew1th, 1t 1s brought to birth in the soul on a sudden ... "
. At this ~01~t, a brief digression about ancient historiography seems relevant.
Ltke Pl?to s dialogues, ancient historical works such as those of Herodotus and
Thucydides are much closer to poetry and rhetoric than is sometimes
36
acknowledged. For example, the speeches in Thucydides' history are
p~rhaps bes_t understood_ not as reports of what was actually said, but as the
ldnds of things that fillg~t have been said. Similarly, Xenophon, a close
contemporary of Plato, g1ves assurance in his Cyropaideia (I. 1, 6) that he
~ants to r~port_ what is know~ about Cyrus. Yet what one finds in this work
~s not an histoncal account of the Persian ldng, but Xenophon's portrait of an
ideal monarch. ~ _Cícero _remarked: "that Cyrus sketched by Xenophon was
not for the cred1b1h~ of h1story, but for the ideal of just rule" ("Cyrus ille a
Xenophonte non a~ ~storiae fide~ scriptus, sed ad effigiem iusti imperii, " Ad.
Q. fr. 1, 1'. 2~). S1ID1larly, there 1s no reason to believe that Plato's dialogues
ar~ transcnpt1ons of the Socratic elenchus. He was not a courtreporter but a
thinker. very much influenced by the world of Athens with its litigation
sympos1a, and dramas. '
Plato's dialogues continue to be read, and they are often difficult to interpret
partly because he seems to. reach no final conclusions in them, but Jeaves his
readers and students to think things out for themselves. As v. Tejera and
37
others have observ~d, Plato 's use of the dialogue form may suit his
reluctance t? dogmat1ze or systematize in any obvious fashion. Quite recently
A. T. Cole_m The Origins ofRhetoric in Ancient Greece, has emphasized h~
the d.ramattc form of Plato's dialogues excludes any single voice of authority
even that of Socrates. He is not, according to Cole, a completely reliabl~
spo~sm~ 1?r ~e author, a_nd what emerges is "the author's ability to replace
anything m it w1th something better should need arise-in the form of new
36. :º~ a rece~t good discussion of these and other matters, see A . J. Woodman,
Rhetonc m Class1cal Historiography: Four Studies (Portland Oregon· Areopagitica
Press, 1988). ' ·
37. V. Teje~a, Plato 's Dialogues One by One: A Structural /nterpretation (New York·
lrvmgton Publishers, 1984), 4-5 and 75. ·
 Toe Third Way
36
evidence, new insights, new arguments not envisioned when the original piec,e
was produced. " 38 It is also important to observe that not only do ~latos
dialogues avoid reacbing "final" conclusions, but th~ also suggest to~1cs for
future discussion and further hypotheses to be exannned, none of which ~e
fully pursued within a given dialogue. Moreover, the vi~s ~resented m
various dialogues are not always consistent: the human psyche 1s discusse~, for
example, in the Phaedo and Republic and yet Socrates p~esents ~eellllngly
incompatible views of it.39 In brief, unlike a drama (or tnal), which has a
beginning, middle, and end, Plato's dialogues are often openended and allow
the reader (or listener) to reflect further about the topic(s) discussed, perhaps
connect this discussion with other works in the corpus, and come t~ the
realization that the written dialogues need "help." In contrast to other wnters,
philosophers may realize that their works need discussion in order to be
understood and, as T. Szlezák observed, the dialogues are the~selves
examples of discussion (written) that need oral or verbal ass1stance
("mündliche Hilfe"),40 and are thus models for _what Pl~to w?ul~ consider ~e
"help" necessary for following and understanding the mvest1gat1ons begun m
the dialogues. Along lines similar to those of Cole and Szlezá~, M. Fre~e has
recently argued for the inextricability of pbilosopbic teaching and hterary
form.41 For Frede, Plato thought that the only way to present arguments (and
arguments do bave a prominence in Plato's conception of pbilosophy ~s found
in the dialogues) was in the form of fictional dialogue. Althou~h the d1~ogu:s
are supposed to teach a pbilosopbical lesson, they "are not p1eces of d1dact~c
42
dialectic with Plato appearing in the guise of the questioner. " Much of ID:eu
didactic quality consists not in what is argued or said, but in making one think
about the arguments presented.
43
•
My preceding reflections are not meant to be an exhaus~1ve a~count_ of the
origins of the Platonic dialogue, but given that Plato hved m a time of
transition from orality to literacy, and that the dramas, lawcourts, and
symposia of Athens influenced bis choice of the dialogue form, were not more
38. The Origins of Rhetoric in Ancient Greece (Baltimore: Johns Hopkins University
Press, 1991), 124-25.
39. Various scholars such as R. S. Bluck, P/ato's Phaedo (London: Routledge & Pa~l,
1955), 3ff., have noticed that the soul is regarded as having three p~rts in the Rep_ubllc:
intellect or reason, the passionate or spirited part, and the appet1tes, whereas m the
Phaedo Socrates speaks of the soul as if it were intellect or reason alone.
40. "Gesprache unter Ungleichen," 105. . .
41. M. Frede, "Plato's Arguments and the Dialogue Fonn," in Oxford Stud1es m
Ancient Philosophy, eds. J.C. Klagge and N .O. Srnith (Oxford: Clarendon Press, 1992),
201-219.
42. /bid., 219.
43. /bid. , 216.
Orality and Literacy of Plato's Dialogues 37
pbilosopbical or theoretical considerations involved in bis use of this Iiterary
genre? Studies by Cole, Frede, Szlezák, and others strongly suggest that there
were, and although the Phaedrus is an important source for Plato's views on
literacy and orality, many of bis other works, for example, the Meno and Book
X of the Lo,ws, must ~lso be c?~idered. ~ Study of these works, including
the Seventh Letter (the mauthent1c1ty of which remains unproven) and perusal
of an ever-burgeoning secondary literature on Plato, suggest that bis views on
the spo~en an~ written ~ord cann?t be state~ in definitive or summary fasbion.
Szlezák s stud1es espec1ally call mto quest1on what he considers the modern
theo~ of the dialogue, a theory traceable back at least to Friedrich
Schle1ermacher's first printed translation of Plato (1804). According to
Szlez~k, the "minimal version" (Minima,lform) of this "modern theory" is that
the dial_ogues have the function of either reporting actual, living discussion
(le?endig~s Gesprilch), or of condemning learned, systematic presentation of
philosophical thought (the uú·y-ypa.µµ<:x, or scholarly treatise) and thus Plato
chose the dialogue forro. 45 To be sure, Plato condemns the uú~-ypa.µµa. in the
Seventh Letter and Phaedrus, but contrary to advocates of the "modern
theory,~• Szlezá~ obs:rv:s that Plato's own dialogues were often designated as
uv-y-ypa.µµa.m m ant1qmty, and even in Plato's own usage uú-y-ypa.µµa. did not
mean ~ learned, systematic treatise, but anything written. So in condemning
uv-y-ypa.µµa.,a., Plato seems to reject written works as a means of presenting
46
truth. _Furtherroo~e, if Szlezák and others are correct, Plato nowhere wrote
d°':"~ bis most senous thoughts, and bis negative views on the usefulness of
wnttng may explain bis choice of the dialogue forro: written works need
"h~lp" from verbal disc~sion, oral debate, and a lively exchange of ideas.
Wntten wor~ cannot easlly be defended, because the views presented in them
are not readlly retractable or revisable, and a reader may not be stimulated to
further ~efl~ction, especially that brought about by skillful cross-e,camination
or quest1omng. 47
Accor~ing to _Szlezák, Plato's written works exist, as it were, in graduated
fo~: wntten dialogues composed to aid bis readers in thinking more about
senous ma~ters, and so composed that they often suggest or refer to one
another unul the read:r finally casts them aside (somewhat like Wittgenstein's
ladder), and engages m personal reflection on what has been read. In brief,
44. See Platon und die Schriftlichkeit for discussions of these and other dialogues
relevant to the whole matter of Plato on literacy. ·
45 . See especially ibid. , 331-375, where Szlezák presents the main assumptions of the
"modero theory of the dialogue fonn."
46. /bid., ~76-385, for a full discussion of aúnpcxµµcx as used by Plato, and in
reference to his own works by other ancient authors such as Proclus and Toernistius.
47. Many a p~rson has, I suspect, fallen asleep while reading a book or listening to
a dull lecture. Dialogue can be far more stimulating.
38 Toe Third Way
Plato's dialogues may have helped bis friends and students to understand ~s
oral teachings, and to lead tbem to think pbilosopbically. In a Platomc
dialogue, a reader may ultimately be reminded of tbe power of tbe spoken
word and tbat no written work can ever adequately present the trutb.
Pbilo~ophy for Plato was a nomen actionis, and so tbe dialogues seem to be
full of unresolved problems, bints at, or suggestions of the trutb. They are
48
protreptic or "parainetic" compositions, as K. Gaiser well a~ed, and tb.ey
give advice, pbilosopbical direction, or exhortation to tbeir readers wbile
guiding tbem to tbe pbilosopbical or tbeoretical life. They may even have b:en
composed as a forro of "self-communication," like tbe works of Goetb.e, wbich
he undertook "for my own sake and for myself' ("um mein selbst willen und
für mich selbst"). 49
That Plato's dialogues, reminiscent of Socrates' treatment of a slave-boy
(Meno), were designed to "remind" a reader of what was already known, to
actualize latent knowledge, is also possible. 50 At Republic 523b, Socrates
remarks on how sorne sensory reports "invite tbe intellect (ri¡v vo~o1v) to
reflection " and a similar purpose is discernible in Plato's dialogues: they are
a forro of áv&µvr¡cnr;, recalling or recollecting Socrates, tbe teac1:rings Plato
has assigned to bim, and Plato's own philosopbical ~eliefs. ~ebin~ Plato's
choice of tbe dialogue forro, tbere probably lurk all kinds of mtenuons, and
it would be a mistake to focus on only one. As Szlezák perhaps ironically
reminded bis own readers, 51 Plato was not anxious to "rush into print" - he
was not under duress to publish or perish, and so Plato could easily withhold
bis most important or valuable doctrines from tbose not initiated in ~s
pbilosophical way of life. But whatever views one may have of tbe Plato~c
dialogue, W. K. C. Guthrie was on target when he suggeste.d tbat a Plato':11c
dialogue is "no ordinary written work (e1ú-y-ypaµµa) purport1ng to summanze
final conclusions. " 52
In conclusion Plato's choice of tbe dialogue forro was inf!uenced by tbe
drama, symposi~m. and Socratic elenchus, well known phenomena of ancient
Atbenian life. Toe dialogue forro probably took its origin from these.
Moreover, Plato's dialogues reflect a transition from orality to literacy insofar
48. See K. Gaiser, Protreptik und Pariinese bei Platon, Tübinger Beitriige 40
(Stuttgart: Koblhammer, 1959).
49. From his letter to Chr. G. Korner on 26 November 1812, quoted by F. Jürss,
"Platon und die Schriftlichkeit," Philologus 135 (1991): 168. Jürss's article is an
excellent survey of sorne of the problems connected with Plato and literacy.
50. See Jürss, "Platon," 168-169, K. Gaiser, "Platons Menon und die Akademie,"
Archiv far Geschichte der Philosophie 46 (1964): 241 ff., and Szlezák, Platon und die
Schriftlichkeit, 179-190.
51. Szlezák, Platon und die Schriftlichkeit, l.
52. See Guthrie, A History IV , 65 .
Orality and Literacy of Plato's Dialogues 39
as he believed that the written word was not wholly adequate for pbilosopbical
discussion. This requires leisure or <1)(.oM , tbat leisure tbat was available to
Plato's aristocratic and well educated friends. Toe dialogue forro has a long
bistory after Plato, and its beginnings are easily traceable to bim and to tbe
essentially oral culture in wbich he lived. To "publish a work" was to read it
aloud before an audience, and if scholars such as Havelock, I. Düring, and E
Kenyon are right, it is especially witb Aristotle that the Hellenic world passed
from oral instruction to tbe habit of reading. 53 And so Aristotle is the
"reader" (áva-yvwe1TT]r;) of tbe Academy, but even in tbe Hellenistic world
"reading" seems to have been "reading aloud," not silent reading. Toe habi~
of orality were (and are) slow in passi~ away, perhaps because every
tboughtful reader (and writer) realizes tbat tbe written letter "kills" wbile tbe
54
spirit ~akes .ªl~ve. Th~ genius of Plato's dialogues may, in the last analysis,
be tbe1r subhnunal renunder that neitber the written nor tbe spoken word is a
wholly adequate guide to tbe trutb. Finally, as anyone who has atténded tbe
theater knows, an often asked question is: "what do you tbink tbe play
mean!?" Plato's dialogues, like tbe dramas of ancient Athens, raise similar
quest1ons. Whetber Plato had an esoteric or unwritten pbilosophy becomes
somewhat irrelevant if M-yor; in its written and spoken forros is inadequate to
convey the trutb.
53. See l. Düring, Aristoteles. Darstellung und Jnterpretation seines Denkens
(Heidelberg: Winter, 1966), 8 and F. Kenyon, Books and Readers in Ancient Greece and
Ro1:1e, 2d ed. (Oxford: Clarendon Press, 1951), 25: "it is not too much to say that with
Aristotle the Greek world passed from oral instruction to the habit of reading. "
54. 11 Corinthians 3.6.
 3

Plato's Audiences, or How Plato Replies to the

Fifth-Century Intellectual Mistrust of Letters

Elinor J. M. West

"What does it matter who is speaking?" Beckett

"It is Plato's voice we must listen to, not our own disguised as Plato's." Tigerstedt

To this day, in the city of Leuven-known far and wide, inside and outside
of Belgium, for its ancient and distinguished university-is the cast-iron statue
of a student. He stands on a pedestal, with open book in hand. At this he
stares, through holes on either side of his nose, reading. In his other hand is
a glass, held just above a third opening in the top of his head. From this glass
water is continuously poured into that head, the excess overflowing and
splashi~ down ali over the student, finally to spill into a lruge pool around the
bottom of his pedestal. Whether this fountain stood in the city square before
or after Leuven had also become famous for brewing beer I do not know, but
when I first saw it there, the statue was dressed in a very wet tuxedo jacket.
It was shirtless but wore a small, black bow tie-a mascot obviously beloved
by the current body of students, especially, it might be supposed, around the
time of their final examinations and farewell parties.
Perhaps this same statue (so humorously condensing into one visual idea the
consumption of letters and the consumption of beer) would also have appealed
to that first generation of students who flooded into Leuven when the
university was founded in 1431, perhaps before Leuven had become famous
42 Toe Third Way
for beer. If so, these students may bave shared the attitude toward
acculturation characteristic of many of today's large universities: namely,
education consists in a process of pouring the words of authorities, famous as
men of letters, into the hollow interior of readers who no longer hear the
sounds of great ideas, much less question what is being disseminated. Whether
this same attitude could be as appropriately applied if one were to set out to
describe what was going on in fifth-century Athens, where Socrates spent d~y
after day, not writing, but conversing with young men on the streets of bis
city, is worth asking. Toen one's attention wil~ at once b~ drawn to. ~e place
letters occupy in the lives of those brought up m oral or bterate tra~t!ons and
to how one is to characterize such differences. Yet before detenmmng that,
one must first recognize a prejudice often inculcated together with letters and
expressed without shame by the British historian Edward Gibbon when he
writes:
... the use of letters is the principal circumstance that distinguishes a civilised people
1
from a herd of savages incapable of knowledge or reflection.
That this attitude should still haunt our English has recently been recognized
by two ancient bistorians. Toe first begins a book on ancient literacy by calling
2
attention to the synonymity of the adjectives "uncultured" and "illiterate. "
Toe second, quoting Gibbon, recognizes the need for challenging what Gibbon
asserts, before then setting out to characterize the culture of fifth-century
Athens as a mixture of oral and literate elements. 3 Curiously enough, what
each then describes, for the most part independently the other, are conflicting
attitudes toward writing on the part of native Athenians, if not on that of
literate foreigners who had come to this ancient Mecca to revolutionize what
was believed and taught. 4
Socrates, of course, knew each of these teachers. Indeed, he is found talking
with Protagoras, Gorgias, Prodicus, and their students in Plato's dialogues.
We know that Socrates could read. We also know that, unlike the Sophists, be
l. Edward Gibbon, Decline ami Fall of the Roman Empire, ed. Bury, vol. 1 (New
York: Heritage Press, 1946), 190.
2. W. Harris, Ancient Literacy (Cambridge: Harvard University Press, 1989), 4. Also
below n. 11.
3. R. Thomas, Oral Tradition ami Written Record in Classical Athens (Cambridge:
Cambridge University Press, 1989), 1-14. Also, see her Literacy ami Orality in Ancient
Greece (Cambridge: Cambridge University Press, 1992), 3-5.
4. Toe sophists are often held responsible for corrupting older Athenian values, but
few scholars have related such charges to the Athenian mistrust of letters, save perhaps
Thomas, Oral Tradition, 18, n. 6; 20, n. 17.
Plato's Audiences 43
did_. ~ot write. 5 But .whether be did not publish because be was only
sellllhterate or had dehberately set out to converse and teach within the ancient
oral tradition of bis ancestors is worth examining. 6 Otherwise, bow will we
know whether it is appropriate or inappropriate to conceive of the book in the
band of that silent reader in the city square of Leuven as a copy of Plato's
Socratic dialogues?
1. Tradition and Transmission
An oral tradition can be distinguished from a literate tradition not so much
through the absence or presence of letters, as througb a peculiar ambivalence
exhibited in men's attitudes toward letters. 7 Are written marks temporary
bousing for ideas; shelters in wbicb thoughts may be stored to be picked up
later, when there will be time to discuss those ideas that bave only been
temporarily silenced or housed? Or are letters authorial omaments, signatures
owned and co~picuously displayed, verbal jewels to be borrowed only
between quotat1on marks, or to be bougbt, sold, and copyrighted for
consumption by an anonymous public or readers who resemble that nameless
5. W.K.C. Guthrie submits (Socrates [Cambridge: Cambridge University Press, 1971],
6), that "there is, and always will be a 'Socratic problem' ... since he wrote nothing."
At Phaedo 97c Socrates listens to someone reading from the book of Anaxagoras; at
Phaedrus 230e, he is led from the city he !oves to the country, where he learns nothing
from tree~,. because of a book then read aloud to him by Phaedrus. In Xenophon's
Memorab1lla 1.6.14, Socrates talks of unrolling scrolls left behind by wise men of old
so that he and his friends can pick out what is good, but at Memorabilia 4.2. lff
Euthydemus is wamed that from reading he may only acquire the appearance of being
wise. Also below, n. 35.
6. My attention was first drawn to this problem by E. Havelock, "Toe Orality of
Socrates and the Literacy of Plato" in New Essays on Socrates, ed. Eugene Kelly
(Latham, Maryland: University Press of America, 1984), esp. 78 and 85, where it is
suggested that Socrates clung to traditional orality and most probably could not write.
'f!1ºs~ who _treat :lato's dialogues as fictions are notas sensitive as they might be to the
histoncal dunens10n of this problem.
7. B. Gentili, Poetry ami /ts Public in Ancient Greece, trans. T. Cole (Baltimore:
Johns Hopkins University Press, 1988), 4: "To be called oral, poetry must meet one or
more o_f ~ee cond'.;ions: (1) oral composition ... (2) oral communication .. . (3) oral
tra~~ss10n . . . . M. Parry's work is an example of all three, below n. 10; Ryle's
pos1t10n on the oral publication of Plato's dialogues an example of two (see Plato 's
Progress [Cambridge University Press, 1966], 21-54); see also n. 6 above. But it is
Gentili's third criterion that no one to my knowledge has fully explored. Perhaps it is
best approached not only through relating Plato to Homer but through considering with
:niomas, Oral Tradition, ch. 1, esp. p. 19, the rich mixture of oral and literate elements
m fifth-century Athenian society.
44 Toe Third Way
cast-iron image of a reader in the city square of Leuven? Just what are letters?
Or more precisely, how are men's attitudes toward unseen sounds, present
only in time, or their unmouthed literate counterparts, present only in space,
to be distinguished from one another?8 Or more provocatively, what are men's
attitudes toward letters during a period of crisis, when long-established
customs are threatened by a change from oral to literate practices or vice
versa; that is to say, times like our own when the face of the world is being
refeatured through the invention of fresh manners of communication, such as
talking books, voice mail and touch-tone phones?
That early in this century Milman Parry should have heard, through the
marks in those books customarily thought of as Homer's, vestiges of sounds
suggesting oral composition is remarkable, especially in light of the fact that
this very same bard had already been represented as a precocious genius with
a book roll in hand, three centuries before Christ was born. 9 It is also not
peculiar, in view of Gibbon's prejudice, that Parry, after putting forth his
hypothesis, should have complained of wandering ghostlike through the annals
of time, at home neither in Homer's epoch nor his own. 10
That Parry's thesis, with the help of Lord, should then have become more
widely accepted after his death is perhaps not so remarkable. For between the
1930s and the 1980s the mass media had ploughed upa vast area of heretofore
silent linguistic terrain, so that scholars and their students were beginning to
recognize through their own ears and experiences that noisy revolution in
communications that the electronic media had wrought, and hence were
becoming more sensitive to those temporal parameters governing the oral
production of Homeric song. 11 In any event, philosophers must attend to the
8. Unless a scholar is also a musician, it may be difficult to appreciate how temporal
attitudes govem oral thinking or to remember, as does M. Parry in The Maldng of
Homeric Verse, ed. A. Parry (Oxford: Oxford University Press, 1971), 321, how
remembering depends not upon first seeing something in print and then visualizing it, but
upon first having heard it and then developing one's sensitivities to differences in sound.
See W. Ong, Orality and Literacy (New York: Methuen, 1982), esp. chs. 1-3.
9. E. Hoffman, "Exploring the Literate Blind Spot: Alexander Pope's Homer in Light
of Milman Parry," Oral Tradition 1/2 (1986): 381-396.
10. M. Parry, Homeric Verse, lix. Also below, n. 24.
11. It is well to remember the pioneering efforts of those earlier scholars who faced
the unpleasant task of calling to the attention of their colleagues the fact that on occasion
so-called illiterates may be highly civilized and hence the importance of exploring the
problem of literacy in ancient Greece. Portunately, K. Robb, in bis introduction to
Literacy and Paideia in Ancient Greece (Oxford: Oxford University Press, 1994), recalls
for us not only the history of these wintery days but also how E. Havelock had predicted
the stampede that would eventually take place after the scholarly community realized that
oral culture was far from a contradiction in terms. lndeed, Robb suggests that Harris'
conclusions conceming the restricted leve! of literacy in fifth-century Athens not only
Plato's Audiences 45
Greek sensitivity to time and to what is appropriate at the moment for a
particular audience to hear. Otherwise, they will not hear the oral dimension
of _Socrates' philosophizing, resounding through the temporal gaps Plato
dehberately opens up in the architectonic structure of those lettered
conversations into which he breathes the spirit of Socratic talk. 12
So, taking Homer and ~s audiences as our paradigm, let us inquire why
Greek poets are so often said not to compose new songs and stories as much
as to find fresh ways of telling the old. 13 It is a sentiment one hears when
Aeschylus notoriously speaks of his tragedies as slices of Homer's great feast
and then acts out tensions arising during his own century between Athenian
males and females when confronting Homer's Agamemnon with his consort
and queen, Clytem.nestra. It is detected as well when Euripides without
actually mentioning contemporary events, alludes to the brutal treatm~nt of the
Meli~ by the Athenians, through starkly reenacting the sack of Troy and the
suffe~ng and enslavement of Trojan women. It may also be present when
Gorg1as composes rhetorical defense speeches for archaic figures such as
Hele~ ~d. Palamedes in order to attract native Athenians as his students. 14
Yet, 1f It 1s sometimes thought that Plato reechoes this tradition when he
substitutes his philosophical hero, Socrates, for Homer's Odysseus or Achilles,
few hav~ observed how Plat~ h~s ~orrowed elements of oral structuring from
Homer. Here may be a s1mdanty that suggests why neither Homer nor
Plato are authors, in our sense of the word.
At~e~ding first, then, to Homer's relation to his audience, let us discover
why 1t 1s ~ot those familiar myths, such as the Cyclops or the Trojan horse,
usually said to be Homer's that are fabricated first but, on the contrary, it is
conf~ what had earlier been thought of as shocking when reponed by Havelock, but
may rn fact be more stringent.
. 1:. Resisting_ any such temptation, C. Kahn warns us not to be taken in by an "optical
illus10n," that 1s to say, "Plato's success in re-creating the atmosphere of the fifth
century" ("Did Plato Write Socratic Dialogues?" in Essays on the Philosophy of
Socrat:s, ed. Hugh Benson [Oxford: Oxford University Press, 1992), 39).
13. Better Homer restaged for yet another time / Toan matters unessayed in prose
and rhyme" (Horace, Ars Poetica, 129-30).
14: R. ~- Winnington Ingram, "Clytemnestra and the Vote of Athena," in Oxford
Readmgs m Greek Tra~~dy, ed. E. Sega! (Oxford: Oxford University Press, 1983). G.
Murray frnds that Eunp1des composed bis Trojan Women right after Melos had been
captured and her meo put to death and her women enslaved (Eunpides and His Age
[Oxford University Press, 1965), 63-8).
. 15. S~arities ~~tween Ho~er and Pla~o are discussed by Notopoulos, "Mnemosyne
lllOral L1terature, Transactwn Proceedmgs of American Philological Association 69
(1938): 4~5~493. See als~ ~- H~lperin, "Plato and the Erotics of Narrativity" and L.
Ko~man, Silence and Imitat10n rn the Platonic Dialogues" in Oxford Studies in Anciem
Ph1losophy, ed. Klagge and Smith, suppl. vol. (1992).
46 Toe Third Way
conversation that is prior to myth.
Recall how Demodokos, the bard of the Phaia.kians, responds to the request
of an unknown guest to sing about the wooden horse. It occurs right after he
has sUDg of the Trojan war and been rewarded for the excellence of his soog
of the Akhaians with a fine cut of meat and that sort of lavish praise one
reserves for a bard suspected of being divinely inspired. 16 Indeed, hinting that
the skill to shape his soog may have been given to Demodokos by the Muses
or Apollo, the disguised Odysseus actually suggests that either the bard had
been at Troy himself or had heard what had happened from the mouth of
someone who had been there. 17
Hence, as early as Homer, the truth of a soog was judged by those who had
witnessed the events of which a bard saog, and since the Muses. as the
daughters of Memory, are ever-present, it is they who breathe truth into the
soog of a blind bard, like Demodokos, who offers his voice to them as their
instrument. Or put otherwise, what the witness, Odysseus, discovers in
Demodokos' siogiog is the blind bard's ability to provoke again within his own
breast pain like that he had suffered at Troy. 18 If the bard can recreate with
equal skill his own wooden-horse stratagem, then it is the disguised Odysseus
who will confirm the divine origin of Demodokos' sioging.
Toe request for a soog about the wooden horse comes from a man whom the
Phaia.kians have yet to recognize as Odysseus. Their unnamed guest, on the
other hand, is painfully aware that, with the death of his men on the island of
the sun, he has lost his identity as a sacker of cities, if not his renown as the
inventor of the wooden horse. Aware of this, we, as Homer's audience, unlike
the Phaia.kians, discem the need of this stranger to hear again of what has been
lost together with his men in order that, by coping with his past, he might
come to grips with what is yet to come. 19 Or why in an oral tradition it is
conversation that gives rise to story, even as here it is the request of this
16. C. Sega!, "Bard and Audience in Homer," in Homer's Ancient Readers, ed.
Lamberton and Keaney (Princeton University Press, 1992), 8: "the song ... wins from
Odysseus the most extravagant praise that a bard receives in Homer (8.474-88)."
17 . For the muse, see A. T. Cole, The Origins of Rhetoric in Ancient Greece
(Baltimore: Johns Hopkins University Press, 1991), 34; Sega!, "Bard and Audience," 8
and 17, and most esp. Thomas. Literacy and Orality. 115-7.
18 . Sega!, "Bard and Audience," 8-9, distinguishes two different audiences within the
poem: the Phaiakians , who are distanced from the events of which the poet sings and
hence take pleasure in what they hear; Odysseus, who is not distanced and hence feels
pain at what he hears.
19. See below, n. 51.
Plato's Audiences 47
unknown guest that provokes the tale told by Demodokos.20
With such oral exchange in mind, perhaps one can discem, in fifth-century
mixtures of oral and literate elements, attitudes characteristic of the time
duriog which Socrates philosophized, attitudes so distinct from our own that
we must puzzle over why the Athenians would take the time to read aloud in
their assembly a letter sent them by Nicias. 21 Or why Nicias, who was sick
and under siege in Sicily, would trouble to communicate with them in writing?
Why not dispatch a messenger to inform the Athenians of his own and hence
their grave and imminent danger? Perhaps Nicias was beginning to appreciate
the clarity and accuracy of the written word, in contrast to the suspicion and
mistrust earlier associated with Ietters, as one scholar recently argued. 22 But
if that were the case, why would Thucydides represent this incident in the
history he writes in such a way that we, as his readers, seem to be listening
together with the Athenians to what he writes, as it is read aloud, unless he is
thinking of us as living, breathing, aural witnesses who, together with the
Athenians, are to remember in the future the significance of what he writes?23
Indeed, it is feasible to ask whether it is not still this ancient admixture of
sight with sound and lettered with unlettered that will testify to the greatness
that was Athens.2A Or to ponder how those who continue an oral tradition by
recreating events from the past within themselves might exhibit different
attitudes toward time, composition, and remembering than those of us who are
ruled by the spatial conventions that govem a literate tradition. It is a question
that can fortunately be raised from within this concrete fifth-century context.
When Themistocles rebuilt the north wall of the Acropolis, which had been
destroyed by the Persians, he did not use pristine marble, freshly mined from
the stone quarries, but fragments of those column drums Ieft behind when the
20. When analyzing the logic of orality, one can not treat words in isolation either
from men or the world, as do logicians. lndeed, in order to exist at all, words must be
in time either through being spoken by someone or heard and remembered by someone
else. Put in modero locution, it is not enough that p knows s, p must say s to q; and q
must hear and remember.
21. Thuc. vii.8.2-16.1.
22. Harris, Ancient Literacy, 78.
23. See Thuc. i.22 ("ktema es aei") and Ryle: " ... it was intended for listeners.
Thucydides must have expected [his audience] to remain constant through many, many
sessions" (Plato's Progress , 54). Cf. Thomas, Literacy and Orality, 103-4 and n.9, for
those who interpret differently.
24. Reporting his father's attitude toward the historical method in literary criticism,
Adam Parry (Homeric Verse, !xi) found M . Parry unable to. prevent a certain feeling of
fear for the past as he saw it perceived by his contemporaries: "That is to say, the
Greeks through their literature turned the past into the present and thus made it into a
m_irr~rfor themselves." (My italics.) It is this which Parry thought of as showing up the
flimsmess of the hold our past literature has on ourselves. See also below, n. 25.
48 Toe Third Way
old walls were pulled down and destroyed by the Persians. Was this because
the Athenians were short of materials, or Themistocles short of time? Or did
he, fine statesman that he was, embody the shape of the past within the
present, so that the Athenians going about their business below in the agora
might look up to the Acropolis and observe these fragments from their past
and recall their near defeat by and ultimate victory over the Persians?25 If so,
how do elements from the past provoke remembering from within, so that men
may relive what had occurred all over again in the present? And how different
is such an experience from looking up a document that records past events so
that one might read in a text about what had occurred, reconstructed
chronologically, but wholly from outside of the particular time period
represented?26
Perhaps if one asks the right sorts of questions, it will be possible to release
from within a tradition dissimilar from our own, the reason why many an
intelligent Athenian might still have harbored ambivalent attitudes toward
letters even as late as the late fifth century, indeed, even after Herodotus had
recognized and celebrated their preservative powers in this self-conscious
paragraph from the very beginning of a history, where this reason is given for
his authorship.
These are the researches of Herodotus of Harlicarnassus, which he publishes in the
hope that the memory of what men have done should not be destroyed by time, and
that the great and wonderful actions of the Greeks and Barbarians should not lose
their due fame. 27
Why did the Athenians not follow his lead? Why not simply accept letters
and carve what is to be preserved for all time into marble or stone? Why did
the Athenians instead remain suspicious of writing and letters?
Thomas has recently confirmed that the Athenians never really trusted letters
unless what had been written could also be confirmed by the voices of
25. J . B. Bury puts Them.istocles' use of such materials down to haste (A History of
Greece [New York: The Modero Library, 1937], 315). On the other hand, J. Miliadis,
in The Acropolis (Athens: M. Pechlivanidas; distr. Atlantis, 1960), 13, writes: " . . .
great architectural fragments oftemples destroyed by the Persians have been incorporated
in the wall in an orderly fashion. It is said that Themistocles wished this to be done, so
that the forgetful Athenians m.ight see them from the Agora and be reminded of the
past. "
26. Contemporary choices between the synchronic or diachronic neither duplicate nor
substitute for what is reported above in ns. 24-5. Nor is it to be recaptured by adopting
A. E. Taylor's assignment of dramatic dates in Plato: The Man ami His Works (London
& New York: Methuen, 1986), 46 or those who recreate Plato's development.
27. Herodotus, Persian War, Bk. l. lt is likely that Herodotus published his book by
reading it aloud. Nevertheless, unlike Plato, he put his name into his text.
Plato's Audiences 49
eyewitnesses. She is not concerned with merely collecting evidence in order
to measure when and how the Athenians may finally have come to embody our
literate values. On the contrary, she instead seeks to discover when and how
new attitudes toward letters were replacing the old. This she discovers in legal
practices.28
When a written contract was presented in court, it was not enough simply
to submit the document itself as evidence. What was required in addition was
the word of witnesses who could testify to what had first been decided by
word of mouth. Nor was this demand made only by those who were illiterate
and who might therefore be "written off' as prejudiced. It was also Iiterate
Athenians who mistrusted letters, as is evident, for instance, when the avant-
garde tragedian Euripides, who was notorious for his collection of books,
dramatized such mistrust in his Hippolytus.
~at condemned Euripides' intelligent hero, Hippolytus, and his books,
cunously compared with smoke up in the air, was a short letter written by his
stepmother Phaedra to her husband Theseus, informing him that she had taken
her own life because of the attempted rape by his bastard son, Hippolytus. 29
Toe chorus of women know that Phaedra's written letters lie, that a discreet
secret is hidden in and by théir silence. 30 But these women have been worn
to secrecy by Phaedra. Therefore, they are unable to testify to the falsehood
of what was written, even though they can and do warn Theseus that he will
be making a serious mistake in too hastily trusting such silent marks. 31 Nor
is the voice of Hippolytus any longer to be trusted, even though he would have
relished witnessirg what Phaedra might say when cross-examined by her
32
husband. Hence, in the absence of those who can be questioned and the
presence of a son whose voice is mistrusted, what Theseus tragically believes
28. Thomas, Oral Tradition, ch. 1, esp. 29.
29. The comparison at 1.953 of writing with smoke may strike us as odd unless the
inroads made by letters into the lives of fifth-century Athenians is recalled. Por then one
reco~~s ?ºw sounds, which had been invisible, are becoming visible, although perhaps
not mtellig1ble to those not fully literate. That may be why Theseus, wbo can read
Phaedra's note, compares his son's books at 953 with smoke (kapnos) .
30. In 1953 B. Knox suggested that the situation that brings the four principal
characters of this play into significant relationship with one another and makes an artistic
unity of it is silence and speech ("The Hippolytus of Euripides," in Oxford Readings in
Greek Tragedy, 313). When recent studies ofthe relation of orality to literacy are added,
I believe Knox's insight can be confirmed and expanded.
_31. On Phaedra's choice of honor as the good (or at least her good) see Knox, "The
Htppolytus," 315: "May it be my lot not to pass unnoticed when I act nobly, and not to
have many w1tnesses when my acts are disgraceful (403-04)."
32. O. Taplin, Greek Tragedy in Action (Berkeley & Los Angeles: University of
California Press, 1978), 115.
50 Toe Third Way
are those silenced sounds that resemble his wife's corpse not simply by being
bound to her dead hand, but in their inability to reply when questioned. 33
Toen, just before Hippolytus dies, the victim of his father's curses, the
goddess Artemis descends from Olympus on the machine to straighten out the
false perceptions wrought by Phaedra's lying hand. Your son is guiltless,
Theseus is told. Too hastily have you condemned him. But unable to save
Hippolytus' mortal life, Artemis promises rewards: Young maidens about to
be wed will cut their hair in mourning and remember Hippolytus' purity, as
they sing of Phaedra's love for him.
Why should letters be prized instead of those songs that have long served
to commemorate, in the minds of those who come after, the noble actions of
ancestors worthy of remembrance? During a period of crisis, the literate
Euripides struggled with conflicts arising from the inroads made in a
heretofore basically oral culture, through the use of written letters. Nor is this
example singular. Almost a century later in his Phaedrus, Plato also
acknowledges a similar mistrust. 34
Toe marks of sounds that have been silenced being likened to the painted
images of living creatures, letters are blamed either for behaving like
paintings, which say nothing when questioned, or for saying, when read aloud,
one and the same thing over and over again. 35 Nor does their immovable
corpse-like insensitivity to oral exchange elude notice. For after being removed
from time and sheltered in space, these mute marks are said to resemble
33. Taplin, ibid., 95, says of the tablet on which Phaedra writes: "lts impersonal,
indirect and immutable message is typical of the lack of proper communication in the
Hipp." He also asserts (117) that "Hippolytus has been trapped in a conspiracy of
silence."
34. Euripides' Hippolytus was perforroed in 432 BC. Plato's Phaedrus has been
variously dated. Hackforth, Phaedrus (Cambridge: Cambridge University Press, 1952),
3-7, will only "guess" on the basis of interna) evidence that Plato perhaps composed it
about 370 BC. Puzzled by Plato's criticism of the written word, Harris, Ancient Literacy,
92 & n. 128, thinks that Plato may be telling us what Socrates used to say sorne forty
years earlier at the dramatic date of the dialogue. But Thomas attributes the criticism to
Plato (Oral Tradition, 32-3).
35. At Protag. 329a., after Protagoras' long speech, Socrates raises the same objection
against the spoken word. Here is a good reason for seconding Griswold' s idea that at the
heart of Socrates' criticism and Plato's refusal to assert anything in his own name is an
unwillingness on the part of either to actas an authority. Here as well may be Socrates'
reason for using irony rather than blame in the face of those who act as if they already
know and hence are wise. See C. Griswold, Self-Knowledge in Plato's Phaedrus (New
Haven: Yale University Press, 1986), 220-1 : "Plato's anonymity or silence .. . is an
indispensable element of his use of the dialogue forro and so his response to Socrates'
criticism of the written word. In these ways the Platonic dialogue forro goes sorne way
toward countering Socrates' fear that writing will serve as a pretext for dogmatism."
Plato's Audiences 51
books, which also are unable to answer when questioned. How can written
marks become sensitive to dijferences in audiences? How can they know to
whom to speak and before whom to remain silent? Or still more provocatively,
how can these insensitive inscriptions immortalize the open spirit of Socrates'
questioning, without at the same time killing off that spirit? Toen it was not
Socrates, but Plato who was confronted by this paradox: Why not philosophize
as did Socrates? Why write?
2. Plato's Audiences
The look of letters on the pages of a book said to have been authored by
Plato are of no more concem to him than the empty margins surrounding the
text or the empty places deliberately left to signify a book's end or beginning.
Nor is Plato interested in those letters and words that are somehow displaced
in the process of printing; those typeproofs an author must correct in order
that his loosely targeted reading public, silently assimilating wholly by eye,
will not be misled by strange spellings or garbled grammar. Here is one sense
then, in which Plato is not an author in our sense of the term; that is, a writer'.
prizing and perfecting his own ·marks or word-things. Indeed, those marks too
fr~quently taken for Plato's might better be compared with tones of music,
w1th sounds only temporarily housed in musical notation, awaiting the moment
when they will be restored to time and the world to be completed or heard.
For when remembered aloud for our benefit, it is Plato who is heard, even as
Homer, not only mimicking the timbre of other men's voices but replicating
their verbal choices and favorite patterns of argument. 36
But it is important to ask whether Plato's own voice will also be heard. For
in recognizing that he rarely, if ever, speaks in his own person, one comes
face to face with the problem of his authorship. 37 Nor will it suffice to
dismiss thi~ difficulty by suggesting that, unlike Herodotus, Plato simply did
not letter his name as author into his text, especially since it is not so much his
signature as his voice that an ancient Greek would have identified with his
36. Recently, Plato's preference for a mixture of epic and dramatic styles at R ep.
392D-394c, has been applied to interpreting his dialogues, esp. by Halperin, "Erotics of
Narrativity," 94.
37._ The problem is discussed at Iength by L. Edelstein ("Platonic Anonymity,"
American Joumal of Philology 83 [1962): 1-22) who explicitly asks why Plato refuses
to take r~sponsibility ~or any of the views in his dialogues. Griswold (Self-Knowledge,
222) re~lies that the dialogues are intended to serve as mirrors. "They allow the reader
to_ see hnnself in the text." Also see E. Tigerstedt, Srockholm Studies in rhe History of
Luerature 17 (1976), ch.7, and below n. 47.
52 Toe Third Way
person. 38
That there would be no Platonic compos1t1on without Socrates seems
plausible. Yet to come to grips with Plato's enigmatic silence, one must first
distinguish those interpretative attitudes that treat such silence as
philosophically significant from those that either do not recognize Plato's
failure to speak in the first place or, after reconstructing his authorship, act as
though it were philosophically irrelevant. Since the greatest number of scholars
has adopted the latter opinion, let us begin with their method of reading Plato,
only then to consider what sort of author Plato might be if he had deliberately
decided not to use his own voice or to appear in his works as our authority on
Socrates.
Studying the language and allusions implicit in Plato's texts, most scholars
begin by reconstructing the dates at which he can be thought of as authoring
each of his dialogues. What is taken for granted when his corpus is divided
into early, middle, and late "Plato" is that the author, Plato, matured to
develop his own philosophy independently of Socrates. Toen it is the Socrates
of the early dialogues who is identified with the Socrates of history and the
doctrines (if any) that Socrates espoused. In the middle dialogues, on the other
hand, Socrates is treated as Plato's mouthpiece. Here, it is Plato who not only
masks his own features with those of Socrates, but then breathes his theories
through the mouth of Socrates. AH of this implies that Plato's failure to say
which doctrines are his presents little difficulty for those who read within this
system. Assumi~ that Plato developed, it is those who can most convincingly
date Plato's literary remains who can then act as our authorities and say in
which dialogues Socrates speaks and in which he is only Plato's philosophical
pseudonym.39
For the words of this Plato to appear in the book held in the hand of that
cast-iron statue in the city square of Leuven, a secondary source,
authoritatively defending this theory of Plato 's authorship, would at the same
time have to be poured into the hollow interior of that student with the water
from the glass in his other hand. Yet how unfortunate it is that a sleight of
hand should testify to how this seminal distinction has not only been adopted
but disseminated. For it assumes what it sets out to prove: namely, Plato
38. We think ofwhat is written and then signed as carrying authority, but the Greeks
did not share our attitudes toward signatures, even if they did recognize that an
individual's handwriting was distinctive. See Harris, Ancient Literacy, 73, n. 38; 198,
n. 117.
39. Por the most important recent affinnation, see G. Vlastos, Socrates: Jronist and
Moral Philosopher (lthaca: Cornell University Press, 1991), ch.3. The view is also
assumed by Guthrie, Socrates; N. Gulley, The Philosophy of Socrates (New York: St.
Martin's Press, 1968); P. Friedlii.nder, Plato, vol. 1-3, trans. Meyerhoff (New York:
Bollingen Poundation, 1958); and H. Benson, Essays. See below, n. 40.
Plato's Audiences 53
developed. 40 Indeed, only lately have scholars who read in this way been
willing to consider other ways questioning such orthodoxy.41
Those, on the other hand, who take Plato's philosophical silence seriously
have at least two reasons for interpreting his texts differently. In the first
place, what is questioned is that attitude toward time that is premised not on
Greek temporal attitudes, but on a nineteenth-century European concept of
change in time, authored by Charles Darwin. 42 Hence, remembering how
Greek tragedians generally and ancient oral epic poets more especially set out
to recreate the past again in the present, those who would establish a fresh
distinction between Plato and Socrates set out to inquire whether similar
temporal patterns might not also be discovered in Plato and then buttress their
reading of discrete conversations with a Greek etymology that literally suggests
what is true as not forgettable, a-lethes. 43
Further, wishing to avoid Latín anachronisms, those who believe that Plato
would never have been the philosopher-poet he became if it were not for
Socrates hesitate to talk about what Plato has made as factor fiction. Orto put
the point interrogatively, how can one prejudge what may still be in time by
thinking of it in absolute cate~ories as not yet available even to Aristotle?44
40. In her review article, "Platonic Chronology Reconsidered," Bryn Mawr Classical
Review 3 (1992): 4, n. 3, D. Nails points out how virtually ali contemporaries who treat
stylometric or other approaches to chronology worry over the circularity problem in
connection with the supposed evolution of Plato's thought. After comparing G. Ledger's
Re-Counting Plato: A Computer Analysis of Plato's Style (Oxford: Oxford University
Press, 1989) with H. Thesleff's Studies in Platonic Chronology (Helsinki: Societas
Scientiarum Pennica, 1982), Nails concludes her comparison with a chart illustrating the
very different conclusions drawn by these two scholars. She points to the incompatibility
of their solutions and submits that "the chronology of the Platonic dialogues has never
been so spectacularly at issue."
41. See J. Klagge's account of why analytical philosophers have been reticent to
consider philosophy in light of literary style (Oxford Studies in Ancient Philosophy , supp.
vol. [1992]: 1-4). He goes on to acknowledge two different altitudes toward Plato's
writing, the one described above n. 39; another recently generating this controversy:
Does Plato expound views at ali? Also consult D. Graham, who recognizes the
importance of treating Socrates as an oral philosopher ("Socrates and Plato." Phronesis
37 (1992): esp. 147ff.)
42. See J. Randall, Plato: Dramatist of the Life of Reason (New York: Columbia .
University Press, 1970), 34: " . . . the chronological order of the dialogues had no
relevance at ali until the dominance of the idea of evolution .. . . "
43. Por other scholars who make use of the etymology of aletheia, consult Thomas,
Literacy and Orality, 115.
44. Par from having our Latín literary classification, the Greeks even lacked our Latin
word "literature" (Aristotle, Poetics, 1447b). Curiously enough, Clotho, the Spinner, is
one of the Greek fates. Her name is derived from from the verb clothein, to spin.
54 The Third Way
Or again, how is one to conceive of what Plato has made as "fabricated''. or
untrue when what disappears through the use of this Latín is the anc1ent
Hellenic idea of what is fated as woven?
One has only to recall how the Fates in Hesiod and Plato are imaged as
immortals weaving the fabric of what is to be, in arder to understand how the
activities of weaving and spinning had profaundly influenced Greek attitudes
toward events in time and the world. 45 Indeed, the women in Homer's /liad
are surely not fabricating in the sense of the Latín when they work into their
weaving scenes from their own lives and those of their warrior husbands. Nor
do our grandmothers falsify what has occurred when they deliberately cut bits
and pieces of clothing once wom by members of their families and then sew
these brightly colored bits of the past together to make quilts of temporal
textures to evoke memories from those who had wom such fabrics during their
lives. It is because the Latin contrasts fiction with fact to separate time from
the world, not to mention present from future, that it is best put aside.
Otherwise, how is one to recognize the difference between passirg down
another man's word-things after surrounding what he has said with quotation
marks, and recreating the oral spirit of Socrates' pbilosopbizing? Or make t?
use of an earlier temporal paradigm, how is one to recognize that Themistocles
was not honoring himself but the Athenians when he put them in mind of the
past again in the present by constructing a new wall far the Acropolis not from
pristine materials but from fragments of the old column drums pulled down by
the Persians?
Or put otherwise, recognizing that far Plato remembering may not consist
in memorizing another man's word-things or even in picturing in one's mind's
eye a series of sentences near the bottom of the page of a book written by a
colleague, but in hearing again in the world the sounds of a friend's talking so
that in making use of marks outside of oneself one will be reminded from
"Rhapsodist" is from rhaptein, to stitch together.
45. Oxford Classical Dictionary, 2nd ed., 432-8 . Fate comes from the Latinfatum
meaning "what is said," and is probably derived from a belief in spoken magic rather
than prophecy. At 430-2, the OCD describes the Greek gods and goddesses as spinning
the great realities with a thread around man, as though he were a spindle. From this
image is derived the "harsh spinner," Clothes (Od. 7 .197). Toe three Moirai are first
imaged in Hesiod (Theog. 217ff) under their familiar names Clotho (the Spinner);
Lachesis (the Dispenser of Lots), and Atropos (She who cannot be turned). At Rep . 617,
as recounted by W. Greene in Moira: Fate, Good, and Evil in Greek Thought
(Cambridge, Mass.: Harvard University Press, 1944), 16 and 315ff., these three women
assist Ananké (Necessity) in whose lap is found not only the spindle governing the
universe, but the choices of life. Although the lots are dispensed by Lachesis, from these
each soul selects his own life, which is then woven into the present by Clotho and turned
into fate by Atropos, who ftxes Ananke's spindle so it cannot be turned back.
Plato's Audiences 55
within of the sounds of something once said in the world, it will be clear why
a few scholars still think of Plato and bis brothers, Adeimantus and Glaucon,
as noc simply having known Socrates but as having conversed with bim.
Hence, attending to this relation of world to words, it may seem plausible to
recall how Plato had bumed bis tragic dramas and tumed bis artistic skills to
recreating in letters the pbilosopbical sounds of Socrates' conversations, but
in such a way that those of us listening would be reminded through bis
inimicry of what it felt like to be cross-examined by Socrates- an attitude
provoked when Plato is conceived of not as accurately representing Socrates,
but instead as evoking the past again in the present.
Perhaps it was what Socrates had done befare he was convicted that led
Plato to put the sounds of bis talking again in the world. 46 Plato must have
been impressed by the fact that Socrates refused to go into exile befare bis
trial, knowing that bis absence from Athens would insure Anytus of the silence
he required. Further, Plato was surely no less impressed by Socrates' refusal
to name exile as bis penalty, because then bis life would most probably have
47
been spar:ed. ~e~haps this also explains why Socrates emphasizes during bis
defense bis unwdhngness to ever stop pbilosophizing. Indeed, why, in lieu of
ex.ile, Socrates propases a fine as bis penalty, even though not unaware that
the jury, in order to insure bis silence, might then vote for bis death.
The idea that Plato should recreate the paradoxical manner of Socrates'
talking, so that bis polyphonic nuances would continue to be beard and those
in Athens who had desired to put an end to bis talking might be awakened and
forced to examine what they falsely believed true about Socrates is not of
48
course new. But that Plato must be silent in arder that Socrates'
pbilosophizing is rightly interpreted or heard is a fresh way of talking about
why Plato should direct the well-known paradoxes of Socrates to us.
Or put again differently, first one must accept the fact that Plato had
.46. Socrates promi~es those who voted for his death in the hope of being relieved in
this ~ay from rendenng an account of their lives (Ap . 39c-d) that others younger than
he will. take up w~at he has left incomplete. Here is my reas on for confirming the
sugge~t10n of J. Klem, A Commentary on Plato's Meno (Chapel Hill: University of North
Carolina Press, 1965), 7: "We have to be serious about the contention that a Platonic
dialogue, being an imitation of Socrates .. . actually continues his work. "
47• At Apol. 29c Socrates suggests that Anytus' death sentence is the result of his not
leaving Athens before his trial. Guthrie agrees: "Anytus and his associates would have
been content-probably more than content- if he had simply removed himself . . . . "
(~ocrates, 63). At several junctures, Socrates shows that he is well aware that it is his
silence ~at is a.t issue. See 37e, for example, where Socrates explains why he will not
°'.1me exile as his penalty and then why he simply can not agree to be quiet. Toere is also
bis prediction at 39c mentioned in n. 46 above.
48. See the self-knowledge theme in C. Griswold, Self-Knowledge, 2-16.
 Toe Tbird Way
56
recognized that he would be unable to speak as an authority on Soc_rates and
rightly communicate the truth about bis philosop~zing,_ a~d that th_at 1s why he
confounds us instead with Socrates' pbilosopbical 1diosyncras1es, so that
provoked by the sounds of this talking we will be force~ to unp,ack that
double-dialogue structuring that has again been detected m Plato s letters
during our century.49 But in order that one may readily hear the tempo:al
dimensions within Plato's conversations, it may be helpful not only to conce1ve
of a fresh way of attending to bis writing, but also to show how what has been
made is rooted in Homer.
3. A Fresh Distinction Between Socrates and Plato
There are three heuristic principies, all of wbich may be of sorne help in
defining the right way of listening to and hence talking about "P~ato."
Knowing that he would be unable to speak in bis o~n voice and _nghtly
communicate the truth about Socrates, Plato, as poet-pbilosopher, dec1ded to
sh<JrV us instead how Socrates would behave with respect to particular
respondents, for then we would be in a position to witness wh~t was shown
through asking and answering questions. Or expressed d1fferently, by
deliberately incorporating discontinuities into a text, Plato woul_d. turn thos_e of
us who had been listening into aural witnesses, capable of rais1ng quest1ons
about bis texts and explaining what had been shown about Socrates through
explaining such discontinuity.
Here are examples of the sorts of questions one might put to "Plato." Why
does Socrates adopt this form of argument when talking wi~ th: famous
sopbist Protagoras? Why a dilemma?50 Or why, after represent1ng bi~self as
an expert on matters of love, does Socrates invite bis host, the young dil~tta_nte
Agathon, to say whether love is a god? What is it about Agathon tha~ 1nv1tes
Socrates' question?51 Or why after listening to Glaucon and Ade1mantus
49. see how A. Bowen expresses Tigerstedt's idea of Plato's double dialogue
structure: " .. . the reader is obliged to attend a conversation . . . to which he is a silent
listener and in which he has no part; ... on the other hand, he is challenged to reflect
on this conversation and address its questions" ("On Interpreting Plato," in Platonic
Writings, Platonic Readings, ed. C. Griswold [New York and London: Routledge. 1989],
55ft).
50. Por Socrates' use of Protagoras' two logoi, consult Randall, Plato, 106-7. In "The
Mask of Socrates' Protagorean Argument in Plato's Euthyphro," The Philosophy of
Socrates, ed. K. Boudouris (Athens: lnternational Center for Greek Philosophy and
Culture, 1991), 424-435, I argue that the dilem.ma is Protagoras' favorite form of
argument and can be expressed in the form of two logics.
51. Por Socrates' question to Agathon, consult M . Stokes, Plato's Socratic
Conversations (Baltimore: Johns Hopkins University Press, 1986), 118ff.
Plato's Audiences 57
construct arguments challenging bis own position on justice, would Socrates
admit that without prior knowledge of their characters, he would have been
taken in by what they had said.52
If such questions are pertinent, Socrates is not well-conceived in isolation
from bis respondents nor Plato in isolation from Socrates. 53 E ven as the
Muses breathe words into the mouth of the bard, so Plato breathes life back
into those marks left bebind on paper or papyrus when bis dialogues are heard
as genuine exchanges. This is why "Plato" is best represented as the relation
between Socrates and bis respondent(s). Yet since "Plato" is more than those
whom he impersonates, this principie is to be augmented with two others.
Earlier, when listening to Homer, we became sensitii.ed to how one knows
far more through listening to one conversation in light of anotlier, than by
listening to one conversation in and by itself. For example, when Demodokos
entertained the Phaiakians, they only heard the song of their bard. We, on the
contrary, recalling how it was Homer who was using bis own voice to
impersonate Demodokos, perceived far more than the Phaiakians. We knew,
for example, why bis singing would bring pleasure to those who were
acquainted with Odysseus only .through song and why their unknown guest, on
the contrary, would be pained at being reminded of bis own wooden-horse
stratagem. Yet for this reason we also knew why it was Odysseus who could
confirm the truth of the events related in song, to praise Demodokos as taught
by the Muses. Indeed, when King Alkinoos heard in bis guest's breathing the
sounds of bis pain and stopped the bard's song, tactfully requesting bis guest
to tell them bis name, but still promising safe passage home, we, unlike the
Phaiakians, were able to catch Homer in the act of weaving distinct temporal
parameters into bis song.
That is to say, we noticed and asked why King Alkinoos should all at once
recall a prophecy told bim sometime ago by bis father: namely, how Poseidon,
jealous of the privilege of the Pbaiakians to return atIY and ali men home, had
promised that one day, after one such guest had been safely delivered to bis
native land, he would wreck their ship and turn it into a mountain range in the
middle of their harbor. That King Alkinoos should recall this prophecy just
after he promises to take his guest home, and then belittle and dismiss wbat
he recollects seems curious. But if the king is deaf to the impending threat,
we, recalling Poseidon's grudge against Odysseus, will, on the contrary,
52. Rep. 368a-b. One of the first to match logos and ergon in Plato was M. O'Brien,
"The Unity of the lAches," Essays in Ancient Greek Philosophy, ed. Anton and Kustas
(Albany: SUNY Press, 1971), 303ff.
53. Here is one reason why I find it misleading to put the rules of logic prior to those
of oral com.munication. Por a sensitive discussion of the differences between those who
read in light of rhetorical questions and systematic philosophy and those who hear what
is genuinely puzzling in what Socrates asks, see M. Stokes, Socratic Conversations, 1-35.
58 Toe Third Way
suspect that what has been remembered at this time will occur immediately
after Odysseus has been put ashore on lthaka. But that is because we do not
listen only to the king but to Homer impersonatiog the King as he remembers
this prophecy. Indeed, it is because we have gathered together in memory
events scattered in time that we are able to catch Homer in the act of weaving
the past together with the future even as he creates bis soog. Only then can we
54
inquire how Plato creates similar temporal structures.
In the Futhyphro, for instance, Socrates' respondent prides bimself upon
never having prophesied anything that has not come true. Tactfully testiog the
powers of wbich Euthyphro brags, Socrates invites bim to say what will occur
as the result of Meletus' charges. That Euthyphro should assure Socrates that
he will be successful in makiog bis defense and that nothing will come of it is
not a prediction that can be judged true or false at the time Socrates and
Euthyphro converse. It is only after we read Plato's dialogue to recall how
Socrates was put to death as the result of Meletus' charges that we grasp what
Plato sho.vs. Or put otherwise, it is up to us to ask questions about what "
Euthyphro predicts in order to grasp what we are invited to witness about
Euthyhro's vaunted powers.
To have found Plato weaving temporal parameters into a text does not
suggest that he has been caught, as was Homer, in the act of oral composition.
Wbat is implied, rather, is that from the perspective of a later conversation
with Plato we can witness that self-deception that Socrates would put
Euthyphro in a position to recognize, especially since Euthyphro is the only
person whom Euthyphro will accept as an authority. 55 That those who briog
different questions to this dialogue will come up with different interpretations
is imperative to recognize. Indeed, Socrates and Euthyphro are often assumed
56
to have failed because neither comes up with a definition of piety. Yet what
54. See E. Sega!, "Bard and Audience," 11: "This kind of narrative self-
consciousness, typical of the Odyssey, forces the hearer to become aware of the work's
construction ... ." A similar observation is made about the relation of Plato's readers
to the characters in the Symposium by Halprin, "Erotics of Narrativity," 100.
55. At 4d-c, Euthyphro explicitly rejects other men's opinions. See R. Robinson,
Plato's Earlier Dialectic (Oxford: Clarendon Press, 1953), 16, for a good comparison
of witnesses in a law court with Socrates' respondent as his own witness.
56. Since many scholars interpret Plato's so-called early dialogues in light of
Aristotle's testimony, they assume that Socrates not only seeks universal definitions of
ethical tenns, but also ethical theories like those found in science. What I assume, on the
contrary, is that Socrates would help Euthyphro know himselfby getting Euthyphro to
identify and acknowledge a belief he is unaware of as belonging to himself. Or put
otherwise, Euthyphro is actually in a state of belief, and that explains why he mistakes
a belief attributed to the world for knowledge. lndeed, that is why Socrates asks
Euthyphro at 11c3 and again at 15b8 to identify whose "works in words" are moving
Plato's Audiences 59
is taken for granted with the latter interpretation results from assuming that
what Plato made is no longer in time but has been permanently housed in
space and letters. 57 That is to say, it is because of the space left empty at the
bottom of the last page of Plato's text that a reader concludes that he has come
to the end of what has been represented and, hence, that neither Euthyphro nor
Socrates has successfully defined piety.
But if Plato is assumed to be deliberately mute or not speaking directly to
us in bis own voice about Socrates because that would spoil the open spirit of
Socrates' questioning, one may then grasp how the aporía at the end of a mute
set of letters does not represent Socrates' failure to find a definition so much
as it shows us how Plato has chosen to recreate the spirit of bis
philosopbizing. This is to point out how Plato deliberately stops the sounds of
Socrates' talking, without bringing what has been said to a satisfying close,
and why those of us who have been listening are meant to feel frustrated
enough to put questions directly to Plato's letters: To ask, for example, why
Socrates now ironically reiterates, as at the start, that Euthyphro surely knows
what piety is and to ponder whether Euthyphro has not finally felt bis own
contradiction reflected back through Socrates' stinging irony even though he
could not have heard it earlier when he believed he was rightly charging bis
old father with impiety. 58 Or put otherwise, it is after the manner of Socrates
that Plato would leave us in aporía, forcing us to think our own way through
such a paradox and forcing us to decide who is actually responsible for that
"work in words" that on two separate occasions Socrates characterizes as too
readily moved in a circle. 59 Yet how can it be Socrates who is called to
account when Euthyphro no longer denies that these words are his and runs
away, as do bis statements? Hence, only by matcbing words with men in the
world can one grasp why Socrates is not seeking an impersonal definition but,
on the contrary, has set out to help Euthyphro recognize how false are bis
beliefs about piety.
Two heuristic principies have been assumed in offering this solution. First,
there is an analogy between Socrates' relation to bis respondent and Plato's
relation to us. Second, "Plato" is the temporal unity of each of bis texts. What
is expressed through the first is not Plato's intention to live up to the letter of
around in a circle.
57. Indeed, that is why an analytical philosopher will deliberately rewrite what
Socrates says in Plato in the fonn of a timeless sentence or proposition.
58. For a discussion of the difference between Socrates as a midwife of false beliefs
and someone who simply tests the consistency ofbelief-statements, see E. J. West, "Toe
Mask," 423-430 and 433, n.10.
59. See above n. 56.
60 Toe Tbird Way

Socrates' philosophizing, but its spirit; through tbe second, what one can hope
to accomplish by interpreting Plato's letters. . ..
That Plato should deliberately structure wbat be creates w1lli: an abibty to
respond when questioned suggests not only hbw,. even ~o~gh us1ng letters, ~e
is not unfaithful to the oral spirit of Socrates' philoso~hiz1ng, but ~so h?w, m
good oral style, be would pass on that spirit. Nor is 1t a small sat~sfactto~ to
bave found bow mindful Plato was of that mistrust of letters, which anc1ent
historians bave only recently discovered in fiftb-cen~ Athens, ?r how Plato
would remedy that mistrust through mixing oral with h~erate practtces. Indeed,
it is in Plato's decision not to write as an authonty on Socrates but to
communicate instead the spirit of his talking directly to us that ~l~to 4
successfully seeds his garden of letters with enigmatic paradox reqwnng
resolution before one can explain to others in times to come bow they may talk
together in order to substantiate what Plato is sh(}Mng about Socrates.
Neither Published Nor Perished:
The Dialogues As Speech, Not Text

Joanne Waugh

In Plato 's Progress Gilbert Ryle observes that the traditional assumption that
Plato's dialogues were composed for individual readers of the written word is
just as speculative as Ryle's own view that the dialogues were composed to be
orally delivered, i.e., recited or read before an audience (though not always
for the same type of audience). 1 Ryle's aiguments about which dialogues were
composed for which audiences, and his intriguing suggestion that the Apology
was a defense not of the historical Socrates but of Plato himself continue to be
highly speculative hypotheses, but the claim that the dialogues were orally
delivered is no longer as speculative as it once was. In the nearly three
decades since Plato's Progress appeared, classical scbolarship has fully

l. Gilbert Ryle, Plato 's Progress (Cambridge: Cambridge University Press , 1966), 53
and 52. According to Ryle sorne of the dialogues were wriuen for a public audience,
while others were composed for private or academic audiences. Specifically, he claims
that most of the dialogues were composed to be delivered orally as part of dialogue-
competitions at Games beld at Olympia or Athens. Toe Timaeus and part II of the
Parmenides, however, were wriuen for oral presentation to the members of the
Academy; this explains why they lack the dramatic and eristic character of the early dia-
logues. Toe mammoth dialogues, the Republic and the Laws, were composed to be read
aloud to the middle-aged and elderly members of private clubs in Athens. Purther
references to this work will be noted in parentheses following the citation.
62 Toe Third Way
realized the consequences of a series of revisionary-some might say
revolutionary-events in earlier twentieth-century classical scholarship, and in
so doing, has required us to revise our assumptions about the reading and
writing habits of fifth- and fourth-century Greece and, pari passu, our
assumptions about the audience for whom Plato composed the dialogues and
the means by which the dialogues reached their audience. 2
Indeed, the claim that Plato wrote for individual readers of the written word
cannot be made without argument if we accept William Harris' conclusions in
Ancient Literocy that even in Athens there was only a very small circulation
of written literary works from 479 to 323 and that "in the main the authors of
prose literature could not and did not have to depend heavily on individual
reading for the diffusion of their works. " 3 Toe means by which people
2. Toe events to which I am referring include Milman Parry's demonstration of the
oral and traditional character of Homeric Verse, Michael Ventris and John Chadwick's
decipherment of Linear B, and Rhys Carpenter's dating of the Greek alphabet. Por
further information regarding these achievements, see Milman Parry, The Making of
Homeric Verse, ed. Adam Parry (New York: Clarendon Press, 1971); Michael Ventris
and John Chadwick, Documents in Mycenaean Greek (Cambridge: Cambridge University
Press, 1950), as well as John Chadwick, The Decipherment of Linear B (Cambridge:
Cambridge University Press, 1970); Rhys Carpenter, "Toe Antiquity of the Greek
Alphabet," American Journal of Archaeology 31 (1933): 8-29; "Toe Greek Alphabet
Again," American Journal of Archaeology 42 (1938): 58-69.
3. William V. Harris, Ancient Literacy (Cambridge: Harvard University Press, 1989),
86. In the chapter devoted to classical literacy and its limits, Harris argues that from 479
to 323 writing was increasingly used for economic, legal, and civic purposes, but that
legal practice, like the administration of large-scale business, remained to a considerable
extent oral and independent of documents. "Toe law of sale set little store by documents;
live witnesses to oral contracts were what counted. Receipts were still unknown. While
documents gained authority, they remained subject to a lot of oral control: when an
Athenian witness's evidence was read, he had to be there consenting. Toe reason for ali
this was not that Greek writing was inherently inappropriate for signature-writing.
Rather, the oral way of doing things was established and universally understood, while
letters and sorne other types of document had a certain reputation as a source of fraud,
a reputation which was periodically reinforced by the discovery of forgeries" (72-73).
W orries about fraud and forgeries ma y strike us as peculiar if we focus on the fixity that
writing provides. Lacking the speakers, however, it is as easy for writing to lie, i.e. , to
be false, as it is for it to tell the truili.. Truth is opposed to falsehood, and one cannot
judge whether the speaker is telling the truth if the speaker is not present.
Harris suggests that the ability to read and write may have been identified with the
propertied classes, who could afford not only instruction for their children and members
of their household, but also the materials necessary to make widespread use of writing.
Thus, he observes that "it is obvious that by the fifth century anyone who wanted to play
a leading part in the public affairs of Athens or of practically any other Greek city, with
the possible exception of Sparta, had to be literate and be able to make use of the written
Neither Published Nor Perished 63
became acquainted with literary texts was still oral,4 even the dissemination
of prose works, Harris observes, could be partially oral: "Protagoras read his
works aloud at Athens, or had them read out for him; Anaxagoras was read
out aloud; and in the fourth century, though these habits may have changed
soroewhat, we are told that Isocrates sometimes read out his productions . . . "
(85-86). .
Even before Ryle published Plato 's Progress, Eric Havelock had argued in
Preface to Plato that a proper understanding of Plato's dialogues requires that
we consider the audience for whom they were composed, but Havelock's
argument, like Ryle's, did not get the type of reception that one might
expect. 5 It is an interesting question in the history of ideas and in the
sociology of reception why Ryle's and Havelock's works met with such resis-
tance. It seems likely that Havelock's conclusion that Greece was not a literate
society until at least the closing years of the fifth century was hard for many
to accept, especially since so rouch of what we value as Greece's legacy to
western civilization comes from the period when Greece was ex hypothesi pre-
literat~.~ Ryle foresaw that sorne of his readers may be reluctant to give up
the divme Plato for a human one, suggesting that the compensation for
transforming Plato "from the sage who was boro at his destination to the
philosopher who had to search for his destination" is that "we lose a Nestor
but we gain an Ulysses" (192). Ryle's point that Plato was a flesh-and-blood
Athenian who no less than the rest of us was formed to a great extent by the
time and society in which he lived, is one that, perhaps, needs even more
emphasis than is suggested by the comparison with Nestor. For western
philosophy at least, losing the divine Plato is rather more like losing Apollo,
a loss that can hardly be compensated by gaining Odysseus, his godliness and
his many devices notwithstanding.
word" (74). Still, after examining the uses to which writing was put by the fourth-
century democracy at Athens, Harris concludes that "it was never essential for the
functioning of the Athenian democracy-or, we may assume, of most other Greek
democracies-that practically every male citizen should be literate or even semi-literate"
(79). In summing up his chapter on classical literacy, Harris writes that the notion that
eve~ male ~itizen should know how to read and write made its appearance during the
class1cal penod of Greek culture, but carne nowhere near to realization even at Athens.
4. In support of this conclusion, Harris notes that the author of the Rhetorica ad
Alexandrum picks out two poets who are anagnostikoi, suitable for reading, not for
performance; they are the secondary figures Chaeremon and Licymnius, and they are
exceptions (85-86).
5. Eric Havelock, Preface to Plato (Cambridge: Harvard University Press, 1963).
6. If we accept Harris's conclusions in Ancient Literacy, it is questionable whether
Greece was a literate society even by the close of the fifth century, or indeed, ever,
depending on how one characterizes a "literate society."
64 Toe Third Way
Doubtless, there are explanations of why scholars of Greek philosophy
might resist Havelock's and Ryle's conclusions that are better left to
sociologists and ethnographers of academe. However, I want to a.rgue that
sorne of the resistance to the notion that a proper understanding of Plato's
dialogues requires consideration of their audience results from a conception of
philosophy and its history that, although true to the spirit of "Platonism," is
nevertheless at odds with the fact of the Platonic dialogues. That this
conception of philosophy and its history is largely and ironically derived from
Plato is the result, I will suggest, of reading the dialogues as philosophical
texts rather than as representations of speech of a kind associated, at least once
upon a time, with philosophers. From the fact that Plato wrote dialogues, it
follows, of course, that if they were to be known to anyone other than their
author they would have to be read, but it <loes not follow that the dialogues
were written to be read silently by individuals, as opposed to being read aloud
or being performed before an audience. Nor <loes it follow that the dialogues
were or even should be read as-or for- philosophical texts, i.e., logically
ordered sequences of statements, the knowledge of which constitutes
theoretical knowledge, what Ryle characterized as "knowledge that. " 1 Reading
the dialogues as philosophical texts is, of course, the procedure one would take
if he or she were trying to determine what Plato's positions were on the so-
called "problems of philosophy," a procedure that takes for granted, among
other things, that these problems existed for Plato as they do for a
contemporary reader, and that solutions to them were offered in the
dialogues. 8 Toe notion that Plato's dialogues could or should be read as
containing philosophical texts is part of philosophy's enduring conceit-and of
its conception of itself as a discipline-that the "writing" of philosopby has no
effect on the "thinking" of philosophy, that writing philosopby is merely
recording or inscribing philosophical speech. "Philosophical speech" is not
ordinary talk, even when it is about "ordinary language;" rather, it is the kind
of "speech" that wbile grammatically tensed is logically tenseless; speech, the
meaning of which is held to be largely independent of the occasion of its
utterance or inscription; speech, in short, that consists of logically ordered
statements that used to be known as propositions.
Philosopby, then, claims to be not a "genre" of writing nor a kind of
7. See Gilbert Ryle, The Concept of Mind (New York: Harper and Row, 1949),
chapter II. "Knowing How and Knowing That."
8. "Analytic philosophers," James K.Iagge writes in the "Editor's Prologue" of the
volume he edited in the prestigious Oxford Studies in Ancient Philosophy, "have tended
to view the dialogue formas little more than a (dispensable) vehicle for the conveyance
of Plato's substantive philosophical theories" (James C. K.lagge and Nicholas D. Smith,
ed., Methods of lnterpreting Plato and His Dialogues, Oxford Studies in Ancient
Philosophy, suppl. vol. [Oxford: Clarendon Press, 1992), 3).
Neither Published Nor Perished 65
literature, but a type of language, and on this conception of philosophy it
really matters not whether the language is spoken or written. What matters is
tbat it be exact, scientific, and rigorous, and that it take the form of necessary
truths. These qualities of philosophical language and its formulation as
necessary truths it owes to being about a special kind of object, an object with
certain characteristics or powers that compel parts of the mind or parts of the
body to believe true propositions about the object. That a necessary truth
locates its certainty in its cause rather than in the a.rgument one can give for
it is a notion that Richard Rorty, following Dewey, sees as the "fruit" of the
Greek, and especially Platonic, analogy between knowledge and perception.
This notion is nicely illustrated in the following passage.
The essential feature of this analogy is that knowing a proposition to be true is to
be identified with being caused to do something by an object. Toe object which the
proposition is about imposes the proposition's truth. Toe idea of "necessary truth"
is just the_id~a of a proposition which is believed because the "grip" of the object
upon us 1s rneluctable. Such a truth is necessary in the sense in which it is
sometimes necessary to believe that what is before our eyes looks red-there is a
power, not ourselves, which compels us. Toe objects of mathematical truths will
not let themselves be misjudged or misreported. Such paradigmatically necessary
truths as the axioms of geometry are supposed to have no need of justification, of
ar~ent'. of discussion-they are as undiscussable as the command of Zeus shaking
the lightnrng, or of Helen beckoning to her bed. (Putatively rational ananke is, so
to speak, just a sublimated form of brute bia.)9
Thus, the assumption is that philosophical language simply gives voice to
statements that mirror sorne externa! or transcendent reality that causes these
statements to be true. Our giving voice to them or writing them down is
nonessential or accidental; the truth or necessity of these statements has
nothing to do with their utterance or inscription by us. Mathematical truths and
the axioms of geometry are thus exemplary instances of the exact scientific
rigorous kind of language that constitutes philosopby; and the ~bjects tha~
cause us to be~ieve the propositions of mathematics and geometry, models of
the sort _of obJects that could cause us to believe philosophical propositions
necessanly true.
Yet. as Jacques Derrida has pointed out, while the necessary truths sought
by philosophy do not depend on our uttering or inscribing them, we cannot
9. Richard Rorty, Philosophy and the Mirror of Nature (Princeton: Princeton
University Press, 1979), 157-158.
66 Toe Third W ay
belp but conceive of them as inscribed somewhere, at least metaphorically, º
1 of philosophy, given philosophical developments since the Enligbtenment. In
for these necessary truths are the inscribed Platonic Forms that the soul recol-
lects, or the clear and distinct innate ideas written in the soul, or the Laws of
N ature written in "N ature's Own Vocabulary," or the a priori conditions regu-
lating reason or grammar or thought written in the brain or mind of the
knower, or the set of true propositions that are written in ali possible worlds.
Writing "fixes" language; an inscription preserves a statement or utterance for
others to find. Writing thus enables statements about what is fixed, as things
are fixed when Zeus nods; statements about what is universal, intelligible, and
necessary; statements about things that are not the result of our doing, but
come from a power other than human. Philosophical "speech" does not appear
to be parasitic on writing, only because our philosophical statements are seen
as merely tokens of sorne original inscription of which we are not the authors,
but "see" or "discover" or "read" in Reality, Nature, or the Soul. What is
"read off' or deciphered is the "message" or "proposition" that Reality causes
us to believe; whoever writes this language at which time for what audience
or, indeed, whether this language is written, is irrelevant.
Though it may be of sorne historical interest who wrote what at what time,
and it may even be of sorne philosophical interest to identify where past
thinkers failed to be sufficiently "philosophical" and thus mistakenly identified
the wisdom of their day for the One True Vocabulary, on this conception of
philosophy the study of philosophy is independent from the study of the history
of philosophy. Insofar as philosophy consists of "perennial or timeless prob-
lems," one need not know bow others bave addressed these problems, one
need only address them using those techniques of mathematics, logic, and
analysis that are alleged to be similarly timeless. 11 Toe notion that philosophy
consists of timeless problems permits and encourages the view that if one does
the history of philosophy, it should be done as "intemal history," i.e., as a
study of the philosophical discourse that arises in relation to the language,
concepts, and methods of previous philosophical discourse, rather than
"extemal history," i. e., as an historical study that explains philosophical dis-
course in terms of extraphilosophical factors,-for example, the occasion of
its inscription or utterance.
Yet even an "intemal history" of philosophy cannot sustain this conception
10. Jacques Derrida, Dissemination, trans. Barbara Johnson (Chicago: University of
Chicago Press, 1981), especially Section l. "Plato's Pharmacy," and On Grammatology
(Baltimore: Johns Hopkins University Press, 1976).
11. Thus. it is not surprising that insofar as philosophers see philosophy as consisting
of timeless problems, what they regard as the history of philosophy is not really an
historical study at ali, but the process of "rationally reconstructing" tbe writing of past
philosopbers into the language and problematic of contemporary philosophy, which, in
tum, is identified with the "problems of philosophy."
Neither Published Nor Perisbed 67
the absence of the belief that the nonhuman reality that exists besides us is the
cr~ation of a . "Thinking, Spe~ng Being," or is itself a Thinking, Speaking
Be1ng, there 1s no reason to think that the world is split up, as Rorty says,
"into sentence-shaped cbunks called 'facts, "' that inscribed "out there" are
Platonic Forms or sorne language that the world proposes for us to speak. If
we are not the creation of sorne Thinking, Speaking Being, or identical with
such a Being, there is no reason to think that in the Mind are recollected
Forms, innate ideas, universal structures, or a priori concepts-i. e., sorne
Ianguage or vocabulary that transcends the languages or vocabularies of
particular societies at particular times. If the languages we speak are not
i°:'cribed in na~re or in the mind, they must be our invention. i.e., they must
anse out of particular forms of life in particular societies. This means that our
ideas of the self, like our descriptions of the world, seem to be "created by the
use of a vocabulary, rather than being adequately or inadequately expressed
in a vocabulary. " 12
Of course, it is just these changes in the "intemal history" of philosophy
that have led sorne scholars to conclude that attempts to reconstruct bow the
his!Orical background shaped the interpretation of a philosophical position are
fut!le and, ~onsequently, that rational reconstructions are not only desirable but
also unavo1dable. Attempts to acquire historical knowledge are futile if one
assumes . that it m~st take the form of a set of historical facts or necessary
~ruths. Smce .ther71s no "innocent eye" or "immaculate perception," the claim
1s that our histonc~ ~escri~ti?ns, n_o less than any other type of perceptual,
conceptual, or cogmt1~e acuv1ty, w1ll be a function of the language, theory,
?r set of structures or mterests we bring to them. Moreover, it is argued that
1f our languages, theories, and interests do not mirror what is "out there " or
"in here," .but instead place it under sorne description, our accounts of ;hat's
out there, m here, or what was said "back then" will not be easily, if ever,
separated _fr~m ~; languages. ~eories, s~ructures or interests that lay behind
our descnpt1ons. The conclus1on that 1s thought to follow is that we can
~2. See Richard Rorty, Contingency, irony, and solidarity (New York: Cambridge
Uruversity Press, 1989), 7.
13. Rational reconstructions and other interpretations, whicb as Richard Robinson says
~Pk:to 's_ Earlier Dialectic, 2nd ed. [Oxford: Clarendon, 1953]) are for the sake of
msmuatmg the f'.t1ture, ~ay not be as innocent nor as high-minded a practice as their
adherents so~et1IDes cl~IID, especially ~ben they discuss tbese historical works in light
of the. foundat10ns of philosophy, the sc1entific method, or the universal moral principie
~o wb~~b the a~erents sub_scrib~. "Toe indispensable value of studying the history of
ideas, accordmg to Quentm Skinner, is that we leam "the distinction between what is
~~cessa:Y and what is the product merely of our own contingent arrangements"
( Mealllilg and Understanding in the History of Ideas," History and Theory 8: 52-53;
68 The Third Way
never arrive at the correct or true version of past events because of the
conceptual and cognitive baggage we bring from the present. However, it does
not follow from the fact that giving an account of something means placing it
under a description, that we cannot arrive at historical descriptions, for there
is no reason to believe that speakers of different languages have different
conceptual schemes, rather than different vocabularies. 14 Vocabularies, unlike
conceptual schemes, are not fixed, and even when these vocabularies are
written or inscribed, the meaning of the vocabulary is neither fixed nor
exhausted by these inscriptions. A vocabulary, after all, is a list of words and
meanings that are abstracted from the speech acts of individuals. Although one
learns how words have been used in order to communicate with other speakers
of one's language, words may be used differently or altered by individual
speakers, although if these words are assigned radically different meanings it
is extremely unlikely that one will be able to communicate with other speakers,
15
unless, of course, it is understood that one is speaking "poetically. " From
the fact that speakers have a vocabulary or language different from our own,
it does not follow that it is impossible to learn this vocabulary. This remains
true even in the case where we cannot observe these speakers directly-e.g.,
fifth- and ·fourth-century Attic Greek speakers-but must reconstruct it from
written records. 16
quoted by Rorty in "The Historiography of Philosophy: Four Genres," in Philosophy in
History, eds. Richard Rorty, J.B. Schneewind, and Quentin Skinner [Cambridge:
Cambridge University Press, 1984], 51). lt will be far easier to mistake our own
contingent arrangements for what is necessary or universal if we don't examine the
possibility that philosophers of other epochs did not conceive of, let along subscribe to ,
what we take to be necessary.
14. Davidson has pointed out that the very notion of a conceptual scheme as that which
gives "form" to the "content" of experience is unintelligible: "The Very Notion of a
Conceptual Scheme," in Donald Davidson, Inquines into Trnth and lnterpretation
(Oxford: Clarendon Press, 1984), 183-195.
15. If the poet's listeners or readers "get" something out of the poetry, if only the
desire to understand it, they will judge the poetry successful, or at least worth their
attention. What weight such judgments have will, however, depend on who these
listeners or readers are.
16. It is interesting to note that the earliest Greek inscriptions seem to be wa ys of
making the objects on which they appear "speak": the Greek inscription on the Mantiklos
Apollo translated into English reads "Mantiklos dedicated me to Apollo the far darter;"
the inscription on the Diplyon oionchoe says: "for he who all the dancers sports most
playfully." Por discussion of these inscriptions see, L. A. Jeffery, The Local Scnpts of
Archaic Greece (Oxford: Clarendon Press, 1961); Kevin Robb, "The Dipylon Prize
Graffito," Coranto 1, no. 1 (1971): 12-14.
Por a discussion of the difficulties in trying to establish what would be true of a
society that is nonliterate from its later written records. see Eric Havelock, The Muse
Neither Published Nor Perished 69
lt is, of course, Havelock's claim that the notion of philosophical discourse,
as well as the notion of the mind that employs it, is "created by the use of a
vocabulary," a vocabulary that depends on a quite specific forro of life, viz.,
the intellectual revolution in fifth- and fourth-century Greece made possible by
alphabetic literacy. This extraphilosophical factor is not recognized as
providing for philosophical discourse insefar as one fails to see that the
"timeless" chara.cter of its language results from the fact that this language is
fixed or inscribed in an externa/ object. That it is written or inscribed is the
cause of its fixity and independence from a particular speaker or reader; the
chara.cteristics of fixity and independence are not properties of the language
itself, or of any "objects" that this language is supposedly about, or to which
it allegedly refers, and thus not reasons for believing that this language sta,tes
a necessary truth. How we know-through language that has been fixed by
being inscribed-is mistaken for what we know-a "timeless" proposition or
necessary truth. The inscription of language allows language itself to be the
object of study and, in turn, language becomes philosophical. How Ianguage
became philosophical was the central focus of Havelock's work; thus, it comes
as no surprise that I cannot find better words for describing this process than
those that he provided, and so I quote bere at length from Havelock's "The
Orality of Socrates and the Literacy of Plato":
There arrived a dawning recognition that language existed or could exist as a
subject of inquiry identifiable as separate from the speaker who wrote it. It could
become like the cosmos, a subject of analysis in its own right. Since it was now an
artifact, or visualized as one, it was seen to have components which when analyzed
became separate words. . . . Their ways or habits of linkage could be further
classified . . . . This language thing, now identified, was recognized as being of
two modes: the traditional one, rhythmic, fluid, imagistic, dynamic, a language of
human action and reaction for which the traditional label was and remained epas
and its plural epe . . . The new one was a counting, calculative use of language
akin to arranging units of speech in sorne new and unrhythmic order, a rationalized
language. Por this the preferred term became lagos . . . . A key characteristic of
this logos-speech began to be recognized as a matter of syntax, which crystallized
itself in the "is" statement . . . in sharp contrast to the narrativised statement,
which describes things as "becoming and perishing." If the verb "to be" was used
Leams to Write (New Haven: Yale University Press, 1986). Although written Janguage
is different from spoken language, both are a "way of knowing our way around in the
world," to echo Donald Davidson, although he does not discuss this distinction in this
~ontext. As such, neither is necessarily beyond us, even though we might have to engage
man extraordinary amount of historical study. Davidson characterizes language as a way
of knowing our way around in the world in "A Nice Derangement of Epitaphs," in
Inquines into Truth and lnterpretation: Perspectives on the Philosophy of Donald
Davidson, ed. Ernest LePore (New York: Basil Blackwell, 1986), 445-446.
70 Toe Third Way
to state the existence of principies, categories, relationships, ideals, and the like,
nice questions were bound to be raised about how exactly it should be employed,
how it should behave in this conceptual way, and with what distinctions; indeed,
what in fact this verb thus used signified, what its meaning was. Or did it have
one? Could its implementation be regarded as a negative factor rather than as a
positive one?
Toe new "is" syntax required a new type of subject-impersonal, non-active,
abstracted as we might say from any particular action or transient event. This called
for and encouraged a use of vocabulary which could isolate this subject in its
abstraction and its unchanging uniformity. Toe names Achilles and Odysseus
enjoyed a personal identification, as names to which were attached things that these
names did. But here is a new set of names doing nothing, but, in the eyes of their
users, very important names, more significant than Achilles or Odysseus, ephemeral
persons who were boro and died. Toe new names symbolized things like justice and
virtue which never died.17
Toe questions asked in and through this new syntax about these new objects
are familiar to us through the poem of Parmenides and the dialogues of Plato,
especially the Republic, lbrmenides, Theaetetus, Sophist, and Statesman. Toe
timeless character of the inscribed statement that allowed for the recognition
or invention of the propositional function, the "is" that though grammatically
tensed is logically tenseless, allowed also for the recognition or invention of
18
the concepts or categories that are related timelessly through the "is. " The
recognition or invention of these timeless concepts or categories permitted and
encouraged the recognition or invention of the activities of knowing them, and
of the agency that performs these activities of knowing them, and because of
this relation, must be, like them, timeless. 19 Havelock observes that existing
17. Eric Havelock, "Toe Orality of Socrates and the Literacy of Plato: With Sorne
Reflections On Toe Historical Origins of Moral Philosophy in Europe," in New Essays
on Socrates, ed. Eugene Kelly (New York: University Press of America, 1984), 76-79.
Further references to this work will be notes in parentheses following the citation.
18. Havelock, "Socratic Orality, Platonic Literacy," 79-80, provides a discussion of
how the third-person pronouns (he, she, it), which in English are used as subjects of
verbs to "identify the agent in separation from his action" but that are contained in the
verb in Greek, are employed to name, as it were, abstract concepts. "In order to identify
the very existence of these new abstract subjects of lagos, the simplest thing to do was
to add to them the third personal pronoun: 'the just, yes, it' (auto to dikaion). This had
the effect of isolating the word, and the idea behind the word, as an object of thought.
Toe device was then improved by extending the idiom into ' yes, it, according to it' (auto
kata auto) or even 'it, according to it, yes, it' (auto kata he auto)."
19. See Havelock, "Socratic Orality, Platonic Literacy," 81 : "The new effort to deal
with abstractions seemed to require separate signification, a terminology of what we
might call pure intellection. Thinking and thought as a conceptual process was recognized
as replacing, or at least as supplementing the activities of sensing, of noting, of looking
Neither Published Nor Perished 71
wo~ds lik~ "noein, to be ~ware of or sensible of, phronein, to have wits,
togzzes'.haz, to ~lly, skopem, to look at, epistasthai, to get on top of (in
roaste~ng a skill) we_re converted to the senses of thinking, reasoning,
analyz1~, ~derstand1ng scientifically, and the corresponding nouns,
phronesz~, epz_steme, nous, dianoia, (thought, science, mind, intellect) began
to turn_ mto mdexes of sheer thought and abstract intellection" (81). Toe
Homenc poems attest to a tradition that allowed the operations of human con-
sciousness to be objectified as separate things or agents which could speak to
or act on a person, and to whom a person could speak or on whom a person
could act. They were, as Havelock says, "only loosely differentiated as
between what we would call 'spirit,' 'will,' 'wish,' 'desire,' 'decision '
'sense,' 'heart, ' 'wit,' and 'mind.' But there they were, and there you wer;.
Presumably they were inside you or part of you in sorne way, and yet they
were not you" (81-82).
Thus, the logos-speech of the pioneers of philosophy recognized or invented
the self that c_ould be the obje_ct of its own Ianguage, and thus inspected in the
same way as ~ts _other new obJects of discourse- justice, virtue, beauty. In the
absence of extst1ng words for ~elf or personality, Socratic dialectic, according
to Havelock, made use of existing vocabulary. 20 In the dialogue between
Socrat:s and his interlocutors, "the ' me, ' the 'you, ' the ' him,' could become
reco~mzable, objects of discourse," as well as things that could be examined
and 1nspected (83). Moreover, since these pronouns also refer to the speaker
they nam~ him as the "the source of the logos and of the thoughts that ar~
produced m the course of exercising the logos" (83). The name for the agent
th~t ~as the source of the logos yet somehow separate from it also carne from
exisung vocabulary. The psyche known from at least Homer onward as the life
for~e, ~~t traditionally a thing that thinks or speaks, nor the consciousness of
an md1v1dual man but what animates him as long as he is alive, became the
self. To quote Havelock:
at, of perceiving and feeling."
20. Havelock argues that Socratic dialectic accomplished this by, once again, using the
Greek pronouns that we would describe as "understood" in the verb form· "but 'me •
'you,' •~.' symbolizing a person considered as object of action or spee~h. id:ntified
the person m separation from his ac_tion. In constructing a speech which could recognize
the speaker as a separate personality, you could begin with the pronominal objects: I
speak ª. lagos conceming 'me': you speak a lagos conceming 'you.' But this is
nonsens1cal: . how c~~ ' you' do something to 'you.' Toe second 'you' has to be thought
~f as_ a~ obJ~ct, d1stmct from the first 'you.' So, it became objectified by the same
~gu1st1c dev1ce that was used to fonnulate the 'it yes it. • Toe third personal pronoun is
, dded to the_first personal, and the resultant symbol becomes 'me yes him' (eme auton),
~ou yes him'(se auton), 'him yes him' (he auton)" ("Socratic Orality Platonic
L1teracy," 83). '
72 Toe Third Way
since it is this "me myself" which develops the method of the logos and the power
of intellection, it follows, in the case of the human being, that to think is to be alive
and to be alive is to think. The true virtue of the psyche-its function-is to think
or to produce thoughts. To stop doing this is to stop living or at least to be living
a half life like a sleepwalker (87).
Toe "speechless ghost" of Homer becomes the self from wbich pbilosopbical
language originates until Ryle reasserts its ghostly status in bis characterization
of mind as "the ghost in the macbine. "21
Toe notion of the "ghost in the macbine" is intertwined with the idea that
"theorising is the primary activity of minds, and is a private, silent, or interna[
operation" (27, emphasis added). From this follows the "intellectualist
doctrine" in wbich intelligence is defined in terms of truth rather than the
other way around, that is, intelligence is conceived of as the apprehension of
fixed truths (27). What this doctrine ignores, according to Ryle, is that
theorizing is an activity people learn: "the trick of talking to oneself in silence
is acquired neither quickly nor without effort: and it is a necessary condition
of our acquiring it that we should have previously learned to talk intelligently
aloud and have heard and understood other people doing so. Keeping our
thoughts to ourselves is a sopbisticated accomplishment" (27).
Toe notion that theorizing is a private, silent operation also supports the
belief that knowing how to theorize is a matter of "knowing that" because it
requires that one observe rules or apply criteria. On the basis of this
assumption, it is inferred (wrongly) that any activity that is to be described as
intelligent must be preceded by
an intellectual acknowledgement of these rules or criteria; that is the agent must
ftrst go through- the interna! process of avowing to himself certain propositions
about what is to be done ("maxims," "imperatives," or "regulative propositions"
as they are sometimes called); only then can he execute his performance in
accordance with these dictates. He must preach to himself before he can practice
(29).
This intellectualist legend is, of course, identified with Plato as the author
of the philosopbical texts that appear throughout the dialogues. However, even
if viewed as representations of philosopbical speech, the dialogues appear to
be attempts to make explicit the rules and criteria that one must know in order
to theorize about ideas like justice or piety or knowledge. But to say that these
rules and criteria must be made explicit through philosopbical discussion is not
the same thing as saying that knowing how to theorize requires that one first
21. Gilbert Ryle, The Concept of Mind, esp. 11-24. Further references to this volume
will be noted in parentheses following the citation.
Neither Published Nor Perished 73
k:now tlult these rules or criteria are the necessary conditions of theorizing, nor
is it to say that the agent has to "preach to himself before he can
practice"-that he must "go through the intellectual process of avowing to
himself certain propositions about what is to be done before he can execute his
performance." It is true, of course, that in the Theaetetus Socrates character-
izes thinking (dianoeistluli) as a talk (logon) that the soul has with itself (189E-
190B). Learning pbilosophy, if it is learning how to think, is also learning how
to talk, even if this talk is one that the soul has with itself, granted, of course,
that it has previously learned to talk intelligently aloud by hearing and
understanding other people doing so. Learning how to talk intelligently entails
learning how to reason, and neither is this activity, at least initially, a private,
silent, or interna! operation. "Underlying all the other features of the
operations executed by the intelligent reasoner," Ryle observes, is
the cardinal feature that he reasons logically, that is, that he avoids fallacies and
produces valid proofs and inferences, pertinent to the case he is making. He
observes the rules of logic, as well as those of style, forensic strategy, professional
etiquette and the rest. But he probably observes the rules of logic without thinking
about them. He does not cite Aristotle's formulae to himself or to the court. He
applies in his practice what Aristotle abstracted in his theory of such practices. He
reasons with a correct method, but without considering the prescriptions of
methodology. The rules that he observes have become his way of thinking, when
he is taking care; they are not externa! rubrics with which he has to square his
thought (48).
One learns this sort of thinking by listening to those who are engaged in
such c01wersations and then by engaging in them oneself. Plato's dialogues
represent such speech in order that their listeners might learn to speak and
think philosopbically, for there appears to be no other way in which they could
learn to do this. 22 Whether or not Ryle was right about the individual
occasions and specific audiences for which the dialogues were composed is
still a matter of controversy, but it is hard to deny bis claim that Plato saw that
the way to teach people how to think and speak pbilosophically was to
compose dramatic examples of such discussions that they could hear and in
wbich they could, at least vicariously, participate. Speech is the example par
excellence of Ryle's "knowing how;" learning how to speak is clearly not a
matter of first learning the rules and then applying them with the end
22. This observation does not commit one to the esoteric position, though this
composition is compatible with this observation. See Klagge, Methods of Interpreting
Plato and his Dialogues, who remarks that "if the esoteric view of Plato were correct,
then the dialogues would constitute something more akin to philosophical exercises for
initiation into the kinds of problems and the style of thinking that interested Plato" (8).
74 Toe Third Way
consequence being that one correctly speaks a language. Even m?re
importantly, the rules for speaking are not fixed, no matter how much füm~
23
language through inscription might lead us to think otherwise. Thus 1s
speech also the example par excellence of Ryle's po~nt _that i°: contrast to
"habitual practices in which one performance is a rephcat1on of 1~ pred~ces-
sors, it is the essence of intelligent practices that one performance 1s mod1fied
by its predecessors. Toe agent is still learning" (42). . . .
That the intellectualist legend fails to recognize that ne1ther thinking nor
talking is initially a silent, private, or internal operation is, perhaps, the result
of language itself becoming the object of study once it is has been inscribed.
Theorizing about language may thus take the form of arriving ata knowledge
of the rules or criteria on which the ability to communicate is [thought to be]
based, as well as answering the sort of questions that Havelock noted were
raised by the discovery of the propositional syntax. But viewing written
language as simply a token of speech is misleading, for written language,
though "fixed" in sorne enduring substance, is ambiguous, Thomas Cole notes,
in a way that speech is not. 24
23. This point has been driven home by Donald Davidson, who after decades of
studying the philosophy of language has concluded that the ability to communicate by
speech is not a matter of mastering what we usually mean by a "language." Davidson's
conclusions are worth quoting at sorne length:
It is only when we look at the structure of this ability that we realize how far we
have drifted from standard ideas of language mastery. Por we have discovered no
learnable common core of consistent behaviour, no shared grammar or rules, no
portable interpreting machine set to grind out the meaning of an arbitrary sentence.
We may say that linguistic ability is the ability to converge on a passing theory
[i.e., a theory that the hearer actually uses in understanding the speaker, anda
theory that the speaker intends the hearer to use] from time to time ... But if we
do say this, we should realize that we have abandoned not only the ordinary notion
of a language, but we have erased the boundary between knowing a language and
knowing our way around in the world generally. Por there are no rules for arriving
at passing theories, no rules in any strict sense, as opposed to rough maxims, and
methodological generalities. A passing theory really is like a theory at least in this,
that it is derived by wit, luck, and wisdom from a private vocabulary and grammar,
knowledge of the ways people get their point across, and rules of thumb for
figuring out what deviations from the dictionary are most likely. There is no more
chance of regularizing or teaching the process of creating new theories to cope with
new data in any field- for that is what this process involves (Donald Davidson, "A
Nice Derangement of Epitaphs," 445-446).
24. Thomas Cole, The Origins of Rhetoric in Ancient Greece (Baltimore: Johns
Hopkins University Press, 1991), 44. Further references to this work will be noted in
parenthesis following the citation.
Neither Published Nor Perished 75
If a written message is much less subject to distortion when a series of
transmissions is involved, any single transmission by means of writing involves
certain ambiguities from which oral communications are largely free. It usually
occurs apart from any situational context, without the clarification that intonation,
phrasing, gesture, and other aspects of delivery bring, and without the presence of
the transmitter to explain, rephrase, and repeat in an effort to minimize or correct
misunderstanding.25
Precisely because speech is public, because the "message" is delivered with
intonation, phrasing, gesture, and other aspects of delivery that clarify its
meaning, because the speaker is present to explain, rephrase, and repeat in an
effort to minimize or correct misunderstanding, one can learn how to speak
intelligently or to speak philosophically. Philosophy must be public discourse,
i.e., speech, before it becomes private, and philosophical thinking, like any
other intelligent practice, is one in which the agent is still learning because
each performance is modified by its predecessors. Thus, we make a mistake
when we fail to see the difference between speech and written language, tak:ing
written language as simply a token of speech, and also when we see
philosophy as a type of language, assumi~ that it really matters not whether
the language is spoken or written.
In the Phaedrus, Socrates claims that it does matter, that written
communication is inferior to speech. Toe majority of Plato's dialogues are
representations of philosophical speech, discourse in which the speakers are
present and indeed are required by Socrates to explain, rephrase, and repeat
in an effort to minimize or correct misunderstanding. 26 Toe notion that being
sophos consists in having expertise in a k:ind of discourse was a familiar one;
as Bruno Gentili remarks: "it is precisely in such terms-as expertise in public
discourse, whether poetry or prose pertaining to history, religion, or the nature
of the physical world-that the activity of the 'wise man' (sophos) is conceived
in Greece from the earliest period down to the end of the fifth century. " 27 In
the fourth century, being a philosophos might have meant having a passion for
the kind of discourse that one best has with one's own soul, but even this kind
25. Cole acknowledges Derrida's On Grammatology as launching this line of inquiry
(Origins of Rhetoric, 44).
26. This may also explain why the Theaetetus, Sophist, and Statesman resemble the
early dialogues in returning to the Socratic elenchus, though in the latter two dialogues
it is not Socrates who is doing the speaking. Of course, if one does not accept the
division of the dialogues into the early, middle, and late periods, and the alleged
systematic development of Plato's philosophical thought-both of which are becoming
increasingly problematic-this feature of the Theaetetus, Sophist, and Statesman may not
require explanation.
27. Bruno Gentili, Poetry arui its Public, ed. Thomas Cole (Baltimore: Johns Hopkins
University Press, 1989).
76 Toe Third Way Neither Published Nor Perished 77
of discourse must first be learned in public. to all who love sophia, but also knowing how and what to love. What is
What one learns in the public discourse that is philosopby is not simply how happening in the process of learning philosopby is the education of desire.
to talk and think in a certain way. If the Platonic Socrates is meant to tell us I submit that it is the identification of philosopby with "k.nowing that p"
anything, it seems to be that it is not simply sophia, but also the passion for where p stands for a set of philosophical texts that leads contemporary ironists
sophia that enables the philosopher to effect change among those who are not like Ro~ to believe that such transformations are more likely to be
yet sophos, including himself. Toe way to do this, Socrates shows us, is first accomphshed by works of art than by philosopby. It would seem that this
to teach people that they must have doubts about their claims to knowledge, passio~ for p~osopby, a passion that transforms the philosopher, is difficult,
or what Richard Rorty calls their final vocabulary, i. e., if not 1mposs1ble, to convey in a philosophical text, a fact that has not been
Jost on the contemporary ironist. Still, this passion can be captured in Plato's
the set of words which they employ to justüy their actions, their beliefs. and their
lives . .. the words in which we fonnulate praise of our friends and contempt for
epistemological dramas, which, given Republic ll, III, and X, might strike the
our enemies, our long-term projects, our deepest self-doubts and our higbest hopes contemporary philosopher who writes in the shadow of Freud and Nietzsche
... words in which we tell sometimes prospectively and sometimes retrospectively, as a t~agic ir?ny, thoug? Plato's Socrates suggests that there is, finally,
the story of our lives. 28 something COllli~ about philosopby's dependence on drama in which he appears
~ the hero. ~es1des, the~e dramas are not just any acts of mimesis conveying
Coming to have such doubts is becoming what Rorty calls appropriately JUSt any pass10n; a pass10n for philosopby comes from neither the epics of
enough, an ironist. Toe contemporary ironist differs, of course, from Socrates Homer nor the tragedies of Aescbylus but the dialogues of Plato.
the original ironist, but resembles him in a crucial way: "the ironist-the
person who has doubts about his own final vocabulary, his own moral identity,
and perhaps his own sanity-desperately needs to talk to other people, needs
this with the same wgency as other people need to make love" (73).
Rorty's ironist needs philosophical talk with the same urgency as other
people need to make love because such talk allays her anxieties about who and
what and where she is; Socrates needs philosophical talk because that is how
he makes love-as Alcibiades shows us in the Symposium. To raise doubts
about one's final vocabulary is a dangerous business unless one finds herself
transformed or, at least, capable of being transformed, through the very same
activity that generated the doubts in the first place. What Socrates, the expert
in matters of love tells us-and Socrates' only claim to expertise is made about
love-is that it is through philosopby that one is transformed into desiring what
is good and fine and noble, and that one who loves such things might become
good and fine and noble by virtue of pursuing this desire. Toe fate of
Alcibiades discourages the thought that such a transformation occurs easily; the
fact of Plato writing the Symposium with its comic rather than tragic vision
shows that it can happen nevertheless.
Philosopby, as the Platonic Socrates practices it, meets Ryle's
characterization of an intelligent practice, one in which the agent is still
learning because each performance is modified by its predecessors. What one
is learning, however, is not just how to talk and to think philosophically, not
just the candidates for the "final vocabulary" that belongs not only to him but

28. Rorty, Contingency, irony, and solidarity, 73. Further references to this work will
be noted in parenthesis following the citation.
 II

Philosophy and Rhetoric:

Between Persuasion and Proof

 5

Socrates' Rhetorical Attack on Rhetoric

David Roochnik

Toe standard story is this: Plato the philosopher is the mortal enemy of
rhetoric, which he condemns on both epistemic and ethical grounds. Toe
philosopher denies that rhetoric is real knowledge: since it is not a legitimate
techné, its teachers are frauds. Furthermore, he accuses it of being ethically
bankrupt: it aims only to gratify the ignorant many and enhance the power of
its expert practitioners. By contrast, Plato champions a purified rationalism and
aims for Knowledge of The Changeless Good. His Philosopher is impervious
to and contemptuous of the ephemeral contingencies of ordinary life. The line
between Platonism and rhetoric, according to this story, could not be more
sharply drawn. 1
A pivotal text invoked by proponents of the standard story is Gorgias 462a-
466a, where, in bis argument against Polus, Socrates seems to level both these
chaiges against rhetoric. It is, he says, not a techne, but only an empeiria and
tribe (462c, 463b), a knack based on experience. Furthermore, as a species of
flattery that aims for the production of gratification and pleasure (462b),

l. Brian Vickers, in his In Defence of Rhetoric (Oxford: Oxford University Press,

1988) is both a good representative of the standard story, as well as a useful guide to the
literature. There have, of course, been numerous rebuttals of the standard view. See, for
example, Livio Rossetti, "Toe Rhetoric of Socrates," Philosophy and Rhetoric 22 (1989):
225-238, for a representative view as well as a brief survery of sorne nonstandard
literature.
82 Toe Third Way
rhetoric is ethically degenerate as well (463b, 501b).
This passage has galvanized the supporters of rhetoric for centuries and so
the two prongs of the attack bave each met with numerous responses. In
antiquity, Cicero, Quintilian, Philodemus, and Aristides ali explicitly offered
vigorous rebuttals to the Socratic charges, and as recently as 1988 Brian
Vickers did the same in bis forceful defense of rhetoric. 2
I will argue that the standard story is flawed because Gorgias 462a-466a has
been read out of context. More specifically, it is flawed because the standard
reading fails to take into account the fact that tbe Gorgias, like virtually ali of
Plato's work, is a dialogue. If this passage is considered within the larger
framework of the dialogue as a whole, it can be shown that Plato's hostility
to rhetoric is ambiguous, and the line drawn between rhetoric and pbilosophy
is not nearly as sharp as most commentators think.
l.
First, to the locus classicus of the "old quarrel" between rhetoric and
pbilosophy, the debate between Socrates and Polus. Polus begins by
complaining that Socrates has unfairly refuted bis teacher, Gorgias. Gorgias,
Polus says, was shamed into admitting that "the rhetorical man" both knows
and teaches what is just and fine and good (461b), and this is what led bim to
contradict bimself. After ali, who could possibly deny that be both knows what
is right and fine and good, and is willing to share such knowledge with those
who do not? Polus's point is that if Gorgias had refrained from making such
a claim about teacbing "virtue," he could have saved bimself from refutation. 3
In a formal sense, Polus's diagnosis of the argument is accurate, for it was
Gorgias's admission that he could and would teach what is right and fine and
good (at 460a) that eventually led bim to contradict bimself. Toe contradiction
arises because Gorgias bad earlier stated (456d-457c) that rhetoric is like
boxing; it is indifferent to, and not responsible for, the uses to wbicb it is put
by its students. Since "use" represents the domain of human value, rhetoric is
"value-neutral," a statement that directly contradicts what Gorgias says at
460a.
As we will see in Part II below, wbile Polus may be right in a strictly
formal sense in locating the contradiction, he is far from diagnosing the real
2. See Cicero. De Oratore, Book 1, Quintilian, Institutio, 11.xviiff., Aristides, In
Defense of Rhetoric, each of which has been published in the Loeb series. Por
Philodemus, see The Rhetorica of Philodemus, trans. Harry Hubbell (New Haven: Yale
University Press, 1920). See also Brian Vickers, In Defence of Rhetoric.
3. Toe historical Gorgias apparently did not claim to teach virtue. See Meno, 95c.
(Throughout I have used Burnet's Oxford edition of Plato. Translations are my own.)
Socrates' Rhetorical Attack on Rhetoric 83
cause of bis teacher's demise.
After persuading him to refrain from long speeches and to engage instead
in question and answer, Socrates invites Polus to ask the first question. Polus
complies: "What do you say that [rhetoric) is?" he asks (462b). Rather than
answering, bowever, Socrates reformulates the question: "Are you asking what
tecbne I say that it is?" (462b). Polus agrees to the reformulation.
It is important to note that, precisely as he did at the very beginning of the
dialogue (447c-448b), it is Socrates who here seems to assume that rhetoric is
a techne (a critica! point I will discuss at length shortly). Here with Polus,
however, Socrates seems to retract the assumption immediately upon asserti~
it: rhetoric is not a techne at ali, he says, but merely an empeiria that produces
gratification and pleasure (462c). In order to clarify, Socrates compares
rhetoric to cooking. It too is not a techne but an empeiria for producing
gratification and pleasure (462d). Both rhetoric and cooking, neither of wbich
"has any share in what is fine," (463a), belong (together with sopbistry and
cosmetics) to the category of "flattery. "
[Rhetoric] seems to me, Gorgias, not to be a technical practice (epitedeuma
technikon), but instead a quality of an intuitive and manly psyche, one that is clever
by nature in dealing with human beings. In general I call it flattery (463a).
What exactly is "flattery?" To explain, Socrates offers the following
scheme:
BODY
Apparently Good Condition Good Condition
Flattery Therapeutic Techne
Cooking Cosmetics Medicine Gymnastics
PSYCHE
Apparently Good Condition Good Condition
Flattery Political Techne
Rhetoric Sopbistry Justice Legislation
. Socrates asserts, but does not prove, that both body and psyche bave an
mherently good condition (euexia: 464a2) that can be securely known and
towards wbich body and psyche each can be moved. Four technai- medicine
and gymnastic for the body, justice and legislation for the psyche-know the
goo~ .con~iti~n of their objects and can move them toward it. Toe pairs
medicme/Just1ce and gymnastic/legislation differ in that the former correct
deficiencies and return their objects to their euexia, wbile the latter create
abidingly good and stable conditions. In contrast to these genuine technai,
flattery:
84 Toe Third Way
perceives, I do not say that it knows but that it intuits (srochasmené), that there are
four branches [of techne] that always exercise their objects, namely the body and
the psyche, toward what is best, and then it [f!attery] divides the parts, pretend~
that it is that which it has insinuated itself into. lt does not care at all for what 1s
best, but it always hunts out foolishness by saying what is most pleasant an~ it
deceives those who are foolish by making it seem as if it is the most valuable thing
of all (464d).
A cook, for example, may pretend that he is a doctor and say ~at ~ cookie
is good for the stomach.4 Toe cookie is delicious and hard to res1st. Srnce the
cook merely gratifies his audience without contributing to the actual excellence
of their bodies, he is like the rhetorician.
Toe entire scheme can be expressed "geometrically": as cosmetics is to
gymnastic, so sophistry is to legislation, and as cooking is to medicine, so
rhetoric is to justice (465c). With such a scheme in place, Socrates can
condemn the forms of flattery as shameful.
F!attery, I call it, and I state that such a thing is shameful, Polus- for l'm sayin~
this to you-because it intuits what is p!easant without the best. And I say ~at ~t
is not a techne, but an empeiria because it has no account (logos) of the things 1t
applies, what sort of nature they are, and so it cannot state the cause (aitia) of each
thing. I refuse to call anything that is irrational (alogon) a techne (465a).
Toe intent of the passage seems obvious: to condemn rhetoric. Toe
particulars of the charge, however, are surprisiq?;ly ambigu?u~. First, exactly
why, according to Socrates, is rhetoric shameful? Because 1t 1s not a techne.
But why is it not a techne? Because it intuits what is pleasant and does not
know the appropriate eue.xia of the psyche. But where does the emphasis in the
charge líe? On the fact that rhetoric aims for the pleasant without the good, or
on the fact that it is only "stochastic," i.e., epistemically deficient? Even the
first sentence of the passage cited above is ambiguous: rhetoric is shameful
because it "intuits" (stochazetal) the pleasant without the good. But is it
shameful because it merely intuits, or because its end is pleasure and not the
good? ..
Toe key point is this: the locus classicus of Plato's attack on rhetonc 1s tw~-
pronged and comprises both an epistemic and ethical dimension. 5 There 1s,
4. It is hard to imagine a cook actually doing this. Terence Irwin, in his commentary
on the Gorgias (Oxford: Clarendon Press, 1979), says, "But surely Socrates is wrong to
say that cookery pretends to offer healthy food" (134).
5. Toe two prongs of Socrates's objection to rhetoric became enshrined, I believe, in
the standard Stoic definition of a techne (as reported by Sextus Empiricus, Against the
Professors, 11.10): "every techne is a body (sustema) consisting of items of knowledge
which are mutua!ly cohesive and having reference to one of the ends which are useful
Socrates' Rhetorical Attack on Rhetoric 85
however, no obvious connection between the two prongs. Toe latter pan of the
passage cited above (465a6-8) seems strictly epistemic (and quite
Aristotelian):6 the various branches of flattery are not technai because they
cannot give a logos of the aitia. Does this inability make rhetoric shameful?
Not in itself, for that would imply that anything that is not a techne is
shameful. A child's ability to walk is surely not a techne, but it is hardly to
be blamed for that. Why, then, does rhetoric's "irrational" (alogon: 465a6)
and therefore "nontechnical" character make it shameful? Presumably because
the rhetorician, unlike the child, líes: he pretends to possess a techne while in
fact he only has an "image" (eidolon: 463d) of it.
Second, rhetoric is condemned on ethical grounds for aiming at the pleasant
rather than the good. But is this aim in itself a matter for blame? Only given
certain assumptions. If, for example, the rhetor aims only for pleasure, and if
pleasure is totally divorced from the good, then he could be fairly blamed.
But if "what is pleasantest over ali is also best over ali," then "if the rhetor
aims at what is pleasantest, will he not aim at what is best over all?"7 In other
words, Socrates does not make it clear why aiming for pleasure is in itself
shameful.
Third, Socrates seems to assume here that because rhetoric aims for the
pleasant rather than the good, it cannot be a techne. He seems to assume that
there can be no techne of or aiming for pleasure. 8 But why? Why can't there
be a techne of cooking or cosmetics whose goal is pleasure? Such a goal
would be quite different from that of medicine or gymnastic, but why can't the
cook or "cosmetologist" apply real knowledge?
Finally, Socrates's scheme asserts that all technai aim for the eue.xia of their
objects. But the eue.xia of the body, the goal and epistemic content of medicine
and gymnastic, seems quite different from that of the psyche, the goal of
legislation and justice. Medicine is concerned with the proper functioning of
the body and unconcerned with whether the well-functioning body is used to
perform just or unjust actions (see 456d). As such, it is strictly a value-neutral
affair. By contrast, the good condition of the psyche, and therefore the technai
of justice and legislation. are by definition value-laden. What, then, are we to
make of the single word eue.xia being used to label two apparently different
conceptions? Does Socrates simply assume that values are a matter of fact that
can be straightforwardly known? Does he think that the psychic eue.xia is, like
that of the body, something fixed and stable and thus easily apprehended by
in life." Toe translation is that of Jonathan Bames, in his "Is Rhetoric an Art?" DARG
NEWSLEITER 2 (1986): 5.
6. See Aristotle 's M etaphysics, 1.1 .
7. lrwin, Plato's Gorgias, 135.
8. See Aristotle's Nicomachean Ethics, 1152b18 for this point.
86 Toe Third Way
a techne? Or, conversely, does he think that every techne, even that aiming
only for the proper, but value-neutral, functioning of the body, is, or at least
aims for the, good?
To summarize: Gorgias 462a-466a, the locus classicus of Plato's putative
assault on rhetoric, raises several questions: what is the relationsbip between
being epistemically deficient and aiming for pleasure? Why can't there be a
techne of or aiming for pleasure? Is the euexia of the psyche as fixed and
stable and therefore as straightforwardly aimed at, as that of the body? In
gener~. there is a tension between the two prongs of Socrates's attack against
rhetoric, the ethical and the epistemic, that the passage itself does not
explicitly resolve or even address. . body. A question must thus be asked: even if the euexia of the psyche "writ
Perhaps there is an implicit resolution. Assume (as Socrates does m the
Philebus at 27e ff.) that pleasure is inherently indeterminate. This might be
true in two senses. First, people experience pleasure in an indefinite and
unpredictable variety of ways and second, pleasure and pain form a continuum
that can (in the language of the Philebus) always adrnit of "the more and the
less" (24a). Pleasure offers no fixed, stable, and epistemically reliable standard
at wbich to aim. Assume further that a techne is a rigorous form of knowledge
with a determinate subject matter.9 If so, it follows that there can be no techne
of or aiming for pleasure, and if rhetoric indeed aims to produce pleasure, it
cannot be a techne. 10
Such an implicit resolution, even if accurate, would only address one of the
ambiguities provoked by the passage: it would explain why there can't be a
techne of pleasure. But it would not explain why aiming for pleasure is
shameful. Again, presumably the reason that flattery in general and rhetoric
in particular are shameful is not simply that they aim for pleasure, but because
they líe: they pretend to be technai when in fact, because of the indeterminate
nature of their subject matter (i. e., pleasure), they are not.
It must be remembered, however, that (in the Gorgias at least) it is neither
rhetoric nor its representatives that do the lying: it is none other than Socrates,
the truth-loving pbilosopher bimself who, in the course of debating both Polus
and Gorgias, invites them, indeed pressures them, into making the claim that
rhetoric is a techne. This is a vital point to wbich I will return shortly.
9. This would have been a rather traditional sense of techne in Plato's time. See F.
Heinimann, "Bine vorplatonische Theorie der Techne," Museum Helveticum 18 (1961):
105-130.
10. E.R. Dodds makes this point in his commentary to the Gorgias (Oxford University
Press, 1959). Toe pleasant, he says, is not "in each case rationally determinable" (229).
J. C. B. Gosling, in his commentary on the Philebus (Oxford University Press, 1975)
puts the same point this way: "They [emperiai like rhetoric] fail to be technai because
no general account can be (or at least is) given of what pleases people, and so there are
no general canons for ensuring success" (153).
Socrates' Rhetorical Attack on Rhetoric 87
The implicit resolution offered above, namely that pleasure is indeterminate
and therefore cannot be a legitimate object of a techne, does not address
another implicit but significant question: is the euexia of the psyche itself
determínate? Since, according to Socrates's scheme, it is an object of a techne,
presumably it is. But, as mentioned above, this would imply that the value-
Jaden euexia of the psyche is as stable and as epistemically reliable as that of
tbe body. At the very least, this assumption requires defense. 11
There is a related problem. As Dodds points out, there is no place for the
individual psyche in Socrates's scheme. 12 This is all the more puzzling since
the individual psyche would seem to be the natural correlate to individual
Jarge," i. e., the polis, is determinate, is that true of the individual psyche as
well? In Aristotelian terms, can both ethics and politics be rendered technical?
Gorgias 464a would seem to assert yes to the latter. But can there be a techne
whose subject matter is the euexia of the individual psyche? Would such a
techne be pbilosophy?
It should be clear that there is considerable ambiguity in Socrates's charges
against rhetoric. Is this a consequence of Plato's carelessness? 13 Or can the
ambiguities of the passage be shown to fit coherently into the larger pattern of
the dialogue as a whole? I think the latter.
2.
Consider one of the most striking but least commented upon facts about the
Gorgias: wbile it is true that Socrates denies that rhetoric is a techne in bis
argument against Polus, he asserts that it is a techne in bis argument against
Gorgias. In fact, this assertion proves to be essential to bis argument.
At the outset of the dialogue, when Callicles invites bim to listen to one of
Gorgias's famous displays, Socrates responds, "I want to find out from bim
what the power of the man's techne is and what it is he professes and teaches"
(447c). Callicles says that Gorgias will answer this question; after ali, the great
rhetorician boasts that he can answer "every" (hapanta) question. With this
boast, the principal issue discussed above is broached: Gorgias can answer
every question; bis subject matter is somehow "everything" and as such is
11. It is in the discussion between Socrates and Callicles that, I believe, the answer
to the question of the determinateness of the psyche emerges. In a future paper, I will
develop this theme.
12. Dodds, Plato's Gorgias, 227.
13. Vickers relies heavily on lrwin, and he relishes every opportunity to criticize
Plato. See In Defence of Rhetoric, 83-147.
88 Toe Third Way
indeterminate. 14 Socrates, however, refuses to allow Gorgias to rest content
with such a claim; he demands that he identify a specific subject matter for ~s
profession of rhetoric. This demand, which shapes the ~nti~e argument _w1th
Gorgias, is a consequence of Socrates's initial charactenzatlo~ of rh~tonc as
a techne a move that he makes early in the dialogue. Cons1der this: when
instructi~ Chaerephon as to what he should ask Polus, Socrates ~ells hi~ to
find out "who he [Gorgias] is?" Chaerephon does not understand this que~t1on.
Socrates illustrates by example: if you were to ask a shoemaker who he ~s, he
would anwer a "shoemaker." Now Chaerephon understands and he dut1fully
asks. "if Gorgias happened to be knowledgeable abou~ the techne_in ':'hich his
brother Herodicus is knowledgeable, what would we nghtly call him? (448b).
Toe answer is a doctor.
With these remarks, Socrates foists the "techne analogy" upon the
rhetorician. In other words, he implies that as the doctor is to medicine, so
Gorgias is to X. But what is the X? Rhetoric. But what i~ rhetoric? _Medicine
has a specific subject matter, the health of the body, which_ determme~ what
medicine is and renders it unique among the technai. Accordmg to the dictates
of the analogy, then, rhetoric must also be able to identify such a s~bject
matter. Socrates's strategy is now clear. He wants Gorgias to identlfy a
particular, that is, determinate, subject matter for rhetoric. Tow_ar~s that e~d
he repeatedly asks variations on the question, "About what (pen fl: 449dl) 1s
rhetoric?"
Gorgias's first answer is, it is "about speeches" (logoi: 449d8._) Socrates,
however, immediately negates this response. After all, ~ther technai also make
speeches. Toe question is, speeches about what? Gorg1as•~ next three_ anwers
are (1 paraphrase), rhetoric has a subject matter that, unl1~e others, 1~ p_urely
a matter of logos (450b); it is about the greatest of all affa1rs (451d); 1t 1s the
ability to persuade (452e). In each case, Socrates shows th~t Gorgi_as's ~ers
fail to describe rhetoric in terms that are sufficiently umque to 1dent1fy and
distinguish it. Other technai, notably mathematics, also have subjects that are
a matter of pure logos (450d); all technai seem to evaluate themselves as
positively as Gorgias evaluates rhetoric (452a-c); all technai, insofar as they
teach, also persuade (453d-454b).
Toe rhetorician resists delimiting his subject matter. This was foreshadowed
in Gorgias's boast that he can answer every question: the ability of th~ rhetor
is unrestricted. His is a power that embraces within it ali powers; he 1s able,
for example, to persuade those ~touched by the arguments .~f no~al
technicians (456a-c). As such, rhetonc seems to be, says Socrates, something
daimonic" (456a6). But if the rhetor has this daimonic power to talk about
14. Quintilian comments favorably upon Gorgias's boast: see 11.xxi.21. Book I of
Cicero's De Oratore makes much the same claim.
Socrates' Rhetorical Attack on Rhetoric 89
everything to everyone, how can rhetoric be a techne? He who possesses a
techne is a technites. an expert, and no one can be an expert in everything. It
is because of this need to identify himself as a expert, i. e., it is because of the
techne analogy foisted on the argument by Socmtes, that Gorgias is eventually
pressured into claiming that the object of his expertise is "the just and the
unjust" (454b; see also 460a).
As discussed in Part I, this claim dooms him because Gorgias has already
stated that rhetoric is like boxing. If a teacher teaches a student how to box
and the student then uses his knowledge unjustly, neither the teacher nor th~
techne ought to be blamed, for they are indifferent to and not responsible for
the good or bad applications of the techne. In this sense, rhetoric, as a techne
is application- or value-neutral, and Gorgias has contradicted bimself. '
As also mentioned in Part I, when Polus claims that Gorgias was shamed
into asserting ~at he teaches virtue, he was only partially correct in diagnosing
the cause of his teacher's demise. Toe real impetus bebind the refutation as
we can now see, is Socrates's use of the analogy, as medicine is to the he~th
of the human body, so rhetoric is to an X that is as determinate as health. In
other words, the presuppositio,:i that rhetoric is a techne is essential to the
refutation of Gorgias. Toe opposite assumption. however, namely that rhetoric
is not a techne, governs the refutation of Polus. Which is it? If in fact
rbetoric is not a techne, what sense can we make of the refutation of Gorgias'
wbich presupposes that it is? '
One answer is that the Socrates/Gorgias debate constitutes a reductio ad
absurdum. From his initial assumption that rhetoric is a techne, Socrates
generates ~ contradiction. This leads to the negation of the assumption, i. e.•
the propos1tion that rhetoric is not a techne. wbicb then becomes the first
premise Socrates uses in bis argument against Polus (as well as bis sincere
belief). 15
This i~ ~ very plausi?le in~erpretation, but does not fully explain wby
Socrates 1ns1sts that Gorgias claim to possess a techne. Why does he foist wbat
he ~kes to be a _false statement-rhetoric is a techne-on his opponent? Toe
obv1ous answer, 1.e., the one that straightforwardly accounts for the "action"
of ~e dialogue, is in order to refute him. Socrates demands of Gorgias, who
claims that he can answer "every" question, that rbetoric be formulated as a
techne with a specific subject matter. Doing this forces Gorgias to identify
_15. This was suggested to me by Professor Francisco Gonzalez. Another resolution to
this problem is to describe Socrates's argu.ments against Gorgias and Polus as a
destruc~iv~ dilemma. Toe first assu.mes that rhetoric is a techne, and ends in a
contrad1ct10n. Toe second assu.mes that rhetoric is not a techne, and ends in a criticism
of rhetoric. Rhetoric is either a techne or it is not a techne; in either case, it is in
trouble.
90 Toe Third Way
precisely what it is he professes to know. Once this identification is made,
Socrates can examine and evaluate bis opponent's claim. At the very least,
then, it is immensely helpful for the pbilosopher if the rhetorician asserts that
he has a techne. To reiterate, once Goi:gias does this, Socrates can refer to the
specific subject matter that belongs to the techne ("the just and fine and good"
[461b]), and then use it as a standard by wbich to measure and finally refute
bis opponent's knowledge claim.
There is more bebind Socrates's perplexing use of the techne analogy than
bis desire to refute bis opponent. Toe above, however, should suffice to show
that bis refutation of rhetoric is problematic. By using the techne analogy
against Goi:gias, he seems to invite the rhetorician to make a statement that_he
knows is false, and to do so because it is to bis advantage. When argwng
against Polus, Socrates follows an entirely different strategy. Simply put, then,
Socrates speaks differently to different people. It might even be said that he
wields the techne analogy as somethiog like a rhetorical device. If this is
correct, then the confidence with wbich the standard story draws a sharp line
between Platonic pbilosophy and rhetoric, and declares them to be diametrical
opposites, is misplaced.
There is another quite tempting way in wbich proponents of the standard
story often draw the distinction between rhetoric and philosophy: the
pbilosopher aspires to transform knowledge of the human good into a techne
and thereby to render human affairs determinate and rationally controllable.
By contrast, the rhetorician affirms, even celebrates, the indeterminate, the
contingent, and the unpredictable. Like an open hand, he welcomes all comers
and applauds ali the many goods embraced by ali the many human beings; he
is willing to talk about everything. Toe pbilosopher is like the closed fist,
grasping firmly-that is, technically-bis conceptual content, namely the
determinate good. 16
But does Plato actually recommend that moral knowledge be transformed
into a techne? Does he advocate the firm fist? This is hardly obvious in the
Gorgias. First, as evidenced by bis conversation with Gorgias, Socrates is not
quite firm even on the question, Is rhetoric a techne? If pbilosophy is (or
should become) a techne, then it should be able to identify that forro of flattery
that imitates it. Toe carpenter knows and will confidently assert that the cbild
playing with a hammer does not share bis techne. But against Goi:gias,
Socrates does not make an analogous denunciation of rhetoric.
Second, there is the glaring absence of the individual psyche in Socrates's
16. This famous image is Zeno's and is cited by Quintilian at 11.xx.7. Toe view
presented here is well articulated by Martha Nussbaum in "Toe Protagoras: A Science
of Practica! Reasoning," in The Fragility of Goodness (Cambridge: Cambridge University
Press, 1986).
Socrates' Rhetorical Attack on Rhetoric 91
geometrical scheme depicting the forms of flattery. If pbilosophy is or should
become a techne, presumably this is what its subject matter would be. Its
absence, therefore, is perhaps designed to cause the reader to wonder whether
the individual psyc_he can in fact be stabilized or rendered determinate enough
to become the subJect matter of a techne. 17 Perhaps this is true as well for the
psyche "writ lai:ge," i. e., the polis. E ven if it is present on the scheme as the
subject ~atter of legislation and justice, there is the tacit, but problematic,
assumpt1on that the euexia of (the political) psyche is as stable and as
straightforwardly apprehended as that of the body.
Third, in an important sense Socrates is similar to Goi:gias. While Goi:gias
offers ~ field, and presu~ably answer, ali questions, Socrates spends bis days
~andenng ~e ~ora, asking ali sorts of questions. It is, quite simply, very
clifficult to 1dent1fy Socrates's determinate field of expertise. It is, as bis
interlocutors well know, difficult to force Socrates to provide straightforward
answers evento the questions he bimselfraises (such as, Is rhetoric a techne?).
And what else is a techne if not a series of straightfoward answers?
Considerations such as these should militate against the standard view that
Plato conceived of knowledg_e (specifically, moral knowledge) as strictly
analogous to a techne, and that this conception is the critical difference
between him and the rhetoricians. 18 At the same time however these
considerations might seem to fly in the face of Socrates's sta~ling asse~ion at
521d: "I thi~ I -~ one of th_e_ few Athenians, if not the only one, who really
attempts (epzchezrem) the polincal techne." Two responses: Socrates does not
say that he actually possesses a techne, but that he "attempts" one. Techne
may well be a kind of ideal at wbich the pbilosopher aims, but to aim for is
not th~ same as to P?ssess. Second, I suggest that Socrates' assertion may well
be a ~nd of reducno. In o~er words, if, of al~ people, Socrates is the only
Atheman to attempt the poht1cal techne, that might be good evidence that no
~ne _actu~ly. possesses such a techne. Socrates, after ali, spends bis days
. wh1spenog m a comer ~ith three or four lads" (485d), rather than fighting
m the Assembly for the v1ews he advocates. In sum, it simply is not clear that
th~ stark and ~tandard opposition (embraced by pbilosophers and rhetoricians
al1~e)-rhetonc (bad [or good], indeterminate, nontechnical, open) versus
philosophy (good [or bad], determinate, technical, closed) is viable. As a
17; ~~dds fails to expl~in. the absence of the individual psyche. He only refers to
Plato s deeply held conv1ct10n that the basic task of the statesman is educational"
(Pla1o's Gorgias, 227).
18. Toe standard reading typically assigns this conception of moral knowledge only
to the early Plato. Aga~, it is ~-ell articulated by Martha Nussbaum in The Fragility of
Goodness. Por a competmg pos1t10n see David Roochnik, "Socrates's Use ofthe Techne-
Analogy," Joumal of lhe Hislory of Philosophy 24 (1986): 295-310.
92 Toe Third Way
result, then, far from being two manifestly different activities, rhetoric and
pbilosophy actually become quite difficult to distinguish. I propose that simply
understanding that this task is difficult is a significant advance beyond the
standard story. To conclude this paper, I suggest in somewhat general anct
speculative terms how, in the absence of the straightforward diametrical
opposition, the distinction between rhetoric and pbilosophy can be conceived.
A single concept has lurked bebind Socrates's refutations of Polus and
Gorgias: that of determínacy. Implied by the scheme used against Polus, the
point seems to be that since pleasure is the goal of rhetoric, and pleasure is
indetermínate, rhetoric cannot be a techne. It is precisely this feature of a
techne, namely that is has a determínate subject matter, that Socrates exploits
when he uses the techne analogy against Gorgias. There he attempts to identify
and thus make refutable Gorgias's otherwise vague and hence irrefutable claim
to teach the "daimonic" power of ali powers. In both cases, Socrates puts the
notion of techne, and therefore its foundational concept, determínacy, to good
use. Specifically, he uses it to refute bis opponents.
There is another sense in wbich the pbilosopher finds the techne analogy
useful. Against Polus, Socrates draws a geometrical scheme-as cooking is to
medicine, so rhetoric is to justice- not only to locate rhetoric, but also to
denigrate it by showing its place in a bierarchy. But in much the same way as
he does with the "divided line" of the Republic, by drawing such a
bierarcbical scheme, he exhorts bis listeners to seek its pinnacle. In short,
Socrates uses the analogy both to negate and affirm: to negate the claim of the
rhetorician to have a techne, and to exhort the uncommitted listener to seek
knowledge of the good. Neither use implies that the pbilosopher himself
possesses a techne with a determínate subject matter. Instead, they imply that
he thinks determínacy, the notion that undergirds bis use of the techne
analogy, is useful and good.
By contrast, the rhetorician's view of determínacy is almost reversed. On
the one hand, he praises indetermínacy. He boasts of bis ability to talk about
"everything," he praises doxa and democracy, he affirms contingency and
change, and insists that bis speeches are situated within the specific
circumstances from wbich they emerge. 19 On the other hand, he claims to
have so sufficiently gained specific expertise that he can teach students and,
very significantly, charge a large tuition. This claim implies that rhetoric is in
sorne sense a techne, wbich, after all, is the paradigmatic form of teachable
19. Neither Gorgias nor Polus actually makes many of these claims, but their
defenders regularly bave. An excellent exposition of this view of rbetoric can be found
in Stanley Pisb, "Rbetoric," in Doing What Comes Natural/y (Durbam: Duke University
Press, 1989).
Socrates' Rhetorical Attack on Rhetoric 93
20
1cnowledge. But how can the indetermínate "everything" be sufficiently
ctelimited to function as the subject matter of a legitimate and teachable techne?
This is, I would rugue, the fundamental epistemic problem that has never
ceased to haunt the rhetorical tradition. 21
Socrates claims no techne. He lives in the midst of indetermínacy,
wandering the agora, picking up conversations as he happens along. But he
praises determínacy and evaluates it positively; he is animated by the goal of
rigorous knowledge when he describes bimself as the only Athenian to
"attempt" the political techne. There is thus a tension within Socrates's attitude
wwards techne. This tension is, I suggest, extremely illuminating, so much so
that it can help to account for the fact that Plato chose the dialogue form to
express bis thinking.
As an imitation of characters both speaking and acting, a dialogue is infused
with indetermínacy and the irremediable particularities of the human speeches
and deeds that it imitates. Socrates does not, for example, offer a theoretical
refutation of rhetoric. Instead, he argues against Gorgias at a very particular
time and place. Because the dialogue is written thus, the reader can never
extract any single argument Socrates might happen to make and then declare
it "Platonic. " Toe argument may well be tailored to meet the dialogical needs
generated by the particular context in wbich it is spoken. A good example is
Socrates's vacillation on the question, Is rhetoric a techne? Against Gorgias,
the answer, at least initially, seems to be yes. Against Polus, the answer is
unambiguously no.
Socr~tes's P?Sitiv: evaluation of determínacy, clarity, intelligibility, and
techne 1s thus mextncably situated within the contingencies of the life that is
the dialogical context from wbich they are spoken. From within that turmoil
the pbilosopher makes determínacy and techne look good and seems to offe;
them as the final goal of inquiry.22 But Plato never allows that desire to be
completed. Simply put, he never writes a treatise or a techne. He only writes
dialogues.
Like ~e good rhetor, Socrates speaks differently to different people and
thereby h1mself manifests "an intuitive and manly psyche, one that is clever
20. See P. Heinimann, "Eine vorplatoniscbe Tbeorie der Techne," for this point.
21. Two examples of ancient rbetoricians treating this problem are Cícero in Book I
of De Oratore and Quintilian in Book 11.xvii-xxi of Institutio. Por a contemporary
u~?'1t_e, see Renato Barilli, Rhetoric (Minneapolis: University of Minnesota Press, 1989),
VW -XJ.
22. A crucial Iine for my interpretation is found at 508a, wbere Socrates claims that
the differencc between bim and Callicles is tbat the latter has failed to absorb the great
!esso~. of "geoi_netrical equality," namely tbat tbe world is informed by
mtellig1ble/dctermmate structure, tbe articulation of wbicb can become a meaningful
goal.
94 Toe Third Way

by nature in dealing with human b~ings. ~ _Dnlike th~ rhetorician,_ however,

socrates remains steadfast, first, in his pos1t1ve evaluatton of det~rminacy and,
in his denial that he himself possesses any determinate form of
second , . thi · ha
knowledge. In this paper, I have only begun to discuss how s tens1on s pes
the Platonic conception of philosophical knowledg~. What I hope I ha~e shown
is how this conception emerges from the confrontatton between_the ~hilosopher
and the rhetorician, a confrontation that has been badly oversimphfied by the
standard account given of it.

Apodeiknusthai, Dialegesthai, Peithein:

A Reconstruction of Plato's Methods
of Argument in the Phaedo

P. Christopher Smith

Cebes is always tracking down sorne argument or other (logous tinas) and is not in
the least wont to !et himself be persuaded (peitheistha1) by wbat someone just
bappens to say (Phaedo 63a).

In what follows I propose a via media between those who argue that the
dialogues of Plato are to be treated as the medium in which he chose to malee
his esoteric doctrines available to a wider public and those who argue that, on
the contrary, the dialogues are to be reduced to literary dramas, a self-
contained play of words that points to nothing further than the play itself. My
thesis, contrary to these extremes, is that the dialogues, if viewed in retrospect
from Aristotle's Organon, prove to be arguments, logoi, urging us to a
prohairesis tou biou or choice of life the opposite of sophism's; they are, that
is to say, a blend of dialectical examination and rhetorical exhortation, best
understood in terms of Aristotle's account of each of these in his Tapies and
96 Toe Tbird Way
1
Rhetoric, respectively. However, precisely to counter sophism's skills at
confutation Plato sees himself constrained at the same time to explore the
possibility ~f founding his dialectical-rhetori~~• argume~tatio~ in a forro _of
argument beyond the inescapable controvertab1hty of all d~~ecttc and ~hetonc.
And since mathematicians display the kind of invulnerab1hty to sophism that
Plato seeks, the argumentative paradigm that he tries out for ~s foundational
role becomes mathematics' apodeixis or demonstration, a kind of argument
whose nature- and presumed superiority- is displayed in Aristotle's Prior and
Posterior Analytics.
Let me be specific about these two different kinds of argument or Logos, the
tension between which, as I see it, drives the Phaedo dialogue. In retrospect
from Aristotle we may say that, on the one hand, there are arguments that
infer, deduce, prove that something is necessarily so. As in geometry or
number theory, for instance, these arguments either start, themselves, from
self-evident, noetically intuited first principies (archai) or start from ~e
conclusions of other arguments that do (Analyt. Post. 72a 7f.). And proceeding
by conceptual incompatibility or compatibility, they reach nec:ssary
conclusions about which there is no further question. Such arguments Anstotle
calls apodeiknusthai haplos, demonstrating in the strict sense (see, for
example, Analyt. Post. 72b 26, et passim). . .
Opposed to these are dialectical-rhetorical arguments, which begm fro~
what Aristotle variously calls an erotesis endoxos or an erotes1s
antiphaseos- expressions not easily translated (!opica 104a9, 1na!yt. Pr.
24bl). Toe first is even an oxymoron. Literally, 1t means a quest1orung that
is generally accepted and believed; let us call it a "belief in question." Toe
1. In regard to recovery of the rhetorical, protreptical dimension of Plato's dialogues
we owe the most, of course, to Konrad Gaiser. See his Protreptik und Paranese bei
PI.aron (Stuttgart: Kohlhammer, 1959), especially 150-55 and 192-94 on p~rainesis_ or
"wooing" (Werben) in the Phaedo. However, rather than Gaiser's Fonngeschzch~e, which
traces the rhetorical elements in Plato back to their precedents, my approach will be the
opposite Wirkungsgeschichte: Plato will be interpreted retrospectively _sta_rtin~ from his
historical effects in Aristotle. And here Hans-Georg Gadamer is the pnnc1pal influence;
see his Wahrheit und Methode (Gesammelte Werke 1), 305-12, for the concept of
Wirkungsgeschichre. Por the Wirkungseinheit or unitary effect fusing Aristotle with Plato,
see Gadamer, The Idea of the Good in Pl.atonic-Aristotelian Philosophy , trans. P.
Christopher Smith (New Haven: Yale University Press, 1986). .
Por my general understanding of Plato, however, I rely above all on W~~rhell und
Methode, 368-75, "Das Vorbild der platonischen Dialektik," and the expos1t1on of ~e
Cratylus, 409-22. (Compare my "Plato as Impulse and Obstacle in Gadamer s
Development of a Hermeneutical Theory ," in Continental Philosophy IV: Gadamer and
Hermeneutics, ed. Hugh Silvennan [New York: Routledge, Chapman & Hall, 1991), 23-
41).
Plato's Methods of Argument in the Phaedo 97
second would be something like a "questioning of a contradiction" or more
felicitously, perhaps, "a contradiction in question." Toe point of both
expressions is that though generally believed (endoxos), the premises of
dialectical-rhetorical argument are still open to question. Hence, they will have
the form, "X is so, is it not (am ge)?" (Topica 101b30, 32). This kind of
argument will proceed by "topics" (topoi), say, "similarity and difference,"
(homo~otes ~i heterotes) or '_'~e mo~e and the _Iess" (to meizon kai to elatton),
and w1ll arnve at only prov1s1onal, mconclus1ve "conclusions." In short its
proced~e is dialogical and discursive, and its rorro is dialegesthai or talking
so~ething thro~gh. C_on~aries are entertained simultaneously and the question,
which of them 1s so, 1s sull open at the end. Hence, here there is apodeixis or
demonstration only in a loose sense, and in rhetoric, at least, the apodeixis is
absorbed_ in ~ pistis or "convincer," whose results are not necessary truths at
~1, but hkelihoods. Toe Logos here is thus an eikos logos or likely argument.
Ft?'11l_y, contrary to the goal of demonstration in the strict sense, the goal of
this ki~d of a¡gument, at least in its rhetorical application, is not to bring some
theor~?cal onlo~k:r to ~n insight (mf:thema) of which he or she says, "Yes,
I_ see. Rather, 1t _is to mduce a ch01ce or decision (prohairesis, krisis) in a
hstener expressed m "Yes, I will." And since not assent but consent is sought
he~e, th~ logos must be ~ased on pathos and ethos, i. e., on the disposition
(d1athes1s) to be effected m the audience and on the character of the speaker.
For not only how someone sees something but how he or she feels about it
must be changed (see Rhet. 1356al-20).
In what follows I will provide an exposition of this tension in Plato between
~ialectical-rhetoric~ and apodictical paradigms as it is specifically displayed
m_the Phaedo, for 1f any dialogue is focused on argument as such, surely it is
~s one.. Indeed, I will argue that its supposed "doctrines," e.g., of the
1~ortahty o~. the s?ul or the existence of eternal forros, are-contrary to
Plotmus et alu-bes1de the point, and that its purpose is to provide both
~ample ~ents and, in meta-a¡gumentative reflections, a theory of a¡gument
1tself. It will e~e~e in our investigations that Plato's apodíctica! project,
whateve~ h_opes It_Illlght have held out for securing argument against sophistic
word-tw1st1~, ult1ma~e~y- fails: in th~ end, apodeiknusthai or demonstrating by
conce~tual mcompat1b1hty must g1ve way again to the very dialectical-
rhetoncal peithein or convincing that it was meant to replace. In the end
com:iction, ?~stis. will be achieved only in talking through (dialegesthai) ou;
received op1ruons (endo.xa), and the only security against sophism will be in
the ethos of the speaker, in this case Plato's "Socrates. "2
As a matter of fact, the first of the sample a¡guments advanced in the
Phaedo (63e-69e, resumed 78b-84b), namely, that one who has lived well
2. Compare Aristotle, Rhetoric 1355b19.
98 Toe Third Way
should die willingly, is explicitly said to be rhetorical insofar as it is called an
apología, an apología or defense tbat, it is hoped, will be more convincing
tban tbe one Socrates had made to his judges (see 63b). This would place it in
tbe second of tbe major divisions of rhetoric specified by Aristotle (Rhet. 1.2.
1358a35 ff.), namely logoí díkaníkoí or judicial arguments, whose principal
concerns are tbe right and the wrong, and whose subject matter is actions
taken in tbe past. Clearly tbe concern here, however, is much more a matter
of tbe "beneficia! and harmful," and with actions to be taken in tbe future,
albeit in Socrates's case here, a very immediate future. Hence, properly
speaking, this argument will not be an apologia at all but ultimately a
protrepsís or exhortation characteristic of what Aristotle terms logoí
sumbouleutikoi or advisory arguments. But whatever the case may be, we are
in the realm of rhetoric here and in a form of rhetoric tbat, unlike Aristotle's
third division, epideictical or "display" argument, urges a krisis, a decision or
judgment, upon tbose listening (see Rhet. I, 3). Consequently the argument
will be a pístis or inducement of conviction, whose task is persuasion to
choose a certain way-this as opposed to an apodeixís, which aims at a mere
insight (mathema). In this first argument, tben, the point is not so much that
death is not to be feared, but rather that we should choose a life such that we
will not have to fear it.
Furthermore, as a lagos sumbouleutikos, the argument's subject matter will
be ontologically different from the subject matter of an apodeixís. For it will
consider things tbat, far from being necessarily what they are, are "susceptible
of being otherwise;" "No one," Aristotle says, "takes counsel (bouleuetai)
about things that he holds to be incapable of either having been otherwise,
being otherwise in tbe future, or being otherwise now" (Rhet. 1357a5-7).
Consequently, the argument will not "prove" what must necessarily be true,
but will "show, " by an eíkos logos or "likely argument," what is likely to be
so but might not be. Any apodeiknusthai here will be apodeiknusthai in the
loose sense only. This meaos, of course, that the argument will always be
open to a counterargument that things were not, will not be, are not, as they
have been presented. And indeed, precisely this vulnerability is confirmed by
what occurs in the Phaedo.
Now if we focus on the logos itself here, and not yet on the ethos and
pathos that are, as Aristotle emphasizes, indispensable "convincers" (pithana)
in any such advisory argument (Rhet. 1356alff.), we can see that contrary to
demonstration, which proceeds by necessary logical entailment, this logos
proceeds by topics, specifically, by "similarity" and "tbe more and the less"
(see Topics 114b25ff. and Rhet. 1363bff; also 1397b12ff.): Death is like the
practice of philosophy in that both involve the separation of the mind from the
body. Philosophy abandons care for the adornment of the body; philosophy
purifies itself of the body's influences, in order, like mathematics, to arrive at
undistorted insight into its objects, the just (to dikaíon), the fair (to kalon), and
Plato's Metbods of Argument in the Phaedo 99
the good (to agath_ on), and to obtain its life of reasonableness, temperance, and
justness (phronészs, sophrosune, díkaiosuné). These things, the philosopher
fínds, are to be chosen (haireton) . Now death is the greater form of these
separations of the soul from the body, indeed the perfect form, so-a minori
or "from the lesser" - the greater, perfect form is better still and is to be
cbosen over an ignoble escape.
In characterizing the form of this logos, we should note that, again, quite
!n contrast.to ~e. ma~~tical l~gos apodeiktikos, which, we note in passi~,
1s already 1mphc!tly pnv1leged m the subtext within this argument, we have
actually begun w1th an erotesis endoxos or received opinion still in question
i.e., that "deatb is the separation of the soul from the body." Furtbermore'
since the_premises_here, unli~e the premises of a mathematical argument, ar;
"susceptible of be1ng otherw1se, " the conclusion too is still open to question
and _has the form,. "Death, the~, is to be hoped for, is it not (am ge)?"
Obv1ously, then, this argument, hke all arguments by the topics of similarity
and the more and the less, remains subject to controversion. Indeed Socrates
impl~ci~ly grants this possibility and allows tbat consequently the p~ithein or
conv1cuon sought may not have been achieved: "If, then I am more
convincing (pithanoteros) in this apology," says Socrates, "tha~ I was to the
Athenian judges, it is well," to which Cebes responds, "In the other matters
you seem to me to have spoken well, but what [you said] about tbe soul
generates a lot of mistrust (apistia) in human beings" (70a). Hence, this
~ment must be buttressed by others, which follow immediately upon it:
~d~e~ no small amount of encouragement (paromuthia) and convincing
(p1stzs) is needed that the human being's soul is immortal" (70b). Importantly,
howev~r, ~ese next arguments too are intended only to convince the audience
of a hkelihood and not at all to demonstrate an insight. Peitheín and not
apodeiknustf:ai is th: task: "Is it your will," Socrates asks, "that we pursue
ou_r recounung of ~ngs to the end (díamuthologomen)- whether it is likely
(e1kos) or not that things are as we have said?" (70b) (my emphasis).
The_ first of these corrob_orative arguments (70c-72e) is clearly dialectical-
rhet?ncal and would convmce us by the topic of contraries (enantia) (see
Top1c~ l 12b2_6ff.) that the soul is not scattered at death but remains to generate
new hfe: as m general less comes from more, and vice versa, so too slower
comes ~om quic~er, worse from better, heating from cooling, combining from
sep~rat1on, genes1s from corruption and vice versa. So too with sleeping and
waking and so too with life and death, etc.
Howev~r, the second corroborative argument, from being reminded or
ana~észs (72e-76e), does come close to a demonstration. Indeed, Simmias
spec1ficall~ asks at 73a, "Of what sort were tbe demonstrations (apodeixeis)
of these things?" , ~d, as we_~ill see, he will affirm at 92d that the argument
made_ h~re w'.15 a ~tnct ap~ems ex hupotheseos in the geometrical mode: Any
new ms1ght 1s be1ng rernmded of a previous insight. Previous insight in the
100 Toe Third Way
soul entails preexistence of the soul. Therefore, by logical deduction, the soul
preexists. To be sure, the argument for the first premise (73a-75b) is not in the
Ieast apodictical. Indeed, it is an eikos logos by the dialectical-rhetorical topic
of similarity: Socrates proceeds by drawing an extended analogy between
being reminded of a conceptual insight by physical things and being reminded
of a physical thing by other physical things. -But we should note that the
similarity and dissimilarity that Socrates establishes between being reminded
of a physical thing by another physical thing and being reminded of a
conceptual mathematical thing by physical things, i.e., of equality itself (auto
to ison) by two seemingly equal sticks, does prefigure precisely that
mathematic-ontological transition from the empirical to the conceptual that
Socrates will mal<e in bis attempt to secure the foundations of argument against
sophistical controversion-as do the earlier references to the fair (kalon) and
the good (agathon) (65d) repeated here (77a). Clearly, the project already
emerging at this point is to put argument about these "goods" to be chosen on
the same foundation as argument about mathematical equality.
Nevertheless, an indication of troubles to come is already given, for the task
of convincing (peithein), wbich, as rhetorical, involves not only a convincing
by logos but also a change in the listener's pathos or feeling-here from fear
of dying to hope-has obviously not been accomplished. Simmias remains "the
most adamant of human beings in bis failing to be convinced by arguments
(apistein tois logois)" (77b). Hence, a third fortification of the original thesis
that death is not to be feared is now advanced (78b-84b). And once again the
argument has the forro of an eikos logos, here by the dialectical-rhetorical
topic of sameness and difference. And even if its subtext once again privileges
the mathematical, this argument, in resuming the "apología" of 63b-69e, is
clearly intended as a rhetorical pistis and not as an apodeixis: there are
perceived visible things that change by composition and decomposition, and
there are conceived invisible things, like "the equal itself," that do not change
at ali. Toe soul is more like (homoioteron, 78b, 79b, 79e) the invisible and
more different from the visible, so, very likely, the soul is neither composed
nor does it decompose. So death is not to be feared. Or (79b-80b): the divine
rules and is immortal; the soul is like the divine and rules; therefore it is
likely, is it not, that the soul is immortal. So-and here we have the mythos
of the diamuthologomen at 70b-
This soul, then, invisible and on its way to another place which, like itself, is
worthy and pure and invisible, on its way as if to the true Hades with the good and
sage god, the true Hades to which, if God wills my soul is soon to go-is it ours
to say, as most do of this soul, that being of such kind and such nature, it is blown
away forthwith and destroyed when it is separated from the body? Far from it my
dear Cebes and Simmias 1 (80d).
Plato's Methods of Argument in the Phaedo 101
And now that the protreptical "apología" has been resumed, it emerges that
tbe conclusion of the three previous arguments, i. e., that the soul exists after
death, was really only a premise for the main rhetorical logos: the soul exists
after death, so choose the life of purity now that will assure that its existence
then will be with "the good god," and flee the life of sopbistical pleone.xia.
As powerful as this exhortation is, however, it fails as a pistis, for it fails
to convince (peithein). And precisely this is what Plato's brilliant dramatization
here-the dead silence, then the murmuring, the doubts expressed and, later
(87b ff.), the transitional switch to the narrator, Phaedo-is meant to bring
home. 3 Again, Simmias: "In regard to such matters, it seems to me, as
perhaps it does to you, that clear knowledge is either something impossible in
this life or something exceedingly difficult" (85c).
Even so, Simmias, every bit the hound on the trace of arguments that Cebes
is, is anxious to go on. If, he says,
it is impossible to gain insight (mathein) into these matters, orto find out how they
stand, one should take the best of human arguments (ton anthropinon logon), the
one most difficult to refute, and borne a!ong on it as on a raft, risk navigating one's
life-if, that is, one cannot maké one's way more securely and less perilously borne
by a more unshakeable divine argument (epi bebeoterou logou theou tinos) (85cd).
In this proposal we have the entire dialogue in a nutshell: only "divine
arguments" based on unshal<eable insight would be entirely secure against
sopbism's counterargumentive strategies and its ability to destroy any position.
Ideally, then, we should make these "divine arguments" the foundation of our
ethical arguments, just as mathematicians have done for their mathematical
arguments. But in matters of life and death, in matters of precisely what the
good, fair, and right life is that we are supposed to choose, these infallible
"divine arguments" are in fact not available to us. Risking instability and
confutation, we must therefore rely on the best of "human arguments," those
that we have put to the test and subjected to refutation, those that have proven
strongest so far.
In keeping with this principie, Simmias and Cebes now proceed to put
Socrates' "human arguments" for the irnmortality of the soul to the test (85e-
88b). Simmias's counterargument, like the arguments it would refute, is by the
~opic of simila~ty: fo~ ali one ne ds to do to defeat an argument by similarity
7
~s c~ange the snrule: 1f the soul 1s to the body as the harmony is to the lyre,
1t w1ll cease to exist when the body does. And so too with Cebes. "It seems
_3. Gadamer,_ in his seminal "Toe Proofs of Immortality in Plato's Phaedo" (in
Dialogue and Dc~lec~ic, trans . P. Christopher Smith [New Haven: Yale University Press,
1980), 21-38), h1ghhghts the dramatic importance of this shift in the scene and the
interlocutors. See p. 30.
102 Toe Third W ay
as if I, just like Simmias, am in need of a likeness," he says (87b): if the soul
is like a cloak that lasts longer than the body, it will perish eventually, even
if it outlasts a number of bodies in the meantime.
As the reflective interlude between Phaedo and Echechrates now drives
home, the obvious meta-argumentative point here is that any argument by
similarity can be refuted. And thus Plato has skillfully led us to the crux of the
dialogue: In what argument can we still put our trust? "For the argument
Socrates argued was powerfully convincing (sphodro pithanos)," says Phaedo,
"but now has collapsed into mistrust and disbelief (eis apistian)" (88d).
Plato now gives us a first indication of the kind of argument that in fact
would be trustworthy amidst all this mistrust. Regarding the argument from
anamnesis, Cebes is made to say: "I was wonderfully convinced (epesthen) by
it-more than by any other argument-and remain so," in which opinion
Simmias wholeheartedly concurs (91e-92a). And when Simmias is pressed to
choose between this argument and his harmony argument, it is made clear why
he concurs: the harmony argument, he says, was aneu apodeixeós and meta
eikotos, without demonstration and by likelihood, and, he continues,
I lcnow full well that arguments making their "demonstrations" by meaos of
likelihoods are deceptive, and, should one not guard oneself against these, one
might well be fooled-in geometry and in ali other matters. But the argument about
being reminded and understanding was demonstrated by means of a worthy
hypothesis (di ' hupotheseós a.xias apodexasthm) (92dc).
Toe reference to geometry makes clear that demonstrations by likelihood are
not demonstrations in the strict sense. Only demonstrations by means of a
hypothesis, as in geometry, are real demonstrations, and only these are
immune to sophistical refutation. lndeed, we know from the Theaetetus that
mathematicians were challenged by sophistical counterarguments based on
appearances- appearances, for instance, that the circle coul.d be squared-and
that they were immune to these arguments' deceptiveness.4 Simmias tells us
here what the basis of their immunity was. Their demonstrations were from
a "worthy hypothesis" in the form of let X be Y, and they proceeded not by
means of topics based on apparent likenesses, differences, and opposites, but
by means of logical entailment and contradiction, conceptual compatibility and
incompatibility: let X be Y, Y implies Z, therefore X implies Z. Or,
negatively, let X be Y; Y implies Z, but X excludes Z, therefore X is not Y
(see Analyt. Post. 87a7-12 on the apodeixis eis to adunaron).
Significantly, precisely the latter is the form of Socrates's first arguments
in response to Simmias's harmony argument: let Soul be Harmony; Harmony
4. See the Theaetetus 164e and 169c and Gadamer, The Idea of the Good, 40.
Plato's Methods of Argument in the Phaedo 103
iroplies Contemporaneousness with the thing harmonized, but Soul, as
preexisting the body (the result of the anamnésis demonstration), excludes
Contemporaneousness with the body; therefore, Soul is not Harmony (92a-c).
Or: let Soul be Harmony. Harmony implies More or Less, but Soul excludes
More or Less. Therefore Soul is not Harmony (93b-94b). Thus, it emerges
that the argument b~sed on the "hypothesis ... that the soul is a harmony"(94b)
has been demonstrated, in the strict sense, to be false and that the hypothesis
roust be rejected. With that we are prepared for the grand meta-argumentative
interlude that will develop the theory of a lagos by an apodeixis ex
hupotheseós, an interlude occasioned here by Cebes' "cloak" argument (95e-
105b). Toe exemplary problem to be addressed here is whether Soul implies
Destructibility or excludes Destructibility.
To lead his readers up to the theory of an apodeixis ex hupotheseós, Plato,
again with great pedagogical skill, has Socrates begin with a kind of lagos that
we might best call an account or explanation. In this lagos, as in the initial
argument for the immortality of the soul, the subject matter is stated in terms
of process between enantia: things come to be one thing, having been the
opposite. Having been cold, they come to be hot; having been small, they
come to be large. And all this coming into being and passirg away, it is
proposed, is to be accounted for in terms of a Heraclitian process "down, "
which is combination or addition, and "up," which is separation or division
and substraction. The point is that, as stated, all such logoi are vulnerable to
sophistic confutation: on the one hand, "two" is said to result from
combination or addition, but this is contradicted when "two" is said to result
from the opposite, namely separation or division. Thus is one's mind thrown
into instability : " ... and oftentimes," says Socrates, "I was thrown around, up,
down, investigating the likes of these things" (96b). Precisely this dismay
epitomizes the natural reaction to a sophistic word-twisting that makes one
seem to have said the opposite of what one did say. How then are we to
defend against sophists making our heads swim in this way? Certainly not on
their own level-as ali but two of the previous arguments in this dialogue have
made definitively clear.
Plato now has Socrates take another run at it, so to speak, the famous
deuteros plous or "second voyage," "kataphugon eis tous logous," fleeing into
arguments and accounts as such (99de). Here in argumentative theory, it is
proposed, a refuge from sophism might be found at last:
I hypothesize each time an assertion (logos) that I judge to be strongest, and the
things that seem to me to be consonant with it I, posit as true, be this in regard to
causal explanations (peri aitias) or all manner of other things. And what is not
consonant with my hypothesis, I posit as not true (100a).
Furthermore, he says to Simmias and Cebes,

104 Toe Third Way
. .. holding on to that secure foothold of your hypothesis, you would answer in this
way. And should someone attack your hypothesis itself,. yo u would first let tha~ g?,
and not answer him until you had investigated the things that follow from 1~ 1Il
regard to which of these, in your view, are consonant with each othe~ and_which,
dissonant. And when you are constrained to provide an argument (d1dona1 logon)
for the hypothesis itself, you would posit another hypothesis tha: appears to be the
best of those above it until you carne to the one that was suffic1ent (lOlde).
All this it should be noted, is said in opposition precisely to the "wiser ones,"
the sophoteroi (101a), who "delight in themselves when they, in their wisdom
(hupo sophias), are able to conflate everything" (lOle) ..
In short, to defend against sophism, vague metaphoncal arguments about
physical process will be shunned, and r~c?~e will b~ had to arguments_ by
conceptual compatibility and incompat1b1hty, for which the mathema~1cal
apodeixis ex hupotheseos provides the paradigm. To use _the example g1ven
bere (l0lbc), if we argue about "two," the argument w1~l be abou~ what,
exactly, "two" is, what the concept of "two itself' log1cally entatls and
excludes, and not about how "two things" carne to be. Loose talk about
"combination" and "separation" of things will be abandoned for the number
theory behind addition and subtraction, multiplication and divi_sion. .
Toe issue, of course, is whether this method can be umversal1zed ~nd
extended to matters like auto to kalon, namely beauty itself, if we are speaking
aesthetically as we seem to be here (100d), or the fair, noble, and ~ecent
itself, which becomes the object of our ethical deliberation (~ouleuesthat). Can
this method be paradigmatic, that is, not only for Logm as a_ccounts and
explanations in the mathematical realm, whe~e things _are necess~l~ what they
are, but for logoi as arguments for an ethical chmce and dec1s1on, where
things are "susceptible of being otherwise"? Here, r~m~kably, th_e answer
seems to be "yes": Apodeiknusthai haplos, demonstratlng m the stnct sense,
appears to become the paradigm for dialectic and rhetori~. . .
Perhaps this is the point to note that a crucial reversal m the relat10nship of
language and things would be required if ethical argumen~ ~ere ~eall~ to be
made apodictical, a reversal that the Pythagorean mathemat1c1ans Smuruas and
Cebes bave already executed even if it is only now that they are made
explicitly aware of it. To have apodeiknusthai in the s~rict. sense, ~nomata
must bave become semeia. Word-names that lend a thing 1ts meamng and
being in the very naming of it, much as personal n~mes tell us ~ho_ someone
is, must bave become signs designating exactly a thing whose be1ng 1s already
seen. Onomata, that is to say, must have become eponumiai, eponyms named
after the eidos or "visible" forro of the thing. Toe mathema~ical terms
Largeness and Smallness are readily grasped examples of th1s reversed
relationship of name and thing, for any mathematician sees what eac~ ~f th~se
is, sees its ti estin, before assigning a name to it (102bc). And so 1t 1s w1th
Plato's Methods of Argument in the Phaedo 105
every mathematical term, with one, two, three, even, odd, and so forth, to cite
the examples used here.
The advantage of this reversal for securing arguments is obvious. For once
arguments are couched in such eponyms signifying what is already seen clearly
and distinctly to be what it is, they are indeed immune to the sophists'
exploitation of the power of word-~e~ to confound us. _Plays on ~ ~ord's
ambiguity will not impress a mathemat1c1an, who sees prec1sely what 1t 1s that
"one" or "two" are named after and signify, and none of these clearly and
distinctly discemed entities can be made any longer to look like the opposite
of itself by sorne twisting of the word we use for it. For what it is and what
it is not, have always already been stabilized.
This, then, is the strategy Plato proposes in order to rob sophistic
controversialists, the "antilogikoi" of 90c, of their toehold: to the unknown
person objecting that before opposites were said to come from opposites
(103a), Socrates responds, "Toen, my friend, we spoke of things that have
opposite qualities, naming the former after the latter with the eponym of the
Iatter (eponomazontes auta tei ekeinon eponumiai). But now we are speaking
of tbese opposites themselves, by the indwelling of which these things have
their eponyms" (103b). A large thing could come out of a smaller thing, for
instance, but Large itself could never come out of Small itself. For once we
have seen precisely what it is tbat the eponym "large" is named after, once we
have seen Large itself for what it is, no twisting of the word we use for it can
talk us into believing that it itself is somehow the opposite of itself. In this way
the fear "that one lnight meet up with an opposite argument (enantios logos)"
(101a) has been successfully banished.
Socrates can now proceed on this new foundation for argument: once we
have seen what the things we call "two" and "one" are in themselves, we
could never be seduced into thinking that Two is consonant with One or that
it could be One. But not only is Duality incompatible with Unity; each of these
is incompatible with the "next highest hypothesis" above the other one; each
is incompatible, that is, with the other's genus; One is incompatible with Even
and Two is incompatible with Odd. And now the method of argument by
apodeixis ex hupotheseos is clear for all to see: we will set down, hypothesize,
a genus (genos) whose being we have clearly and distinctly discemed, and
infer from what this genus itself is, which species (eide) may be combined
with it and predicated of it without contradiction, and which may not: Our
hypothesis that Soul excludes Death having been attacked, we will recur to the
next hypothesis above Soul, namely Life. And having clearly seen what Life
itseif is, we will set down the genus Life, knowing that as such it excludes its
opposite, Death as such. Soul itself, we will then infer, excludes Death itself.
And if this proposed argumentative strategy actually works, and if word-names
like "just," "fair," and "good" can be converted into eponyms for concepts
of things whose nature we have discemed clearly and distinctly, then we
106 Toe Third Way
indeed have found an argument immune to sopbistic controversion by
opposites. However, Plato knows fuat even for things like "soul" this mefuOd
does not work, let alone for protreptical argument about fue "just," "fair," and
"good."
For no irrefutable "divine" logos about a soul, mine or yours, has been
provided, as Plato has Simmias point out: "Despite what has been said," says
Simmias, "holding human weakness in such contempt, I am constrained by fue
huge dimensions of fue subject matter fuese arguments have treated, to harbar
distrust (apistia) regarding what has been argued;" to wbich Socrates responds,
"You have said it well, oh Simmias" (107b). 5 It is clear that fue attempt to
secure fue foundations for an "anthropinos logos" about fue kind of life we
ought to choose in a "theos logos" proving fue immortality of fue soul by an
apodeixis ex hypotheseos, has in fact miscarried.
In ofuer words, not even fue premise has been secured for fue main
argument. namely fuat since fue soul is immortal one ought to care for it by
leading a just. fair, and good life. Hence, this main argument must now be
reformulated: "lf fue soul is immortal, it ought to be cared for not only in fue
time we call life but for ali time. For if we leave it uncared for, fue danger
appears terrible" (107c) (rny emphasis). And here fue premise fuat fue soul is
immortal is neifuer established immediately as an arche by intellectual
intuition, nor is it demonstrated from such an arché; rafuer, it is left pending
precisely as what Aristotle calls an erotésis endoxos, a question generally
believed to be answered in one way, but not definitively answered; it is left as
an erotésis antiphaseos or contradiction still in question. In ofuer words, the
contradiction between "Yes, fue soul is immortal" and "No, fue soul is not
immortal" is left unresolved. And consequently in fue argument built upon this
"belief in question," fuere is no trace of apodeiknusthai left; rafuer, we have
a pure rhetorical peithein or convincing, in this case by fue grand eikotés or
likeness of fue myfu of fue real earth and underworld, fue former wifu its
hopes, fue latter wifu its terrors.
As we might expect, fuen, fue conclusion, drawn at 114c, is fuat of apure
rhetorical protrepsis:
On account of these things whích we have passed in review, Simmias, it is
necessary to go to any lengths that in our lives we partake of excellence and
reasonableness (aretes kai phronéseós), for the prize is fair and the hope, great.
Here we are exhorted to flee fue life to wbich sopbism would seduce us and
to choose fue fair life instead.
It is now clear, I think, fuat Plato did not expect after ali to escape
5. See Gadamer, Dialogue ami Dialectic, 36.
Plato's Mefuods of Argument in fue Phaedo 107
sopbism's confutational skills by founding all logoi in a mafuematical apodeixis
ex hupotheseqs and fuat bis "Socrates" in fue Phaedo has, in response to bis
Pythagorean mterlocutors, tested this possibility only to find it wanting. Toe
conclusion to be drawn is remarkable: in fue end Plato agrees wifu Aristotle
that fue only difference between sopbism and rhetoric is in fue prohairesis tou
biou, fue choice of life, wbich fue speaker has already made and fuat fue only
available way to secure a logos fuat would persuade someone else to the choice
of a just life is furough fue ethos of fue speaker who makes that logos, through
his or her character. As Aristotle puts it:
Those who are good people (tois epieikesl) we trust and believe (pisteuomen) more
a~d s?oner, and this strictly speaking in all things, but where opinions go in both
drrect10ns, most of all .... Character (ethos) has the greatest power of convincing
(Rhet. 1356a5-6 , 13).
This is not in fue least to say fuat Plato makes no arguments at ali and fuat fue
Phaedo collapses into sorne sort of literary drama wifu no argumentative
content. Tbe point is rafuer fuat in ethical matters, matters of the just, fue fair
the good, fuere is for us no divine argument. no theos logos, but only hum~
argument, only ho anthropinos logos. 6 Here, fuere is no transformation of
questionable ~eli~f into sorne unshakeable mafuematic insight regarding what
each of fuese 1s "1tself," no eidos "after wbich" fuese words might be named.
Consequently, opinions (doxai) about these will always be divided. And fuus
he~e ."most of ali," the ethos of fue speaker is decisive. It is not i;
c?mc1dencf;, fuen, fuat Plato concludes fue protreptic argument of this
dialogue, his argument to bis readers, with Phaedo saying,
This was the end, Echecrates, of our companion, a man, as we might say, who of
all those we encountered then was the best, most reasonable and most just (118a).
6. See ~y Hermeneutics ami Human Finitude; Toward a Theory of Ethical
Umierstand'.~g (New York: Fordham University Press, 1991), 188-209 et passim, on the
non-availab1hty of mathematic eide for ethical reasoning.
 III

Philosophy as Myth, Drama, and Enactment:

Between Imagination and Reason,

Plot and Argument
 7

Methodology in the Reading of

the Timaeus and Politicus

T. M. Robinson

In this paper I shall be dealing with the question of methodology in regard

to a highly specific aspect of Plato's works, the myths, and even more
precisely the two great cosmological myths of the Timaeus and Politicus
respectively. In looking at them, I sball be guided by the following principies:

1. Toe dialogues being dialogues, they cannot be expected to affirm hard

and fast philosophical positions in the way a treatise of, say, Aristotle can. On
the other hand, they do seem, pace John Randall, 1 to be rather more than just
the free-floating conversations of more than usually literate Greeks. So the
view that oudeis hekon harmarta,nei, to cboose one example, is consistently
affirmed and defended throughout the dialogues, up to and including the Laws,
by the lead-figure in any given dialogue, be that figure called Socrates or
Eleatic Stranger or Athenian, and sucb consistency I take to be reasonable
proof that Plato himself, and very likely Socrates, was strongly committed to
sucb a view. More generally, I infer from this and analogous positions that
legitimate stress on the overall propaedeutic role of the dialogues is fully
compatible with the claim that there are a number of clear philosophical

l. John Hennan Randall, Jr., Plato, Dramatist of the Life of Reason (New York:
Columbia University Press, 1970), 144 and passim.
112 Toe Third Way
affirmations running throughout them (such as, for example, the real
distinction between mind and body) that can be fairly described as Platonic.
In so inferring, I am distinguishing between commitment and argument;
commitments notoriously antedate arguments elaborated in their support, and
often survive despite the fear on the part of the proponent that no argument so
far has proved fully adequate to demonstrate their cogency and perhaps,
indeed, none ever could.
2. Turning to the various mythoi in the dialogues, I take it as a basic
principie of investigation that a mythos, however else it is to be understood,
invariably describes something consonant with, never contradictory of, what
is elsewhere in the same piece of writing unequivocally affirmed to be the
case. If one wishes to name the principie, one might call it a variant on the
principie of hermeneutical generosity: while a given author may in the end
turn out to have been flouting the principie of noncontradiction, hermeneutical
generosity compels us to assume at the outset that such was not that author's
conscious intent.
3. As far as specifically cosmological mythoi are concemed, I shall atgUe
that they are in fact Plato's device for the exercise of that special type of
imagination that lies at the heart of significant cosmological speculation.
Returning to my second principie, and turning specifically to the Timaeus
in illustration of it, I wish to defend the view that the affirmation that the
universe gegone ("has come into being" 28b7) that one finds trenchantly
affirmed in the opening statement of metaphysical/cosmological principies
cannot reasonably be overridden by an interpretation of the subsequent mythos
that makes it affirm- despite its clear prima facie intent to corroborate that
opening statement-that the cosmos in fact ou gegone. This, it seems to me,
would be the case even if one conceded the highly unlikely possibility that
every time he employs the noun genesis or the verb gignesthai in the dialogue
Plato intends the reader to understand it in the sense of "process" (which I
shall call genesis2) rather than "beginning" or "coming to be" (which I shall
call genesisl). For the sense "process" is not a sense that operates pari passu
with the sense "beginning" or "coming to be," but is rather a sense wholly
parasitic upon such a sense. Toe world of "process" (to use Whitehead's
terminology) is such because each of its components across time necessarily
has a beginning ( genesisl) (as well as of course an end); the translation of tbe
totality as a "world of process" (genesis2) is acceptable precisely because tbe
term genesis has not severed its links with its focal meaning of "beginning,"
and because that focal meaning is readily appreciated in context to be
shorthand for the more cumbersome phrase "beginning and end."
Toe linguistic point is, it seems to me, reinforced by a critical philosophical
Methodology in the Reading of the Timaeus and Politicus 113
argument that Plato himself employs in his statement of metaphysical ,
cosmological, and epistemological principies that form the basis for his account
of the world's formation (27d6 ff.). It is, briefly, as follows:
l. A truly real object (i. e., a Forro, TMR), which is apprehensible by
reason in conjunction with an account, has no beginning of existence, but is
an eternal reality; a sense object, by contrast, is something that does come into
existence and is never (let alone is etemally) a truly real object (we recognize
at once a repetition of the argument of Republic 475e ff).
2. Anything that comes into existence does so thanks to the agency of sorne
cause.
3. Sense objects are defined as objects that are seeable, touchable, and have
bulk.
4. Toe world around us is seeable, touchable, and has bulk, and is therefore
such a sense object.
5. It has consequently come into existence (gegone) and, what is more, it
must have done so thanks to the agency of sorne cause.
Given the nature and structure of the argument leading to this conclusion,
as well as the linguistic evidence already adduced, it seems clear to me that
gegone can only have been meant in its temporal sense. But for residual
skeptics Plato has as so often a further and final fail-safe technique to clarify
his intentions, in this case the careful use of the perfect, rather than the present
tense. Let us suppose, for the sake of argument, that it was still a matter of
doubt whether gegone was to be understood in terms of beginning or of
process. Given the way in which the perfect tense operates in Greek-i. e., to
describe the present result of a past event or action-one is driven by such a
usage on Plato's part to conclude either that the world is in a present state of
having at a time in the past come into existence or that the world is in a
present state of process having at a time in the past come to have been
characterized by such process. Either way, we finish up with a trenchantly
clear statement that the world of genesis had a genesis.
What is important about this claim, as Hackforth saw,2 is that it is made
before the account of the world's formation by the Demiu.rge begins; there is
no suggestion that it is an eikos mythos. On the contrary, it is propounded as
2. R. Hackforth, "Plato's Cosmogony (Timaeus 27dff.)," Classical Quarterly 9 (1959):
19ff.
114 Toe Third Way
the direct logical result of arguments elaborated with such care in Republic 5,
and will be repeated as though self-evident in the Politicus. lt is, therefore, it
seems to me, acceptable as a fairly solid basis for our concluding that, at the
time he wrote these dialogues, Plato felt there was rather more reason for
believing the world had a beginning than that it did not. And the point seems
amply corroborated by the mythoi of both the Timaeus and the Politicus, in
which only the details, never the fact, of the world's temporal formation are
the question at issue.
In a word, the two cosmological myths corroborate, as one would expect in
view of the second of the three principies just mentioned, an antecedent
statement that has every appearance of being a strong claim on Plato's part to
be in fact as close to the truth or, in Plato's terminology, as "likely" (eikos)
as the constraints of the subject allow. As for the third principie, a look at
sorne of the trends in contemporary cosmology suffices to show that the
philosophical and scientific imagination Plato manifests in the Timaeus and the
Politicus is, despite the efforts of Xenocrates, Speusippus, and a number of
Neoplatonist successors to account for it in any terms other than literal,3 in
fact that form of imagination characteristic of ali significant cosmological
speculation, past and present.
We can begin with the affirmation, central to the Timaeus, that the world
of sense began at a point in time that is the beginning of time, and that
"antecedent" to this point (a concept Plato is compelled to use,faute de mieux,
48b3, 52d4) whatever existed in the physical realm was characterized possibly
by duration but certainly not by time, since time is the measure of duration of
formed objects. Paced with the same problem, contemporary cosmologists
have arrived for the most part at a very similar conclusion, the only major
difference being that what Plato could only infer on grounds of metaphysical
argument they can make sorne show of corroborating in terms of empirical
testing and observation. I am of course talking about the so-called Big Bang
theory, according to which the world of sense began sorne ten to fifteen billion
years ago with the explosion of a dense atomic structure. Like Plato, modern
cosmologists also talk of time as a feature of the real only after the explosion:
what "happened before" it is as puzzling to them as it was to Plato, but they
3. Aristotle (De Cae/o 279b32- 280al) attributes to Xenocrates the view that the
creation-account of the Timaeus was written didaschalias charin only, a view that may
also have been held by Speusippus and was left open as a possibility by Theophrastus.
For estimates of the evidence, see A. E. Taylor, A Commentary on Plato's Timaeus
(Oxford: Clarendon Press, 1928), 67ff.; H. Cherniss, Aristolle's Criticism of Plato and
the Academy 1 (Baltimore: Johns Hopkins University Press, 1944), 423, n. 356; and L.
Tarán, "Toe Creation Myth in Plato's Timaeus," in Essays in Ancient Greek Philosophy,
ed. John P. Anton and George L. Kustas (Albany: SUNY Press, 1971), 404-405, nn.
140-143 (he adds Crantor to the list and possibly Heraclides Ponticus).
Methodology in the Reading of the Timaeus and Politicus 115
are as satisfied as he was that it is out of the question to describe events
"antecedent" to it in terms of temporal sequence.
Much the same can be said about Platonic and contemporary views on the
nature of space. Until relatively recent times, most philosophers and scientists,
a}ong with any believer in the general reliability of common sense, felt
reasonably comfortable with Newton's description of space in terms of pure
extension. Toe notion was subjected to significant attack by Kant, and his
arguments impressed many, but it was really only in our own century, with the
advent of the theory of relativity and the concept of four-dimensionality, that
a concept of space evolved that would radically revise that of Newton and also
perhaps prove immune to the attacks of Kant. According to this theory, space
is a reality whose primary characteristic is tensility. So characterised, it
expands with its own expanding contents, though not of course into anything;
given four-dimensionality, the universe is finite but unbounded, and the
expansion of space is consequently a curved one, not rectilinear.
What is particularly startling to me about this theory, now pretty well
orthodoxy (for the moment at any rate) amongst cosmologists, is that they
arrived at it by a powerful act_of philosophical and scientific imagination;
years needed to pass by before significant evidence would emerge to reinforce
it. And this is where Plato, too, on reexamination, starts once again to Iook
particularly brilliant. For in a fine flight of cosmological imagination, he too
posits space as, not pure emptiness (as the atomists believed) but an entity in
eternal motion, an entity forever causing the movement that characterizes its
own contents and forever being itself affected by that movement (52e 3-5). He
also affirms strongly, as if anticipating Kant's criticisms, that the entity that
is space is not an entity like standard objects of sense perception, but one that
enjoys an ontological status peculiar to itself, and is the object only of what he
calls a "bastard" form of apprehension (52b2).
The details of the theory need not be discussed here. More important, for
the moment, is the fact that a theory of this subtlety, and one so close to a
theory of our own epoch that has to date withstood the test of a great deal of
empirical observation and experiment, was elaborated long before the notion
ot four-dimensionality was imagined, and indeed at a time when most if not
ali other thinkers, including Aristotle, took space to be characterized by
nothing if not causal neutrality. It is not, of course, the same theory: without
benefit of the concept of four-dimensionality, Plato could arrive at most at the
notion of space as tensile, not as possibly expanding, since of course that
which for him surrounds the universe is by definition not a reality, actual or
potential, into which such expansion might ever have taken place. But to have
!magined even the first of these things-that is, that space is tensile- puts him,
1t seems to me, into the ranks of cosmologists of the first order.
But the surprises are not over yet. If Plato, given the nonexistence in his
day of the concept of four-dimensionality, could not have imagined the
116 Toe Third Way
possibility that the world, after its appearance in time at a moment that was the
first point of time, could have been in a state of expansion, it seems on the
face of it highly unlikely that he could have imagined the contemporary
hypothesis that one possible future for the universe is that it eventually reach
a Iimit to its expansion and begin the process of contraction to something like
its original state. Remarkably, however, he does seem to have thought of
something very analogous to such a hypothesis, and for the evidence I tum to
the myth of the Politicus.
In this myth Plato, after reasserti~ that the world began in time as the
result of demiurgic causal activity, tries to imagine the implications involved
should its soul-a soul that is fully rational, as in the Timaeus"-tum out to
be in fact more bodily-influenced than he had suggested in the earlier dialogue,
and should what we now call the second law of thermodynamics be more
obviously at work in the matter of the world's rotation than he had there
suggested. Of two great metaphors-those of the spinning top and the
spindle-that he had first alluded to as early on in his writings as the Republic,
the first was the one he favoured when composing the Timaeus. With this as
bis guide, he was able to say that the universe would spin on its axis forever,
if the Demiurge did not actively intervene to prevent its doing so (34b, 32c).
In the Politicus, by contrast, he employs the other metaphor and, in a feat of
philosophical imaginativeness in many ways equal to that employed in the
writing of the Timaeus, he talks of the periodic rever-se rotation of the
universe, in the way a spindle will, at the appropriate moment, unwind under
the impetus of its own stored momentum.
This, on the face of it, is just about as far as imagination can go if that
imagination, as Plato's and everybody else's was, is constrained by the concept
of three-dimensionality. What is particularly noteworthy about it, however, is
the manner in which-by contrast, apparently, with the Timaeus-it takes
candid and unconditioned account of the operation of the second law of
thermodynamics. Whether he calls this law themis or heimarmene or to
prosekon, it is still the law that affirms that natural systems, left to themselves,
have a tendency to run down, and a law that he now affirms to be true of the
universe itself, believing as he does (as in the Timaeus) that the world is as
much a physical object as are its constituent parts.
Usi~ this law as his guide, a law beautifully exemplified by the operations
of the spindle, he imagines a moment when the world's spinning motion has
4 . Toe reference to its symphytos epithymia (272e6) I now take to be a reference to a
rational, not an irrational drive within it that causes it to impel world body in a particular
direction. What brings about an eventual change in that direction is to somatoeides,
conceived here by Plato as characterized by forces analogous to the coiling and uncoiling
of a spindle, rather than sorne supposedly irrational element within world soul itself.
Methodology in the Reading of the Timaeus and Politicus 117
used up its momentum for spinning in one direction (described in the Timaeus
as spinning to the right) and at the same time accumulated enough momentum
tor unwinding and spinning in the opposite one. World soul's epithymia, a
rational epithymia for circular motion in a rightward direction (see above, o.
4), understandably resists what is happening, but is only partially successful
in its resistance, in that the world will still continue to spin in circular fashion
but must now for a while spin backwards, or to the left. When the momentum
tor backward spinning is in turn exhausted, the process will begin ali over
again, and so on ad infinitum.
For aoyone acquainted with that variant on the Big Bang Theory koown as
the Oscillation Theory, the affinity between it and what Plato is imagining in
the Politicus is remarkable. According to the Oscillation Theory, the world
expands for a particular period of time till its original momeotum for
expansion is exhausted, and then contracts to its original densely compacted
atomic state, from which in turn it may well expand again, and so on ad
infinitum. During the contraction, the most extraordinary things can be
imagined as happening, including the reversa! of historical events, our growing
younger not older, and so on. Plato the great imaginer imagines all this and
more, to the astonishment of commentators, who have consistently refused to
take it seriously.
But why not take it seriously? In our own century a cosmologist as
prominent as Stephen Hawking thought nothing of speculating that, on one
interpretation of the Oscillation Theory, time itself might be reversed during
periods of cosmic contraction, and he became famous for his images of the
broken cup flying together again and reconstituting itself, and of ourselves
indeed actually growing younger again during the process. s Other
cosmologists disagreed, and eventually persuaded him to change bis mind on
the matter, 6 but all to my koowledge felt the need to take his hypothesis
seriously and replace it if possible with another one more plausible. No one
did what is customarily done to Plato, that is, write off exactly this same
notion (Poi. 270d-e) as being clearly not bis intention.
Let me try to sum up what I have been driving at. Thanks to the
well-intentioned but perverse efforts of such as Xenocrates and Speusippus,
Plato has been effectively robbed of one of his little adverted to but truly great
claims to fame, that is, his brilliant cosmological imaginativeness. We are
fortunate indeed, it seems to me, to live at a time when the analogous exercise
of such imaginativeness by thinkers of the caliber of Einstein, Heisenberg, and
Hawking, and the extraordinary progress in our understanding of the world
5. Stephen W . Hawking, A Brief History of Time (New York and Toronto: Bantam
Books, 1988), 149-150.
6 . /bid., 150-151.
118 Toe Tbird Way

that such imaginativeness has led to, finally allows us. to appreciate how
clearly Plato, unshackled from the chains placed upon him for so long, can
with good reason claim to be their forebear.

How to Read a Platonic Dialogue

James A. Arieti

In discussicg Plato, it is possible to explain Plato's theory of literary

composition-as if he had written a treatise about it-by extracting from the
dialogues all references to the subject and then straining here, squeezing there,
omitting here in arder to fabricate sorne kind of system. I shall not attempt
this. lt is also possible to look at the dialogues themselves to try to explain
Plato's practice of bis literary craft. Such is what I plan to do. First, however,
I must say a few words about what I think Plato was doing and where most
readers of Plato have gane wrong.
Most of the problems in understanding Plato arise from studying the
dialogues as if they were a part of the tradition in which Plato did not
participate. When readers try to find systematic, consistent, straightforward
positions in the dialogues, with sound arguments and clear, unambiguous
meaning, they knock against an iron wall-the dialogues are anything but
systematic. How could they be? After ali, they imitate the rambling and
disjointed talk we find in living conversations. They are anything but
consistent: we find contradictory conclusions in the dialogues-even on matters
like the theory of forms, supposedly the ethereal bedrock of Platonic thought. 1

1. It is obvious that the Parmenides and the Republic-if both are taken literally- are
incompatible. In my book Jnterpreting Plato: The Dialogues as Drama (Savage,
Marylaod: Rowman & Littlefield, 1991), I argue that we are not to take such dialogues
literally. I do not think it is possible to know from these dialogues what Plato's views
 Toe Third W ay
120
The dialogues are anything but straightforward: look at ali the wildly v~ing
ways in which critics handle the metaphors of the cave and of the ~v1ded
line.2 And very often the dialogues contain arguments that are anything but
sound: recall all the equivocations, logical errors. and intellectual games that
all the characters engage in. 3 Scholars have always found Plato a gold mine
for papers because each reader can find a new twist to explain sorne difficulty
in the text.
When various astronomical phenomena presented difficulties to ancient
astronomers, the astronomers, using what they believed to be sound
fundamental principies, worked out the theory of epicyclic motion to explain
the apparent irregularities in celestial and especially planetary and lunar
motion. They carne up with the theory in part because they took as a fixed
principie of nature the necessity of perfectly circular motion. Plato sc~olars,
like those who created the Eudoxan and Ptolemaic system, took as the1r own
fixed principie of nature that no philosopher would knowingly and willfully put
into his philosophical text views that were unsystematic, inconsistent, unclear,
and- perish the thought!-unsound. Thus, to deal with the apparently aberrant
features of the Platonic universe, they invented a sort of analytical theory of
epicycles, that is, the theory of the cycles in Plato's life. Thus arose th~ theory
of the early, middle, and late dialogues-with suitable transitional penods for
the dialogues that could not fit neatly into any period. . apart, and one play makes no reference to the other. Moreover, it would seem
Toe chronological theory to explain difficulties seems to me defect1ve on
severa! grounds: first, as any survey of the various chronologies will s~ow,
especially those based on stylistic variations, it is impossible to be certam of
any proposed order. 4 That Plato wrote the dialogues sequentially and not
simultaneously is undoubtedly true- but every proposed chronology depends
on the caprice of the proposer-from those who would place the Phaedrus as
Plato's first dialogue "because Plato could not have written so passionately
about Jove unless he were a young man himself' 5 to those who would put it
among the late dialogues because it mentions the world of ideas. My view is
that we can no more put Plato's dialogues in order than we could
Shakespeare's plays or the films of Cecil B. DeMille- if we lacked externa!
concerning the forros are.
2. E.g., R.C. Cross and A.D. Woozley, Plato 's Republic: A Philosophic Commentary
(London: Macmillan, 1964), though they collaborated on an analysis, disagree between
themselves on the interpretation of the metaphors of the divided line and of the cave
(196-230).
3. These are discussed at length in lnterpreting Plato.
4. Por a full discussion of chronological confusions and the relevant scholarship, see
Jnterpreting Plato, Ch. l.
5. See, for a discussion of the dating of the Phaedrus, W.K.C. Guthrie, A History of
Greek Philosophy, vol. 4 (Cambridge: Cambridge University Press, 1978), 396, n. l.
How to Read a Platonic Dialogue 121
evidentiary data. Second, the theory of chronology ignores the crucial question
of audience. lf we don't know who Plato's audience was for any particular
dialogue, how can we presume to know why he did or did not include various
arguments? Perhaps an analogy would clarify this point. lf Albert Einstein had
been asked late in his career to speak to a group of boy and girl scouts, he
would surely have spoken in simple, even simplified, language about simple
or even simplified subjects. If many years later we found the speech among
bis papers, it would be erroneous to assume an early date on account of
simplicity or simple topic. So with Plato. Short, seemingly simple dialogues
Jike the La.ches or Crito may have been written after longer, difficult
dialogues, like the Theaetetus or Fbrmenides. We simply do not know.
lt also seems to me a fundamental error in interpretation to require the
understanding of one dialogue to depend on what is said in another, except,
perhaps. for a very few that refer to preceding dialogues. like the Theaetetus
and the Sophist, or the Republic, Timaeus, and Critias. But even in these
cases, there are many puzzles: for example, if the conversation of the Timaeus
is really to be understood as following that of the Republic, why is there so
violent a change in the cast of cbaracters?6 This point seems obvious when we
tbink about other authors. For example, it is not necessary to understand the
Antigone to understand the Oedipus lyronnos. They were written many years
absurd to suppose that an earlier text requires a later one in order to be
comprehended: if the author bad died before completing the later one, would
comprehension be lost forever? Scholars of an author can, perhaps. gain sorne
insight into an author's mind by reference to another work, but such
understanding cannot rely on the other work. In fact, sucb references are as
likely to contaminate understanding as to promote it. And yet bow often bave
we heard that such and such a point in Plato can be understood only by
reference to an argument in another dialogue?
I should like to propose an altemative theory to the philosophical difficulties
whicb we find in the dialogues, and to do so, I would like to do just one thing.
I would like to toss out the premise of virtually ali work on Plato: that be is
writing the kind of philosophical work in which the philosopher writes as
clearly, as straigbtforwardly, and as soundly as he can. Instead, I should like
to take as my prernise that Plato is not writing this kind of philosophy at
all-the kind we have come to know since Aristotle. Instead, I should like to
assume that he is writing works of drama-works whose intention is
principally to inspire- and that the inspiration in the dialogues is to
engagement in a life of the mind, to the doing of philosophy with other people,
and not with dead or even lively texts. Plato, I would venture to say, has been
6. See Jnterpreting Plato, 19-20, and especially n. 4.
122 Toe Third Way
enonnously successful in bis purpose, for very many who have chosen an
impoverished life devoted to pbilosophy have done so because of the
inspirational capital he has bequeathed to us.
Long after we have forgotten the wandering maze of arguments in the
Phaedo, for example, we still remember the hernie figure of Socrates
cheerfully drinking hemlock, and we think to ourselves: I wish I could be like
that! What has moved us has been the courage of Socrates in pursuing the
arguments about the immortality of the soul wherever they led. We were
moved because unlike most people, who are dismayed when the arguments get
tough, Socrates persevered even though he didn't know whether he would
succeed. And in the end of the Phaedo no absolute proof is found-how could
it be? Socrates knows that no proof has been found and that he lacks
knowledge about bis soul's immortality. Thus, he is able to be heroically
courageous, as he could not have been had he proved bis immortality-after
ali, what courage would be needed if you are sure you're going to Paradise?
Let us take another dialogue written to inspire-the Grito. In my view, what
matters here is the drama-not the arguments-though to be sure, the
arguments need to be understood to grasp the dramatic point. Toe dramatic
situation tells ali: the sbip sent to Crete for the annual religious rite is
returning, so executions can again take place. Crito tries to persuade Socrates
to escape. It is difficult for us to imagine a situation more stressful than that
Socrates finds bimself in. He is to be executed though he has a sure sense of
bis innocence, and he now knows the precise day of bis execution. His death
is no theoretical far-off possibility but a certainty in three days. How does he
respond? Despite all the temptations to live, despite the easy out of granting
the legitimacy of Crito's arguments, Socrates inquires into the pbilosopbically
correct course of conduct. Even in the midst of the urgent crisis, he engages
bis mind in order to determine the most rational conduct. Most of the
scholarship on tbis dialogue deals with the question of whether the arguments
of the imaginary Laws are valid or not. But the arguments of the Laws don't
seem to me to be the heart of the dialogue's teacbing. Their truth or validity
is not the issue. What matters in the play is that the character Socrates cannot
refute them, that Socrates believes them to be the best arguments available. At
the end of the dialogue, Socrates challenges Crito to come up with a better
argument if he has one. He doesn't, and Socrates decides to follow the best
argument he has. He does not say, ''1'11 escape; maybe I'll come up with a
better argument tomorrow." No, the drama shows him acting on the basis of
the best argument he has at the moment when he has to make bis decision.
And here lies the inspirational and dramatic point of the dialogue: a person of
the mind, no matter how stressful and unnerving the situation, decides- even
in matters of bis own life and death- on the basis of the best rational argument
that he has at the moment when he must decide. This is the point of the
dialogue, and it does not matter whether scholars find loopholes in Socrates'
i
How to Read a Platonic Dialogue 123
aiguments for this point to be true. 7
Next, I should like to discuss brietly the intellectual milieu in wbich Plato
was writing. In the beginning and middle of the fourth century a great many
schools of rhetoric were thriving in the Greek world. Toe power of speech to
effect one's will in the increasingly democratic political assemblies and in
other public gatherings, the ability to stir citizen-armies to a bigh pitch of
roartial enthusiasm, the skill to manipulate the truth in legal disputes before
juries-all made facility in speaking a very valuable and sought-after quality.
Schools opened and actively competed with one another. Isocrates, for
example, began a school to teach the liberal arts and to turn out roen capable
of writing speeches for all occasions. Alcidamas, a major rival of Isocrates,
promised to turn bis students into first-class extemporaneous speakers. 8 These
and many other schools competed with each other for students. It is my view
that a number of Plato's dialogues have as a fundamental, if not primary,
purpose, to show that those who emolled in the Academy-Plato's
school-could compete with and surpass the students of the rivals. This is why
so many of the dialogues involve oratorical competition and, in fact, why such
competition is the focus of so much drama in the dialogues. Let me support
roy observation by referring to two very famous dialogues, both of wbich are
about oratory: the Symposium and the Phaedrus. Now it is only after a lot of
though~ and interna! tunnoil that I say this. Toe Symposium is the first dialogue
I read m Greek as a sophomore at Grinnell College, nearly thirty years ago·
after reading it, I fancied myself an expert a la Socrates about love, and
heaven knows how many lines I borrowed from the dialogue in discussi~
romance. But that's ali to the wind now, for I no longer think that Iove is a
serious doctrine in the Symposium or in Plato.
Toe Symposium takes place at a party, where the partygoers ali agree to
have a rip-roaring good time competing with each other in an oratorical contest
in wbich they all are to deliver speeches about eros-sexual desire.
Eryximachus introduces the discussion by pointing out that wbile Prodicus has
written an outlandish praise of salt, no one has ever praised sexual desire. (Let
me add that it has long been a mistake to translate "eros" as "love." Toe
word, as Kenneth Dover points out in bis edition of the Symposium, means
"sexual desire. " 9 If it meant "affection" or something of this sort, the claim
7. In fact, the more loopholes they find, the more likely it is that Plato wants us to see
~e drama~ic point. When we see a drama, we understand that the characters say what
1s a~propi:iate for the~ to say as characters in the drama; the "truth-value" of what they
say 1s of mterest as 1t enhances their characterization. Only when it comes to Plato's
Philosophical drama do readers simply assume that Socrates is to be understood as
speaking in perfect arguments.
8. In an essay "On the Writers of Written Speeches."
9. K.J. Dover, Symposium (Cambridge: Cambridge University Press, 1980), 1.
124 Toe Tbird Way
that no one had ever praised it would be self-evident nonsense, as from Homer
and the lyric poets are numerous praises.) Toe various guests then deliver their
speeches, in which the most noteworthy quality of each is the way the god
"Eros" resembles each speaker. Thus, for example, Dr. Eryximachus' Eros
is a healing doctor, Aristophanes' Eros is a comic division and comic
searching for an original half; Agathon's eros turns out to be a tragic poet with
tender feet; and Socrates' Eros, Socrates' "sexual desire," is revealed to be a
barefooted philosopher! Who wins the oratorical contest? Why Socrates does!
He leads us along slowly, step by step, plausible step after plausible step, until
we get the magnificent conclusion that sexual desire-sexual desire!-is
philosophy. Surely many philosophers (if not ali) are sexual dynamos-but are
we to believe that sexual desire is philosophy? And yet the dialogue is so
beautiful that we are convinced that it is so! Here lies a warning: aesthetic
beauty is not truth, despite what Keats might say. Why should sexual desire
be a barefooted philosopher a la Socrates rather than a tragic poet with tender
feet a la Agathon? Toe last great scene, when Alcibiades enters, is, I think,
meant to reinforce this point. In Alcibiades we see the failure of Socrates'
scheme of sexual desire. If sexual desire is bringing forth in the beautiful,
what was more beautiful than Alcibiades to bring forth in? Must we not see
in this portrayal of Alcibiades Socrates' failure? Is not the appearance and
speech of Alcibiades a sign of how empty are even Socrates' comments about
sexual desire? In a broader sense, perhaps Plato is suggesting that the kind of
relationship Socrates had with Alcibiades is insufficient to render such a one
a decent citizen, much less a philosopher. Perhaps he is suggesting that a much
more systematic relationship, such as might be found in his new academy, will
be more effective.
Or, consider the Phaedrus-a dialogue that contains three competing
speeches. Two are on an outrageous theme, like the speeches in the
Symposium: here on the hideous proposition that you ought to grant sexual
favors not to your lover, but to someone else. Phaedrus first reads Lysias'
speech on this repulsive theme. But we are not to think too ill of Phaedrus: he
doesn't really care about sex; his interest is in speeches; and in Plato he
represents a sort of stock character- the speech enthusiast. What enchants
Phaedrus is the terribly clever way in which Lysias defends his thesis. After
hearing the speech, Socrates criticizes it as a speech: it lacks organic unity and
a decent architecture, and he proceeds to deliver a much more artful and
balanced speech, still on the repulsive theme. Midway through he stops to
deliver a better speech on eros, a speech that he analyzes, together with the
first two speeches, in the second half of the dialogue. Toe second half of the
dialogue contains a rather splendid discussion of how education in rhetoric
ought to take place and in fact is remarkably similar to Aristotle's
Rhetoric- giving, I think, a lot of weigbt to the idea that Aristotle taught
How to Read a Platonic Dialogue 125
rhetoric in the Academy. 10 Toe dialogue ends with a reference to Isocrates,
0 ne of Plato's rival school directors. Embodied in the dialogue is another
controversy concerning rbetoric- the controversy raging in Plato's day, the
debate between Alcidamas and Isocrates on whether speeches best should be
written in advance- Isocrates' view- or extemporaneous- Alcidamas' view.
The debate is dramatically portrayed in the Phaedrus: we have first the written
speech of Lysias- read aloud to Socrates by Phaedrus- and then two
spontaneous, extemporaneous speeches delivered by Socrates. Again, I think,
(he dialogue is meant to persuade Plato's contemporary audience that it too can
get a better education in oratory at the Academy than at either the school of
Isocrates or that of Alcidamas.
Another dialogue that does the same sort of thing and also perhaps makes
a comment on the Alcidamas-lsocrates debate is the Menexenus, in which we
find Socrates- whom I take here to be the stand-in for the one taught at the
Academy-delivering an extemporaneous funeral oration. Socrates' point
seems to be that a knowledge of the topoi-of the standard sort of
commonplaces made at such occasions-renders the job very easy: no genius
is required if you know the rules of the genre. This is the point of the
dialogue. Toe speech is so good, however, even though it is put together from
ready-made parts, that many readers-mesmerized by its beauty- have taken
it as a serious work. But the dialogue's point, as I see it, is to persuade those
listening to it that they too could become good speakers if they enrolled in the
Academy.
Other dialogues also contrast Plato's school with those of other rivals, as we
see confrontations with the schools established by the students of Protagoras
Gorgias, and various others. '
In the space I have left I should like to look at two dialogues in a little bit
of detail to see how Plato practices his art in the matter of characterization. I
should like first to look at a few lines from the Meno in order to examine how
Meno is sketched by just a few words spoken by himself and by Socrates and
how this portrayal fits into the dialogue as a whole. I should then like to do
the same for the liuthyphro. In the Meno, I see Plato's purpose as competitive:
Plato wants to expose the insipidity of Meno's education-an education
obtained in Gorgias' school. In the case of the liuthyphro, I see Plato's
purpose as defending Socrates from the charge of corrupting youth. Toe
defence, however, comes not so much from the argurnents as from the drama.
First, the Meno. Meno, I think, is fully characterized in the first two
_10. See W. Jaeger, Paideia: The ldeals of Greek Culture. vol. 3. Translated by Gilbert
Highet (New York: Oxford University Press, 1944), 185 ; also his Aristolle:
Fundamentals of the History of His Development, translated by R. Robinson (London:
Oxford University Press, 1962), 37.
 Toe Third Way
126
Stephanus pages. Let us look at the opening question and Socrates' response.
Can you tell me, Socrates-is virtue something that can be taught? Or does it come
by practice? Or is it neither teaching nor practice that gives it to a man but natural
aptitude or something else?
Toe question is startlingly abrupt- a sophisticated and complicated question
about virtue. Toe very formulation of the question shows a substantial training
in philosophy- Meno knows ali about the controversy between no""!s and
physis and is open to the possibility that something else may be respons1ble for
a man's acquisition of virtue. And so the question is designed to make Meno
appear to think himself intelligent. But to ask a question that is so complete,
that so well covers ali the possibilities, shows that Meno must have worked out
the intricacies of the matter. Toe question-so fulsomely formulated- has as
its motive a desire to show off, to initiate a debate that will illustrate one's
learning and ability in eristic. What Plato is portraying in Meno's opening
speech, then, is a stock character of a sort we still find around us, the
trained- how shall I put it- the first-year graduate student in philosophy, who
knows ali the answers and is confident in his ability to acquit himself well in
any debate. Toe question Plato gives to Meno is not an eamest request for
understanding, it is a challenge to a fight.
Now my view is that Plato' dramatic point in this dialogue has nothing to
do with the precise question of whether virtue can be taught- or, for that
matter, with virtue. What Plato is doing is showing the ineffectiveness of his
rivals' teaching methods. What they are producing are students who know set
arguments with set questions and set answers. They are like those guides you
run into on tours who have memorized a certain amount of information about
their particular site but are unable to answer any kind of question that is j~t
a little bit different from standard. They have no real knowledge because the1r
knowledge depends on mere memory, not understanding.
Socrates' response to the question completely trips up Meno: it also shows
the satyr-play quality of the dialogue. Socrates is going to have fun with
Meno!
Socrates at first seems wholly beside the point, as he talks about the old
Thessalian reputation for wealth and horsemanship. Now, he adds, they are
philosophers too. Socrates shows that he was not taken in by Meno's question.
Gorgias taught the Thessalians and particularly Meno's friend Aristippus to
answer confidently any question they were asked-the kind of thing Mr.
Memory did in Hitchcock's The Thirty-Nine Steps. This, says Socrates, is what
Gorgias himself does in claiming to answer any question put by anyone in the
Greek world. And now Socrates makes his most shocking and indeed
ridiculous claim of ali: he attributes to all Athenians his own claim of
ignorance: "Here at Athens there is a dearth of wisdom. At any rate, if you
How to Read a Platonic Dialogue 127
put your question to any of our people, they will all alike laugh and say, 'You
must think I am singularly fortunate, to know whether virtue can be taught or
how it is acquired. Toe fact is that far from knowing whether it can be taught,
¡ have no idea of what virtue itself is.' " Socrates claims this same Athenian
ignorance and modesty for himself. How, he asks, can he know any of the
properties of virtue if he doesn't know what virtue is; how can he know what
Meno is like if he is totally ignorant of Meno?
How shall we say that Plato has characterized Socrates in his opening
remarks? He too is portrayed as a sophisticated fellow- one who knows how
to avoid answering the kind of question put to him and who, like any skilled
politician, knows how to reformulate the question in his own terms. But no
one will grant his claim about all the Athenians: it is a mere maneuver
intended to throw Meno off guard. Toe lie does not, I think, reflect well on
Socrates here, but it shows his agility in discombobulating his adversary. It
works admirably well. Meno is thoroughly discombobulated. Instead of
challenging the wild assertion, Meno asks, "Is this true about you, Socrates,
that you don't even know what virtue is? Is this the report we are to take home
about you?" Meno has thus already lost track of the strategic course of his
argument. He is now lost in the morass of the idiosyncratic nature of Socrates.
And so on. As I point out in the discussion of the dialogue in my book
Interpreting Plato, Socrates over and over again slips out of the various
set-traps Meno has laid to catch him. In fact, I believe the famous episode
with the slave boy, in which the lad learns how to double a square, indeed, the
introduction of the entire passage concerning the theory of recollection, takes
place in order to elude Meno's trap.
It doesn't take long, then, for Plato to sketch Meno's persona. And, as we
have seen, that depiction of the essential elements of his character, his
complete reliance on prefabricated philosophy, memorized set-pieces, is key
to understanding what the play Meno is ali about.
Now a brief glance at the F.uthyphro and the way in which the title character
is portrayed. Let me say at the outset of the discussion that I do not believe the
dialogue to be about piety or holiness. This usual view, that the dialogue is
about holiness, arises predictably because Socrates and Euthyphro discuss
piety. It is, however, as much a mistake to assume that any of the dialogues
is about what the personae dromatis discuss as it would be of Shakespeare's
plays. Lear is not about how much the old king's daughters love him: it is
about self-delusion, madness, and nature. Toe F.uthyphro is also about
self-delusion, about the self-delusion that drives a man to act with absolute
certitude; it is about self-delusion so intense that it thrives even when the
hollowness of its foundation is absolutely manifest.
Toe little play begins as Euthyphro expresses surprise to see Socrates at the
royal porch, a law court, instead of at the Lyceum, where he could talk to
youngsters as usual. Note, by the way, how different is Euthyphro's question
128 Toe Third Way
from Meno's. Euthyphro's is a question that actually seeks information:
This, Socrates. is something new? What has taken you from your haunts in the
Lyceum, and makes you spend your time at the royal porch? You surely cannot
have a case at law, as I have, before the Archon-king.
Toe characterizing element here is the contrast Euthyphro draws between
himself and Socrates: he uses his own situation to define Socrates' situation.
We shall see him do this over and over again.
Socrates explains that he has been indicted by Meletus, a young man he
doesn't know on the serious charges of corruption of youth. Toe very fact of
the charge is' evidence of Meletus' excellence, says Socrates, for to bring such
a charge he must care a lot about the young. Surely, he adds, Meletus will
have a great future.
Euthyphro suspects that the charge arises from ~ocrates' re~erenc~ to his
special "sign," the voice that sometimes keeps him from do1ng ~ngs he
should not. Euthyphro scoffs at the charge even while acknowledg1ng how
easy it is for such things to be misrepresented in court; he says h~ suffers a
similar fate: when he makes predictions in the assembly, the Atheruans laugh
at him, even though his predictions are always right. Toe character Euthyphro
presents himself as a man infinitely confident of his abilities. A moment lat~r,
Euthyphro predicts that Socrates' case will come out well (3E)-and we real1ze
from the dramatic irony how good a prophet he is.
It turns out that Euthyphro has come to court to bring an indictment against
his father for murder. Although his family has scolded him quite severely, he
knows he is doing the right thing. Socrates expresses amazement and says that
Euthyphro must be far advanced in wisdom to do so unusual a thing as
prosecute his own father (4A). Yes, says Euthyphro in another self-praise, he
is very far advanced indeed (4B).
Euthyphro begins his account with sorne very Socratic-sounding statements.
We must not consider whether or not a man is related when making
prosecutions, only whether or not the prosecution is right. We learn, however,
that the case is not a simple one. As Euthyphro tells the story, a laborer
became drunk and killed one of the domestic slaves. Euthyphro's father tied
up the culprit, threw him into a ditch, and sent a servant to A~ens to ask what
to do. In the meantime, the man died. Toe delay, presumably, 1s accounted for
by the fact that the incident occurred on the family farm in Naxos, an
Athenian possession. Those, Euthyphro says, who criticize him for prosecuting
his father surely do not know what divinity is or its relationship to holiness!
Socrates asks Euthyphro whether his knowledge of divine things is so excellent
that in the circumstances he has described he can rightly accuse his father
(4E). And again (5A) Euthyphro boasts about his special knowledge.
Toe play begins, then, with a young man absolutely convinced of the
How to Read a Platonic Dialogue 129
rightness of his case. He is plagued with not a single doubt, not a single
rooment's hesitation, not a single question. Toe case he has described,
however, is by no means an obvious one; it certainly cannot have been
self-evident to Plato's audience that Euthyphro is acting rightly. On the surface
it seems as though the drunken laborer had death coming to him. Surely it
does not seem as though Euthyphro's father murdered him; at worst, he might
be guilty of negligence. Of course we recognize that a commonsense view and
a legal view, especially in litigious societies like our own and ancient Athens,
are quite often diverse. No matter; however one looks at it, the case is
complex and would give any reflective man pause. But not Euthyphro! He is
totally sure that he knows just what the holy thing to do is.
What we have, of course, is a reflection of Socrates' situation. Meletus, a
young man like Euthyphro, is prosecuting Socrates, an old man like
Euthyphro's father. Both youngsters must be very sure of themselves to take
such action upon themselves! What we sball see in the play, I think, is an
arrogance in Euthyphro that we can assume to be present in Meletus. Even
when Euthyphro will be given every cause to question his own judgment, he
will persist in believing in the rightness of his accusation. Socrates will effect
no change whatsoever in his young interlocutor. Toe audience will be left with
this question: if Socrates has had no effect at all on Euthyphro, how then could
be have dented the consciousness of any youth enough to have corrupted him?
Toe true danger is the other way around: it comes from youth, who, without
benefit of wisdom, experience, or knowledge, cavalierly prosecute their elders.
Plato's point is this: watcb out for the young! If they're not educated properly
(as they would be in the Academy), look at all the trouble they can cause. We
also see, of course, how very difficult it is for Socrates to convince anyone of
anything. If he cannot convince the well-meaning if confused Euthyphro, how
will he ever convince a jury full of people like him?
Over and over again, Euthyphro is compelled to admit that he is confused
or cannot follow the argument. None of these defeats shakes his confidence in
himself. His confidence is expressed as he compares himself to both Cronos
and 2.eus; again when he boasts that he knows that the gods approve of what
be is doing; again when he says that he knows what art serves the gods
because he knows more about divine things than any other man; and finally
again, at the end, when he tells Socrates be will tell the whole story about
piety at a later meeting.
And so, it seems to me, the play is about Socrates' complete failure to affect
Euthyphro's erroneous self-confidence. Not only does Euthyphro not know that
h~ is ignorant, even when he has admitted his confusion, he persists in trusting
his knowledge. In a sense, then, I would link this dialogue with the
Symposium, for both it and the Symposium show the cavalier presumption that
dorninates people's pronouncements about the gods. If the results were good
fun in the party at Agathon's house, we see here how dark is the flip side of
130 Toe Third Way
such presumption: it can convict and execute a man.
It may appear from the preceding remarks that I do not believe that Plato
had a philosophy or any philosophical positions. Such is not the case. Surely
Plato held positions. Toe question is whether we can discem those views with
any certainty in the dialogues. 11 I think we can see hints of sorne positions
through the mist of the dialogues, just as we can see hints in the works of
Shakespeare, Sophocles, and Euripides of what they thought. But it is not
easy. We can surely tell many of the issues to which they directed their
attention; then we can apply our minds to try to deduce what the plays as a
whole tell us, if anything.
Perhaps it would be useful to give examples of the kinds of teaching to be
derived from the dialogues when they are looked at as whole plays. Let us
remember, however, that the kind of teaching that one receives from art is
very different from that which one receives from a philosophical treatise; art
works primarily by appealing to our emotions, and this will be true of the
dialogues insofar as they are works of art, albeit intellectual works of art.
When we read or see a tragedy, for example, these are the lessons that we
leam: to honor the gods, to reverence the gods, to believe oracles, to bury the
dead, not to behave hybristically. These are not lessons that are startlingly
new, but when they unfold in the various plays, the audience sees their truth,
sees that human life and the order of the cosmos depend on their being true.
Toe lessons do not dazzle with breathtaking novelty, but they convey lessons
of significance nevertheless. Toe teaching in the Platonic dialogues will be
more Iike the teaching in drama than like that in discursive works of
philosophy.
A few of the dialogues seem to have as their lesson the inspiration of
Socrates' memory. Thus the Phaedo shows how courageously Socrates met his
death; the Grito shows him sticking to his post and choosing to stand by the
best argument he has, even when such a choice meaos death. Toe Futhyphro
shows how little Socrates corrupted youth. Even if his conversations with the
various characters in these dialogues are aporetic and leave the youths
bewildered, how has bewilderment harmed them? In the case of Euthyphro,
the conversation seems to have had no effect at ali, and if there has been no
effect, there has obviously been no corruption.
Sorne of the dialogues seem to have attacked presumption. No doubt Plato
was continuing the Socratic mission of testing those who proudly trumpeted
11. In the Seventh Letter, if the letter be genuine, Plato denies ever committing to
writing his serious doctrines . Toe letter's authenticity is with remarkable cyclical
regularity either affirmed or rejected. I do not know whether the letter be spurious or
not; its sentiments seem to me valid. Por a discussion of the letter, with references to the
relevant secondary material, see /nterpreting Plato, 14-15.
How to Read a Platonic Dialogue 131
their wisdom to the world. Thus the Symposium mocks those who presume to
1cnowledge about the gods; it teaches us to beware of those who make up
stories about the gods and expect them to be believed: people tend to create
gods in their own images or to promote their own interests. This surely is a
teaching usefully leamed in ali generations. The Phaedrus is an attack on
uncritical enthusiasm for Iistening to speeches. Clevemess like Lysias's is no
guarantee of truth or sense. Almost the same teaching is found in the
Menexenus, where Socrates shows how easily speeches on set themes may be
roade up.
In other dialogues, like the Crotylus and (perhaps) the Symposium, we Ieam
to follow the mean. In these dialogues, various extreme positions are taken,
by Socr~tes or the interlocutors, and the absurdity of each extreme is depicted.
The audience or reader goes away from the dialogue realizing the folly of the
extremes and thinking that perhaps wisdom Iies in the mean. Thus in the
Cmtylus we leam that the positions of excess naturalism and excess
conventionalism are alike unable to explain language; in the Symposium, we
see perhaps the sorrows that come from the excessively sexed life of
Alcibiades and the deficiently sexed life of Socrates.
These are sorne of the teachings that emerge from a dramatic reading of
12
Plato. I admit that they are devoid of the metaphysical, epistemological,
ontological, logical, political, ethical profundity that philosophers have found
in the dialogues. Yet they are positive teachings and the kind of teachings that
art provides. They are also teachings that would be good to absorb into our
consciousness. Toe history of the world-the dismal record of theological
cbarlatans, of demagogic propagandists, of bleary-eyed philosophers and
mindless politicians, of scientific hucksters, and of people who abandoned
argument whenever it threatened sorne practical advantage- shows how
important to the world were the lessons Plato had to teach in his dialogues.
Now what seems to me the best way to ascertain Plato's views about the
precise philosophical issues that his dromatis personae raise is to look at what
Aristotle says about them. Here, let me be bold: I maintain that to find out
wbat P!ato's views generally are, we should look, not at what Aristotle says
tbose v1ews to have been, but at what Aristotle himself says about the matters!
Two simple examples will, I think, illustrate my notion. When we look at the
latter part of the Phaedrus, we find a scheme for education in rhetoric that
seems to serve as the model for what Aristotle actually does in his Rhetoric·
when we !ook at the idea of a good life that is argued in many dialogues (fo;
example, m the Gorgias) we find a definition of happiness that is essentially
tbe same as that worked out in the first book of the Ethics. And the
correspondence seems to me repeated over and over again. Furthermore, as
12. Por more, see lnterpreting Plato, 247-250.
132 The Tbird Way

I have argued in Interpreting Plato, Plato very often is arguing dramatically

against views put forth by bis interlocutors, rejecting the _view~ o~ both
Socrates and bis disputant, sometimes arguing that the solut1on hes m the
mean sometimes suggesting that the method the disputants are using to pursue
a co~clusion is fallacious or even silly. When Aristotle argues against a
position found in the dialogues, he is arguing discursively against a position
13
against which Plato is arguing dramatically. Thus, the two actually agree!
Let me conclude with this last thought. When I read Plato with my students,
we are, to be sure, interested in what Plato had to say-to the extent that we
can make sense of it-but we are still more interested in the truth of the
matter. Amicus Plato, sed magis amica veritas: we love Plato, but we love
truth more. We continue to argue where the dialogues have left off. We too 9
ask what justice, what piety is; we look at what Thrasymachus and Euthyphro
and Callicles and Meno and Socrates had to say, but then we keep on arguing
among ourselves. Toe dialogues thus become not ends in themselves, but a Plato's Dialogues as Enactments 1
means to further pbilosopbical reflection. Such a use, I hope, is true to Plato's
spirit.
Gerald A. Press

It is common, and commonsensical, to assume that the dialogues, however

peculiarly, contain and assert Plato's pbilosophical doctrines. 2 It is assumed
~at philosophers have doctrines that they present in their treatises, and so the
dialogues come out of much scholarship looking like treatises. Tbis means that
their primary mode of operation is conceived as saying; they are only
secondarily thought of as perhaps doing something. 3 Toe problem is that what

13. Why Aristotle would misunderstand Plato, even when he had studied for many
years at the Academy, is a question I try to answer in lnterpreting Plato, 9-10.
l. This paper has benefitted from the comments of Debra Nails, Holger Thesleff,
Joanne Waugh, and Elinor West. Toe errors that remain are my own.
2. E.g., l. M. Crombie, An Examination of Plato 's Doctrines (London: Routledge,
1962). By "doctrine" I mean, as the word itself indicates, a settled teaching, a definite
view about something that is or can be communicated to others in words. Examples
would be references to "Aristotle 's doctrine of the mean," meaning his view of the
nature of the moral virtues, or to "Christian doctrine," meaning the theological and
moral beliefs that constitute Christianity. It is consistent with this usage that ones
tea~hings may change from time to time, but it is not to be assumed that anyone who
wntes prose, even that anyone who writes "philosophical prose," ea ipso has doctrines
?r that such doctrines are being communicated in that prose. These are propositions that,
1t seems to me, are to be proveo if they are not evident prima facie.
3. On uses of language as doing rather than saying, see J. L. Austin, How To Do
Things With Words (Cambridge: Harvard University Press, 1962) and John R. Searle,
Speech _Acts (London: Cambridge University Press, 1969), Expression arui Meaning
(Cambndge: Cambridge University Press, 1979), Speech Act Theory arui Pragmatics
134 Toe Third W ay
they say seems variously equivoca!, contradictory, or aporetic, and their author
anonymous. 4
One altemative has been to take the dialogues as performances5 or mimes, 6
or as instantiating their philosophical doctrine rather than directly asserti~ it. 7
There is a philosophical doctrine that can be put into propositional or assertoric
form on this view, but Plato has left it to us to observe, infer, and state it. 8
Toe dialogues are mimetic in the sense that they act out a point that is
essentially propositional or could just as well be expressed in propositional
form. 9 This view is an improvement on the previous one, because it
(Dordrecht: Reidel, 1980).
4. Toe logic of these problems is often forgotten. They begin fromfacts about the
dialogues, e.g., Platonic anonymity. It is a fact that Plato nowhere speaks; there is no
character called Plato, and Plato wrote no treatises. This entails that, if you want to
attribute a claim oran argument to Plato, as distinct from the character into whose mouth
he has put it, then you must have a justification other than tradition, comfort, and habit
for the belief that this character speaks for Plato. It might seem so obvious as to be in
no need of proof that Socrates speaks for Plato, but then on what principie can it be
justified to assign that role to the Eleatic Stranger or the Athenian Stranger or Timaeus?
It must be a principie and, indeed, a defensible one, if the attributing of claims and
arguments to Plato is not to be merely a matter of habit and preference, but to show
something about the dialogues. Toe question is addressed directly in a recent article by
Michael Frede ("Plato's Arguments and the Dialogue Form," in Methods of lnterpreting
Plato and His Dialogues, Oxford Studies in Ancient Philosophy, suppl. vol. [1992], ed.
James C. K.lagge and Nicholas D. Smith, 201-219). I think Frede is correct to conclude
that "the dialogues are not philosophical treatises in disguise" (219), but his assumption
that "Sometimes we are confident that an argument ... is Plato's argument, i.e., an
argument Plato himself endorses" (203) begs the very question about the impact of
Platonic anonymity that he himself raised so well on the previous page.
5. Gilbert Ryle, Plato's Progress(Cambridge: Cambridge University Press, 1966) and
Mitchell Miller, The Philosopher in Plato's Statesman (The Hague: Martinus Nijhoff,
1980) and Plato's Parmenides: The Conversion of the Soul (Princeton: Princeton
University Press, 1984). Toe question of the dialogues' performance remains an open
one.
6 . Jacob Klein, A Commentary on Plato's Meno (Chapel Hill: University of North
Carolina Press, 1965).
7. Robert S. Brumbaugh, Plato for the Modem Age (New York: Crowell-Collier,
1962) and "A New Interpretation of the Republic," Joumal of Philosophy 64 (1967):
661-70.
8. This would also be consistent with Kitto, POIESIS: Structure and Thought
(Berkeley: University of California Press, 1966), 243-56.
9. K.lein (Commentary, 17) makes it quite clear that he believes a doctrine is being
acted out: "Answers can be given in a written text by the very action it presents. This
is what usually happens in Platonic dialogues and what constitutes their dramatic or
mimetic quality." Toe idea conflicts, however, with the assertion of the well-known
Plato's Dialogues as Enactments 135
recognizes that the dialogues are dramas or doings, 10 but it still sees them
aiso and ultimately as sayings. Toe proponents of this view are right about the
importance of the performative and dramatic aspects of the dialogues, but the
problem remains that these aspects are, in the end, subordina.ted to the
traditional assertoric and propositional ones. So recent scholarship seems to
offer two altematives: either the dialogues are disguised treatises that directly
assert Plato's philosophical doctrines or else they are plays or mimes that
indirectly express his doctrines by somehow acting them out. 11
I do not mean to assert that these are exclusive altematives, but only to
indicate sorne broad lines of recent scholarly approach to the dialogues. Toe
same may be said more generally: that modero Plato studies have been plagued
by a number of recurrent problems that can be stated as altematives or
oppositions and that resist resolution. Let me mention sorne of them briefly.
■ Much scholarship seems to make us choose between an emphasis on
doctrinal-dogmatic, i.e., on the doctrines of Plato (or Socrates) and an
emphasis on literary-dramatic aspects of the dialogues, such as plot, setting,
and characters. 12
■ Similarly, but not identically, what are sometimes referred to as the
philosophical versus philological approaches; 13 studies focus on the arguments
and their logic or else on traits such as language and structure.
■ Toe doctrinal approach, in tum, raises the problem of dogmatism versus
"philosophical passage" of the Seventh Letter (341c-d) that the kind of knowledge or
philosophy that Plato is after "cannot be put in writing like other subjects," but must,
instead, be generated in the soul.
10. On drama as a doing, see H. S. Thayer, "Meaning and Dramatic Interpretation,"
in Plato's Dialogues: New Studies and lnterpretations, ed. Gerald A. Press (Savage,
Maryland: Rowman & Littlefield, 1992), 47f.
11. One could try to escape the dilernma by claiming that the dialogues are both; that
is, sometimes assertoric and treatise-like, other times (e.g., the myths) indirect and
mimetic. Thus, Glenn Morrow, "Plato and the Mathematicians: An Interpretation of
Socrates' Dream in the Theaetetus," Philosophical Review 79 (1970): 309-33; Amelie
Rorty, "A Speculative Note on Sorne Dramatic Elements in Plato's Theaetetus "
Phronesis 17 (1972): 227-38; and Berel Lang, "Presentation and Representation ln
Plato's Dialogues," Philosophical Forum 4 (1973): 224-40. But this strategy <loes not
escape the dilernma. It turns out either that Plato's philosophy reduces to dogma after ali
(Morrow, Rorty), or else that it is truly a coincidence of form and content, hence a
philosophy of a kind other than dogmatic, as is maintained here.
12. Por examples of doctrinal or dogmatic studies, see "Introduction," n. 13 supra.
13. Michael Stokes (Plato 's Socratic Conversations [Baltimore: Johns Hopkins
University Press, 1986), 10) presents this as the essential characteristic of recent Plato
studies. But see my cornments in Plato 's Dialogues, 1-4.
 136 Toe Third Way
skepticism and leads to the quest for a "third way" 14 that underlies the
present volume. Is Plato a dogmatist, as we are implicitly taught for the most
part today, or is he a skeptic, as might be suggested by the lack of a definite
conclusion to many dialogues and as he was thought to be in the New
Academy of Arcesilaos and Carneades and by a long tradition of "academic"
skepticism?
■ Inconclusive versus conclusive. How are we to interpret the "aporetic"
dialogues in wbich, at least primo, facie, no definite conclusion is reached, no
doctrine is taught? Is there a doctrine bidden or binted at in them? Or do they
really have no doctrine to teach? This is, perhaps, a more serious problem
than many appreciate today; of the twenty-four dialogues generally accepted
as genuine, the vast majority do not, at leastprimo,facie, end in an agreement
between Socrates and bis interlocutor(s) about the answer to the question they
have been discussirg. 15
■ Literacy versus orality. Are the dialogues essentially literate texts, that is,
texts written for an audience of readers as is assumed by those who interpret
them as Platonic treatises? Or do they represent an essentially oral culture in
wbich teacbing (paideia) proceeded in a very different way? 16
■ This leads to the problem of the written versus the unwritten. While we
1
tend to assume that the dialogues are, like other philosopbical texts, the written
!
form of the author's teacbings, the critique of writing in the Phaedrus suggests
to sorne that Plato did not intend the dialogues as expressions of bis true
doctrine at all. 17
■ Socrates versus Plato. Even if we were sure that the dialogues are
14. E. N. Tigerstedt (lnterpreting Plato. Stockholm Studies in the History of Literature
17 [Stockholm: Almqvist & Wiksell, 1977]), followed by Watson (187-208 infra.) shows
how the history of Plato interpretation has oscillated between dogmatic and skeptical
poles. Luigi Stefanini (Platone [Padova: CEDAM, 1932-1935; 2nd ed. , Padova:
CEDAM, 1949; repr. 1992]) tries to find a middle way in the idea of "constructive
skepsis." He cites (I.364) with approval comments on the Republic by V. B. Brocchieri,
La dottrina dello stato nella Cultura Ellenica (Milan, 1934), 167, to the effect that the
dialogue constantly swings from the empirical to the transcendental, from the ideal to the
practica!. See also n. 25.
15. It is worth noting that ofthe twenty generally accepted dialogues in which Socrates
leads the discussion only five are nonaporetic or conclusive in the required sense:
Phaedo, Phaedrus, Philebus, Republic, and Symposium.
16. See the discussions by Hershbell (25-39), West (41-60), and Waugh (61-77). Por
the state of the question on orality and literacy in the history of Greek culture and its
application to Plato's dialogues, see Kevin Robb, Literacy and Paideia in Ancient Greece
(New York: Oxford University Press, 1994), esp. chs. 6-9.
17. In the latter half of the twentieth century, this view is most closely associated with
the "Tübingen school." See "lntroduction," n. 17.
Plato's Dialogues as Enactments 137
doctrinal, it is a problem to distinguish whose doctrines these are since the
roain character in most of the dialogues is "Socrates," and we know that there
was a bistorical Socrates who was a widely influential pbilosopbical figure in
the latter half of the fifth century.
■ This is connected to the problem of the bistorical versus the "Platonic"
Socrates. How can we tell the one from the other? Whose views are actually
presented in a given dialogue? This, in turn, leads into the whole "problem of
Socrates" and the reliability of Plato's characterization in relation to
Xenophon, Aristophanes, and other sources. 18
■ Unitarian versus developmental accounts of Plato's pbilosophy. Did
Plato's pbilosophy remain the same throughout bis Iife? 19 Or did bis
philosopbical beliefs, as seems more likely primo,facie, change and develop?
■ Problems of chronology and stylometry. The developmental view that has
dominated almost all Plato scholarship since the nineteenth century has
depended in large part on stylometric analysis to determine the chronological
order in wbich the dialogues were written, from which, in turn, the doctrinal
development is traced. But serious and fundamental criticisms have been made
of stylometric analysis and the.Iogic of the arguments supporting the generally
accepted chronology.20
■ This leads to an opposition, little appreciated, between chronological and
dramatic orderings of the dialogues. Apart from whether any precise
chronology can be established, is it more illuminating of Plato's pbilosophy to
consider the dialogues in their compositional order or in the order indicated by
18. See n. 60.
19. The most famous of Platonic "unitarians," those who oppose the widespread
acceptance of the idea that Plato's philosophy developed from an early Socratic to later
critica! and Platonic positions, was Paul Shorey; see esp. The Unity of Plato's Thought
(Chicago: The University of Chicago Press, 1904) and What Plato Said (Chicago: Toe
University of Chicago Press, 1933). Por the background of the position, see
"lntroduction," 4-5.
20. The sources of stylometry are Lewis Campbell, The Sophistes and Politicus of
Plato (1867) and Lutoslawski, The Origin and Growth of Plato's Logic (London:
Longmans, Green and Co., 1897). The locus classicus for the history and criticism of
scylometry and chronology is Holger Thesleff, Studies in Platonic Chronology (Helsinki:
Societas Scientarum Pennica, 1982); his "Platonic Chronology," Phronesis 34 (1989):
1-26, is not to be read apart from the ful! monograph. Thesleff's arguments are now
becoming more widely accepted. See, e.g., Jacob Howland, "Re-reading Plato: Toe
Problem of Platonic Chronology Reconsidered," Phoenix 45 (1991): 189-214 ; Richard
A. McNeal, Law and Rhetoric in the Crito (Frankfurt: Peter Lang, 1993), 46-57; Debra
Nails, "Platonic Chronology Reconsidered," BrynMawr ClassicalReview 3 (1992): 314-
27 and "Problems with Vlastos' Platonic Developmentalism," Ancient Philosophy 13
(1993): 273-91, and "Plato's Middle Cluster," Phoenix 48 (1994): 62-67.
138 Toe Third Way
such dramatic indicators as Socrates' age?21
■ Toe problem of myths versus discursive accounts or arguments.
Extending to fables, tales, protreptic sermons, and other non-discursive forms,
the problem is how to understand these if we assume that the dialogues are
22
meant to tell us Plato's doctrines and give us bis arguments for them.
■ Toe problem of indirection versus directness. As suggested earlier, if we
assume that the dialogues express Plato's doctrines, a problem is posed by the
fact of Plato's anonymity. Toe doctrinal approach may be appropriate for texts
23
whose authors address us directly, but Plato evidently does not.
■ Finally, the problem of form versus content. lt is a familiar move to
distinguish sharply between the literary style and dramatic form of the
dialogues on the one hand, and their philosophic content (their doctrines and
arguments) on the other. But the question may be asked whether it is
appropriate, legitimate, or even possible to do so. 24
In the literature, these problems are usually approached individually or a
few at a time in connection with each other, but it is reasonable to suspect that
they are all connected. A clue is provided by the oppositional form in which
they are often expressed, which suggests that the solution will be or involve
21. In her recent edition and translation of the Gorgias, Stefania Nonvel Pieri takes
a middle ground, discussing first the chronological and then the drarnatic position of her
dialogue: Pl.atone. Gorgia (Pilosofi antichi, Nouva Serie 1 [Napoli: Loffredo Editore,
1991)), 29-34.
22. Toe rnyths have long been felt as a problern and, in consequence, there has been
a rninor industry in explaining their role in Plato's philosophy. Studies in the modern
period begin with Henke's De philosophia mythica Pl.atonica (Helmstedt, 1776) and
Huttner's De Mythis Pl.atonis (Leipzig, 1788) and include, among the rnost important
rnonographic studies of this century, J. A. Stewart, The Myths of Pl.ato (London, 1905;
repr. Carbondale: Southern Illinois University Press, 1960); Karl Reinhardt, Pl.atons
Mythen (Bonn, 1927); P. Prutiger, Les mythes de Pl.aton: Etude philosophique et
littéraire (Paris: P. Alean, 1930), and Paul Priedlander's chapter, "Myth," in Pl.ato: An
Introduction, trans. H. Meyerhoff (New York: Bollingen Poundation, 1958), 171-210.
Por a fuller account ofthe discussions, at least up to the early 1980s, see Kent F. Moors,
Pl.atonic Myth: An Introductory Study (Washington, DC: University Press of Arnerica,
1982), ch. l.
23. A point appreciated by Kierkegaard in The Point of View far My Work asan
Author and The Concept of lrony, with Constant Reference to Socrates.
24. Susan Sontag, "Against Interpretation," 13-23, and esp. "On Style," 24-45, in
Against Interpretation (New York: Dell, 1969) is still worth reading. Similarly, R. G.
A. Buxton (Persuasion in Greek Tragedy [Cambridge: Cambridge University Press,
1982], 2): "Every play worth the narne is a unique imaginative creation, a complex of
words, silence, gestures. movement and (sometimes) dance and song; all of which, taken
together, constitute the work's total effect or 'meaning. "'
Plato's Dialogues as Enactments 139
a way of transcending these oppositions.25
I propose a solution in sorne respects similar to Kant's "Copernican
Revolution. " 26 On the prevailing assumption that a philosophy is a matter of
assertions, doctrines, content, and treatises, we have n~t been able to solve the
problems of Platonic scholarship, since this assumption always leaves aspects
of the dialogues unexplained, excluded, or unintegrated into what we take
Plato's philosophy to be. In fact, we are confronted by a set of opposing
altemative orientations. Let us, therefore, try looking at matters from a
different point of view. My proposal also owes something to Jaeger's
description of how a history of paideia would have to be written. 27 Instead
of beginning with a definition of what philosophy or a philosophical text is
(i.e., doctrines supported by arguments), we may have more success by
beginning with the assumption that the dialogues are philosophical documents
and then make our concept of Plato's philosophy conform to what the
dialogues are. 28
In other words, I propose that we rethink or reconceive the dialogues as a
different kind of document or text: neither treatise nor mime, but enactment.
This may offer the beginning or hope of a solution to the persistent problems
we have had by enabling us to bridge the gap between the opposed orientations
discussed above. At the same time, I am painfully aware of what an early
stage I am at in being able to explain what I mean by "enactment."
By referring to the dialogues as enactments, then, I mean to say what kind
of documents they are: I do not think that they are wholly or principally
25. See Walter Watson, "The [?ogrnatic Plato, the Skeptical Plato, and the Dialogic
Plato," 187-208.
26. Second Preface to the Critique of Pure Reason.
27. Jaeger, Paideia, trans. Highet (New York: Oxford University Press, 1939-45),
vol. 3, 300, n. 3.
28. This is actually more consistent with the broad ancient meaning of the tenn
philosophia, which, in any case, was not in wide use before Plato. Indeed, he seems to
have invented the narrow and technical notion of philosophy as a special discipline in
contradistinction frorn that of the sophists, as Jaeger (Paideia [New York: Oxford
University Press, 1944], vol. 3, book 4, chs. 2, 8) and Havelock (Preface to Pl.ato
[~arnbridge: Harvard University Press, 1963]) already saw. More recently, Monique
Dixsault [Le naturel philosophe (Paris: J. Vrin, 1985)] has considered the invention of
philosophy in the dialogues frorn a variety of points of view; see also Alexander
Neharnas, "Eristic, Antilogic, Sophistic, Dialectic: Plato's Dernarcation of Philosophy
frorn Sophistry," History of Philosophy Quarterly 7 (1990): 3-16 anda paper delivered
to the Society for Ancient Greek Philosophy by Edward Schiappa, "lsocrates and
Canons. ~
140 Toe Third Way
treatises nor that their primary mode of philosophizing is assertoric.2 9 I do
not mean to say that they are nowhere or never assertoric, but that such
assertions as may be made cannot be abstracted from their full context without
falsification. 30 A simpler but probably just as mysterious way of saying the
same thing is that the dialogues do and make as well as say.
Toe word "enact" means both to act or play a part and to make to be, as a
legal enactment; thus, it articulates two things that I think are primary: (1) the
by now less controversia! point that the dialogues are plays, whether they were
performed, read aloud, or read silently, and (2) the perhaps more controversia!
point that-beyond saying things-they actually make things.
Albert Cook writes: "A play enacts. In doing so it not only makes a
statement; it also creates an ejfect (my emphasis), in a given context, on an
idealized (Greek) spectator. Toe effect unites with the statement, as the mythos
with the logos. " 31 I emphasize the phrase "creates an effect" in contrast with
"makes a statement" because I think it expresses something true and essential
about Plato's dialogues: whatever statements they may make, whatever
doctrines they may assert, they also create effects that are distinguishable from
those statements and may even contradict them. 32 Toe most obvious example
29. Victorino Tejera (Plato's Dialogues One by One: A Structurallnterpretation [New
York: Irvington, 19841) follows Bucbler (Toward a General Theory of Human Judgment
[New York: Dover, 1951), Nature and Judgment [New York: Columbia University
Press, 19551) in distinguishing among "modes of judgment" and argues that the
philosophy of the dialogues is in the "exhibitive mode." I agree that it is important to
distinguish different modes in which philosophy may be written or done, and in that
sense I agree that the dialogues are exhibitive. I also think, however, that Buchler's
modes ofjudgment reduce to different ways of saying something propositional and to that
extent our analytic too! kit needs to be supplemented. This is what I attempt briefly and
without detall here by speaking of "modes of philosophizing." Philosophers assert or
claim things, but they may also philosophize in other modes.
30. I merely restate here what might be called Toe Schleiermacher Principie: that
statements made by characters in the dialogues can only be properly understood in their
full context of character, setting, and moment. Schleiermacher's Introductions to the
Dialogues of Plato, trans. William Dobson (Cambridge & London 1836), 14.
31. Albert Cook, Enactment: Greek Tragedy (Chicago: Swallow Press, 1971), xv.
32. Daniel Anderson (The Masks of Dionysos [Albany: SUNY Press, 19921) argues
that, following the hints in the Phaedrus and the Seventh Letter, Plato has screened his
audiences by putting inco~istencies or contradictions into the dialogues that will indicate
to the attentive reader that s/he should move to a different leve! of analysis or
interpretation. Similarly, David Halperin argues that "Plato . . . systematically goes
about undermining and subverts the very theories that his philosophical personae
propound" precisely in order not to "leave his readers with a body of dogma" ("Plato
and the Erotics of Narrativity," in Methods of lnterpreting Plato and his Dialogues,
118f.) .
Plato's Dialogues as Enactments 141
of this would be how often Socrates' statements that he doesn't know seem to
us and even to other characters in the dialogues to contradict the true state of
affairs.
If statements operate primarily on the mind, on the intellectual or rational
level of experience, these other effects that the dialogues create operate
through the imagination and emotions. I speak as if the intellect, imagination,
and emotions were separate and distinct, but I don't believe that and I don't
believe that Plato believed it. In any case, in speaking of the dialogues as
enactments, I mean that they create effects in and through the imaginations and
erootions of the audience or readers as much as, perhaps even more, than
through reason or intellect.
How do the dialogues do this? Let me appeal briefly to the examples of
Dante, More, and Cervantes, who deliberately and artfully break down the
barrier between "reality" and "fiction" 33 • Toe principal character of the
Divine Comedy is Dante himself, who "narrates" what he "experienced" in
hell, purgatory, and paradise; in consequence, it becomes difficult or
impossible to distinguish between Dante the poet and Dante the pilgrim. Of
course, rationally we know that Dante the poet, a historical person about
whom we possess reliable factual information, wrote the lines we read. Yet
when we say something like, "Dante says that .. . " or "Dante describes . .
. " or even "Dante writes . . . ", which Dante do we mean? And how can we
be sure?
In Utopía, More makes himself a character-or, rather, like Dante, creates
a fictional character called More-who is present at the description by Raphael
Hythlodaeus of the island of Utopía and its customs. While Raphael is a true
believer in the utopian system, the character More is and remains dubious
about the value or possibility of importing utopian ways. Certainly this
insulates More the author against possible attack for expressing such opinions,
but doesn't it also make us unable to tell what that More believed?
Throughout Part I of Don Quixote, Cervantes repeatedly refers to his work
as a "truthful account" and as "history." In Part II, partly in response to the
counterfeit Part II of the work by a certain Avellaneda that had been in
circulation, 34 Don Quixote encounters persons who have read about his
exploits and are anxious to meet him. But we are at this very moment reading
about his exploits, so what is the difference between these fictional "readers
33. Or history and poetry. Elinor West rightly reminds me that the word "fiction" is
dangerous in these contexts. Toe text I was teaching when I realized that Plato effaced
this distinction was the Phaedo. There, the narrative structure cooperates with the cast
of characters, the mythological revision, the dramatic action, and the conversation to
accomplish the immortality of Socrates.
34. See, e.g., Cervantes, Don Quixote (New York: Signet, 1979), 526, n. 2.
142 Toe Third Way
of Don Quixote" and us? Cervantes, Iike Dante and More, uses a variety of
strategies to blur our usual sense of the distinction between reality or history,
on the one hand, and an imagined world of fiction or poetry, on the other.
Plato does that too, and so successfully that scholars are still trying to decide
where the "historical" Socrates leaves off and the "Platonic" Socrates
begins. 35 Unlike More and Dante, Plato does not malee himself a character
in bis own fiction. Like Cervantes, who carefully attributes authorship to a
certain moor named Cide Hamete Benengeli, Plato sometimes constructs
elaborate fictions about the origin of the text or the narrative we are
reading. 36
But unlike in the Divine Comedy, Utopía, and Don Quixote, the barrier that
is effaced is not only between "reality" and "fiction," but also- and more
importantly for philosophical purposes-between the ordinary or popular view
of reality and a very different one that is Plato's. 37 In other words, the effect
that is created by Plato's dialogues is the reality-for-the-audience, the actual
experience of the world as he sees it, not only intellectually entertained
through our thinking about claims and arguments, but also imaginatively
entered into and emotionally charged or felt. I will suggest sorne of the
dimensions of this in a moment.
Toe effect is created by various techniques used in compound and complex
ways. One is the retelling and revision of stories (legendary, poetic) that are
defining or formative ones for the Greeks individually, socially, and
cosmically. 38 Because these stories have the cultural stature that they have for
Plato's audience, they will involve not only reason, but also memory,
35. Perhaps a testimony to this is the enormous modero "Socrates" literature. See
Richard D. McKirahan, Plato and Socrates: A Comprehensive Bibliography (1958-1973)
(New York: Garland, 1978) and Andreas Patzer, Bibliographia Socratica (Freiburg:
Alber, 1985). Toe most recent works in English are: Gregory Vlastos, Socrates: lronist
and Moral Philosopher (lthaca: Coroell University Press, 1991); Luis Navia, The
Socratic Presence: A Study of the Sources (New York: Garland, 1993); Socratic
Questions, ed. Barry S. Gower and Michael C. Stokes (London: Routledge, 1992) and
Thomas C. Brickhouse and Nicholas D. Smith, Plato's Socrates (New York: Oxford
University Press, 1994). Ali except Navia believe that it is readily possible to make the
distinction.
36. E.g., Theaetetus, Parmenides, Symposium, Phaedo.
37. By "view" here I do not mean Plato's doctrine of reality, but his vision, to which
we will turo below.
38. Stories from the Homeric poems, for example, frame the Crito and Protagoras.
On the intellectual and social role ofthe Homeric stories, see W. J. Verdenius, "Homer,
the Educator of the Greeks," Medelingen der Koninkl,jke Nederlander Akademie van
Wetenschappen 33 (1970): 9ff. and Eric Havelock, Preface to Plato. On Socrates as a
"hero," see Robert Eisner, "Socrates as Hero," Philosophyand Literature6 (1982): 106-
18.
Plato' s Dialogues as Enactments 143
imagination, and feelings. lrony and anonymity also have not only the
intellectual effects that are usually discussed, 39 but also, and perhaps more
significantly, effects on feeling and imagination. 40 Irony is not merely a trick
that, once solved, becomes a key enabling us to decode Plato's encrypted
message. Other techniques with similar effects are the use of metaphorical and
allusive language,41 of characters talcen from the historical experience of the
audience, 42 and of the numerous types of humor. 43
39. Toe general studies by J. A. K. Thompson, Jrony: An Historical Jntroduction
(London, 1926), D. C. Muecke, The Compass of Irony (London: Methuen, 1969), and
Wayne Booth, The Rhetoric of lrony (Chicago: University of Chicago Press, 1974)
should be supplemented by the more specifically Platonic research of Gavin Ardley, "Toe
Role of Play in the Philosophy of Plato," Philosophy 42 (1967): 226-44, W. Boder, Die
sokratische Ironie in den plaronischen Fruhdialogen (Amsterdam: Gruner, 1973), Charles
L. Griswold Jr., "Socratic Irony and Platonic Irony" in Plato's Dialogues: The
DialogicalApproach, ed. V. Tejera and R. Hart (forthcoming), Paul Plass, "Philosophic
Anonymity and Irony in the Platonic Dialogues," American Joumal of Philology 85
(1964), C. Rowe, "Platonic Irony," Novus Tellus: Annuario del Centro de Estudios
Clásicos 5 (1987): 83-101, and especially G. J. de Vries, Spel bij Plato (Amsterdam:
North-Holland, 1949), unfortunately published only in Dutch with a summary in Prench.
40. Preud's account of irony (Wit and its Relation to the Unconscious, ch. 6; The
Basic Writings of Sigmund Freud, trans. and ed. A. A. Brill [New York: Modero
Library, 1966]. 757) could have been written with Plato's dialogues in mind. "Toe
essence of irony consists in imparting the very opposite ofwhat one intended to express,
but it precludes the anticipated contradiction by indicating through the inflections,
concomitant gestures, and through slight changes in style-if it is done in writing-that
the speaker himself meaos to convey the opposite of what he says. Irony is applicable
only in cases where the other person is prepared to hear the reverse of the statement
actually made, so that he cannot fail to be inclined to contradict. As a consequence of
this condition, ironic expressions are particularly subject to the danger of being
misunderstood. To the person who uses it, it gives the advantage of readily avoiding the
difficulties to which direct expressions, as, for example, invectives, are subject. In the
hearer it produces comic pleasure, probably by causing him to make preparations for
contradiction which are immediately found to be unnecessary."
41. Por the ancient recognition of Plato's literary art, see P. Walsdorff, Die antiken
Urteile über Platons Stil, Klassische-philologische Studien, 1 (Bonn: Scheur, 1927). On
matters of style, see Holger Thesleff, Studies in the Styles of Plato (Helsinki: Societas
Scientarum Pennica, 1967).
42. On the significance of the characters, see, e.g., Mary Whitlock Blundell,
"Character and Meaning in Plato's Hippias Minar," in Methods of Jnterpreting Plato and
his Dialogues, 132-72, and Lucinda Coventry, "Toe Role of the Interlocutor in Plato's
Dialogues," in Characterization and /ndividuality in Greek Literature, ed. C. Pelling
(Oxford: Oxford University Press, 1990), 174-196.
144 Toe Third Way
To suggest what I mean, briefly: in reading fiction we are prepared and we
expect to mentally (imaginatively) inhabit a fictional world; this is, perhaps,
the principie source of the pleasure fiction provides. There is a whole-minded
entry into a fictive world, a willing suspension of our disbelief. But this is just
what the strategies of Dante, More, and Cervantes and, I am arguing, Plato
keep us from doing at the same time that they draw us into a world that we
intellectually know to be imaginary. Toe result is that poetry and bistory fuse,
with consequences that transcend Aristotle's distinction between them in the
Poetics (9; 145la36-bl2). Toe measure of Plato's success in this is that he is
the only one whose fictional world has so regularly and so persistently been
mistaken for the bistorical one.
Like other dramas, as Albert Cook noted, 44 Plato's dialogues "create an
effect." What is the effect that they create? I aro arguing that unlike treatises
or mimes, Plato's dialogues actually create in the reader or audience the
experience of the world as Plato envisions it. lt is important to note that they
do this, at least sometimes, apart from any doctrine that could conceivably be
inferred; it is at least arguable that sometimes they create an experience in
conflict with doctrines that seem to be taught, or at any rate have been put into
the mouths of Plato's characters.
On my Kantian or Jaegerian strategy, this implies that the philosophy to be
found in the dialogues just is this experience; in other words, it is
comprehensively what is presented to the audience or reader intellectually,
emotionally, and imaginatively. This means, first of all, that Plato's idea of
philosophy is not a matter of doctrines: neither positively (so that he becomes
a neo-Pythagorean or Neoplatonist) nor negatively (so that he becomes a
skeptic). What needs to be rejected is the supposition that a pbilosopher must
be either a dogmatist or a skeptic, the supposition that these are the
comprehensive and mutually exclusive options. But Plato never speaks;
therefore, in the plain sense of the words, there are no Platonic doctrines in
the dialogues nor does Plato assert that dogmatic assertions are not possible.
This is not to say that Plato has no pbilosophy, but that what he means is
not of the same ldnd as what we usually mean. There are recurrent principies,
beliefs, attitudes, and values that we may-indeed, we feel compelled
to-attribute to bim. I agree. But I deny that they are settled doctrines that
could be written down or taught orally either in the unitarian or the
43. On Plato's use of humor, see Otto Apelt, "Über Platons Humor," Neue Jahrbuch
für das K/assische Altertum (1902): 247-66; Gavin Ardley, "The Role of Play in the
Philosophy of Plato," Philosophy (1967): 226-44; H. D. Rankin, "La ughter, H umour and
Related Topics," C & M 28 (1967): 186-213 and "A Modest Proposal about the
Republic," Apeiron 2 (1968): 1-22.
44. See n. 31.
Plato's Dialogues as Enactments 145
ctevelopmental sense, being persuaded instead of Tejera's and Nails's "open-
endedness. "45 I would include among them such beliefs as that there is a
soul, that it is more important than the body, that there are such things as
virtue, courage, wisdom, and the rest, that pbilosophy is the best way of Iife,
and so on. Two comments: (1) these are general beliefs and I note that wbile
Socrates often states bis conviction of them, he equally often denies any
knowledge of the specifics (e.g., what exactly virtue is) and (2) Socrates'
certainty is moral rather than logical. He never professes to prove any one
them; rather than being conclusions, they are usually the premisses of
arguments and propositions to wbich Socrates insists on getting bis
interlocutors' agreement. If I seem to be denying to Plato what is the
distinctive mark of pbilosophy by reducing bis beliefs to these generalities, I
have several replies to make.
First, recall that I aro explicitly suggesting that our expectations about what
philosophy must have been for Plato may be anachronistic. 46 Let me
emphasize this point. I do not say that no claims or arguments are made; nor
do I say that we should ignore the claims and arguments and any doctrines that
are proposed or discussed. 47 We should, indeed, study them closely, examine
the arguments, reject the faulty ones, and look for better. This is to do what
Socrates repeatedly insists that he is doing and tells bis interlocutors that they
should do. In so doing, we are pursuing the knowledge and truth, the pursuit
45. Tejera (Plato 's Dialogues One by One, 7) well describes the dialogues as having
"literary closure while remaining intellectually openended." Debra Nails takes this idea
one step further in her concept of "The Double Openendedness of Plato's Method"
(Society for Ancient Greek Philosophy, October 1993). It is consistent with positions
recently taken by Anderson (1992), and Kosman, Halperin, and Sayre in Methods of
Interpreting Plato arui his Dialogues.
46. "Retrojections," Cherniss might have said. See The Riddle of the Early Academy
(Berkeley: University of California Press, 1945; repr. New York: Russell & Russell,
1962), 61ff. He was arguing, of course, against assumptions, (still common) about what
the early Academy was and how it operated. See my The Development of the Idea of
History in Antiquity (Montreal: McGill-Queens University Press, 1982), 126-28, 135,
142-43.
47. E.g., the doctrines that may be hinted at of piety (Euthyphro), courage (lAches),
friendship (Lysis), poetic inspiration (Ion), justice (Republic), or moderation (Channides) ,
that knowledge is recollection (Meno, Phaedo) or right opinion with a lagos (Theaetetus) ,
or the various accounts of the soul (Phaedo, Phaedrus, Republic). Even, I would go so
f~r as to say, the theory of Ideas; for my point is not that Plato does or does not (did or
d1d not at sorne moment in his life) believe them to be true, but that the dialogues are not
meant to teach them to us as Plato's doctrines.
146 Toe Third Way
of which is philosophy; we study them qua truth. 48 Wbat we should not do
is study these claims, arguments, and doctrines qua Plato's settled doctrines
that we may remember and repeat as Plato's. 49 Toe question of Plato's own
adherence to any of them has been carefully, deliberately, and consistently
made indeterminable.
Second, I would agree that the dialogues do actually and are intended to
teach tbeir readers and audiences. However-a point that has been stated by
Jaeger and Havelocicl° but tends to be lost sight of-while the dialogues do
"teach," tbe way in which they do so is not what we, conúng later in a long
tradition of professional and technical philosophy, expect. That is to say, they
are paideutic in tbe ancient and general sense, not in a modero and distinctly
philosophical one. 51 What is taught in this sense in the Protagoras, for
example, is no doctrine of virtue (unified or otherwise) nor of hedonism, but
exactly what the franúng interaction with Hippocrates suggests: that the
sophists are variously foolish, vain, competitive, self-contradictory, ignorant
of the very things about which they claim to be teachers, and therefore
dangerous both to the soul of the student and to tbe society in which they ply
their trade. But these are not the sorts of claims or doctrines in which
subsequent philosophers have been interested or have thought to be the special
province of philosophy.
Third, with Thesleff, 52 rather than speak of doctrines, I believe it is more
accurate to call this all a "vision." There are severa! reasons for doing so. For
one thing, it is consistent with the language of psychic seeing that Plato
sometimes makes Socrates use. 53 For another, the terms theoria and theasthai
as indicating what philosophy is and what philosophers do are used in
48. What might be called the "analytic tum" given to the study of ali ancient
philosophy since the mid-century has been true to this aspect of Plato's dialogues. Toe
conviction of scholars from Ackrill and Bambrough in the 1950s to Bames and Williams
in the 1980s that the study of ancient philosophers could be productively combined with
present philosophizing, that what is of interest is not just what they may have thougbt,
but what the truth is, coincides with the view often stated by Plato's Socrates.
49. And in this respect the analytic tum has erred. Por it does not follow from
Socrates' having said in tbe dialogues that philosophy is concemed to know the trutb tbat
the dialogues actually state or are intended to state what Plato thinks the trutb is.
50. By Jaeger, Paedeia, and Havelock, Preface to Plato, n. 28 supra; we have
recently been well reminded of this by Robb, Literacy and Paideia in Ancient Greece.
51. Sometbing recognized by Miller (The Philosopherin Plato's Statesman and Plato's
Parmenides) and Alexander Nehamas, "Eristic, Antilogic, Sophistic, Dialectic," 3-16.
52. Holger Thesleff, "Looking for Clues: An Interpretation of Sorne Literary Aspects
of Plato's 'Two-Level Model,"' in Plato's Dialogues, 17-43.
53. E.g., Phaedo 99b.
Plato's Dialogues as Enactments 147
significant contexts in severa! dialogues. 54 Moreover, it captures the
peculiarly Greek sense of theoria as a cognate of theatron. Toe dialogues are
certainly, from one point of view, full of theories, but theories more as visions
tban as doctrines.
The term "vision" also seems more appropriate to the kind of writing
Plato's dialogues are, their use of myths, the mystifying critique of writing in
tbe Phaedrus, and the language of the Seventh Letter about philosophic
1cnowledge as something not amenable to expression in writing, but rather like
a flame kindled in the soul. What he has to show us is bis vision, and it is also
a vision with respect to us, who "get it"-in plain words-only as a "vision."
I think this is just facing the facts: about the dialogues as fictions, whether
fictional speeches, dramas, or narratives and about how little agreement Plato
scholars have been able to come to about Plato's doctrines in the kind of detail
and precision that we expect on the assumption that doctrines and systems are
what philosophy is all about.
So a first point about Plato's philosophy is that it is a vision rather than a
roatter of doctrines. As important, though, that vision, those beliefs even
insofar as they are expressible iµ words, are not the only point or focus of the
dialogues. What the dialogues show us is that for Plato, philosophy is also a
roatter of attitudes and practices; philosophy is a way of life, as the Hellenistic
schools correctly saw. It is not only thoughts and ideas; philosophy is also a
roatter of feelings and images, doubts and hesitations, moral and political
purpose, continuing engagement with others, perpetua! reflection, dialogue. 55
54. E.g., Phaedrus 250b (justice, temperance), 250e (beauty), Symposium 211d
(beauty), Republic III, 402d (beauty), Republic VII, 517d speaks of tbe descent from
divine theoria to tbat of human things, Republic V, 486a speaks of tbe philosophic nature
as "a mind habituated to thoughts of grandeur (megaloprepeia) and contemplation
(theoria) of ali time and ali existence" (Shorey). Shorey (Loeb Republic ad loe) compares
the passage witb 500b-c, Theaetetus 173e and l 74e where similar tbings are said about
the philosopher. His references to similar sententiae in Marcus Aurelius (7.35), Livy
(24.34), and Cicero, De Natura Deorum 2 suggest that it became a commonplace in
ancient discussions of philosophy.
55. As for Peirce, so, I would say, for Plato, tbe Pina) Truth is the goal sought, but
as a guiding idea ratber tban as sometbing we really expect to attain. Similarly, John
Dewey writes in "Prom Absolutism to Experimentalism": "Plato ... still provides my
favorite philosophic reading. Por I am unable to find in him that all-comprehensive and
overriding system which later interpretation has, as it seems to me, conferred upon him
as a dubious boon. Toe ancient skeptics overworked another aspect of Plato's tbought
when they treated him as tbeir spiritual fatber, but they were nearer tbe trutb, I think,
Iban tbose who force him into tbe frame of a rigidly systematized doctrine.. . . Nothing
could be more helpful to present philosophizing than a 'Back to Plato' movement· but
it would have to be back to the dramatic, restless, co-operatively inquiring Plato of the
Dialogues, trying one mode of attack after another to see what it might yield; back to tbe
148 Toe Third Way
What the dialogues create in us is the experience of this as philosophy, and
of Plato's vision: of human life as lived in a reality through which we catch
enrapturing glimpses of the ideal; 56 but frustratingly, the ideal remains
beyond our grasp just because we live in time and space. Precisely because of
their disconcerting combination of the eternal and the ephemeral, the ideal and
the real, the dialogues actually embody (wbat sorne of them try to articulate)
bow the eternal and ideal is glimpsed but never really grasped in the
epbemeral real world in whicb alone we philosophize. As I indicated earlier,
I think the dialogues do and make mucb more than they actually say.
They cause us to experience both stability and change; as they embody
Plato's vision of reality, so they themselves both remain wbat they are and are
always becorning. What is stable is the morally and intellectually serious
practice of improving oneself, one's neigbbors, and one's fellow citizens,
whicb is the pursuit of truth, knowledge, wisdom, or, in other words,
philosopby. Toe dialogues contain stable doctrines and a stable doctrinal
system only in the sense that tbese are (said to be) tbe aims of philosophizing.
What cbanges, on tbe otber band-botb in tbe dialogues and about tbem-is
tbe propositions tbat one finds oneself compelled to affirm-including no
proposition-as trutb or knowledge at different times, under different
circumstances, and in conversation witb different interlocutors.
Plato's dialogues presenta vision of human life as a troubling but pregnant
interpenetration of tbe stable and tbe evercbanging and of philosopby as
conversational interactions tbat are dramatic, as are tbe kinds and
circumstances of human lives. To put it metapbysically, tbe dialogues create
in us tbe experience of essence-in-existence, of forms-in-things. 51
If I am on tbe rigbt track in what I bave been saying, tbe idea tbat the
dialogues are enactments provides a way to transcend tbe oppositions, to solve
many of tbe problems tbat have persisted in modera Plato scbolarship. Again,
I can only briefly suggest at the moment wbat I think the solutions are.
■ Dogmatism versus skepticism is a false dicbotomy, at least as applied to
Plato. Plato has beliefs, attitudes, principies, but tbey are not asserted
dogmatically. It is a philosopby, but not one wbose defining characteristics are
doctrines or tbe solutions to philosophical problems.
■ Toe "aporetic" dialogues are not a problem unless one expects or
demands tbat a philosopher's writings contain doctrines. But tbere are many
Plato whose highest flight of metaphysics always tenninated with a social and political
tum, and not to the artificial Plato constructed by unimaginative commentators who treat
him as the original university professor." I thank my colleague, Charles Landesman, for
bringing this observation to my attention.
56. Like Alcibiades ' likening of Socrates to the busts of Silenus (Symposium 215aff.).
57. See Watson, 187-208.
Plato's Dialogues as Enactments 149
reasons to believe tbat this is not tbe enterprise of tbe dialogues. Por, as
Socrates states in tbe Meno and implies in many otber dialogues, aporia is
actually an important and philosophical experience.
■ Des pite tbe critique of writing in tbe Phaedrus, we do not bave to believe
either tbat Plato's philosophic doctrine is found in tbe written dialogues or tbat
he taugbt unwritten doctrines orally. Toe Tübingen school (and otbers) are
right to tbe extent tbat Plato doesn't put his doctrines in tbe dialogues, but
wrong tbat he has doctrines expressed orally or that such doctrines are more
important tban what is in the dialogues. In otber words, his true philosopby is
written in tbe dialogues, but it is not something doctrinal or dogmatic.
■ Toe distinction between Socrates and Plato is a problem in tbe dialogues
only if one's primary interest in reading tbe dialogues is to establish doctrines
and tberefore to distinguish between tbose attributable to tbe former and tbose
attributable to tbe latter.
■ Sirnilarly, distinguishing between tbe historical and the Platonic Socrates
is important to tbose who want to establish tbe views of tbe historical Socrates.
But it ceases to be a problem in the interpretation of Plato's dialogues if we
look at tbe dialogues as dramatic enactments in which tbere is a chamcter
called Socrates who, like many otber characters in tbe dialogues, bears sorne
resemblance to historical cbaracters, but wbose appearance bere is part of tbe
broader literary-philosophic or paideutic strategy of tbe autbor.
■ Toe opposition between unitarian and developmentalist approacbes to tbe
dialogues dissolves if we give up tbe assumption tbat Plato's philosopby must
consist of sorne doctrines that eitber do or do not develop and change over
time. I agree witb Shorey tbat tbe main lines of what I would call Plato's
vision remain tbe same. At tbe same time, I think it is likely, tbougb not
finally very important, tbat Plato's interests and literary practices changed.
■ Toe problem of mytb and other nondiscursive forms is a problem only if
you expect pbilosophic utterances to be assertoric or propositional, and a
philosophic text to be expository prose, a treatise. In Plato's dialogues,
however, botb mytbs and arguments are actually tools for getting us to have
an experience and to form us intellectually and morally.
■ Sirnilarly, tbe problem of anonyrnity, like mytb, irony, and aporia, is not
a problem to be solved, but a technique tbat Plato used to accomplish wbat he
wanted, whicb was not tbe construction or communication of a body of settled
doctrine. By keeping bis own beliefs and argument hidden, be encouraged his
audience to confront their own beliefs and arguments, tbat is, to become better
pbilosophers. 58
■ As for form versus content, seeing tbe dialogues as enactments suggests
58. See Kosman, Halperin, and Sayre in Methods of Interpreting Plato and his
Dialogues.
150 Toe Third Way
how, rather than an opposition, they actually coincide. In the essay previously
mentioned, Susan Sontag writes,
A work of art encountered as a work of art is an experience, not a statement or an
answer to a question. . .. works of art ... give rise not to conceptual knowledge
(which is the distinctive feature of discursive or scientific knowledge-e.g.,
philosophy, sociology, psychology, history) but to something like an excitation, a
phenomenon of commitment, judgment in a state ofthralldom or captivation. Which
is to say that the knowledge we gain through art is an experience of the form or
style of knowing something, rather than a knowledge of something (like a fact or
a moral judgment) in itself (30).
Sontag may distinguish a Iittle too carefully between works of art and attempts
at "discursive or scientific knowledge," or she may just not be thinking of
Plato, but what she says here about "a work of art encountered as a work of
art" applies to Plato's dialogues encountered as enactments, as what they are,
rather than as disguised ("discursive or scientific") treatises. They do not give
rise to conceptual knowledge, knowledge of a fact or moral judgment. Rather,
they create a kind of excitation, commitment, and captivated judgment, and the
knowledge that we derive is in the "forro or style of knowing," rather than in
an object or judgment known. To understand the dialogues and our experience
of them in this way seems to me to explain better than can be done on the
treatise and doctrine approach, both why they are so endlessly fascinating and
why we never come to clarity about what Plato's philosophic doctrines are.
Ali of this indicates how considering the dialogues more in terms of what
they do and make than in terms of what they appear to say might help us get
beyond a lot of persistent problems. Finally, let me give sorne idea of what it
would mean to look at sorne particular dialogues as enactments, as I have been
suggesting.
First, two general examples. Without reiterating the interpretive problems
about Plato's alleged metaphysical dualism, I would point out that the vision
of a two-level reality (one-many, same-different, immutable-changing, etc.) is
enacted in my sense, that is, brought into being in the experience of the reader
in virtually every dialogue, by the contrast between Socrates and his
interlocutors with respect to character, behavior, lives, and attitudes, apart
from any claims or arguments made. Another way of saying this is that
virtually ali of the dialogues embody and operate within the two-level vision,
but they vary in perspective, emphasis, and leve!.
Another, perhaps more striking, example is that of philosophy as a distinct
discipline. There is no persuasive evidence that anything like what we and
Plato call philosophy existed before his time. I would attribute its existence to
the dialogues, creating something to be called philosophy as distinct froro
Plato's Dialogues as Enactments 151
soroething else called rhetoric or sophistry. 59
More specifically, the Protagoms can be read as "Socrates' nekuia "
cteveloping the Homeric theme of a journey to the underworld (315c-d); i.;.,
an enactment, as I have already mentioned, of Plato's vision of the teaching
of virtue, actually a revision of the traditional idea, and of his judgment of the
sopbists as teachers. Toe drama of the Protagoras is an exhibition, for young
Hippocrates' benefit, of the sophists' ignorance and the psychic danger that
poses _for ~eir ~ould-be pupil. Socrates' fallacious arguments and seeming
hedomsm, hke h1s extremely aggressive and contentious tactics, are meaos of
bringing about tllat exhibition at the more personal, particular, and phenomenal
level at the same time that they offer us glimpses of and inducements to the
higher leve!.
The Phaedo is certainly about the immortality of the soul, but most or ali
of the arguments allegedly put forward for tlle claim that the soul is immortal
seem poor and unsatisfactory, though scholars never tire of tinkering with
them to make them work. I suggest that we try reading it as an enactment of
the immortality of Socrates, and I want to point out only two laige-scale
aspects of that. First, of course, is the insoluble confusion of the historical
Socrates and the character in the dialogue; second, the effect of the narrative
~aming. Phaedo narrates the discussion to Echecrates apparently a Iongish
ume after Socrat~s death. But the effect of this is that, within the narrative,
Socrates speaks m the present tense; he remains alive. More often than in
many other narrated dialogues, Plato even reminds us of the narrative frame
by r~turning to i~ momentarily, and yet while we read and even after we stop
reading, we contmue to speak of Socrates in the present tense· so Socrates has
become immortal in the souls of us who read tllese dialogue~ and from there
he60 has become an immortal figure in Western culture.
. 59. ~s position is adumbrated by Alexander Nehamas ("Eristic, Antilogic, Sophistic,
~ia~ecttc )_- It was suggested earlier by Jaeger (Paideia) and Havelock (Preface to Plato).
Similarly, 1t has recently been argued by A. T. Cole (The Origins of Rhetoric in Ancient
Greece [Baltimore: Johns Hopkins University Press, 1991)) that Plato should be
understood as the inventor of rhetoric.
60: Toe "he" who, I argue, has become immortal is Plato 's Socrates, the character in
~e d1~logues. This is independent of whether or to what extent that character is like the
histo~cal Socrates. I perso~ally _do not believe that it is rationally possible to distinguish
Platos character from the histoncal Socrates, and I believe that, closely examined, many
of tho~e who work on "Socratic philosophy" base their arguments on a combination of
the w1sh to know what Socrates really believed and assumptions that are untestable.
How_ever, fo_r the present purposes, which are to understand and interpret the dialogues
cons1stent w1th accepted philological and historical evidence and methods, that doesn't
~atter. If ~cholars are interested in ascertaining what the historical Socrates believed and
if they believe that Plato's dialogues are a useful source of information on that subject,
152 Toe Third W ay

Toe Phaedrus can be read as enacting the philosophical seduction of

Phaedrus, attempting to carry him away from the level of ephemeral
persuasion and everchanging opinion, on which Lysianic rhetoric remains, to
the quest of eternal knowledge and proof that is philosophy as embodied by
Socrates, an enactment with cultural ramifications-mythological, sexual, and
intellectual.
I have only been able to sketch here positions and interpretations that will
require a good deal more explanation and argument. But I hope to have made
at least a prima facie case for studying the dialogues as enactments, a kind of
text essentially unlike the familiar treatise-furm of Western philosophical
writing, and embodying a vision of philosophy that is, likewise, essentially
different from the more traditional Western notion of philosophies as a matter
of doctrines and systems. I hope to have made it at least plausible that Plato
IV
is neither a dogmatist nor a skeptic, but someone for whom, on the contrary,
philosophy is a matter of vision and character and a way of life.
Dialectic and Dialogue:

Between Skepticism and Dogmatism

that is fine. But it is a conclusion more extreme than mine-and therefore requiring more
and bener evidence-that the claims and arguments of Plato's Socrates were those of the
historical Socrates.
 10

Self-Knowledge, Practica! Knowledge,

and Insight:
Plato's Dialectic and the Dialogue Form1

Francisco J. Gonzalez

Toe question of why Plato wrote dialogues is, like most questions in
Platonic scholarship, an old and venerable one. Toe discussion of the question
in an anonymous Prolegomena to Plato 's Philosoplry dating from around the
first century A.D. shows that it was much discussed already in antiquity: the
author lists a number of explanations that were clearly current at the time,
ranging from the fantastical claim that Plato wrotes dialogues in imitation of
God's creation of the cosmos to the more prosaic suggestion that he wrote in
this forro to prevent his readers from dozing off. 2 Despite its ancient pedigree,
however, this question has only recently emerged into the vanguard of modero
Platonic scholarship, especially of the Anglo-American variety.
Schleierroacher's famous assertion that the forro and content of Plato's thought

1. I wish to thank David Roochnik, Charles Griswold, and Joanne Waugh for their
helpful comments on earlier drafts of this paper, though they of course are not
responsible for any errors that remain. I also wish to thank participants in discussions at
lhe Catholic University of America and Northern Illinois University for their challenging
questions.
2. Anonymous Prolegomena to Plato's Philosophy, ed. L. G. Westerink (Amsterdam:
North-Holland Publishing Company, 1962), 28-30.
156 Toe Third Way
are inseparable has for the most part gone unheeded during the last hundred
years of English-language scholarship in which debate has focused on the
content of Plato's philosophy in isolation from the forro in which he chose to
express it.
Today, however, increasingly more books on Plato address at least
grudgingly and in passing the question, "Why dialogues?" This question is
even beginning to divide scholars into two opposed camps: those who see the
dialogue forro as essential to the content of Plato's philosophy and thus believe
that Plato had to write dialogues, and those who, explaining Plato's decision
in terros of reasons ultimately extraneous to his philosophy, see only an
accidental relation here. This dispute is potentially a very fruitful one.
Unfortunately, because both sides often talk past one another and do not
properly articulate their opposed views, the dispute has not yet boro much
fruit. What too often happens is that the first camp accuses the second of
completely ignoring the dramatic and literary character of Plato's works, while
the latter simply denies doing this. 3 Yet the important issue here is not
whether in interpreting Plato's works one should pay attention to the context
of concrete setting, character portrayal, dramatic action (both in word and in
deed), humour, myth, etc. Presumably few, if any, interpreters would dare
claim that whatever cannot be translated into syllogisms should be completely
ignored. Toe debate should instead concero whether or not the mentioned
context is peripheral and accidental to the actual content of Plato's philosophy.
Toe real question that divides scholars and that will prove fruitful if properly
addressed can therefore be forroulated as follows: did Plato believe the
dialogue, with its dramatic and literary character, to be the only way of
expressing philosophical truth in writing, or did he see it as simply one way
of expressing a philosophical content that could equally, though perhaps less
vividly, be expressed in a treatise?
3. What happens when the debate between the two approaches is characterized too
simplistically is well illustrated by the interchange between David L. Roochnik and
Terence Irwin in Pl.atonic Writings, Pl.atonic Readings, ed. Charles L. Griswold Jr. (New
York: Routledge, 1988), 183-99. Roochnik attributes to Irwin the following interpretative
principie: "Toe context that surrounds such arguments, be it dramatic, rhetorical, mythic,
or humorous, should be dismissed in the search for correct analysis of isolated
arguments" (184). When the principie is expressed in this extreme form, Irwin can of
course deny that he subscribes to it by pointing out that he does sometimes make mention
of such context. Toe real dispute, however, is that for Irwin, and not for Roochnik, this
dramatic, rhetorical context is peripheral and entirely accidental to the philosophical
content: when Irwin replies to Roochnik's charge by claiming that in Pl.ato's Moral
Theory he does in fact comment on dramatic structure and on the characters of
interlocutors, all of the references he gives are to notes! (194)
Self-Knowledge, Practica! Knowledge, and Insight 157
But if the dispute between the two approaches is defined in this way, how
is it to be decided? Clearly the decision rests on Plato's conception of
philosophy. To determine whether or not the relation between the forro of the
dialogues and their philosophical content is a necessary one, we must examine
what Plato took the nature of this philosophical content to be. Our answer to
the question concerning the dialogue forro depends on our answer to the
question concerning the nature of philosophical knowledge and philosophical
method (what Plato calls "dialectic"). 4
Toe two traditional and opposed interpretations of Plato's conception of
philosophy have both failed to demonstrate a necessary connection between the
content of Plato's thought and the dialogue forro. Toe "skeptical"
interpretation (New Academy, modero "nondoctrinalism" in its extreme forro),
while sensitive to the literary, dramatic, and aporetic character of the
dialogues, denies them any positive philosophical content. Toe "dogmatic"
interpr~tati~n (Old Academy, Neoplatonism, unitarianism, developmentalism),
in cons1denng the airo of Plato's philosophizing to have been the systematic
exposition of philosophical doctrines, necessarily cannot see the dialogue forro
as anything more than accidental and even distracting oroamentation. ~
Therefore, the desire to understand the connection between the forro of Plato 's
writings and their philosophical content must lead us to search for an
alteroative interpretation of Plato's conception of philosophy.
l. A New Alternative: Three Characteristics of Plato's Philosoplzy
In this paper I argue that a conception of philosophy can be worked out
which, if attributed to Plato, would allow us to see a necessary connection
between Plato's philosophy and the dialogue forro without denying this
philosophy positive content. This alteroative would thus allow us to reconcile
the virtues of the skeptical and dogmatic interpretations. Toe conception
proposed here involves attributing to philosophy the following three
characteristics: 1) it is "reflexive" in the sense that its content is not
objectijiable as a result separable from its method; 2) it is mainly practical
~knowl~dge-~ow) rather th~ "theoretical" (knowledge-that); 3) the knowledge
1t pr?v_ides 1s nonpropo~1t1onal. These three characteristics are mutually
explaimng and therefore inseparable. More specifically, the second and third
~- This p_roject of grounding Plato's choice of the dialogue forro in his conception of
philosophy 1s undertaken briefly by Arthur A. Krentz, "Dramatic Form and Philosophical
Content in Plato's Dialogues," Philosophy and Literature 7 (1983): 34-7. Toe
c~aracteristics that Krentz attributes to Plato's conception of philosophy essentially agree
W1th the ones I explore in this paper, as will be noted more specifically below.
5. See introduction to this volume, 1-13.
158 Toe Third Way
characteristics, wbich have been attributed to Plato's pbilosopby by a number
of scholars, are in my view unilluminating and potentially misleading unless
they are recognized to be grounded in the first characteristic. For this reason,
the first and longest part of this paper will focus on the first characteristic
wbile the other two will be dealt with only subsequently and more briefly. In
the case of each one, I first make sorne general observations on what the
characteristic means, secondly defend the hypothesis that the dialogue can,
while the treatise cannot, express a pbilosopby characterized in this way, and
finally and most importantly, present sorne evidence for the ascription of the
characteristic under discussion to Plato's own conception of pbilosopby. Since
the project envisioned here is gargantuan and extends far beyond the confines
of this paper, it is important to note at the outset the limitations of the present
analysis. I do not argue in support of the view that pbilosopby, absolutely
speaking (and not just in Plato), has the three mentioned characteristics.
Furtherroore, I cannot prove here that there is no other conception of
pbilosopby that is equally capable of demonstrating a necessary connection
between the forro of Plato's writings and the positive content of bis philosopby
(though I in fact do not believe that there is another equally good alternative). 6
6. One might look for such an alternative in Charles Griswold's clai.m that a major
reason why Plato had to write dialogues is that this is the only way he could refute the
critics ofphilosophy: "Plato's Metaphilosophy: Why Plato Wrote Dialogues," in Platonic
Writings, P/atonic Readings, ed. Charles L. Griswold (New York: Routledge, 1988),
143-67; see also 12. While I agree with Griswold's conclusions, the difference between
our two approaches is to be seen in how we understand the word "metaphilosophy." Por
Griswold "metaphilosophy" is essentially the attempt to "justify philosophy." I
understand "metaphilosophy" to address the broader question: "What is the nature of
philosophy?" It seems to me that both types of "metaphilosophical" considerations are
needed to explain the dialogue forro (indeed, they are probably inseparable). In other
words, in addition to showing how the dialogues function as a defense of philosophy, we
also need to demonstrate a connection between the content of philosophy and the dialogue
forro. Griswold does deal briefly with Plato's conception of philosophy by focusing on
a characteristic distinct from any of the three discussed in this article: that philosophy
emerges from everyday opinion and remains deeply rooted in it. Griswold in fact
suggests that this characteristic is the major theme of the dialogues (153). The only
appropriate written expression for a philosophy thus characterized would be the dialogue
which, unlike the abstract treatise, is capable of preserving (through various literary
devices) the concrete context in which philosophy is rooted andas a struggle with which
philosophy defines itself. This i.mportant insight complements, rather than contradicts,
the characteristics discussed in this paper. If philosopbical knowledge is not an
objectifiable result separable from the knower, and if it is practica!, then clearly it is
deeply rooted in our everyday experience and in the awareness that defines that
experience.
Self-Knowledge, Practica! Knowledge, and Insight 159
Finally, I cannot show that the conception of pbilosopby I attribute to Plato is
present in and remains constant throughout ali of bis dialogues (something
clearly beyond the scope of any paper), though what is said here does bring
into question the common assumption of a sharp divide between the
conceptions of pbilosopby in the so-called "early" and "middle" dialogues.
The airo of this paper is a limited one: to stake out a new interpretation of
Plato's pbilosopbizing that promises to be more successful in explaining the
dialogue forro than the traditional ones and to show that it receives much
confirroation from what is said and shown in the dialogues themselves.
2. Philosopky as Reflexive: General Description and Hypothesis
In turning to the first characteristic, I must acknowledge that the word
"reflexive" chosen to designate it is far from being ideally suited to the task.
:nie reas?n i~ that this word can have the connotation of subjective
mtrospec~10n divorced from any external objective reality, a connotation that
has nothing to do with the characteristic I wish to ascribe to Plato's
Another important attempt to provide philosophical reasons for Plato's choice of the
~alogue forro _is ~at of Michael Prede, "Plato's Arguments and the Dialogue Forro,"
m Oxf~rd Stud1es in Ancient Philosophy, suppl. vol., ed. James C. K.lagge and Nicholas
D. SIDJth (Oxford: Clarendon Press, 1992), 201-219. One reason Prede provides is the
extreme difficulty, if not i.mpossibility, of getting oneself "into a position in which one
can speak with authority, with sorne kind of justified confidence, out of expertise and
knowledge about a ce_rtain subject matter" (214). Plato felt this difficulty just as much
as Socrate~ _and the d1al~gue forro allowed him to present bis views without doing so
f:om ~ pos1t1on of authonty (215-6). This strategy for explaining the dialogue forro is in
line w1th the "skeptical" interpretation mentioned above and discussed in the introduction
to thi~ ~ollection. I do not find it very convincing. Why must a treatise put the author in
a pos1t1on of "authority"? Prede 's view is apparently that a Platonic dialogue requires
one to "arrive at the right view by one's own thought, rather than on the authority of
som:body ~!se ... " (217), but can we really deny that a treatise by Hume or even Kant
requ~es this as _well? 11:1 gene~!, I do not think the dialogue forro can be explained by
the kind o~ plat1tude w1th which Prede's article concludes: "Por nothing but our own
~ought gams ~s knowledge" (219). What philosopher would deny this? However, there
IS m~ch more rn ~rede's rich article than this suggestion. The most i.mportant reason he
prov1des for the dialogue forro is Plato's conception of knowledge. Prede argues that "To
know, we lean:_i from the early dialogues, is not just a matter of having an argument,
however good 1t may be, for a thesis" (216). Instead, knowledge involves that the rest
of one's beliefs, an_d even someti.mes one's life, be in line with one's arguments. Prede
therefore ~haractenzes knowledge for Plato as "a highly personal kind of acbievement"
(216). :rns suggestion, which is left rather enigmatic, appears in line with Griswold's
suggestlon above and with my own view that philosophical knowledge is reflexive and
therefore personal.
 Toe Third Way
160
philosophy. By calling philosophy "reflexive," I instead mean that, while it is
not subjective or relative but is oriented towards a transcendent reality and
truth, it nevertheless is not objective in the way that the natural sciences, for
example, are. What is behind this denial of objectivity to philosophy is the
view that philosophical knowledge, however objective and universal it may be
in sorne sense of these words, is notan objectifiable "result" which as such is
separable from the person who knows and the method by which he or she
knows. Whatever might be and has been said about the personal involvement
of the scientist in scientific inquiry (one can think here of Michael Polanyi)7
and about the influence of the way the scientist inquires on the results he or
she attains, the fact is that at the conclusion of such inquiry, there is sorne
specifiable, objectifiable result that can be passed on to other inquirers with no
loss of its value. In philosophy, however, according to the present
characteristic, knowledge is not thus separable from the inquirer and the way
in which he or she inquires. This means that philosophical knowledge depends
on self-knowledge to a degree not paralleled in the natural sciences, to the
degree, that is, that philosophers cannot know the "object" into which they are
inquiring without knowing themselves. lf philosophical knowledge consisted
simply of results obtained at the end of an inquiry (e.g., conclusions deduced
and concepts analyzed), then clearly such results could be known
independently of any self-knowledge, and thus philosophical knowledge would
not be "reflexive." On the present view, however, philosophical knowledge
is instantiated in the inquiry itself (if properly conducted) and not in an explicit
conclusion to the inquiry. This is why philosophical knowledge is neither
purely subjective nor purely objective: it is to be found neither in subjective
introspection nor in sorne objective result, but rather in the very character of
the inquiry carried out, an inquiry that, though oriented towards an externa!
reality, is also inseparable from who the inquirer is and is a reflection of who
the inquirer is.8 In other words, the subject and the object become one in the
7. Polanyi criticizes what he sees as a prevailing conception of purely objective science
that seeks "to eliminate from science such passionate, personal, human appraisals of
theories, or at least to minimize their function to that of a negligible by-play" (Personal
Knowledge: Towards a Post-Critica/ Philosophy [London: Routledge and Kegan Paul,
1958], 15-16). Polanyi opposes to this a conception of science that would bridge "the
disjunction between subjectivity and objectivity" (17).
8. This type of reflexivity has been well-described by Wolfgang Wieland in his
excellent and extremely important article, "La Crítica de Platón a la Escritura y los
Límites de la Comunicabilidad," Méthexis 4 (1991): 19-37. Wieland claims that the
knower for Plato is someone who has already identified himself with his knowledge to
such an extent that he is unable to distance himself from it or to situate himself at its
margins. One cannot therefore speak significantly of knowledge without taking into
consideration the person who knows (27).
Self-Knowledge, Practica! Knowledge, and Insight 161
method which mediates between them. This view makes philosophical method
everything. The method is not simply a tool by means of which we arrive at
sorne objectifiable results which themselves constitute philosophical
kllowledge; instead the method is philosophical knowledge.
Again, it is important to emphasize that this identification of philosophical
knowledge with philosophical method does not divorce philosophy from an
extemal or transcendent reality and render ita self-contained, purely subjective
game: a kind of Rortian "conversatio_n" where philosophical concepts are
tossed back and forth, and different subjects are inquired into, but with no
prospect ?r intention of arriving at such a thing as "Truth. " 9 Toe present view
saves philosophy from this fate by maintaining that the truth that the
philosopher seeks becomes manifest in the very method by which he or she
seeks it. Philosophical knowledge is identified with philosophical method
because philosophical truth is such that it reveals itself not in sorne
propositional result that "represents" or "mirrors" the world, but rather in the
very way in which the philosopher inquires, in his or her orientation.
Philosophical knowledge is, accordingly, not a knowledge of facts that can be
passed on to another, but rather a personal orientation towards the truth in
which this truth can make itself manifest. 10
" 9. In describing ~s own conception of philosophy as "conversation," Rorty writes:
We are not conversmg because we have a goal, but because Socratic conversation is an
activity which is its own end" (Consequences of Pragmatism [Minneapolis: University
of Minnesota Press, 1982]. 172).
.. 10. A s~ar characterization of the philosophy found in the dialogues is given by
Jurgen M1ttelstrass : "In Socratic dialogue (or in the Socratic form of a dialectical
intention) the actual goal of any philosophical activity, a philosophical orientation, is
alr~ady reache_d in the approach itself and not just by acquiring material knowledge.
Domg something better and not knowing something better is what overcomes the
sophistic ~tention. Dialectics in the Socratic-Platonic sense is not just a form of
argum~ntat!on but al~o essentially a (philosophical) form of life (Lebensform)" ("On
So~rat1c D1alo?ue," m _Platonic Writings, Platonic Readings, 131). Krentz similarly
clatms that philosophy m Plato's dialogues "is presented as a process rather than a
product or specific philosophical doctrine" and involves "an existential dimension
e~pressed in the practica!, ethical and socio-political actions of many characters in the
d1~logues" ("D:amatic P~rm, ". 34-5). Marjorie Grene , following in the footsteps of
'.'11chael Polany1, has also illumtnated this characteristic of "personal knowledge": "Toe
tmpersonal aspect of knowledge arises from and returns to personal participation in the
search for and acceptance of the object to be known. Por only the explicit, formulable
c~re o~ ~owledge ca_n be transferred, neutrally, from person to person. Its implicit base
(smc~ 1t 1s not :erbalized and cannot be formulated and so impersonalized) must be the
gropmg, the onentation and reorientation ofsomeone" (Marjorie Grene, The Knower and
the Known [London: Paber & Paber, 1966), 24-5).
162 Toe Third Way
Philosophy is "reflexive," then, in the sense that philosophers can know the
truth they seek only through reflection on what they do and who they are. This
truth is "objective" in sorne sense, but not in the sense of being sorne
objectifiable and expressable content that can be distributed anonymously. It
is inseparable from the method and can never conclude the method; it is
revealed in the very process of philosopbizing. For example, if we inquire
properly into the nature of virtue, our inquiry will itself exhibit virtue; if we
inquire properly into the nature of the good our inquiry will itself mirror the
good (be good-like). Pbilosophy, if reflexive in this way, would turn out to be
a method that shows rather than proves, that manifests rather than describes.
It is primarily neither deductive nor analytic, but rather exhibitive. It does not
conclude that virtue has such and such properties but rather itself exhibits
virtue as a whole.
Hypothesis: Reflexivity and the Dialogue Form
What must now be seen is that if pbilosophy is not "reflexive" in the way
described, if it aims at an objectifiable result completely distinct from itself,
which it either discovers (like a science) or produces (like a craft), then its
relation to the dialogue form cannot be anything but accidental. lf their object
is thus entirely distinct from their method, nothing stands in the way of
philosophers either describing and reaching conclusions about their object in
a scientific treatise or giving instructions for its attainment in a technical
manual. If, on the other hand, philosophy has the reflexivity described above,
then the necessity of the dialogue form becomes immediately evident. To say
that philosophy is reflexive is also to say that it cannot be abstracted from the
philosopher who philosopbizes. This kind of abstraction, however, is what
typically characterizes the treatise in wbich are presented simply arguments
and conclusions and in which the philosopher is present as little more than the
universal voice of reason. 11 A philosophy that is truly reflexive, i. e., a
philosophy in which the truth is revealed in the very process of philosopbizing
and therefore cannot be separated from this process as a result or "doctrine,"
requires a form of writing for its expression that can fully include the
philosopher and show him or her philosophizing in a specific context with
specific interlocutors. In other words, a form of writing is needed that, rather
11. There appear, of course, to be sorne exceptions, such as Descartes' Meditations.
Yet Descartes' personal presence in the Meditations is dueto a characteristic of modero
philosophy: the attempt to arrive at objective knowledge by starting with the subject as
the ultimate ground. Yet because the aim here is in fact to arrive at scientific, objective
knowledge, the philosopher eventually drops out and the Meditations is replaced by The
Principies of Philosophy.
Self-Knowledge, Practical Knowledge, and Insight 163
than presenting a body of doctrines or even an isolated meditation, can depict
the very process of philosophizing in ali of its complexity. This description
appears to fit best Plato's dialogues.
3. Philosophy as 1emperonce: Reflexivity in the Charmides
If this is an adequate defence of the hypothesis, it is still the case that
nothing can be concluded about the necessity of the dialogue form for Plato
until this conception of philosophy as being "reflexive" is shown to be bis
own. It is time, then, to see what evidence there is in the dialogues for the
"reflexivity" I have just described. I will focus on the Charmides, the Laches
and the account of the good in the sixth book of the Republic. I obviously will
not be able to present in full my interpretations of the Charmides and the
La_ches, 12 but I believe, that only an outline is needed to illustrate the present
pomt.
Toe Charmides, as is well known, is about the elusive virtue of temperance
(sóphrosuné). As is also well known, Socrates and bis interlocutors in the
dialogue examine different possible definitions of temperance and find none of
~em satisfactory. We_have here another one of those apparently futile aporetic
dialogues.. However, 1f we look at what happens in the dialogue, if we turn
o~ attention away from "results" and to the inquiry itself, a very different
p1cture ~m.erges._ Let _us look first at what happens near the beginning of
Socrates d1scuss10n w1th the boy Charmides. Socrates has been told by Critias
that Charmides is renowned not only for bis good looks but also for his
temperance. Accordingly, after professing to be able to cure a headache with
wbich Charmides is afflicted (caused by sorne intemperance?), Socrates
extra~agantly praises the boy's family: in bis usual sly way, Socrates is clearly
tempting the boy here to indulge in pride and thus disprove bis temperance.
Therefore, immediately after bis eulogy of Charmides' clan, Socrates asks the
boy. point-blank if h~ is temperate (158c). What makes this question a
part1cularly nasty one 1s that any answer will place the answerer in a bad light.
If ~e boy answers, "Yes, I am temperate, Socrates," then he will be making
an 1mmodest boast that belies bis possession of any temperance. On the other
han~, if he s_ays, "No, I ~m ~ot temperate, Socrates," then he will not only
admit to lacking an essentlal v1rtue, but he will also be calling bis older cousin
a liar, an example of disrespect for his elders that will only confirm bis lack
12. ~ offer detailed interpretations of both dialogues in my forthcoming book, Dialectic
and Dialogue: Plato's Practice of Philosophical Inquiry.
164 Toe Third Way
of temperance. Socrates' question, then, is simply unanswerable. What is the
boy to do? He blushes. 13
What makes this scene of great importance can be summarized in two
points: 1) any answer the boy could have given to Socrates' question would
have concealed his temperance; 2) by failing to answer the question and
blushing, the boy e.xhibits his temperance. Charmides' temperance will of
course ultimately prove to be rather superficial; the modicum of temperance
he has, however, is made manifest in his reaction to Socrates' question. In
other words, in his silent blush Charmides exhibits temperance as he himself
will proceed to describe it: temperance as quiemess (159b) and modesty
(160e).
This kind of reflexivity is of course not unusual on the level of ordinary
experience. Charmides is naive, and so one would expect him to blush at nasty
questions and to describe in his definitions temperance as he actually
experiences it. What makes the present scene so important, however, is that
the rest of the dialogue simply reenacts it on the higher plane of dialectic. At
the dialogue's conclusion, Socrates has proveo unable to answer the question:
"What is temperance?" His reaction is not a blush but a more sophisticated
version thereof: a confession of ignorance. Furthermore, if we look at what
occurs in the dialogue as a whole, we can make the same two points with
regard to the final scene of the dialogue that we made with regard to the
Charmides episode: 1) Any definition with which Socrates might have
answered the question, "What is temperance?" would only have concealed the
nature of temperance and 2) Socrates succeeds in e.xhibiting the nature of
temperance precisely by failing to answer the question.
These two claims could be fully justified only through a detailed
interpretation of the dialogue. Here, however, I will merely show what
direction such an interpretation would take. We saw that the temperance that
Charmides exhibits in his silent blush is precisely the temperance later defined
by him as quietness and modesty. Likewise, the temperance that Socrates
exhibits in his avowal of ignorance is precisely the temperance that is defined
in the dialogue as self-knowledge, where this knowledge is further interpreted
as being a "knowledge of knowledge," i. e., knowing what you know and what
you don't know. At the end of the dialogue Socrates shCMls that he knows what
he knows and does not know and that this knowledge is something beneficia!,
while the inquiry has concluded that such a "knowledge of knowledge" is
neither possible nor beneficia!.
The definition of temperance as self-knowledge is in fact first suggested by
the future tyrant Critias. Toe important point to note, however, is that while
13. Toe boy himself goes on to give the same explanation I give for his failure to
answer the question. However, it is his initial reaction that is most telling.
Self-Knowledge, Practica! Knowledge, and Insight 165
for Critias this definition is an empty formula borrowed from the sophists, for
Socrates self-knowledge is something he exhibits in the very way he inquires.
This contrast between Critias, who simply states in a definition what
temperance is without exhibiting temperance, and Socrates, who exhibits this
temperance by questioning the definition, is particularly clear in one scene of
the dialogue. Critias has just introduced his definition and is angry at Socrates
(accuses him of picking a fight) for beginning to question it (a most
intemperate attitude, exhibiting a lack of self-knowledge besides!). Socrates
replies as follows: "How can you think that I have any other motive in
refuting you but what I should have in examining into myself? This motive
would just be fear of my unconsciously fancying that I knew something of
which I was ignorant" (166c-d). 14 In arrogantly defending his definition of
temperance as a knowledge of knowledge, Critias shows that he lacks
temperance as thus understood; in questioning this definition, Socrates
demonstrates that he has the temperance it attempts to define. In fact, this
definition must be refuted, because to believe, as Critias does, that the nature
of temperance can be captured in a definition is to show an incredible lack of
temperance. In believing that a definition can answer the question, "What is
temperance?" 15 Critias only shows that he has no understanding of
temperance (in the same way that by answering positively the question, "Are
you temperate?" you would only belie your temperance). Socrates, however,
14. Socrates' description at 167a of the man who has a "knowledge of knowledge" fits
perfectly his description of himself in the Apology: such a man would go around testing
others to see what they know and what they only think they know .
15. It is often thought that this is Socrates' belief, a belief that he actually imposes
upon his interlocutors. Yet Socrates' elenchus, if properly understood, does not commit
him to any such belief. Socrates uses the demand for a definition as a way of destroying
the pretensions of those interlocutors who think they have the kind of knowledge of
virtue that Socrates himself never pretends to have, i.e., a knowledge of virtue that is
fixed, determínate, and dogmatic, and who therefore are the ones who assume that they
can "easily say" what virtue is (this is a pretension of both everyday collllllon sense ami
sophistic wisdom; therefore, both are targets of Socrates' elenchus). We can imagine
Socrates thinking something lik.e the following about his interlocutor: "So you believe
that your knowledge of virtue is such that you have no need of philosophy. You do not
spend your time questioning and inquiring into the nature of virtue as Ido, and therefore
you clearly think you fully grasp what virtue is. Well, if you are so fortunate as to have
this kiml ofknowledge, you must certainly be able to define what virtue is. Tell me, then,
what is virtue?" Here what Socrates is doing in no way commits him to the view that
virtue can be defined or that one must have such a definition to know what virtue is and
to be virtuous. Socrates could justas well be trying to show (and this I claim to be the
right interpretation) that one can be virtuous only through philosophical inquiry and self-
examination, something that is possible only once one no longer believes that one has a
fixed, determínate knowledge of virtue that could be expressed in a definition.
166 Toe Third Way
in the very process of refuting the definition of ternperance as a knowledge of
knowledge, exhibits what this definition tries, but fails, to state. 16
We can therefore expect that the problerns encountered by the definition in
17
the dialogue find their "solution" in what Socrates birnself actually does.
Toe rnain flaw in the definition of ternperance as a knowledge of knowledge
is, of course, that such knowledge appears ernpty and therefore of no benefit.
Toe knowledge of knowledge cannot be a knowledge of any of the objects that
are the subject rnatter of a particular craft or science. In fact, the knowledge
of knowledge cannot be a knowledge of anything, perhaps not even of itself:
for how could it be an object to itself? (After all, can there possibly be a
seeing of seeing or a hearing of hearing, as Socrates asks at 167c-d?) Most
devastating for the definition, however, is that wbile ternperance rnust clearly
be good, the knowledge of knowledge as such cannot be a knowledge of the
good. Toe dialogue therefore leaves us with two possible characterizations of
temperance, both of wbich are inadequate: if ternperance is identified with a
knowledge of knowledge, then it is rendered ernpty and useless; if ternperance
is sirnply identified with a knowledge of good and evil, then it ceases to be
self-knowledge, wbich it rnust in sorne sense be. 18
It would seem that an adequate account of ternperance would have to
reconcile these two one-sided characterizations. But is such a reconciliation
possible? Can a knowledge of knowledge also be a knowledge of the good?
This is not possible in theory, but now let us look at Socrates' proctice.19
16. Jacob Klein has given an excellent description of the difference between Socrates
and Critias: "What Critias says is far from wrong perhaps, but the possible rightness of
his statement is at best 'in words' only: his possible being right does not mean, as we see
a short while later, that he, in fact, possesses sophrosyne and understands what he is
saying. What Socrates, on the other hand, has to say about 'fear' manifests 'in deed,'
manifests shiningly, Socrates' own sophrosyne" (J. Klein, A Commentary on Plato's
Meno [Chapel Hill: University of North Carolina Press, 1965], 24).
17. I am here in agreement with Henry Teloh's attempt to find the solution to the
dialogue's aporia in the dialectic that Socrates himself practices (Socratic Education in
Plato's Early Dialogues [South Bend: University of Notre Dame Press, 1986) , 65).
18. Before defining temperance as "self-knowledge," Critias defined itas "doing what
is good" (163d), but this definition was rejected because Socrates showed it to have the
consequence (recognized as absurd by both of them) that one can be temperate without
knowing that one is (164a-c).
19. After taking the argument of the Charmides completely at face value, Stanley
Rosen attributes to Socrates/Plato a result- and object-oriented conception of knowledge
that renders self-knowledge impossible: he allows that the conception of sophrosuné
"raises the possibility of self-knowledge in a nontheoretical sense. But this possibility is
never fulfilled by Socrates and his students, because it contradicts the notion of
knowledge as derived from arithmetic and geometry" (The Ancients and the Modems:
Rethinking Modemity [New Haven: Yale University Press], 105). This tension is further
Self-Knowledge, Practica! Knowledge, and lnsight 167
First of ali, as we have seen, Socrates clearly exhibits a knowledge of
1<nowledge, not only in bis final avowal of ignorance, but throughout the
dialogue. But does Socrates' knowledge of knowledge prove ernpty and of no
benefit? Does it not display an understanding of the good? If we contrast
Socrates with Critias and bis arrogant confidence based on sopbistic formulas,
we see that Socrates in the very way in which he inquires into himself sh(MIS
us what it means to be good. Toe "good" here is not sorne "object" for the
attainment of wbich self-inquiry is only an instrument. lnstead, the nature of
the good is revealed precisely in Socrates' willingness and ability to inquire
into birnself. Socrates' knowledge of knowledge thus has content but this
content is not an answer or a definition; this content is to be found in the
reflexive process itself that characterizes a knowledge of knowledge. Thus, the
dialectic practiced by Socrates, by both being reflexive (a knowledge of
knowledge) and having content (showing us what goodness is ali about),
reconciles the two opposed characterizations of ternperance and thus itself
exhibits this ternperance. 20 Ternperance is not to be found in a set of
definitions or rules that dialectic might one day discover. 21 Ternperance is an
aspect of dialectic itself.
characterized by Rosen as follows: "The dialogues suggest the possibility of self-
knowledge in a practica! rather than a theoretical sense. Yet the analysis of practice they
contain is such as to assimilate it into either theory or production" (100). Yet the
Charmides, with its clearly deliberare contrast between what can be shown in practice
and what can be explained in theory, is nota record of Plato's failure to explain self-
knowledge, but rather his attempt to show that self-knowledge, precisely because it is not
like arithmetic or geometry, eludes direct analysis (which as such is always
"objectifying") and therefore can only be exhibited through that indirect mode of
expression that characterizes the dialogue form. Self-knowledge, in other words, is a
prime instance of what Rosen himself calls the "limits of analysis."
20. Por an excellent expression ofthis point see Bernhard Waldenfels, Das Sokratische
Fragen: Aporie, Elenchos, Anamnesis (Meisenheim am Glan: A Hain, 1961), 75.
21. L. A. Kosman expresses well the same point: "no definition, no matter how good,
can be guaranteed to provide the understanding of that being which it attempts to
articulate, and specifically ... no account can be judged adequate by virtue of its verbal
form alone" ("Charmides' First Definition: Sophrosyne as Quietness," in Essays in
Ancient Greek Philosophy, vol. 2, ed. John P. Anton and Anthony Preus [Albany:
SUNY, 1983), 211). Kosman rightly adds, however, that the definitions suggested in the
dialogue are not rejected as being completely false but are taken up "into the final
understanding meant to emerge from the dialogue" (207). The characterization of this
understanding I provide in this paper agrees with Kosman's view that, while the
understanding emerges from the confrontation between different logoi, it cannot itself be
embodied in a specific logos. "This fact is at the heart of dialectic: the fact that the truth
cannot simply be said, but must be made to speak forth from what is said" (209).
168 The Third Way
4. Philosophy as Courage: Reflexivity in the Laches
When we tum to the Laches, we notice the same reflexivity. In this dialogue
Socrates exhibits courage in the process of searching for it. Thus, courage
proves to be not an answer (definition) to the process of dialectic, but rather
something instantiated in this process itself. This will become apparent through
a brief sketch of what occurs, both argumentatively and dramatically, in the
Laches as a whole.
In this dialogue Lysimachus and Melisias are concemed witb tbe education
of their sons. They bave therefore invited tbe generals Laches and Nicias to
an exhibition of a man fighting in armor in order to ask tbem whetber or not
tbey consider this an art wortby of being pursued by tbeir sons. When tbe two
generals express their opinions, they are found to differ. Nicias maintains tbat
the art of fighting in armor is one wortb learning. Besides its other advantages,
this art can make one braver and more confident in battle (182c). Thus, Nicias
is presupposing that a certain skill or science can in fact make one more
courageous. Laches, on tbe otber band, argues tbat even if this art could be
leamed, it would be of little or no use. We see that the teacbers and students
of it do not prove themselves any braver in the heat of battle than someone
ignorant of the art (183c-184c). Lacbes is thus very skeptical concerning tbe
ability of mere knowledge to make one more courageous. The reason for this
disagreement between the two generals is revealed by what we leam about
tbem througbout the course of the dialogue. Nicias is an "intellectual" who has
often conversed witb tbe sophists and witb Socrates himself and who believes
tbat no forro of knowledge is witbout value (e.g., 187e-188c). Lacbes, on tbe
other hand, is a "four-square" practical man who believes tbat ali talk is cheap
unless backed up by actions (188c-189b).
At 190c Socrates effects a transition from tbe discussion of the particular art
of fighting in armor to a discussion of courage in general. When Lacbes offers
"remaining at one's post" as a definition of courage, Socrates objects that he
is not concemed solely with courage in battle but with all kinds of courage,
for example, courage in enduring poverty and sickness as well as courage in
fighting bad desires and pleasures (191d-e). Toe scope of tbe discussion has
tbus become very broad.
At this higher level of the inquiry the debate between Nicias and Laches
recurs. Laches, as bis distrust of learning and technical skill would lead one
to expect, defines courage as simple endurance (192b-c). When Socrates
objects that stupid endurance is hardly courage (192d), we see tbe flaw in
Laches' position. Because he does not allow knowledge to have anything to do
with courage, Laches ends up in effect identifying it with stupid endurance.
Most importantly, this is precisely the kind of courage that Laches exhibits in
tbe inquiry. He stolidly endures under tbe onslaugbt of Socrates' questions and
criticisms. However, because, as he himself admits, he is unaccustomed to
Self-Knowledge, Practical Knowledge, and Insight 169
philosophical discussions (194a), he makes no progress. Nicias must therefore
eventually join the fray. As his intellectualist bent would lead one to expect,
Nicias advances a position completely opposed to that of Laches: he goes so
far as to identify courage with knowledge (194d), i.e., a knowledge of wbat
is to be feared and what sbould inspire confidence (195a).22 Laches' response
is violent: he angrily asserts that courage and knowledge are two totally
separate things (xwpii;: ofrrrou <10</>ícx É<1Ti11 c'xvopeícxi;:, 195a). He then
proceeds to raise specific objections like those raised by Socrates earlier
against tbe suggestion that courage is wise or prudent endurance. A doctor
koows what is to be feared and wbat should inspire confidence in sickness; is
he thereby more courageous than a non-doctor? Generally, every craftsman
(dérrdourgos) knows what is to be feared and what should inspire confidence
in his appropriate craft (téchné), but no sane person would say that this
knowledge provides the craftsman with courage. As Socrates himself pointed
out, craft-knowledge, as the kind of knowledge that assumes control of
contingencies, seems to leave no room for courage (192e-193c) insofar as
courage clearly involves risk. Nicias replies that he is not talking about this
kind of knowledge. When comered, however, Nicias cannot identify the kind
of knowledge he is talking about. As a result, he sees his definition of courage
evaporate into an empty, abstract "knowledge of good and evil" with no
determínate content. 23 Like Laches, Nicias too exhibits in the very way be
22. This view is of course defended elsewhere by Socrates himself: Protagoras 360d,
Republic 429c, 442c. Perhaps the most interesting aspect of this part of the Laches is that
the views wbich Nicias advances and Socrates refutes are for the most part what we
would call "Socratic doctrines." Thus we see that Socrates has no sympathy with what
are supposedly bis own views when these views are simply affumed by bis students as
doctrines.
23. What is wrong, of course, is that Nicias' definition turns out to identify courage
with virtue as such. Scholars in search of doctrines jump on tbis conclusion and interpret
it to be a statement of the "Socratic paradox" that all the virtues are one. Thus, Terence
lrwin writes: "The Laches concludes with an argument to show that all the virtues are
really a single virtue, knowledge of good and evil; other dialogues strongly suggest the
same conclusion" (Plato's Moral Theory [Oxford: Clarendon Press, 1977], 86). Yet the
ostensive purpose of the argument is not to prove the unity thesis, but to provide a
reductio ad absurdum of Nicias' definition of courage! Even if one accepts Irwin's
reasons in n. 62, 302 for seeing a covert and indirect proof of the unity thesis here, one
must still explain why Socrates in this case, rather than comrnending Nicias' definition
for entailing the unity thesis, instead uses this entailment to refute the definition. The
reason, I would suggest, is that the formula "virtue = a knowledge of good and evil,"
like the claim, "courage is a knowledge of what is to be feared and what should inspire
confidence," is an empty and meaningless formula. In general, the unity of the virtues
is nota "doctrine" for Socrates because it has meaning only as revealed in what Socrates
actually does in the dialogues. All the virtues converge in the reflexive, practical, and
170 Toe Third Way
inquires the kind of courage he expresses in bis definition.24 In the argument
he relies entirely upon a pseudo-knowledge consist~ of definitions and verbal
distinctions wbich he has simply picked up from others. He puts ali bis
confidence in these intellectual weapons, only to have Socrates take them
away.z,
Toe debate here is between two opposed understandings of courage and two
opposed characterizations of thinking. In the one case, courage involves a
knowledge that assumes control of ali contingencies. With such knowledge one
has a certainty and self-assuredness that allows one to face all daogers
laughing. One's skill makes one practically immune to external circumstances.
Yet this kind of dependence on skill makes one unable to deal with the
unpredictable and can therefore easily degenerate into cowardice. In the other
case, courage involves dispensi~ with ali knowledge and skill and hurling
oneself into the unknown, opening oneself to fortune and passively enduring
it. This kind of courage, however, easily becomes mere folly or rashness.
Is there any way of reconciling these opposed characterizations of courage
represented by the two generals? In the dialogue we see them reconciled in
Socrates. 26 Again. the answer is not to be found in what Socrates says, since
this dialogue too ends in aporia. Rather, the answer is to be found in what
Socrates does, i.e., how he gives up a definition when it does not prove
adequate, how he confesses bis ignorance and yet persists inquiring. After the
first attempts at defining courage fail, Socrates encourages Laches with the
following words:
nonpropositional knowledge that Socrates embodies and exhibits in every inquiry.
24. Ironically, Nicias bimself recognizes this reflexivity (i.e., the fact that Nicias
bimself is an illustration of bis defmition of courage and therefore is being examined with
it) to be a general characteristic of Socrates' method: "whoever is close to Socrates and
enters into conversation with bim will neccesarily be led round and round by the
argument until he falls into giving an account of himself, of the manner in wbich he lives
and has lived, even if he began to speak about sorne other subject" (187e-188a).
25. Charles Griswold rightly observes about Nicias that "bis answers are too
discursive, too verbal. Laches is not altogether wrong in suggesting that Nicias speaks
like a sopbist, that is, as someone whose talk is clever but not founded on sound insight"
("Philosophy, Education, and Courage in Plato 's Laches," /nterpretation 14 (1986]: 192-
3).
26. An excellent account of the contrast between Laches and Nicias and of their
contrasting deficiencies is given by Michael O'Brien in "Toe Unity of the Laches," in
Essays in Ancient Greek Philosophy, vol. 1, ed. J. P. Anton and A. P. Preus (Albany:
SUNY, 1971), 303-315. O'Brien also sees the con.flicting tendencies of the dialogue as
coming together in Socrates: "In Socrates the theoretical strands of the dialogue come
together and find their practica! solution" (312).
Self-Knowledge, Practical Knowledge, and Insight 171
If you agree, we too must endure and persevere in the inquiry, and then courage
will not laugh at us for failing to inquire coumgeously into the nature of coumge,
wbich after all may frequently be endurance (194a).
This is precisely what Socrates does: he inquires courageously into the nature
of courage and thus exhibits courage in the very process of inquiring into it. 27
But what kind of courage does Socrates display? One that reconciles the two
extremes represented by Laches and Nicias. Unlike the courage of Nicias, it
is a courage that does not tenaciously hold on to formulas but is willing to
abandon a statement if the truth reveals itself to be elsewhere. It is a courage
open to contingency, i.e. , which knows that it cannot anticípate the truth or
capture it once and for ali in certain rules or definitions. Thus, it is the
courage of confessi~ one's own ignorance, one's own vulnerability before the
truth. Unlike the courage of Laches, on the other hand, Socrates' courage does
not hurl itself into the unknown or give itself over to misology. It is willing
co al'glle and venture a hypothesis, knowing that in its very ignorance it has a
share of wisdom. Both skepticism and dogmatism are forms of cowardice. Toe
tension between knowing and not knowing, the desire to be good without the
possession of any skill that will guarantee goodness, this is the proper sphere
of courage.
These comments only outline what cannot be fully understood without a
detailed reading of the dialogue. Yet here again we can see that the virtue in
question is to be found in the very process of dialectic and not in sorne
"answer" that would end this process. Toe reflexivity is unavoidable.28
27. This self-referential character of the Laches is also noted by Griswold. Throughout
the article cited above Griswold brings to our attention the relation between courage as
~n ~bject of philosophical inquiry and courage as a characteristic of philosophical
mqm.ry; he clearly shows that what is at stake in the dialogue as a whole is nothing less
th~ the ~ture and possibility of philosophy itself: ". . . the effort to defme courage
philoso~hically seems to require the courage to philosophize.... Toe Laches is, among
o~er 1;11"1gs, an _effort to give a AÓ-yo~ of what courage is; the deed (Ep-yo 11) of gjving
lhis Ao-yo~ requrres courage. Thus the definition of courage, and the philosopher's
courage, stand to each other as word to deed" ("Philosophy, Education, and Courage in
Plato's Laches," 178).
28. I believe, but cannot argue here, that this characteristic the Laches and the
Channides have of instantiating what they are about is also a characteristic of the
Euthyphro apd the Lysis. On the latter, see my article, "Plato's Lysis: An Enactment of
Philos_op~cal Kinship," Ancient Philosophy 15 (1995): 69-90. In fact, I would be willing
to mamtam that, to a lesser or greater degree, this is a characteristic of all of Plato's
dialogues. In a recent book on the Theaetetus (Philosophy and Knowledge [London and
Toronto: Associated University Presses, 1992]), Ronald Polansky has shown this dialogue
to ~ave ~e ~en~ioned characteristic. He describes Plato as "contriving to mirror the
top1c of bis mquiry by the action of the dialogue" and shows how "the entire dialogue
172 Toe Third Way
Limitations of Conclusion: "E:arly" versus "Middle" Dialogues
It will of course be thought that the validity of these conclusions concerning
Plato 's conception of philosophy is limited by the supposed fact that the Laches
and the Charmides are early dialogues very unlike the dialogues of Plato's
mature "middle period. "29 Though this paper cannot entirely remove this
limitation, a couple of points can be made. lf the interpretations just presented
are correct, the Laches and the Charmides are very sophisticated dialogues
with a great deal of "positive content." These interpretations thus render
impossible the facile distinction between "early" and "middle" dialogues that
is often made and according to which the former do not really contain Plato's
philosophy (either because they are only negative and refutative or because
they are mere depictions of the historical Socrates) while the latter finally
present this philosophy in a conclusive manner. Any change in Plato's
conception of philosophy between the Laches and the Charmides, on the one
hand, and the dialogues attributed to Plato 's middle period, on the other,
would have to be much more subtle than this. Toe other point that needs to be
made is that the "middle period" dialogues themselves are not, as is often
thought, systematic treatises in disguise. 30 Toe Republic exhibits reflexivity
when Socrates the philosopher introduces himself into the city he is
describing31 and employs in the discussion the hypothetical method he will
acts out what it is about" (245).
29. Toe data of Gerard Ledger's book, Re-counting Plato (Oxford: Clarendon Press,
1989), actually situate both dialogues very close to the middle-period dialogues in what
Ledger calls an "early middle group." Ledger's data have recently been used to argue
for the inclusion of the Charmides in the middle group itself: Debra Nails, "Plato's
Middle Cluster," Phoenix 48 (1994): 62-7. It is also important to remember that
wherever these dialogues are placed, it can hardly be questioned that Plato must have
been at least in his thirties when he wrote them (Plato was about 28 years old when
Socrates died). So it is clear that we can not be dealing here with immature works of
Plato's "youth."
30. Plato's "distance" asan author must be taken into account as much in interpreting
the "middle" dialogues as in interpreting the "earlier" ones: "It seems to me important
to read the doctrines of Socrates or of the Eleatic Stranger in the later dialogues with the
same hermeneutical attention to ironic distancing and mimetic displacement as we bring
to earlier dialogues" (L. A. Kosman, "Silence and Imitation in the Platonic Dialogues,"
Oxford Studies in Ancient Philosophy, supp. vol. [1992), 85). Por the same point see
Prede, 214.
31. After Socrates has introduced the philosopher/king paradox and has concluded that
no present forro of government is suited to philosophy, Adeimantus asks if the state best
suited to philosophy was the one described earlier (books 11-V). Socrates replies: "In
other respects it is that one ... but we said even then that there must always be sorne
people in the city who understand the rational principie on which the constitution is
Self-Knowledge, Practica! Knowledge, and Insight 173
describe in the Divided Line. 32 Yet an examination of the dialectical structure
of the Republic cannot be undertaken here. Instead, I briefly want to indicate
bow Socrates' description of philosophy in the middle books of the Republic
appears to recognize philosophy's reflexivity.
5. Philosophy and the Good: Reflexivity in the Republic
After having claimed in Book VI that we "do not sufficiently know the
good," Socrates illustrates our ignorance by refuting two common defínitions
of the good. One definition, that "the good is pleasure" (505a), is untenable
because one is ultimately forced to admit that there are bad pleasures as well
as good (505c-d). Toe other definition, that "the good is knowledge," runs
into the following paradox: if we define the good as "knowledge" we will
bave to specify what kind of knowledge; yet if the only answer we are able to
give is "knowledge of the good," our definition will be circular (505b-c). This
circularity is the same as that found in the Laches and the Charmides:
temperance and courage are not objects externa! to our knowledge but are
rather in sorne sense to be identified with this knowledge itself. 33 This leaves
based, the same which guided you, the law giver, when you made the laws" (497c-d).
This passage makes clear what the philosopher/king paradox really involves : making the
philosopher (Socrates' "you" is modest) who discusses the ideal city a part of this ideal
city itself.
32. It has often been remarked that the stages of the discussion in the Republic parallel
the stages described in the Divided Line. Por a good account of specific parallels, see
Robert S. Brumbaugh's book, Platonic Studies in Greek Philosophy (Albany: SUNY,
1989), especially 29-37. It is important to note, however, that the Republic as a whole
remains on the leve! of the hypothetical method employed by mathematics (third section
of the Line). Toe reason is simple: Socrates never pretends to provide us with that
knowledge of the Good which characterizes the fourth and highest section of the Line;
instead he simply describes how such knowledge might be attained. In other words,
Socrates only describes the "longer road" without actually taking it.
33. Ironically, what is thus being refuted here is the so-called "Socratic paradox"
which asserts the identity of virtue and knowledge. This does not mean, however, that
Plato in the Republic is putting into Socrates' mouth a criticism of a doctrine that he has
Socrates defend in earlier dialogues. Toe equation of virtue with knowledge, rather than
being presented as a doctrine anywhere in Plato's dialogues, is instead always presented
as a problem. Even in the Protagoras where Socrates presents the most thorough
argument for this thesis, the thesis is rendered problematic by his insistence that virtue
cannot be taught (since it seems that every body of "knowledge" in the normal sense of
the word could be taught). Toe reason why the claim "virtue is knowledge" is
problematic is that it is meaningless unless accompanied by an understanding of what
constitutes philosophical knowledge. Once we see that philosophical knowledge is not for
Plato a "knowledge of definitions" and that it instead is characterized by the reflexivity
174 The Third Way
us with the paradox: if the good is knowledge, then knowledge of the good
would simply be a knowledge of knowledge: this is circular and apparently
empty. Toe problem is the same as that encountered explicitly in the
Charmides.
Does Socrates in the Republic, then, simply avoid this paradox by
completely distinguishing the good from knowledge, by making the good a
mere object of theoretical and scientific knowledge? Not only would such a
"solution" sacrifice what is gained in other dialogues through the presermtion
of this circularity and paradox, but in fact there is sufficient indication in the
Republic itself that the circularity must be maintained. According to the
analogy drawn by Socrates between the Sun and the good, the good is not only
the ultimate object of our knowledge, but is also the cause of our knowledge
of itself and of everything else. However one interprets this claim, it at least
suggests that the good is much more intimately linked to our knowledge than
anything else that is only an "object" (and not also a cause) of our knowledge.
Toe claim also suggests that in knowing the good as the ultimate ground of all
our knowledge, our knowledge must in sorne way turn back upon itself.
Normally the good appears only through our knowledge of "objects" (such as
mathematical objects) by providing the intellectual illumination that makes this
knowledge possible. In order to know the good itself, then, it appears that our
knowledge must turn back upon itself in order to see the source of its own
illumination and thus that knowledge of the good must also to sorne extent be
knowledge of knowledge.
A point that Socrates makes in specifying his analogy gives further
indication of the relation between the good and knowledge and thus of the
reflexivity involved in knowing the good. Socrates claims that in the same way
in which sight is sun-like, so is knowledge "good-like" (agathoeides, 509a).
This claim does not identify the good with knowledge and therefore avoids
Socrates' earlier criticism of such an identification. 34 However, the claim
does maintain a very close relation between knowledge and the good that <loes
not seem to hold between knowledge and any other object (knowledge is not
triangle-like) and that appears to give knowledge of the good the kind of
reflexivity encountered in the Laches and the Charmides. The good is mirrored
in the very knowledge that it causes. Though distinct from our knowledge, the
described in this paper, then we also see that this knowledge is itself problematic. Toe
Republic, lilce the earlier dialogues, deals with the problems inherent in philosophical
knowledge as well as with the accordingly problematic identification of this knowledge
with virtue.
34. Toe good transcends knowledge, justas temperance, though present and revealed
throughout the course of a specific inquiry, is always more than what is to be found in
this inquiry.
Self-Knowledge, Practica! Knowledge, and Insight 175
good is present in our knowledge in the way in which no mere "object" of
1cnowledge can be. In other words, the knowledge by which I know the good
is itself good, just as the knowledge by which I know temperance has been
seen to be itself temperate and the knowledge by which I know courage has
been seen to be itself courageous. The good and its different facets are
revealed not as objects or conclusions of our reasoning but rather as
characteristics of the process of reasoning itself. Thus, the circle just
encountered is not empty: the good is manifest in the very activity of knowing,
just as temperance and courage are manifest in the very way in which Socrates
conducts an inquiry.
6. Further Evidence for Rejlexive Charocter of Plato 's Philosophy
In concluding this discussion of the evidence for attributing to Plato the view
that philosophy has a reflexive structure, note should be made of two claims
about philosophical knowledge frequently found in the dialogues that would
appear groundless without the presupposition that philosophy has this structure:
1) that one cannot possess . this knowledge without being profoundly
transformed by it and 2) that this knowledge is not transferable. The first claim
is found clearly stated in the Protagoros when Socrates warns the enthusiastic
but naive Hippocrates of the danger he faces in entrusting himself to a sophist.
There is little danger in buying food and drink from a grocer because these
things can be taken home in vessels and stored there until one is able to
consult someone who knows what is edible or drinkable and in what quantity.
There is great danger, however, in purchasing knowledge (µ0t8~µ0tm) because
it cannot be stored in a vessel but is necessarily taken into one's very soul by
learning (,o µ&8r¡µ0t i:v odrr& r& v,-vx-6 A0t/3óvm KOtt µ,0t0óvm), so that one
walks away having already been either harmed or benefited (314a-b). This
characteristic of philosophical knowledge is further illustrated in the
Symposium when the philosophical lover who has arrived at a vision of Beauty
Itself is described by Diotima as being so transformed by this vision as to give
birth thereby to true virtue (212a). The second claim about philosophical
knowledge is found in the same dialogue when Socrates remarks to Agathon
how wonderful it would be if wisdom (uo</>í0t) were like water, which can be
poured from a fuller vessel into an emptier one, implying of course that it is
not like this (175d). These claims, then, that philosophical knowledge
necessarily transforms its possessor and that it cannot be simply passed on to
someone else would be obviously false if this knowledge were an objectifiable,
directly expressable body of doctrines. Someone in possession of such
doctrines could easily pass them on to another and "store them away" without
being personally transformed by them. That Plato does not consider this to be
possible with philosophical knowledge shows that he does not believe it to be
176 Toe Tbird Way
objectifiable as a set of doctrines but rather sees it as having the "reflexive"
structure discussed above. 3j
7. Necessity of Dialogue Fonn Confinned
Now that we have seen evidence for the inseparability of method and object
in Plato's conception of philosophy, we are in a position to see the necessity
in Plato's choice of the dialogue form. Because the object of philosophy is to
be seen reflected in the very process of philosophizing, Plato strove in his
works to depict this process as accurately and as vividly as possible, varying
his depiction with each dialogue in order to reveal a different face of truth
each time. To characterize Plato's philosophy in this way is not to equate it
with "empty philosophizing, " 36 since this process as he depicted it has been
seen to be rich with content; this content simply does not have the form of an
objectifiable and universally transmittable result. Plato, then, did not write
treatises on temperance and courage because he did not see the natures of these
virtues (their "forms") as capable of being revealed in sorne explicit
conclusions; instead, he thought they could be revealed only as methods of
philosophy.
35. This reflex.ivity is what Kenneth Sayre appears to have in mind when he claims,
somewhat vaguely, that "philosophic knowledge, for Plato, is a state of mind which is
not a commodity to be transmitted from one person to another" (Review of H . J.
Kramer's Plato and the Foundations ofMetaphysics, Ancient Philosophy 13 [1993): 176).
Referring to the passage from the Protagoras, Marjorie Grene observes that knowledge
for Plato always involves an aspect of acquaintance or a "confrontation with the object
to be known," where this confrontation "at the same time transforms the person himself
who attains knowledge" (Grene, 21). Unfortunately, Grene oddly compromises this
insight when she insists, on the basis of incredibly weak evidence, that Plato ultimately
conceives of knowledge as being necessarily explicit and wholly impersonal (Grene, 30).
36. E. N. Tigerstedt, lnterpreting Plato (Stockholm: Almquist & Wiksell International,
1977), 103. Por a similar critique made of "nondoctrinal" interpretations see H. J.
Kramer's acerbic and, I think, extremely unfair criticisms of Wolfgang Wieland in
"Kritische Bemerkungen zu den j üngsten Áusserungen von W . Wieland und G. Patzig
über Platons Ungeschriebene Lehre," Rivista di Filosofía neoscolastica 74/4 [1982), 579-
92. These criticisms make clear that for Kramer there are only two possible ways of
interpreting Plato's philosophy: as a systematic metaphysics oras an empty and endless
questioning. In other words, the sole options for Kramer are what I have called the
"dogmatic" and the "skeptical" interpretations. Therefore, an interpreter who does not
agree with Kriimer's anribution to Plato of an ax.iomatic, deductive system is seen by him
as making Plato a "skeptic" or at least an "agnostic" (see Kriimer's "Zur Aktuellen
Diskussion um den Philosophiebegriff Platons," Perspektiven der Philosophie 16 [1990),
85-6). Such fruitless debates between two extremes that both fail to do justice to Plato's
thought are what make the working out of a third alternative so imperative.
Self-Knowledge, Practica! Knowledge, and Insight 177
8. Philosophy as Practical: General Description and Hypothesis
Now we can turn to the second and third parts of my initial hypothesis
which, because they are closely related to the first, can be dealt with much
more briefly. Toe second part states that there is a necessary relation between
philosophy and the dialogue form if philosophy is more practica! than
theoret~cal. This hypothesis can be defended by the observation that nothing
stands m th~ w~y of expressirg directly in a treatise a theoretical knowledge
that something 1s the case. That the earth is round, that two plus two equals
four, that form is distinct from matter, are ali knowledge claims which can be
set forth in a treatise. On the other hand, if philosophy is a practica!
knowledge of huw to deal with objects, it cannot be expounded as a set of
doctrines but can only be exhibited, in this case dramatically. "Knowledge
that:· can be,state~'. "kn™'.ledge how" can only be sh(Mlfl. Thus, the necessity
behind Pla~ s wi:ittng of dialogues can be justified only if we move away from
understan~ng philosophy as a purely theoretical enterprise and begin to see it
as a peculiar "know how. " Only if we stop looking for what is concluded and
instead focus on huw the inquiry is conducted, will we be able to read the
dialogues as dialogues. ·
In examining if this conception of philosophy as practica! is Plato's own we
c~ start by n?ting the relation between it and the reflexivity of philos~phy
disc~sed e~her. If the knowledge with which this reflexive philosophy
prov1des us 1s not a knowledge of definitions or conclusions, then what kind
of knowledge is it? lf the knowledge at which philosophy aims is not
completely externa! to the inquirer and objectifiable in the way that the
knowledge provided by the natural sciences is and if it therefore is not
scientific or "theoretical" knowledge, then what is it? Toe answer is that this
knowledg: is to be understood as essentially proctical. Practica!, as opposed
to theoret1cal, knowledge cannot be expressed in sorne result or conclusion
externa! to ~e knower but instead can only be shown or displayed in the very
act of knowmg. A knowledge of how to swim is not to be found in certain
~onclusions and definitions that the swim.mer has reached and possesses but is
1nstead made manifest only in the swimmer's actual ability to swim. Tbis
knowledge is therefore not separable from what the swimmer does and even
who the swimmer is. 1t is for this reason that such practica! knowledge is not
~ro~f~mble. Swi~ can indeed be taught, but only indirectly through
~tatton ~nd practtce. Though the teacher can of course provide the student
w1th certam rules and tips, such transferable information by itself in no way
178 Toe Third Way
constitutes a knowledge of how to swim. 37 Perhaps, then, philosopby itself
is a practical knowledge which as such is is to be found, not in any body of
doctrines which the philosopher possesses, but rather in what the philosopher
does and who he is. Toen it could be said that what we principally gain
through philosopby is not knowledge that certain objects have such and such
properties, but rather what a German scholar (Wolfgang Wieland) has called
Gebrauchswissen38 and what we would call "knowledge how to make use of
things. ,,39
9. "User's Knowledge" in Republic, Cratylus and Euthydemus
This suggestion, however, seems contradicted by Plato's frequent use of a
theoretical model of knowledge in the dialogues. In the Republic a visual
model is used to describe how we know the good, and both here and in other
dialogues the forms are described as being seen. This kind of talk seems to
render knowledge of the forms as objective and as theoretical as it can be.
However, there is a text in the Republic itself that describes this knowledge in
terms of a much more practical model. It is to this text that we should now
turn.
In a well-known part of Book X of the Republic, Socrates attempts to show
the inferiority of imitation at first by contrasting the imitator with the
craftsman and God. God creates the form or nature of a couch; the craftsman
looks to this form in producing a concrete couch on which one can actually
recline; the imitator copies the concrete couch in a painting. Thus, what the
37. See what Wolfgang Wieland has written about the "reflexivity" of practica!
knowledge (Platon und die Formen des Wtssens [Gottingen: Vandenhoeck & Ruprecht,
1982], 241). His main point is that practica! knowledge does not have as its intentional
content objects from which one could distance oneself. Instead, one understands oneself
in this kind of knowledge.
38. A major thesis that Wieland has defended is that the insight provided by dialectic
"weniger dem Typus des theoretischen Wissens als dem Typus des praktischen Konnens
zugehort" ("Platon und der Nutzen der Idee; zur Funktion der Idee des Guten"
Allgemeine Zeitschrift jür Philosophie 1 [1976]: 32).
39. Gilbert Ryle is primarily responsible for introducing the distinction between
"knowledge that" and "knowledge how" into contemporary philosophical discourse (see
The Concept of Mind [London: Hutchinson & Hutchinson, 1949], 25-61). Of course,
there are problerns with Ryle's distinction, and it has not been universally accepted. For
a discussion of sorne ofthese problems, see William Lyons, Gilbert Ryle: An Introduction
to his Philosophy (New Jersey: Humanities Press, 1980), 57-69. However, no matter how
one accounts for the distinction between "knowledge how" and "knowledge that," it
seems to me that it must hold in sorne sense, and that what I say here about the
"reflexive" and nonobjectifiable, nontransferable character of the former adds support
to the distinction.
Self-Knowledge, Practica! Knowledge, and Insight 179
¡nútator produces is the copy of a copy, and he accordingly has no knowledge
of the form from which he is three times removed. Toe craftsman, on the
other hand, in referring to the formas a model, "sees" it and thereby appears
to have knowledge of it. This example, then, reinforces the view that
kJ10Wledge of the forms is theoretical. As the discussion proceeds, however,
Socrates suddenly changes his example: the maker, though superior to the
¡nútator, does not really have knowledge but only true belief; the person who
has knowledge is instead the one who uses what the maker only produces. Toe
craftsman is no longer granted direct vision of the form, but is said simply to
foliow the instructions of the person who knows how to make use of the
form's concrete copy and in that use knows the form itself. Toe general points
that Socrates now proceeds to make deserve our full attention:
And will we not say that the same is true of everything? - What do you mean? -
That corresponding to each thing there are three arts (Téxva<;), the user's art, the
roaker's art and the imitator's art? (xpr¡aoµévr¡v , 1roi~aovaa11, µiµr¡aoµévr¡v) - Yes -
Is it not the case that the excellence, beauty and correctness of each artefact, living
thing or activity concerns nothing other than the use (oú 1rpo<; &1,.1,.0 n ij ri¡v
XPEÍav Éar0 for which it was made or by nature carne to be? - That is so. - It is
necessarily the case, then, that the person who uses each thing will have the most ·
experience (ɵ1rELpÓmro11) of that thing and will inform the rnaker how good or
bad his products are for the use he makes of them . .. . Toen concerning the same
artefact [the example of a flute is being discussed] the rnaker will have true belief
about its excellence and its defects by associating with the person who knows and
being compelled to listen to him, while the user will be the one with knowledge (o
ÓE XPWµEvo<; É1ria~µr¡11, 601c-602a).
Most scholars have either ignored or refused to grant importance to the
identification here of knowledge of a thing's form or nature with a user's art.
James Adam, for example, represents the typical view when he claims that no
metaphysical significance is to be attached to the statement that the user has
knowledge. He proceeds, however, to admit: "There is no doubt a certain
sense in which-if we have regard to Crat. 390B ff. and Euthyd. 288E ff.- o
xpwµevo<; has not indeed scientific knowledge of the Idea, but something
analogous thereto. "40 Adam simply assumes that knowledge of the forms
40. Adam's comments continue as follows: "Dialectic, which is the scientific
knowledge of Ideas, is 1'ar' É~ox~11 the xpwµÉvr¡ e1r,a~µ71, the Science which alone
knows in what respect each thing is good and useful and uses things accordingly (cf.
Euthyd. 290C, Crat. 390C), proving itself thereby the royal or kingly science (VI 505A
n.) • . . . But if Plato had intended us to pursue this vein, he would, I think, have
furnished us with sorne hints in the course of the argument itself" (The Republic of Plato
vol. 2, 2d ed. [Cambridge: Cambridge University Press , 1963], 404). But what hints are
offered in the dialogues for Adam's view that dialectical knowledge for Plato is scientific
180 Toe Third Way
must be understood to be "scientific" knowledge in our sense of the word, in
which case the user's art can at best be very loosely analogous to it. Wolfgang
Wieland, however, has rejected this assumption and claimed that the user's art
described in Republic X reveals what Plato considered to be the structure of
philosophical knowledge per se. 41 Philosophy is not a scientific knowledge
of objects but rather a knowledge of forms that guides philosophical discourse
(i.e., the peculiar use of words, hypotheses, and images that defines this
discourse) without objectifyiog or reifyiog these forms. 42
A few observations can be made in support of Wieland's view. First, it
should be noted that the user's art as Socrates describes it is very broad in
scope: it is said to provide us with knowledge not only of man-made
implements, but also of natural objects and actions. Second, the user's art
clearly provides us with a knowledge of the forms or natures of these thiogs,
since what more can be involved in knowiog a thiog's form than knowiog its
excellence, beauty, and rightness? Tbird, the identification of dialectic with a
user's art like the one described here is not unheard of in the dialogues.43 In
the Cratylus the dialectician is contrasted with the "legislator" who makes
words by beiog identified with the person who knows how to use words
(xpi,CT0m, 390c). What this appears to mean from the context is that the
dialectician is characterized by an ability to use words in such a way that they
fulfill their function of makiog manifest determinate meaniogs (an ability that
is not characteristic of everyday discourse). In the "Euthydemus,44 we see that
the only thiog that distioguishes Socrates from the eristics is the use to which
he puts his arguments. Therefore, in his discussion with the boy Cleinias,
Socrates identifies with happiness a knowledge of how to use correctly (op0wc;
xpi,a0ai) one's possessions (virtues and skills as well as material goods; 290b-
d). That highest knowledge which will make us happy must therefore be a
knowledge of how to use correctly the discoveries and products of ali other
kinds of knowledge. It is implied in the rest of the discussion that this ultimate
and deductive? Toe passages Adam cites come as close as any other passages in the
dialogues to telling us what the nature of this dialectical k.nowledge is.
41. Wieland, Platon und die Formen des Wissens, 293.
42. See Wieland's characterization of what knowledge of the forros would be like if
modelled according to a user's art rather than according to "scientific k.nowledge" (lbid.,
148). ' substitutes for Laches' practical "know how" a purely cognitivist and
43. Detailed interpretations of the Cratylus and the Euthydemus along these lines can
be found in my book, Dialectic and Dialogue: Plato 's Practice of Philosophical Inquiry
(forthcoming).
44. It should be noted that Gerard Ledger in his book Re-counting Plato (Oxford:
Clarendon Press, 1989) has been led by his stylometric studies to place the Euthydemus
in the middle group of dialogues along with the Republic.
Self-Knowledge, Practical Knowledge, and Insight 181
"know how" is philosophy itself. Thus, dialectic is explicitly identified with
a user's art in the followiog passage:
No more belongs to the art of hunters (8r¡pevnK~) than to hunt and capture. When
they capture what they hunt, they are unable to make use of it ()(p~a8c.i). Instead,
two types of hunters, the hunter in the literal sense and the fisherman, give what
they have caught to the cooks. Geometers, astronomers and calculators, however,
are also hunters by virttle of the fact that they not only make diagrams but actually
discover realities (r& bvrn), but since they do not k.now how to make use of what
they discover but k.now only how to hunt, those of them who are not totally
senseless hand over their discoveries to the dialectians for their use (Kc.rc.xpija8c.i,
290b-c).
Of course, a problem that must be dealt with in pursuing this view is that the
lengthy discussion of dialectic in the middle books of the Republic does not
explicitly identify dialectic in this way with a user's art. However, two points
can be made. First, the very identification of the ultimate object of dialectical
knowledge with the good makes "use" central to this knowledge. Toe idea of
the good is introduced by Socrates as that "by the use of which just thiogs and
ali other thiogs become useful and beneficial" (505a). If this is the case, then
dialectic as a knowledge of the good clearly must be what Socrates later in
Book Ten calls a "user's knowledge." Secondly, it should be remembered that
in describing the two top sections of the Divided Line, i.e., mathematics and
dialectic, Socrates describes them almost entirely in terms of what they do or,
more specifically, how they use hypotheses. It is precisely for this reason that
there has been so much dispute about the objects corresponding to these two
sections. (Are there mathematical entities intermediate between forms and
particulars, as seems required if the third section is to have its own objects?)
In whatever way one decides the dispute, it cannot be denied that Socrates
regards the practice of dialectic as much more important than its objects for
distioguishing it from mathematics and other sciences. There seems to be
enough indication, then, that Plato understood dialectic more in terms of
practical than of theoretical knowledge to render fruitful pursuit of this
connection to the dialogue form. 45
Of course, it is not sufficient to characterize philosophy as "practical
knowledge." Even Laches has a practical knowledge of courage, and yet
Socrates is clearly not satisfied with this and seeks to inquire beyond practical
experience. As we have seen, however, this does not mean that Socrates
45. One might also adduce as further evidence the claim in the Lysis (210a-c) that
those things of which we have knowledge will belong to us because we will be able to
use them and profit from them.
182 Toe Third Way
theoretical conception of virtue. Just as Laches exhibits his understanding of
courage in what he does, so does Socrates exhibit his understanding of courage
in what he does. Toe contrast here is not between theoretical knowledge and
practica! knowledge but rather between two kinds of practica! "know how."
Toe difference is made clear by the fact that Socrates' dialectical " know how"
involves refuting the ordinary conceptions that guide everyday "know how,"
with the goal of allowing an understanding of the virtue in question to emerge
through this very process of refutation. It is in these terms, then, that the
dialectician's use of hypotheses and ordinary conceptions would have to be
construed. Toe nature of this "use" can not be further explored here, but it
can be suggested that Plato chose the dialogue fonn precisely for the sake of
expressi~ philosophy as this kind of a practica! knowledge.
10. Philosophy as NonpropositionaJ.: General Description ami Hypothesis
Toe third part of my initial hypothesis focuses on a characteristic of
philosophy that is a necessary complement to the one just discussed: the
relation of the dialogue form to philosophical knowledge is a necessary one if
this knowledge has the character of nonpropositional insight. Toe defence for
this hypothesis lies in the fact that a proposition can always be directly stated
in writing. lf, therefore, philosophical knowledge is propositional, there is no
reason why it could not be directly communicated in the written treatise. If,
on the other hand, the knowledge provided by philosophy is nonpropositional,
it will not lend itself to expression in a treatise but only to indirect suggestion
in the dramatic dialogue. It has been noted by others that whatever truth Plato
chooses to communicate in the dialogues, whether early or late, is necessarily
communicated indirectly.46 This is due to the fact that the dialogues all
portray conversations in which Plato himself does not take part. Even if we
make the questionable assumption that Socrates or the Eleatic Stranger is
Plato's mouthpiece, in this case Plato would still be communicating indirectly
through these characters. In addition to this silence caused by Plato's
"anonymity," there is within the dialogues themselves much that is suggested,
hinted at, and shown without being directly stated. This characteristic of the
dialogues is particularly evident in my reading of the Laches and the
Charmides, according to which the most important truths in these dialogues are
exhibited but never asserted. It is also evident in the omnipresence in the
46. "We shall not be able to read directly out of the dialogues anything that counts eo
ipso as a theory of Plato" (Kosman, "Silence and Imitation," 85).
Self-Knowledge, Practica! Knowledge, and Insight 183
dialogues of that indirect form of communication known as "irony. "47 Thus,
despite the great deal of talk in them, the Platonic dialogues are penneated
with silence, both the silence of the author and the silence that fonns a
backdrop to the conversations themselves. But if Plato understood
philosophical knowledge to be entirely propositional, this cultivation of silence
in the dialogues would appear entirely pointless. What could in this case have
prevented Plato from presenting his arguments and conclusions in his own
person and from leaving no recognized truth unstated? Extraneous reasons,
e.g. , that Plato chose to keep his doctrines hidden for pedagogical or political
reasons, are unsatisfactory and depend, as I show below, on a misreading of
the Seventh Letter. On the other hand, if philosophical knowledge involved for
Plato sorne sort of nonpropositional insight, the silence of the dialogues
becomes apt and necessary. 48
To determine whether or not philosophy for Plato has this third
characteristic, we must first see how it is consistent with and yet fills a gap left
by the second characteristic. Toe consistency lies in the fact that practica!
knowledge appears to be itself nonpropositional, i. e. , not reducible to
propositional "knowledge thaL " 49 However, by referring to the role of
nonpropositional insight in philosophy, the present characterization supplies
something that would be lacking if we were to describe philosophy as being
simply practica!. Even if we grant that philosophy is essentially a " know how,"
we must also grant that it is not the same in kind as knowing how to ride a
bicycle. Toe reason is that philosophy clearly has sorne theoretical content. ~
47. "lrony is a way of speaking (or writing) which is meant to point to what is not
spoken (or written), to what is silent and is kept in reserve, as it were, by its originator"
(Charl~s Griswold, "Irony and Aesthetic Language in Plato's Dialogues," in Philosophy
arut Lzterature, ed. Doug Bolling [New York: Haven Publications, 1987). 78-9). This
article as a whole provides an extremely insightful and helpful account of the different
forms and levels of irony to be found in the dialogues.
48. Toe connection between the nonpropositional character ofphilosophical knowledge
and the necessity ofthe dialogue form has also been recognized by Drew Hyland , "Why
Plato Wrote Dialogues," Philosophy arut Rhetoric 1 (1968): 42-43.
49. That Plato himself believed practica! knowledge to be nonpropositional is suggested
by the fact that in Republic X the verbal instructions that the user communicates to the
maker provide the maker with only true belief, not knowledge.
50. I therefore in large part agree with the objections that have been made against John
Gould's overly simplistic identi.fication of epistémé in the Socratic paradox "virtue is
knowledge" with "knowledge how" (The Development of Plato's Ethics [New York:
Russell & Russell, 19551). Por pertinent criticisms , see Gregory Vlastos, "Socratic
Knowledge and Platonic Pessimism, " Philosophical Review 66 (1957): 226-38 and John
Rist, "Knowing How and Knowing That," in Eros and Psyche: Studies in Plato, Plotinus
arut Origen (foronto: University of Toronto Press, 1964), 115-41. Rist shows the
problem with Gould's view particularly well when he argues that if the knowledge
184 Toe Third Way
In practicing philosophy we do come to understand something (call this a
"forro" or what you will). Socrates' inquiry in the Laches and the Charmides
provides us not only with a skill to be imitated but also wi~ i~si~ht i~to what
courage or temperance is. As we have seen, however, this ms1gh! 1s not a
definition that is stated at the end of the dialogue but rather something that is
"sparked" by what happens in the dialogue as a whole. As was also seen, no
definition succeeds in doing justice to this insight. In this way the third
characteristic, like the second, follows from the first one, i.e., that philosophy
does not aim at an objectifiable result, but is reflex.ive. Through the inquiry
itself we come to see what temperance or courage is even though ali attempts
to say what it is have failed. Plato's frequent vision metaphors are therefore
very appropriate for describing this aspect of philosophy.
11. Nonproposidonal Knowledge in the Seventh Letter
Toe most explicit statement of this characteristic of Plato's philosophy is of
course to be found in the Seventh Letter. There Plato claims that what he takes
identified with virtue is simply a "know how," then one can have this knowledge even
if one "has no intellectual grasp of what is good and what is bad," even if one does not
recognize "that there is sorne thing which may be called good" (122). Where both Rist
and Vlastos go too far, however, is in simply assuming without argument that this
intellectual grasp of what is good must take the form of propositional "knowledge that."
It is this assumption that makes them reject Gould's position altogether by maintaining
that all "knowledge how" in the dialogues is fully reducible to "knowledge that." While
J. Hintikka is also critica! of Gould's view, he is better able to do justice to Gould's
iosight by making a distinction between "knowledge that" and "knowledge what"
(knowledge of a thing's essence) and claiming that epistémé combines the latter with
"knowledge how" ("Plato on Knowing How, Knowing That, and Knowing What," in
Knowledge and the Known [Dordrecht/Boston: D. Reidel. 1974), 31-49; see especially
37). According to Hintikka, then, Gould is right in maintaining that epistémé is not
simply "knowledge that." However, since Hintikka's "knowledge what" is simply a
knowledge of defuútioos, he sees it as simply one species of "knowledge that." N. D.
Smith takes up Hintikka's notion of "knowledge what," but distinguishes it from
propositional "knowledge that" in a way that Hintikka does not. Thus, Smith writes t~at
"knowledge what" differs from propositional knowledge in that it "does not necessarily
take propositioos or sentences as its objects, or known items" ("Knowledge by
Acquaintance and Knowing What in Plato's Republic," Dialogue 18 [1979): 282). Yet
Smith does not appear to remain true to this claim when he, like Hintikka, persists in
identifying "knowledge what" with knowledge of true propositioos about a thing's
essence. Gould's critics, then, have either not at all or inadequately coosidered the
possiblity that the knowledge which guides "knowledge how" and together with it forms
two complementary aspects of Platonic epistémé is that nonpropositional iosight which
also makes its appearance in the dialogues.
Self-Knowledge, Practica! Knowledge, and Insight 185
roost seriously (apparently, the ultimate principies) cannot be expressed in
words as other studies can (jn¡rov -yap oúocxµw~ w~ &)..),ex µcx0f¡µcxm 341c).
Plato of course does not mean that we must begin to stutter whenever we
attempt to talk about the good or that the ink will freeze in our pens whenever
we attempt to write about it. Plato's point is instead that such a lecture or
treatise on the good will necessarily fail to express the true nature of the good
and therefore will be more deceiving than enlightening. In understanding what
Plato says here it is important that we do not make the mistake made by those
who wish to attribute to Plato "esoteric doctrines." Though the fear of
roisunderstanding on the part of the uninitiated is cited in the Seventh Letter
as one reason for not attempting to express in writing the nature of the
ultimate principies, the letter goes on to show that there is a weakness inherent
in the very nature of language that makes impossible any verbal expression of
the principies, whether written or oral. 51 In this way it is shown that the oral
formulation of esoteric doctrines is no more possible than the written
formulation of exoteric doctrines. But what then is this weakness inherent in
the nature of language? As the letter explains, it is the fact that a sentence or
proposition expresses how a thing is qualified (.poion ti) rather than what it is
52
(ti esti). Toe point is apparently that any statement I make, for example,
about virtue, will only qualify virtue in one way or another by saying that it
51. Kenneth Sayre has rightly argued, against those who use the letter to attribute to
Plato "unwritten teachings," that Plato's criticisms here by no meaos leave the spoken
word untouched, but that his contention is the strong one that "neither oral nor written
language is capable of expressing the grasp of being which stands at the end of
philosophic inquiry" ("Plato's Dialogues in the Light of the Seventh Letter," in Platonic
Writings, Platonic Readings, 91; see also 95-6; see also Sayre's review of H. J. Kriimer's
Plalo and the Foundation of Metaphysics, 172-6). Sayre's article as a whole should be
coosulted for a defeose of the attribution to Plato of a "nondiscursive" view of
knowledge, especially since his defeose does what cannot be done within the limits of this
paper: it draws upon evidence from a number of dialogues as well as from the Seventh
Letter. A more detailed presentation of my interpretation of the Seventh Letter can be
found in the last chapter of my book, Dialectic and Dialogue: Plato 's Practice of
Philosophical Inquiry (forthcoming) and in my article, "Nonpropositional Knowledge in
Plato" (forthcoming).
52. "Furthermore, the four make manifest no less (oúx ~TTov) how a thing is
qualified (ro 1roióv n) than what it is (ro ov) due to the weakness of language (Óta TO
rwv 1,.ó-ywv éxu0evé<;)" (342e-343a). Toe words "oúx ~TTov" may suggest that language
expresses what a thing isjust as muchas how it is qualijied. However, these words are
clearly an understatement. Toe weakness of language is that it expresses how a thing is
qualified in place of expressing what it is . As a later passage states, while the soul seeks
the being of a thing and not its quality, each of the four meaos of philosophical
knowledge "offers the soul, both in words and in deeds, that which is not sought" (343b-
c).
186 The Third Way
has this or that property. What the letter assumes, then, is that to qualify
virtue thus in a proposition can never be equivalent to knowing what virtue
itself is. In knowing only that virtue has certain qualities (a knowledge that can
be stated in propositions) we do not yet know what virtue itself is (a
knowledge that therefore is nonpropositional). 53 It is for this reason, and not
for the salce of any obscurantism, that this knowledge cannot be expressed in
words as other studies can be. 54 Thus, the argument of the Seventh Letter
claims unambiguously that Socratic questions such as "What is temperance?"
and "What is courage?" can never be answered and thus the aporia of the so-
called early dialogues proves necessary.55 Any possible definition will
predicate certain qualities of these virtues more or less truly or falsely, but will
necessarily fail to express what either virtue is. The knowledge sought here is
simply not definitional. Though we can talk around this knowledge, we cannot
express it as such in any set of propositions. It instead has the character of that
insight which Plato describes with the metaphor of the "leaping flame."
The authenticity of the Seventh Letter has of course not been proven (if the
burden of proof indeed rests on those who accept itas genuine), though recent
scholarship appears to be leaning in its favor. 56 Yet the criticism of verbal
expression found in -the letter not only appears to be assumed in several
dialogues, but is an explicit theme of the l'haedrus, 51 while the distinction
53. This is why the unity of the virtues and the equation of virtue with knowledge are
never presented as "doctrines" in the dialogues, though scholars misrepresent thern as
such. Toe knowledge that unifies all the virtues and is virtue is not sorne definition or
sorne formula that could be expressed in propositional form; it is instead the kind of
knowledge which can only be exhibited and is exhibited best in the philosophical
enterprise (i.e., in what Socrates is shown to do).
54. Por a good account of this argument, see Rafael Perber, Die Unwissenheit des
Philosophen oder Warum hat Plato die "ungeschriebeneLehre "nicht geschrieben? (Sankt
Augustin: Academia, 1991), 45-52.
55. Cf. Perber, ibid., 47.
56. Atternpts to prove or disprove the letter by rneans of stylornetry have not been
conclusive. Most recently, however, Gerard Ledger's cornputer analyses have shown the
Seventh Letter to be so close in style to the late dialogues (the Laws in particular) that
he has taken its authenticity to be virtually proveo (Re-Counting Plato [Oxford:
Clarendon Press, 1989), 148-50, 199). Given the thoroughness of Ledger's investigation
("a far more cornprehensive survey of the entire corpus of Platonic dialogues than has
ever been atternpted before," 226), the stylornetric evidence at present can be said to be
on the side of authenticity.
57. Sayre sees the critique of writing in the Phaedrus as agreeing with the argument
of the Seventh Letter: "Par frorn extolling vocal over written speech, the primary burden
of these passages in the Phaedrus, like those of the Seventh Letter examined previously,
is to draw attention to a kind of intellectual awareness that eludes expression in speech
and writing" ("Plato's Dialogues in the Light of the Seventh Letter," 98).
Self-Knowledge, Practical Knowledge, and Insight 187
itsel~ b~tween what a thing is (ti esti) and how it is qualified (poion ti esti) is
one 1ns1sted upon at several places in the dialogues. 58 What is unique to the
Seventh Letter is simply the explicit grounding of the criticism on the
distinction. It is also important to note that the Seventh Letter does not make
Plato a "mystic." It is impossible to deny the importance of argumentation and
discussion in Plato's conception of philosophy. However, this characteristic of
phil~sophy is not incompatible with the view that after all the discussirg and
aigUlllg there remains a silence which as such resists all direct verbal
e~press_ion. The fact that the means of philosophical understanding are
discursive does not necessarily make this understanding itself discursive.
Instead, through the "rubbing together" (-rpí{fov) of ali kinds of statements
and a~m.en~ (the P_ractical aspect of philosophy), we have the "leaping
flame of 1ns1ght. It 1s because the understanding provided by philosophy is
to be found in such insight rather than in a body of propositions that Plato had
to choose the dialogue form for its expression.
12. Conclusion
The p~e.s ent stu~y h~, I ho~e, both demonstrated the possibility of
char~ctenz1ng ~lato s philosophy m such a way as to render necessary its
l?
relat1on ~e d~alogue form and shown that each of the three interdependent
charactenst1cs discussed-the reflexivity of philosophy, the practical character
of philosophical knowledge, and the nonpropositional character of
p~losophical insight-can be defended and finds support in the text of the
dialogues. Of course, more evidence needs to be considered for a complete
defense of this wide-ranging thesis. What cannot be doubted however is the
necessity of P?1"5~ing the project begun here. The reason is si~ple: we 'cannot
expect to do Just1ce to the form of Plato's dialogues as long as we bring to
them a conception of philosophy derived from a philosophical tradition in
which the treatise has always been the preferred mode of expression. Like both
th~ Skeptics and ~e Neoplatonists, interpreters tend to come to the dialogues
w1th a P!econceptton of what counts as genuine philosophy. Yet the true value
of the dialogues, and the source of their endless fascination, lies in the fact
that. through both their content and their form they continually challenge us to
rethink the very nature of the philosophical enterprise.
58. Meno 71aff., 88d-e, 100b; Republic 354b; Laches 190b-c; Gorgias 448e;
Prot~g~ras_ 36~c. With the possible exceptions of the Meno and the Republic, however,
the d1stmct1on 1s not as "technical" and as clearly defined in these passages as it is in the
Seventh Letter.
 11

Dogma, Skepticism, and Dialogue

Walter Watson

l. Dogmatic, Skeptical, anti Dialogic Interpretations of Plato

Introduction: The History of the lnterpretation of Plato

Dogmatic interpretations of Plato hold that he has a doctrine; skeptical

interpretations do not. Both have a long history. Cícero in Book I of his
Acaáemics has Varro present Plato's philosophy as a three-fold system
comprising living well, nature, and reason and discussion, which he says was
received from Plato by the Old Academy. Cícero counters this with the claim
that Plato's writings exhibit rather the skepticism of the New Academy. After
giving a brief account of the principies of Arcesilaus, he says, "This they call
the New Academy, which however appears to me to be the old one, if, at
least, we reckon Plato as one of that Old Academy. For in bis books nothing
is affirmed positively, and many arguments are allowed on both sides of a
question; everything is investigated, and nothing positive affirmed. " 1
Augustine retums to the dogmatic Plato by proposing in his Contra
Acaáemicos that the New as well as the Old Academy possessed the true
Platonic doctrine but concealed it behind a skeptical facade. "Arcesilaus very

l. Academica i.12, trans. C. D. Yonge in The Academic Questions, Treatise De

Finibus, and Tusculan Disputations, of M. T. Cícero (London: Bohn, 1853), 21.
 Toe Third Way
190
prudently and ingeniously concealed the complete doctrine of the Aca~em:,;
and cached it as a golden treasure to be discovered sorne day by postenty.
Similarly, Carneades and his associates "knew what truth itself was, but they
approved false things in which they recognized a laudable imitation of true
things. " 3
E. N. Tigerstedt, in his history of the interpretations of Plato, documents the
recurrence of skeptical and dogmatic interpretations and notes occasional
attempts at a "third alternative," as in Jean de Serres, Theophilus Gale, and
Luigi Stephanini. 4
Toe difficulties of both the dogmatic and skeptical interpretations are evident
and well-known.
Toe attempt to construct a Platonic doctrine runs counter to the manifest
form and content of the dialogues, to the place of words as images at the
lowest level of the divided line, to the disparagement of writing in the
Phaedrus, and to the warnings in the Seventh Letter against the attempt to
reduce philosophy to an art.
It does not help to suppose that in addition to the dialogues there were
unwritten doctrines, for these too, if we are to believe what is said in the
dialogues and letters, could not have been given a definitive formulation as
doctrines. They too would require dialogic transmission. Even in the perfect
state, where we may suppose doctrinal knowledge to be perfected, the
instruction of the philosopher-kings in the greatest study is by dialectic, not by
lecture. Toe only way out of this difficulty is to reject the dialogues and letters
in favor of a construction of one's own.
On the other hand, the reduction of Plato's thought to skepticism runs
counter to the definite conclusions the dialogues frequently reach, to their
opposition to the skepticism of the sophists, to the possibility of knowledge at
the upper levels of the divided line, to the rejection of probabilities for truth
in the Phaedrus, and to the claim in the Seventh Letter that the correct
philosophy affords a vantage point from which we can discem in ali cases
what is just for communities and individuals.
Tigerstedt's history of the interpretations of Plato must therefore excite our
wonder. If both the dogmatic and skeptical interpretations are evidently
unsatisfactory, why have so many pursued them? And why have so few
pursued a third altemative?
2. Contra Academicos iii. 17.38 , trans. Denis J. Kavanagh in The Writings of St.
Augustine, vol. I (New York: Cima, 1948), 215.
3. /bid. 18. 40, Writings of St. Augustine, 217 .
4. E. N. Tigerstedt, The Decline and Fall of the Neoplatonic Jnterpretation of Plato,
Commentationes Humanum Litterarum 52 (Helsinki: Societas Scientiarum Fennica,
1974), and Interpreting Plato (Stockholm: Almqvist & Wiksell International, 1979).
Dogma, Skepticism, and Dialogue 191
Viewed in the light of the Apology, the history speaks with the voice of
God. In the Apology the voice of God, and its transmission through the bite
of the gadfly, is heard in contradictions. 5 Contradictions arise when
conventional opposites are applied to Socrates. Is Socrates wise or ignorant,
a cl~er speaker or a plain speaker, a pious believer in the gods or a skeptical
questtoner of the gods, an appeaser of his audience or an irritator of his
audience, serious or joking, arrogant or humble, political or non-political,
conservative or revolutionary, a conformist or a non-conformist, obedient to
the laws or disobedient to the laws, a teacher or not a teacher, a specialist or
a universalist, rational or irrational, this-worldly or other-worldly, contented
or discontented, a winner of his case or a loser of his case? Toe point is not
that the opposites co-exist in Socrates, but that their application to him reveals
that what we ordinarily take to be opposites are not really opposites. It is not,
for example, that Socrates is both innovative and traditional, but that his
traditionalism and his innovation are the same thing. If ali things derive from
a single principie, and Socrates is most in accord with this principie, then he
will exhibit unities where we who are less wise see oppositions. His
p~l~sophy is paradoxical in the literal sense of being contrary to accepted
op1mon. Perhaps the opposition between skepticism and dogmatism in its
application to Plato has similar origins and can be similarly instructive. Is
~ere in the dialogues a unity of conventional opposites which, when
mt~rpreted by those who think in terms of conventional opposites, as we ali
do 1nsofar as we are ignorant, falls apart into opposed dogmatic and skeptical
interpretations? This at least is the hypothesis of the present paper.
Toe primary basis for the skeptical interpretation is the form of the
~ialogues, _as ~e se: in Cícero, while the primary basis for the dogmatic
mterpretatton 1s their content, as we see in Varro's account of the Old
A~adem~. A third way of interpretation must unite the form of the dialogues
w1th their content. I shall call this third way "dialogic" or "dialogical,"
although this name is also claimed, justifiably, by sorne skeptical interpreters
of Plato, notably by my friend and colleague Professor Tejera. Toe shared
~ame cor_responds to our shared criticism of the prevailing dogmatic
mterpretat1ons. Toe usual name for the skeptical interpretations today
however, is "dramatic. " 6 '
5. See Walter Watson, "Toe Voices of the God," in New Essays on Socrates, ed.
Eugene Kelly (Lanham, Maryland: University Press of America, 1984), 173-79.
6 . 1:"ofessor Tejera denies that his reading is skeptical, arguing that the dialogues as
dramatlc make no knowledge claims, neither the dogmatic claim to know nor the
skeptical claim not to know. But this agrees with the true skeptical position of Pyrrho and
the New Academy, which rejects even the claim not to know as dogmatic. Toe
~kepticism of_ Plato's Socrates, on the other hand, unites dogmatism and skepticism, for
1t knows that 1t does not know. However this may be, I shall be using Professor Tejeras's
192 Toe Third Way
Toe characteristics of the dialogues as dialogues enable us to identify the
key points of difference between the skeptical and dogmatic interpretations.
Toe dialogues present the views of individual speakers, the speakers interact
in concrete situations, their interaction takes the form of dialogue, and the
dialogues for the most part take place independently of one another. Toe letters
do not differ essentially from the dialogues in these respects, for they too are
statements made by an individual to others in a concrete dialogic context.
l. Truth: Divine or Human?
Toe statements of the dialogues are attached to individual speakers. Our first
problem is whether the dialogues express the opinions of individual speakers
or truths that are independent of the speakers who express them.
For the skeptic, the Platonic dialogues present the opinions of the individual
speakers, not truths that are independent of these speakers. Plato is not among
the speakers in the dialogues, and we are therefore not justified in saying that
any of them presents his views or, if we athetize the letters, that there is a
Platonic doctrine. Whenever we cite a statement from the dialogues, we should
cite it as the view of the person who makes it, not as the view of Plato
himself. Toe arguments of the dialogues may lead to shared opinions, but they
remain the opinions of those who share them. Speakers may express beliefs
about a wisdom that transcends human individuals, but these beliefs are again
simply the opinions of the speakers, and cannot be taken as an assertion by
Plato that there is such a wisdom.
This view naturally leads us to think of the dialogues as primarily dramatic,
since in dramas the statements of the characters are taken primarily as what
these characters think rather than as statements of what is the case.
Althougb the dialogues present the views of individual speakers in dramaric
form, they do not treat these views merely as individual opinions. They are
continually concemed with a shared evaluation of these opinions, and the
Theaetetus argues against the view that truth is merely personal. It is always
possible to read statements as simply the personal opinions of those who make
them, but here this interpretation is opposed by the whole thrust of the
dialogues themselves, and thus reflects the skeptic's view of truth rather than
the view of truth expressed in the dialogues.
Toe same point applies to the claim that the dialogues are primarily dramatic
· or exhibitive rather than assertoric. Justus Bucbler's distinction between
assertoric, active, and exhibitive judgments, which corresponds to Aristotle's
distinction of theoretical, practical, and poetic sciences, is not found in the
dialogues and is not a Platonic distinction. If one applies it to the dialogues,
reading of Plato as my principal exemplar of the skeptical interpretation.
Dogma, Skepticism, and Dialogue 193
it is evident that they ~re theoretical, practical, and poetic all at once. Toe
!_'haedo ~or example 1s at ~nce a sustained argument concerning the
nnmo~ty of. the_ soul, an action of Socrates and his associates at a critical
~ome~t m the1r hves, and a dramatic sequence of events. In the dialogues,
v1~e 1s knowledge and both are beautiful, and in their upper reaches we
glimpse ~e good as at once a principie of knowledge and of action, as divine
and m~anable and at the s~e time a pattem for action among variable things.
The dt'.11ogu:~ are "ente~mng and dramatic, but to read them as primarily
dramattc, as plays wntten for our entertainment, imposes an alien distinction
upon them.
Th~ situation is no better if we use a Platonic rather than an Aristotelian
meamng_ for "dramatic." Drama in the dialogues is imitative and falls short of
the re~tt?' apprehe~ible by dialogue. To treat the dialogues as primaril
dramat1~ m a Platomc sense is to prefer the imitation to the reality. Y
. Toe mt~mal content of ~e ~alogues is thus at odds with the skeptical
mterpre~t1on ?ased on th7dialo~tc fonn. Toe interna] content requires a view
of ~ruth ~n ~hich assertonc, active, and exhibitive judgments are all united in
thetr denv~t1on fro1? ~ co_mmon source, and in which the primary distinctions
are not ~on~n~ ~stmcttons among human individuals or human disciplines,
but v<:rttcal dtsti?cttons.refle~ting degrees of proximity to this common source.
In bnef, wb~t 1s re~mr d _is. a conception of divine truth that provides a
7
standard agamst which mdiv1dual opinions and human disciplines can be
measured. "God ought to be to us the measure of ali things and not man as
men commonly say." (½ws iv. _716d.) We are not comfortable today with ~lk
about God. We tend to 1g~ore tt and think it irrelevant to an author's merits.
I~ may be'. ho:"ever, that 1f we take seriously the role of the divine in the
º:
dialogues 1t_ w~ll lead us to a conception of the divine that can be for us a
source . 1ns1g~t r~ther than embarrassment. Let us turn now to the
dogmatrst s contnbut10n to such a conception.
. ~~ the dogmatist, it is possible to disengage Platonic doctrines from the
mdividual speal«:rs who express them. Socrates frequently attributes his
state~ents or act1ons to an authority wiser than himself. He attributes his
vocatt~n to_ th~ command of the Delphic oracle, his speecbes in the Phaedrus
to th~ 1~p1ratton of _the local divinities or the muses, his knowledge of Iove
to D1ot1ma, and gmdance on moral questions to his divine sign. In the
arguments themselves, whicb constitute the greater part of the dialogues it ·
the argument that controls the interlocutors rather than they who contr~l ~!
~ment. And when one cons~ders the unity of the whole set of dialogues the
mflue~ce of a power transcending the individual speakers is still more evident
The ~1ews expressed by S?crates, Timaeus, the Eleatic Stranger, and th~
Atheman Stranger cohere w1th one another in a way that suggests a common
source.
194 Toe Third Way
Toe dialogues thus disclose multiple ways in which what transcends us, and
which in its ideal perfection is customarily called divine, ingresses into human
affairs. Toe most important way in which divinity ingresses into human affairs
is through argument itself as an illumination of the mind by the divine, as we
learn in the Republic (vi. 508d). When Socrates suggests that the Eleatic
Stranger is a god in disguise come to observe and expose the weakness in
argument of Socrates and bis associates, Theodorus replies that the man is no
god, but certainly divine, for so he would call all pbilosophers, and Socrates
emphatically approves (216a-c). Pbilosophers are divine; right opinion in the
Meno is a gift of the gods to meo (100a); Ion is inspired by Homer who was
inspired by the Muse (534c); the Delpbic Oracle as interpreted by Socrates
becomes a command to pursue a life of inquiry (23b); the Laws of Athens in
tbe Crito speak with the voice of a god (54e), the laws of Crete and
Lacedaemon are said to have been instituted by gods (i. 624a), and the words
of the Laws appear to the Athenian to have been spoken not without divine
inspiration (vii. 811c); love in the Symposium is a hermeneus that interprets
between gods and men (202e), and in the Phaedrus gives rise to a divine
psychagogy; the universal animal of the Tima.eus is the work of a god (30a).
In fact, ali wisdom and ali goods come to us from the gods.
To the interpreter, according to the dogmatist, is left the task of formulating
the doctrines that are authorized by these ingressions of the divine. One need
not accept them as having divine authority, but it is reasonable to suppose that
Plato held them to be true, that they are bis doctrines.
Toe divine content that the dogmatic interpreter wishes to separate from
individual speakers is, however, always relative to, rather than independent of,
the individual speakers. Who but Socrates would hear in the statement of the
Delpbic Oracle that no one is wiser than Socrates a command to irritate bis
fellow-citizens? These appear to be very different things. Who but Socrates
hears the laws of Athens as a living, rational voice? Not Crito, who also is an
Athenian. Love is a messe~er from the gods, but each person gets a different
message, as we see in the Symposium. Ion is inspired by Homer, but it is not
possible to state the inspiration as a doctrine. Socrates' divine sign has no
formulable content. Toe arguments of the dialogues may be divine, but they
depend on the agreement of the individual interlocutors at each step. It might
be thought that the inspired speeches of the Phaedrus are an exception to the
impossibility of formulating the divine in words that are the same for all. But
it is here that we are warned that written words are not the same fur ali, that
written speeches are like orphans separated from their parents and unable to
protect themselves from abuse. No doctrine, however divine its origin, can as
doctrine be divine, for as doctrine it is separated from its divine source. Toe
endeavor to formulate truth as a doctrine reflects the dogmatist's view of truth,
not Plato's.
Dogma, Skepticism, and Dialogue 195
Both the skeptic and the dogmatist separate the human soul from the divine
truth. Toe one sees i~di~i~ual opinions separated from their truth, the other the
truth apart from the 1Ddi_v1duals who possess it. Toe third way must here unite
tbe transcendent truth w1th the individuality of the human mind. "Por only the
soul that has beheld truth may enter into this our human form" (Phaedrus
249b, trans. R. Hackfurth), and written words can do no more than remind a
person of what he already knows (275c-d). To be human is to share in the
divine tru~, ~d this ~~ cannot be furmulated apart from the human minds
wh? share ID 1t. Our op1mons and beliets are our individual apprehensions of
an ideal or perfect truth that they do not attain but that they manifest more or
less ad~~uately. They are n~t ~imply true or false, but neither are they merely
our op1mons: they are the d1v1De truth expressi~ itself through the individual
soul.
2. Reality: Particular or Universal?
Toe di~ogues are presented as concrete existential occurrences. Our next
pr?ble~ 1s whether the reality they present is the reality of these concrete
existent1al occurrences or of urichanging forros.
Por the skeptic, . the ~alogues present unique occurrences, and such
occurre~ces have umversal _1mport only in the sense that they can be looked to
as P~digms ?r examples 10 future unique situations. Toe interlocutors meet
at part1c~l~ times and places and say particular things. This particularity is
charactenstic of works ~f ~rt, ~s ~istioguished from works of knowledge, and
thus ~nfirms the skepttc 10 bis 1Dterpretation of the dialogues as primarily
dramat1c.
But if we_ turn _again to the interna} content of the dialogues, we find that
they affirm_ ID vanous contexts the ~rim~ reality not of changing things but
of unchanging forros, and the skept1cal 1Dterpretation is again at odds with a
central feature of the dialogic content.
Por ~e dogmatist, _the primary reality of the dialogues is the reality of
unch~t~ forms o~ ideas. Toe doctrine of ideas is the centerpiece of the
dogmatic 1Dte~retat10~ of Plato, fo~ it seems to warrant the attempt to
formul~te Plat~ s doctnnes on ali subJects. Since ideas exist independently of
ali part1c~ar times and places, and since any subject must be understood in
te~s of ~deas, P!ato'~ doctrines can be formulated by abstracting ideas from
their particular dialog1c contexts.
Toe dialogues themselves, however, proceed in the opposite direction from
the abs_tract to the concrete. Compare the accounts of the unjust man 'in the
Republlc as we move from Book I to Book IV to Book IX. Toe well-known
sequen~e of analogies that proceeds from the sun to the divided line to the
cave mmors the mathematical progression toward solidity by the addition of
196 Toe Third Way
dimensions. As we move up the divided line, we move from the abstract to the
concrete, for the reality is more concrete than the image.
Toe same progression toward the concrete is found if we examine what is
said about a particular subject in the different dialogues. Dialectic, for
example, is never treated abstractly or independently of context, as logic is in
many philosophies. Toe investigation of method is a proper part of any
inquiry, and the dialogues frequently investigate their ?Wn ~ethods, bu~ there
is no separate science of dialectic. Toe accounts of d1alect1c often fit 1t both
to the immediate context and to whatever is being done in the dialogue as a
whole. In the Republic, for example, the accounts of dialectic emphasize that
it is a method of grasping what each thing itself is (vii. 532a, 533b, and 534b).
In the immediate context, this refers to the good itself, but the dialogue as a
whole depends on the use of dialectic to grasp the ousia of the just and the
unjust person. In the Phaedrus, dialectic is said to bring a dispersed_ P_lurality
under a single forro and to divide it at the joint as a carver divides an
organism (265d-266b). In the immediate context, this refers to madness, but
the dialogue as a whole is structured as a repeated division of rhetoric, defined
as verbal psychagogy, into its rhetorical and dialectical sides. In the first half
of the dialogue, the rhetorical rhetoric of Lysias is contrasted rhetorically, by
example, with the dialectical rhetoric of Socrates, used either rhetorically for
deception or dialectically to present a likeness of the truth. In the second half,
the dialectical discussion of rhetoric distinguishes rhetorical from dialectical
rhetoric with respect to their nature, acquisition, and use.
Toe accounts of dialectic often contrast it with controversy (antilogia) or
eristic, but the nature of the contrast is again a function of the dialogic context.
In the Phaedo tbe controversialists (antilogikoi) mix up principies and
consequences (lOle). By avoiding this error, the refutations of the Phaedo end
up strengthening the position they at first seem to destroy, for with the
refutation of the refutation, the original position emerges stronger than ever.
In the Philebus, it is the recognition of interroediates between the limit and the
unlimited that makes ali the difference between dialectic and eristic (17a), and
it is on this recognition that the resolution of the problem of the dialogue
depends.
If we want to know what dialectic is, then, we are directed to the dialogues
themselves in their particularity. Dialectic conforros itself to the realities it is
investigating, and its forros are the forms of the real. These forros cannot be
known in abstraction from the real. A full knowledge of the dialectic would
be a full knowledge of the real in its concreteness.
Both the skeptic and the dogmatist separate the universal from the particular.
Toe one sees concrete particulars without universality, the other abstract
universals without particularity. Toe third way must here unite particularity
and universality. Toe ideas are not universals inhering in many individuals, but
are themselves ideal individuals of which many imperfect images are possible.
Dogma, Skepticism, and Dialogue 197
As ideal. cases l?ey in~lude ali that is to be found in less perfect cases. They
thus achieve umversal1ty not through abstraction but through concreteness. It
is therefore precisely in their concrete existential perfection that the dialogues
image the ideas. They are particulars which, because of the unity and richness
of their content, provide paradigms under which other particulars are
subsumed. They _are not simply cases to which other cases may be compared,
nor do they prov1de abstract forroulae applicable to all cases of a given kind.
Th~ represent ~e fii:,ion of the particular and the ideal, and thus, although
~art1c~ar, have m the1r perfe~tion the character of universals. Toe path to the
ideas 1s foun_d n?t by abstracttng from the concrete richness of the dialogues,
but by pursu1ng 1t. Everywhere in the dialogues we have ideal particulars not
particulars without universality nor universals without particularity. '
3. Dialectic: Rejutati,ve or Affinnative?
Toe dialogues_ ar~ dial?gical arguments, and the general name they give to
such arguments 1s dialect1c. Our next problem is whether dialectic refutes ali
claims to knowledge or itself makes positive affirmations.
For the skeptic, dialectic does not make positive affirroations but refutes or
ridicules the pretensions to knowledge of the sophists and others. As Cicero
says, in Plato's books arguments are allowed on both sides of a question and
nothing positive is affirmed.
If ª. di~ogue is not the refutation of one of its interlocutors, it can be treated
as an 1romc attack on an implied opponent. Toe Republic, for example can be
treated as making fun of absolute rule, and the Statesman can be treat;d as an
attack u~on the sophists and on one-man rule. If a writing makes positive
affirmattons that do not admit of an ironic reading, the Laws or the Seventh
Letter, for example, it is always possible to conclude that it is not an authentic
work of Plato's.
As we have already noted, the dialogues distinguish in various contexts
between dialectic (dialektike) and discussion (dialegesthai, dialektos) on the
one ~a~d~ and controversy (antilogia), eristic (eristike), and agonistic
(agóm~ttke) on the other. Toe two are not always easy to distinguish, and the
Socrat1c .d1alo~es are seen by sorne skeptical interpreters as agonistic, as
~ontes_ts m which Socrates is concerned to get the better of his opponents and
m which the refutations of Socrates injure rather than benefit those who are
refute~. It is particularly difficult to distinguish dialectical from antilogistic
refutatton when the refutation is a purification from the false conceit of
knowledge (Sophist 230b-231b). Socrates' refutations of the claims to
knowl_edg~ of his fellow citizens as described in the Apology are of this sort.
1?e h1~to?cal _Socrates was an agonist, and the approximation of agonistic and
d1alect1c m this type of refutation makes it possible for the Platonic Socrates
of the Apology to approximate the historical Socrates.
198 Toe Third Way
There is, however, a great difference between the skeptical dialectic, or
agonistic, and the Platonic dialectic. Toe skeptical dialectic sets arguments in
opposition to each other without resolution and the result is negative, a
suspension of judgment. Toe Platonic dialectic seeks agreement between the
interlocutors at each step of the argument, and the result is positive, even if
it is only the knowledge that one does not know. In the Apology Socrates
cannot question the jurors individually, but he argues from their own traditions
and values, and it is this wbich renders bis defense, or the bite of the gadfly,
effective, as the outcries it provokes show that it is. Agonistic would argue
against the beliefs of the jurors, not in accord with them, and would not have
the same effect.
Toe dialectician is thus likely to fall unawares into controversy when he
attempts to refute a theory in the absence of its author. Socrates says in the
Theaetetus, "It looks as if we were content to have reached an agreement
resting on mere verbal consistency, and to have got the better of the theory by
the methods of the professional controversialist (agónistikos) . We profess to
be seeking wisdom (philosophoi einai), not competing for victory (ou
agónistai), but we are unconsciously behaving just like one of th~se
redoubtable disputants." (164c-d, trans. F. M. Cornford.) Protagoras, speaking
through Socrates, proceeds to accuse Socrates of agonistic rather than
dialectical refutation because he is not refuting what Protagoras bimself means
and would say (166b, 167e-168b).
Similarly, Socrates says to Meno that in replying to a questioner of the
clever, eristic, and agonistic kind who did not understand bis answer, he
would reply "You have my answer. If I am wrong, it is for you to take up the
argument and refute it." But a more dialectical reply would use terms the
questioner concedes he knows (75c-d).
Toe refutations of dialectic, as distinguished from those of skepticism, lead
to positive results. This is strikingly expressed in a well-known passage from
the Seventh Letter. Plato distinguishes the ti.ve classes with wbich instruction
is concemed: "For everything that exists there are three classes of objects
through wbich knowledge about it must come; the knowledge itself is a fourth,
and we must put as a fifth entity the actual object of knowledge wbich is the
true reality. We have then, first, a name (onoma), second a definition (logos),
third, an image (eidólon), and fourth, knowledge (epistemé) ."1 He describes
the process of instruction as follows: "Hardly after practicing detailed
7. Seventh Letter 342a-b. trans. L. A. Post. Comparable language is used in the
dialogues. Toe dialectician of the Republic lacks nous so far as he is unable to give the
lagos of the ousia of each thing (vil. 534b), the Eleatic Stranger and Theaetetus agree
on the onoma and seek a lagos of the thing (to ergon, to pragma) (Sophist 218c, 221b),
and the Laws refers to the onoma, lagos, and ousia of each thing (x. 895d).
Dogma, Skepticism, and Dialogue 199
comparisons of names and definitions and visual and other sense perceptions,
after scrutinizing them in benevolent disputation (en eumenesin elenchois
elenchomena) by the use of question and answer without jealousy, at last in a
flash understanding of each blazes up, and the mind, as it exerts all its power
to the limit of human capacity, is flooded with light" (344b, trans. L. A.
Post).
Even in cases where the dialectic seems inconclusive, a closer look will
show that there are positive results. See particularly Francisco Gonzalez's
readings of the Charmides and Laches in the present volume, bis paper on the
Lysis, Sternfeld and Zyskind's readings of the Meno and Rlrmenides, and
Zyskind's reading of Book I of the Republic.8
Toe skeptical effort to negate the positive results of dialectic reflects the
method of the skeptics, not Plato's dialectic. To treat the Republic and the
Statesman as primarily ironic in intent negates not only their manifest content
but also most of their excellence. To athetize the Laws not only contradicts the
testimony of Aristotle and takes from Plato the authorship of one of bis
greatest works, but also takes from the Platonic corpus one of its essential
components, as I shall argue shortly.
For the dogmatist, dialectic is a method of establisbing positive as well as
negative results, and it is not the positive statements in the dialogues that are
ironic, but the skeptical disclaimers of knowledge. Toe interpreter's task is to
examine the arguments of the dialogues to see if they justify the conclusions
that are reached. This requires the canons of formal logic, and the resources
of modero symbolic logic may be particularly helpful. Toe question and
answer form of the dialogues does not have the importance it has for the
skeptic, and in fact is inessential, since the argument can be expressed as well,
if not better, by a single speaker as by two in conversation. It may be thought
that the interlocutors are sometimes simply yes-men who contribute nothing to
the argument anyway. For the dogmatist, therefore, it is appropriate to
reformulate the logical structure of the dialogues in order to make clear what
is assumed and what really has or has not been proved.
Dialectic is not the same as agonistic, but neither is it the same as logic. Toe
proofs of logic depend on terms having fixed meanings. If the terms of an
argument sbift their meanings, the argument becomes invalid. Dialectic, on the
8. Francisco J. Gonzalez, "Plato's Lysis: An Enactment of Philosophical Kinship, "
Ancient Philosophy 15 (1995): 69-90; Robert Sternfeld and Harold Zyskind, Plato 's
Meno: A Philosophy o/ Man as Acquisitive (Carbondale, IL: Southern lllinois University
Press, 1978); Robert Sternfeld and Harold Zyskind, Meaning, Relation, and Existence
in Plato 's Parmenides: The Logic o/Relational Realism (New York: Peter Lang, 1987);
Harold Zyskind, "Plato's Republic Book I: An Equitable Rhetoric," Philosop hy and
Rhetoric 25 (1992): 205-21.
200 Toe Third Way
other hand, depends on shifts in the meanings of terms that occur as we gain
insight. It is when accepted meanings and beliefs become unsettled that new
insights emerge. Significant advances in knowledge are associated with new
systems of meaning. Formal structures with fixed meanings have their place
within dialectic, but limit its power if used to validate it.
Toe validity of dialectical arguments depends not on fixed forms of
inference but on the agreement of the interlocutors. Toe arguments of the
dialogues therefore require at least two voices, and each step of the argument
must be shared. If it seems to the reader that an interlocutor is agreeing when
he should be objecting, this confirms the need to take the mind of the other
into account. If the reader were the interlocutor, a different dialogue would
result. Logical arguments are separable from the minds that construct them,
but dialectical arguments are not.
Dialectical inquiry therefore cannot be reduced to written words.
Nevertheless, dialectical inquiry takes place by means of words, and these
words can be written down. lt is in this way that the dialogues represent
dialectic.
Toe attempt to reformulate dialogic arguments as logical arguments
independent of minds reflects the dogmatist's method, not Plato 's. Both the
skeptic and the dogmatist separate refutation from affirmation. Toe one sees
arguments opposed to one another and no positive result, while the other sees
arguments whose validity depends on the affirmation rather than the refutation
of their starting points. Toe third way must here unite refutation with
affirmation. Positive results are at the same time the refutation of the beliefs
from which the argument began. This point is stressed in the Hegelian
dialectic, with its labor of the negative. Toe truth is the whole, and partial
accounts are at once improved and destroyed as they are brought into relation
to larger wholes. Anything that is fixed and not subject to change limits the
scope and effectiveness of dialectic. Toe absence of fixities resembles
skepticism, and the shared construction resembles dogmatism.
Nothing is established absolutely by way of argument. Our arguments and
the positions that they establish always depend on what is not established by
argument. Ali the dialogues are partial, and therefore may seem to conflict
with one another. This brings us to our final problem.
4. The Dialogues: lndependent or Unified?
Toe dialogues take place for the most part independently of one another.
Our final problem is whether they are independent or unified as parts of a
larger whole.
For the skeptic, the dialogues are independent in the sense that each has its
own starting points and pursues its own course in independence of the others.
Toe dramatic circumstances naturally impose sorne limits on what a particular
Dogma, Skepticism, and Dialogue 201
cbaracter can plausibly say or do. Toe Theaetetus, Sophist, and Statesma,n are
connected in their characters and subjects, as are the Republic and the
projected series beginning with the Tima,eus. Toe Apology, Crito, and Phaedo
present a connected historical sequence. Since ali of the dialogues except the
Laws are connected by the presence of Socrates, one can conceive of a grand
bistorical sequence ordered by the life of Socrates, beginning with the
R;zrmenides and ending with the Phaedo. Such modes of connection, however,
do not significantly limit the general freedom of the approach. Any subject
may be examined in the light of any principies.
Toe independence of the individual dialogues makes possible the high degree
of interna! coherence and unity that each possesses, and that has often been
admired and explicated. Toe character of the interlocutors corresponds to their
tbought, and the drama of the dialogues supplements the argument. In the
Meno it is argued that there are no teachers of virtue, and yet we see virtue
being taught, and the participants in the Lysis become friends in discovering
tbat they do not know what friendship is. 9 Every detail in the dialogues, no
matter how trivial or accidental it may seem, may be significant in relation to
the idea that determines the unity of the whole. Gerald Press shows that the
Homeric dream in the Crito leads to a far-reaching comparison of Achilles the
hero of passion with Socrates the hero of reason, and Lee Miller's account of
the four Homeric references in the Protagoros adds a new dimension to the
dialogue and to the use of Homer in education. 10 Even the particular words
and language used in the dialogues may be significant. Proper names are a
well-known example: Meletus who does not care, Polus the colt,
Thrasymachus bold in battle, Ion the traveler, Cephalus the head, Phaedrus the
radiant one, and so on. Toe basic ideas of the dialogue are often present in
commonplace form in the opening sentence. Toe opening sentence of the
Symposium, for example, "Oh, if that's what you want to know, it isn't long
since I had occasion to refresh my memory" (172a, trans. M. Joyce), already
has the basic threefold structure of the unknown marvelous, the hermeneus
who is able to communicate it, and the recipients. "Where do you come from,
Phaedrus my friend , and where are you going?" (227a, trans. R. Hackforth)
can apply not only to the immediate situation but to the larger journeys of the
soul. Or one can bring in the end as well as the beginning: Gerald Press has
9. Robert S. Brumbaugh, "The Meno in Secondary Schools," Teaching Philosophy
Vol. I, No. 2 (Fall 1975): 107-15, and Sterofeld and Zyskind, Plato 's Meno.
10. Gerald Press, "A Dramatic and Non-Doctrinal Interpretation of Plato's Crito" and
Clyde Lee Miller, "'If Two Go Together'-Homeric Allusion as Dramatic and Thematic
Aids in Plato's Protagoras," papers given at the New Jersey Regional Philosophical
Association Meeting, Douglass College, Rutgers University, 23 November 1991.
202 Toe Third Way
pointed out that the Apology begins with "I do not know" and ends with
"God."
For the skeptic, then, Plato's use of the dialogue form permits him to
explore diverse philosophic possibilities unconstrained by doctrinal
commitments, and the independence of the individual dialogues makes possible
their marvelous intemal integration. Taken together, the dialogues are a
potentially infinite set, each member of which is essentially independent of all
the others, a grand rhapsody.
For the dogmatist, the dialogues are parts of a larger system. They are
connected not only by dramatic circumstances but also by intellectual content.
What is said about knowledge or rhetoric or love or virtue in one dialogue is
complementary to what is said in other dialogues. By investigating the
interrelations of intellectual content, we can discem the character of the system
within which all the dialogues have a place. Inconsistencies can be dealt with,
at least in part, by viewing the dialogues chronologically, for Plato's thought
developed over time, and we need not expect earlier dialogues to agree
entirely with later ones.
Toe basic tenets of the Platonic system, for the dogmatist, can be discovered
in the propositions that are defended in various dialogues and are therefore
independent of any particular dialogue: Ideas exist; perceived things image
intelligible things; the good is the principie of all things; virtue is knowledge;
no one does injustice voluntarily; it is better to suffer than to do injustice; the
soul is immortal and knows all things; knowledge is recollection; philosophy
exists in living minds and cannot be written down; the soul has three parts,
reason, spirit, and desire; the perfect state follows the maxim that friends have
ali things in common; the rule of knowledge is best and of law second-best;
and so on. Such tenets govem all the dialogues. Individual dialogues conform
to them, and in this sense are not autonomous.
When we try to express Plato's philosophy by propositions that are true in
all or many dialogic contexts, however, the philosophy loses most of its
vitality and significance. Toe propositions are only attenuated remnants of the
philosophy, echoes or shadows of it, and do not express the philosophy itself.
Toe same problem arises with respect to papers such as the present one that
attempt to state general features of Plato's philosophy. No matter how accurate
their account, they cannot compete with the dialogues as presentations of
Plato 's philosophy.
Inconsistencies among the dialogues are for the most part not of the sort that
can be explained genetically. Toe unified science of the Republic, for example,
is not consistent with the separate sciences of the Statesman. Are we to accept
one view as later and reject the other as earlier? If we follow this path, there
will not be much of a philosophy left. We have noted that what is said on aoy
subject is relative to the dialogue in which it is expressed, and it is surely in
tbis way that most of the differences in the treatment of common subjects are
Dogma, Skepticism, and Dialogue 203
to be explained. lt is not by denying the relativity of dialogic statements to
their contexts that the unity of the dialogues can be found, but by affirming
and utilizing it. Individual dialogues differ in their approaches, and the
differences of approach are not functions of the accidents of the development
of the philosophy, but of the philosophy itself. To look for the reasons for
these differences in the chronology is to look for them in the wrong place, in
a place where they are not to be found. Toe attempt to find the unity of Plato's
philosophy in a set of propositions common to different dialogues, and to
account for differences among the dialogues by their chronological order,
reflects the dogmatist's conception of what philosophy is, not Plato's.
Both the skeptic and the dogmatist separate the independence of the
dialogues from their systematic unity. Toe one sees them as independent and
therefore not a system, the other as presenting a single doctrine and therefore
as not independent. We need not suppose either that the autonomy of each
dialogue precludes their unity as parts of a system or that their unity must take
the form of a common doctrine that precludes their independence. Toe third
way must here unite independence and systematic unity. It is through their
independence that they become parts of a systematic whole. Toe statements in
different dialogues cannot be compared directly because each depends on its
independent dialogic approach. But they can be compared indirectly as
different views of the same subjects resulting from the different approaches.
And the principies that determine these different approaches can be examined
in their systematic relations to each other. It is through the independence of
these principies that they are qualified to constitute parts of a systematic
whole.
We have here a situation that is analogous to the pluralism of philosophies
found in all philosophic traditions. Each philosophy, like each dialogue, has
its own perspective on their common subjects. Each is independently
architectonic of a whole that is the same for ali, so that their unity depends not
on a common doctrine, but on the relations of their autonomous approaches to
each other. This relation I have elsewhere called one of reciproca! priority:
each is prior to the others and includes them within its own architectonic. 11
Toe relation of the dialogues to each other thus provides an analogue of the
relation of philosophies to each other within a Platonic approach. Toe problem
of the seeming contradictions among the different dialogues and of the sense
in which they are parts of a single system is analogous to the problem of the
seeming contradictions among different philosophies and the sense in which
they are parts of a single system.
11. Reciproca! pnonty is used to account for philosophic differences in my
Architectonics of Meaning, 2d ed. (Chicago: University of Chicago Press, 1993).
204 Toe Third Way
Each dialogue and each pbilosophy possess a bigh degree of interna! unity
that results from the idea that determines ali its parts. If the whole set of
dialogues, or of pbilosopbies, were to have this kind of unity, they would ali
need to be determined by a single idea such as the idea of the good. But this
kind of unity, as Socrates indicates in the Phaedo, is beyond our grasp.
Nevertheless, the same kind of principie governs both the interna! unity of the
individual dialogues and their externa! relations to each other, although in the
one case it leads to a unified whole and in the other to the openness and
freedom to pursue any approach whatever. Toe best we can do in constructing
our dialogues and pbilosopbies is to leave ourselves complete freedom and
independence in the approach we will follow, and to work out the
consequences of any approach with maximum thoroughness. Each dialogue or
pbilosophy acquires a place in the system of dialogues or pbilosopbies through
the independence of its own inquiry. Their unity as a system depends on their
independence as parts.
Summary
We have sketched skeptical, dogmatic, and dialogical interpretations of the
Platonic dialogues.
For the skeptic, the statements in the dialogues are the opinions of the
individuals who make them, and we are not justified in taking any of them as
Plato 's doctrines, the reality of the dialogues is the reality of concrete
existential occurrences, the arguments of the dialogues refute tbose who claim
to know, and each dialogue is independent in the sense of having its own
starting points.
For the dogmatist, Plato has a doctrine that is independent of the individuals
who express it and that for Plato has a divine source, this doctrine asserts the
reality of the forms or ideas, it is established by the arguments of the dialogues
so far as they are valid, and it is common to the different dialogues.
For the dialogic interpreter, the dialogues unite the contraries by wbich
skeptical and dogmatic interpretations are opposed to one another. ( 1) If the
truth of individual minds and points of view is set in opposition to the divine
truth, then the truth of the dialogues is missed, for each individual shares in
bis own way in a divine truth that is possessed by us only as individuals. (2)
If the reality of the particular is set in opposition to the reality of the Forms,
then the reality of the dialogues is missed, for it is the reality of the ideal case
wbich acbieves universality through particularity. (3) lf skeptical arguments
that refute fixed starting points are set in opposition to dogmatic arguments
that depend on fixed starting points, then the arguments of the dialogues are
missed, for dialectical arguments reach positive conclusions through the
refutation of their starting points. (4) If the independence of the different
dialogues is set in opposition to their unity in a larger whole, then their true
Dogma, Skepticism, and Dialogue 205
relations to each other are missed, for it is through their independence that
they form parts of a larger whole. Toe dialogues exhibit a paradoxical unity
of the human and the divine, the particular and the universal, refutation and
affirmation, independence and unity.
JI. Dogma and Skepticism in the Dialogues
I have argued for a dialogic interpretation of Plato as distinguished from a
dogmatic or skeptical one. Toe argument can be completed by investigating
what happens to dogma and skepticism within the dialogues themselves. If we
reject the claims of the dogmatic and skeptical interpretations to be
arcbitectonic for the whole corpus and propose instead a dialogic interpretation
as arcbitectonic, it is still possible to find a place for dogma and skepticism
witbin the dialogic arcbitectonic. Toe dogma and skepticism thus subordinated
to dialogue will be different from the dogma and skepticism conceived as
arcbitectonic, and if these differences are made clear, it should cast a definitive
light on the whole problem of dogma and skepticism in Plato.
For our purposes it will suffice to consider the Apology as skeptical and the
Laws as dogmatic. Let us first compare the skeptical role of Socrates in the
Apology with the role of the pbilosopher-kings in the Republic, and then the
Laws as dogma with the nondogmatic political dialogues, the Statesman and
the Republic.
The Apology exhibits the function of the pbilosopher in the actual state,
whereas the Republic exhibits the function of the philosopher in the ideal state.
~ the ideal state the function of the pbilosopher is to govern the state, whereas
m the actual state, the pbilosopher cannot function within the government (he
would soon perish), but he can provoke the citizens in nonpolitical contexts
and as it were from outside the state, like a gadfly on a horse. He provokes
or bites the citizens by calling attention to the contradictions in their thought.
His function is to irritate the state in its ignorance and complacency and thus
to get it moving in a way that will resolve its contradictions, contradictions
that in the ideal state are in fact resolved. Skepticism is thus instrumental to
knowledge rather than an irremediable or final state. And here we can see
~ore clea~ly the error of the skeptical interpreters of Plato, who go wrong not
m supposmg that there can be a Platonic skepticism or that it can be
arcbitectonic, but in the limitation that this arcbitectonic imposes on our
understanding of the dialogues by taking skepticism as final rather than
instrumental. We can see in particular the pernicious effect of the dramatic
interpretation of Plato, which would transform the unpleasant but beneficia!
sting of the gadfly into an object of dramatic enjoyment, and thus leave the
reader secure in bis own ignorance and complacency.
Toe problem of the Laws is the problem of writing a doctrine. Plato can be
doctrinal, but to approach a subject with a view to writing down knowledge
206 Toe Third Way
of it as a doctrine is for him a distinct kind of approach to any subject. Here
we can see more clearly the error of the dogmatic interpreters of Plato, who
go wrong not in supposi~ that there can be Platonic doctrines or that
doctrines can be architectonic, but in the limitation that this architectonic
imposes on our understanding of the dialogues by taking an inferior form of
philosophy to be the best form.
Toe problem of the Statesman is to discover a science in a world from
which the gods have withdrawn (272e-273a), and this is the problem of
skepticism, not in the sense in which the skeptic argues that we have no
knowledge, as, in the Apology, but in the sense in wbich the skeptic is a
humanist and argues that we are alone in the world and must form our own
sciences and rule ourselves without divine help.
Toe problem of the Republic is the problem of constructing in argument the
ideal state wbich depends for its excellence on the unitary principie of ali
things.
We have here three degrees of approximation to the unitary principie of ali
things. Formally stated: Political knowledge, or any other knowledge, may be
either embodied in writing, as in the Laws, or it may exist in living minds.
And if it exists in living minds, it may be either as a separate branch of
knowledge, as in the Statesman, or as the knowledge of the pbilosopher for
whom political knowledge is inseparable from all other knowledge, as in the
Republic.
These formal differences can be developed by considering for each leve!
what interlocutors are appropriate, the account given of a particular subject
such as art, and the way each architectonically embraces the other two.
Toe questioner in the Republic is Socrates, in the Statesman the Eleatic
Stranger, and in the Laws the Athenian Stranger. Toe three approaches are so
different in kind that it is not dramatically plausible to have them pursued by
a single inquirer. Questioning by the ideal pbilosopher leads appropriately to
the ideal state with its ideal unity of knowledge. Toe method of division
learned from human teachers by the Eleatic Stranger (216a) is suited to
defining separate human sciences, but is not suited to discovering their ideal
unity. Toe Athenian Stranger relies on bistory and experience, wbich are
appropriate grounds for legislation, but history and experience do not disclose
either the ideal state of the Republic or the science of the Statesman.
Toe respondents in the three dialogues have the characteristics that are
required if the dialogue is to acbieve its purpose and at the same time
exemplify what is being discussed. Plato's brothers, Glaucon and Adeimantus,
have something divine in them, for they are so prejudiced and irrational that
they persist in favoring justice in spite of the powerful arguments that they are
able to bring against it (ii. 368a). This is precisely what is required in the
prospective guardians of the perfect state when they begin their study of
dialectic (vii. 537e-539d).
Dogma, Skepticism, and Dialogue 207
Toe Young Socrates of the Statesman, on the other hand, repeatedly asserts
himself in ways that impede the progress of the argument. He begins by
disclaiming responsibility for the argument (258c), he follows too short a
course in bis eager division of herd-nurture into the nurture of men and of
beasts (262a), and then too long a course in asking for two paths of division
where one would suffice (265b). He turns away from the Statesman in the
direction of the Pbilosopher when he asks for clarification of the distinction
between part and class (263a), and he turns back toward the Sophist when he
objects to rule without laws (293e). He twice declares the Statesman to be
satisfactorily defined before he is satisfactorily defined (267a; 277a), and thus
would bring the dialogue to a premature end. In short, Young Socrates lacks
the art of measure, the science of the mean, and therefore the difficulties he
presents to the Eleatic are of the kind that would be encountered by the
Statesman in his efforts to impose the mean on those who do not have bis
science. Toe Statesman works in a universe in wbich everything tends to run
downhill, and it is only by bis efforts in an environment that tends to work
against bim that the mean can be preserved.
Clinias of Crete and Megillus of Sparta are ordinary old men, without
special talents or abilities, but they have been brought up under good laws (i.
625a) and they also have reasons to feel friendly toward the Athenian (ii.
642b-e). They resemble the old men who are to rule the state for wbich the
laws are being proposed, and their capacity to understand the Athenian's
arguments corresponds to the capacity of the rulers of that state to understand
the justification for their own laws. They are therefore in their very
ordinariness ideal interlocutors for the Athenian, for their capacity to
understand him becomes a test of the viability of the laws that he proposes.
That the different approaches of the three political dialogues extend to any
branch of knowledge may be seen from the vexed example of art.
In the Republic, the pbilosophers are poets and construct a mythic or poetic
universe, govemed by God, in wbich the cbild is educated. By acting the parts
of characters in this grand poem, they become like the characters whom they
imitate. The unity of art and life in early education is clear from the evaluation
of imitation in terms of its effect on the person who is doing the imitating, not
on an audience (iii. 394d-398b). Ali this being the case, the poets who are not
pbilosophers must be expelled from the state unless they can give a philosophic
defense of their work (x. 607b-e).
In the Statesman, the imitative arts are separated off as one kind of co-
causal art, worked up for our pleasure only, and falling under the supervision
of the Statesman (288c).
In the Laws, the whole universe is art (x. 892b), and the laws themselves
are very like a poem, a poem constructed not without the aid of divine
inspiration, and one against wbich ali other poems are to be judged (vii.
811c-812a). Strict criteria are laid down for the imitative arts that are to be

208 Toe Third Way
practiced in the state. Toe imitation must be of a good object, that is, it must
show the just man as the happy man, it must be true in the sense of correctly
imitating its object, and it must be beautiful in its execution in words, melody,
rhythm, and gesture, that is, all the means of imitation must be suited to one
another and to their noble subject (ii. 660d-669b).
Thus, the philosopher's knowledge of all things enables him to construct in
art a beautiful universe, the statesman's knowledge of an independent art of
politics enables him to supervise the imitative arts as cocausal arts, and the
lawgiver's knowledge of how to embody his knowledge in a poem enables him
to lay down strict criteria for the evaluation of art, an aesthetic, or poetic
doctrine. Just as we have in politics a politeia, a politikos, and nomoi, so we
have in art the constitution of a poetic universe, the separable art of the poet,
and, finally, laws of poetry, or a poetic doctrine.
We can confirm the architectonic character of the three approaches and the
autonomy of the dialogues in which they are presented by noting the way in
which each approach includes the other two.
Toe problem of the Repuhlic is whether the just man is the happy man, and
the answer to this question requires the genesis in speech of the one true and
right constitution and of the four constitutions that result from it by
degeneration. lt is thus a complete science of constitutional forms. The
epitactic or commanding art of the Statesman, however, is for the salce of
coming-to-be (261a), and from the standpoint of the Repuhlic it is not a
science at ali, since it is detached from philosophy, but belongs among the arts
of becoming (vii. 533b). As for laws, there is no need for written laws if
philosophers rule'; but lawgivers such as Solon have a place among the poets,
concemed with images at the lowest level of the divided line. The three
approaches are thus ordered by the levels of the divided line.
Toe Statesman operates in a world from which the gods have withdrawn,
and with them goes any possibility of philosophic knowledge. Man is here
autonomous, entirely on his own, giving his own commands, without
knowledge of, or help from, the gods. Toe diviner, who transmits the
commands of another, is here subordinate to the Statesman, who gives his own
commands, for it is only by custom or convention that the diviner and priest
are said to communicate with the gods (290c).
With the loss of the unity provided by the divine, the arts and sciences are
fragmented into a plethora of cooperative and competing arts and sciences.
Toe possibility of these arts and sciences depends on the being of becoming
(283d), justas the possibility of the Sophist depends on the being of not-being
(Sophist 241d), and the possibility of the Philosopher depends on the idea of
being (Sophist 254a). Toe being of becoming depends not on the sophistic or
relative mean, but on the nonrelative mean of the Statesman. This mean is a
sort of fragmented form of the unitary good, which it replaces. On the one
hand, the mean is relative to the particular things with respect to which it is
Dogma, Skepticism, and Dialogue 209
a mean, and so gives rise to the plethora of arts of the being of becoming. On
the other hand, as the mean, it is something common to them ali, and thus
permits the analogies between the arts which the Eleatic uses to define the art
of the Statesman.
Among the separate arts and sciences, the Statesman's science is unique in
its capacity to rule over ali the other arts and sciences. These include the many
cocausal arts, the ancillary arts of the slave, wage-eamer, and civil servant,
the ministerial arts of diviner and priest, the rival arts of the motley tribe of
sophists, and the kindred arts of the rhetorician, the general, and the judge
(287b-395e). Toe Statesman here takes the place of the gods in organizing the
entire world of becoming so far as it comes within the scope of reason, and
his regime is said to differ from those of his rivals as a god from human
beings (303b). Toe Statesman among the Sophists is like a god among men.
In the Statesman, the unified science of the kind found in the true and right
constitution of the Repuhlic appears as the knowledge of a divine shepherd
belonging to another epoch. Toe contrast is not between being and becoming,
but between two fonns of becoming, one controlled by the gods, the other by
the Statesman when the gods have withdrawn from the world. And when the
Statesman must in turn retire· from rule, he writes laws to guide successors
who lack his science, and these are the Sophists. Thus, our three approaches
are here ordered by the distinction between Philosopher, Statesman, and
Sophist as seen from the perspective of the Statesman.
In the Laws, we are no longer in a universe from which the gods have
withdrawn, but in a universe produced and managed by the gods. Unlike the
universe of the Repuhlic, however, it is a universe in which mind is expressed
in its products. The universe itself is the product of the divine mind, and man
is constructed to be a puppet of the gods, which is the best thing about him
(vii. 803c-804b). Although we live in the midst of products of mind mind
itself, true and free and in harmony with nature, the mind that rules the true
and right constitution, is nowhere to be found, or at least only to a small
extent (ix. 875d). Consequently, it is necessary to have recourse to laws as a
second-best.
The Repuhlic and the Statesman as seen from this point of view contain
doctrines. Toe principal doctrine of the Repuhlic, that the just man is the
happy man, becomes the basis for moral education in the Laws. Toe political
doctrines of the community of property and of wives and children, are,
however, unsuited to a state in which law replaces mind, and private property
and separate households are established in their stead (v. 739a-740a; vii. 807b;
ix. 875d). A doctrine of the mean is derived from the history of the
Peloponnesian confederation following the Trojan War, a history that exhibits
the danger of unlimited power (iii. 683c-693c). Similarly, the histories of
P~rsia and Athens show the need for a mean, for the tempering of monarchy
w1th democracy and democracy with monarchy (iii. 693d-701e).
210 Toe Third Way
Thus, in the Republic, all three approaches are ordered by their
approximation to the ideal, in the Statesman by the distinction between
Sophist, Statesman, and Philosopher, and in the Laws by their doctrines. Each
dialogue in its own way encompasses the others.
Our general conclusion, then, is that skepticism and dogmatism both have
a function within the dialogical approach and indeed can take over the whole
of the dialogues, as we have seen happen in the history of the interpretation
of Plato. Skepticism with respect to all knowledge serves as a stimulus to
knowledge, and skepticism with respect to the gods leads to human sciences
and a world governed by the Statesman. Written doctrines are the best we can
do when the knowledge of the living mind is no longer available. Toe best
writings are those that represent in dialogues the verbal aspect of dialectic, and
therefore Plato has left us written dialogues, skeptical and dogmatic as well as
dialogical, toread when bis living mind would no longer be available to us. 12
12. I am indebted to Richard McKeon for bis general approach to Plato, and to Harold
Zyskind for his readings of particular dialogues. Their work on Plato is not as well
known as I think it deserves to be.
12
The Dialogical Composition
of Plato's Parmenides
Victorino Tejera
One assumption that sorne nondialogical readers of Plato make is that, be-
cause the conversations in them are not between or among "equals," therefore
the dialogues are "really monological." But no conversation is or ever was
between interlocutors who are equal in all respects; an exchange between
equals in all respects would be reflexive communication, since self and other
would be indistinguishable. And as reflexive communication, it would still be
dialogical. 1 And were most conversations, in fact, between equals in ali
1. Logicist and formalist approaches tO: language will remain incomplete as long as
they overlook the showing, by Bakhtin and others, that discourse is basically dialogical.
Of course, C.S. Peirce, M. M. Bakhtin, G.H. Mead, Martin Buber, and Justus Buchler
are not claiming that "language," in the formalist's theoretical sense of a grammarian's
construct, is dialogical. Toe system that, following Saussure, is now called "language"
cannot be either dialogic or monologic, for it addresses no one. lt is we who construct
and address ourselves to it, as linguists, and who theoretically seem to presuppose itas
speakers or writers. See, respectively, Collected Papers, ed. Charles Hartshorne and Paul
Weiss (Cambridge: Harvard University Press, 1938-51); Problems of Dostoevsky's
Poetics , ed. and trans. Caryl Emerson (Minneapolis: University of Minnesota Press,
1984) and The Dialogic /magination, ed. Michael Holquist, trans. Caryl Emerson and
Michael Holquist (Austin: University of Texas Press, 1981); Mind, Seif and Society
(Chicago: University of Chicago Press, 1934); Das Dialogische Prinzip (Heidelberg:
212 Toe Third Way
respects there would be much less talking in the world. When conversations
are between equals in sorne respects, the fact that one speaker may emerge as
dominant does not turn the dialogue into a monologue. Nor, despite our
dominance-oriented culture, do we have to assume that sorne speaker must, or
wants to, be dominant. When a dominance-oriented speaker does appear in a
dialogue of Plato's, it seems entirely appropriate that Plato's Socrates should
tame him or match him, as he does respectively with Kallicles in the Gorgias
or Protagoras in the Protagoras. Nor are the Sophist and Politicus any the less
dialogical because Plato lets the sophistic rhetorician from Elea advance the
best cases he can for rhetorical sophistry and one-man rule, and confute or
confirm himself in his own words. For in these dialogues, the reader is
supposed to see and hear him with the eyes and ears of Socrates, the skeptical
gadfly, the conscience of his democratic city. When, however, Plato's Socrates
is catechical, as with Protarchus in the Philebus, he is still not necessarily
being dominant, but rather kindly and pedagogical. Kindly and pedagogically,
indeed, is how, in turn, (and as we shall see) Plato's Parmenides behaves
towards young Socrates in our eponymous dialogue.
J. Young Socrates' Supererogatory Appeal to a Theory about Ideas
Once we take the Itmnenides as a whole,2 it appears to divide quite easily
into four interlocutory episodes of varying extension: Socrates' invocation of
the theory of ideas (127dl30a), Parmenides' rehearsal of difficulties in the
theory (130a7-135c3), Parmenides' and Zeno's discussion of the antilogistic
training method for knowledge-seekers (135c-137b5), and Parmenides'
Schneider, 1962); and Toward a General Theory of Human Judgment, 2d ed. (New
York: Dover, 1975) and Nature and Judgment (New York: Columbia University Press,
1959). As C.S.Peirce said (4.551), "it is not merely a fact of human psychology, but a
necessity of logic, that every logical evolutíon of thought should be dialogic." We must
remember that logic, as semeiotíc, includes for Peirce speculative grammar and
speculative rhetoric (or methodeutic), as well as "critic" or logic in the modero formalist
sense- and is itself subsumable under ethics and aesthetics.
2. Something that doctrinal readers-whether deconstructionist or logicist-are often
reluctant to do. Deconstructionists too often address what they call "the text" instead of
the work, as if this were a license to treat only parts of the work; and this must be
because in the case of works of art "the work" can only mean "the work as a whole."
It is only interpretations of works of art that are deconstructible, not the works
themselves. Por, to experience or understand the work at all, the perceiver cannot avoid
reconstructing it in the light of the cues and clues provided by it. So all experience of
works of art is interpretive and no interpretation (including our own) may be asserted
dogmatically because all interpretations are deconstructible. But this also means that
deconstruction does not give access to the work of art. Access to the work is achieved
through the experience or reconstruction of it.
Toe Dialogism of Plato's Parmenides 213
"demonstration" or version of the antilogistic method (137b-166c). In the first
inteHectual episode or exchange, young Socrates rehearses "the first thesis of
the first discourse" of Zeno as follows (129e2-5): "if what there is (;a óPTa)
is plural (1roAAa), it/ they must (óei) be both like and unlike, but this is
impossible; for neither can the unlike be like nor the like unlike." Zeno agrees
that this is what he means, and that "the purpose of his arguments" (o
{)oúAovm, aou oi Aá-yo,) is to maintain, against aH arguments, that "they/it
are not many" (wc; oú 1roAA& fon).
We notice, however, that the first compound proposal is true only if we are
speaking of what is absolutely "like" and what is absolutely "unlike." Noting
that neuter plural subjects take the verb in the singular, we may, also, not
allow the ambiguity in the subject of esti to go unexplained.
Readers of the metaphysical poem of the original Parmenides will be aware,
as Plato himself must have been aware, that it is just the subject of esti, and
how to characterize "it," that its discourses are about. To make a long story
short, we can say that the Goddess's. business in the poem is to distinguish
between the two references this unstated subject can have: namely,
what-there-is in the sense of $e AH, and what-there-is that we unguardedly
talk about when discussirg the cosmographic universe as a whole. A point she
wants her young auditor to take is that to talk about the AH at aH, we must
speak with strict consistency. Talk about cosmographic nature, on the other
hand, will inevitably be "of two minds" (óí.Kpavo,) , "backward-turning"
(1raAÍPTpo1ro<;), and incomplete; for nature, as the process that it is, is ever
completing itself and had-as the sixth century believed-no beginning.
Young Socrates does notice that Zeno has appealed to "likeness" and
"unlikeness" unqualifiedly, i. e., in the absolute sense; but he does not dissolve
Zeno's paradox by pointing out the equivocation upon which it hinges. Instead,
he capitalizes on the fact that this implies that Zeno is granting that there is
such a thing as "an idea of likeness in itself' (aú;o Ka0' aú;o e!boc; n
oµo,órr,mc;), and such a thing as an abstract idea of unlikeness (128e9-129al).
And don't you think, he adds, that both you and I and the things which we say
are many partake (µera°}...aµ{)&ve,v) of these two ideas? Here we find young
Socrates doing what he is going to do in many another dialogue: he puts words
in the mouth of his interlocutor that lead the conversation into the subjects that
he, Socrates, wants to discuss. Toe subject here is the theory of ideas. This,
of course, implies that the author who is managing the conversational
exchanges also wants that theory to be the subject of discussion.
Toe two older interlocutors exchange a smile, and Parmenides asks
Socrates, did you invent this distinction that separates the idea itself from the
things of which it is the idea, and does it seem to you that there is a
likeness-itself apart from given likenesses, and also a separate oneness-itself
and a separate idea of many? But then, says Parmenides broaching a first
difficulty with the theory of ideas, won't you have to grant that there are
214 Toe Third Way
separate ideas of such things as hair, or mud, or filth _(130c7)? But ~o~rates
is reluctant to admit that there are ideas of such things as filth ( pu1roc;,
130d2-10); he thinks such ordinary, ugly things are as they appear, and only
wants to think about such separable ideas as that of beauty or the one or the
many. And Plato lets young Socrotes get alffl)' with thi~ ~asion of. the
question. Toe latter then dissolves Zeno's paradox by explaimng how ~ngs
"participate" in (the idea ot) "oneness" or "manyness," so that there ~s ~o
contradiction in a given thing's being a unity in one respect and a plurahty m
another respect. As he concludes, Socrates says, "I sh~uld ?e:·· m~re ~azed
if anyone could show in the ideas themselves (ev o:uro,c; ro,~ eweu,),
wbich are extrapolated concepts (ev ro,c; ')\o-y,uµéi,),_ ~s _s~me
multifariousness and perplexing entanglement wbich you descnbed m v~sible
objects (130a)." Fowler's translation _of logisnwi as "int~llect_ual concept10_ns"
blurs the point that the separate ideas are abstracuve (1.e., calculat1ve)
constructions.
My purpose here, however, is not to discuss the logic ?r the details ~f the
theory of ideas, but to bighlight the dramatic uses to wbic~ Plato puts 1t, as
well as what he does with it and to it in the course of our dialogue. We must
remember that if the dialogues are ordered dramatically, as the portrayal that
they are of key encounters in the intellectual life-cycle of a great s~e~r, then
the R:irmenides would have to be the first in that order. And the s1gmficance
of this for Plato's dialogues as a series is that we may not take Socrotes, in
dramatically subsequent dialogues, to be innocent of its refutati~n .by
Parmenides and so as merely expounding it (as in the Republic) or as bnngmg
it up merely for the sake of explaining something (as in the Phl;edo, the l~t
of the dialogues in the dramatic order). That young Socrates kee~ess m
running down (µem0e,c;) and sniffing out (ixveúHc;) the weaknesses m what
people say (ra ')\ex0évm, 128c2) is found worthy of remark by Zen~,
establishes this as a lifelong attribute of Plato's Socrates. It also tells Plato s
readers to be wary of attributing the parologic in some of Socrotes ·~
discourses to naivete, and to look instead for its rhetorical or ethzcal
nwtivation or else for its ground in the culture of oral-aural communicatio~.
Moreover, that Zeno, twice in one paragraph (128b9-c9), tells Socrates bis
perception of what is written (ra -yp&µµo:m) is d~~ective reminds us that
Socrates did indeed grow up in an oral-aural, or semihterate culture.
2. How are We to 1bke Parmenides' Rehearsal of the Difficulties
in the Theory of Ideas?
To appreciate the tone and purpose of the exchange (from 130a7 to 135c3)
in wbich Pannenides takes Socrates through the difficulties in the theory of
ideas, it is best to begin with the way in wbich that discussion closes .at
134elO-135c5. "Yet these difficulties and many more attach necessanly
Toe Dialogism of Plato's Parmenides 215
(ávo:-yKo:,ov) to the ideas," says Parmenides, "if there are ideas themselves
of what exists and each idea, as such, is a separate one" (d do"tv o:urw o:i
ióéw rwv óvrwv Ko:i óp,eirw r,c; 'éKo:urov eióoc;). "Listeners to these
claims," he continues, "make difficulties and argue that neither do such ideas
exist nor could they, by any possibility, be known to human nature did they
exist.... And it'll be someone greatly gifted who can ascertain that for each
thing there is a genus and a beingness-in-itself (oúu,o: o:úri¡ Kcx0 • o:úrf¡v), and
only a most wonderful man (0o:uµo:urorépou) will discover (eúpi¡uonoc;) and
be able to teach another all these things and judge-them-sufficiently
-and-felicitously (ó,euKpm¡cT&µevov) ." Young Socrates agrees. "But on the
other hand," concludes Parmenides,
if anyone, looking back at (ix1ro/3'>-.év,o:<;) ali these and other such difficulties, also
does not permit ideas ofwhat there is (rwv ovrwv) to exist and does not distinguish
an idea of each individual thing, such a one will have no place to which to tum his
thought because he has not allowed that the idea of each existing thing is always the
same, and in this way he altogether destroys our ability to converse (óio:'>-.É'"fE<J9o:,)
with each other.
If these words are about both what the practice of verbal communication
cannot help assuming and the difficulties with those assumptions, then
Parmenides' juxtaposition of the two sides of the question can be seen as a
conversationally balanced outcome that allows the interaction to continue-as
it must, given what Parmenides has still in store to teach the young Socrates.
From the logical point of view, however, the unacceptable consequences
educed by Parmenides from the theory of ideas-not to mention the "many
more difficulties" (1r&vu 1ro>,.>,.&, 135al) he alludes to-have refuted it. But on
the other hand, Parmenides has insisted that it is an inevitable assumption of
intelligible discourse: "What, then, will become of the pursuit of knowledge
(</>tAo<To</>icx), if you cannot know the ideas?" he goes on to say. Young
Socrates confesses that, for the present, he cannot tell. And he cannot, because
the possibility of discourse has been grounded in a theory with unacceptable
consequences to which there seems to be no altemative, given the mind-set of
the fictional young Socrates. We have seen that, at this stage, he is not just
elenctic; he had been eager to share with Zeno the efficacy of "bis" theory of
ideas.
3. Discussion of the Proper Method of Intellectual Troining for
the Pursuit of Knowledge or Philosophia (135c-137b5)
Toe youngster's pet theory in a shambles, Parmenides gracefully lets bim
off the hook by praising bis "drive to ratiocinate" (ópµi¡ . . . bri rouc;
M-yovc;, 135d3) wbich, he assures bim, is noble and inspired, but wbich, wbile
216 Toe Third Way
he is still young, must be exercised and trained beyond mere verbal facility if
he wishes not to miss the truth. Socrates rises to the bait: "what then,
Pannenides, is the manner of this training?" Well, the thing that pleased me
in your exchange with Zeno, answers Pannenides, was your move to discuss
the difficulty in the light of the ideas, as they may be called-of what is
especially conceivable by reason.
Going from there, it is still necessary to do something more.
You must consider not only what happens if a given hypothesis is true, but also
what happens if it is not true. . . . And, briefly, about that which you might ever
hypothesize, supposing that it exists or that it does not exist and whatever else it
undergoes , you must examine what happens in relation to itself and in relation both
to whatever particular things you anticipate, as well as in relation to other things
and to ali of them as a whole ... if you plan to train yourself rightly to see into
the truth (füóy,Ea9cu) completely (136b8-c6).
We note, in passi~. the quasi-jormulaic phrosing of Pannenides' words here;
it is a formulism that-unless and until its meaning is grasped-sounds like
authoritative mumbojumbo.
To this Zeno adds that most people "don't know that without such a wide-
ranging (11),áv17¡;) circuit (oiE~óóov) through everything it is not possible for
the mind to come to the truth" (136el-3). Young Socrates had just asked
Pannenides to take up one such hypothesis and work through it as an
illustration for bim to learn from, and Pannenides had at first refused what he
calls an unmanageable affair (&µ~xc:xvov . . . 1rpc:x-yµc:xuíc:xv). But at the
insistence of Zeno, Antisthenes and the others (as Pythodorus told bim,
according to Antiphon, the narrator of the dialogue) Pannenides agrees-with
an entertaining show of trepidation at having to launch into "an ocean of
words," a show that itself takes ali of ninety words and that compares bim to
the horse past its racing prime in Ibycus's poem about the shock of falling in
Iove in old age, and in wbich he also calls what he is about to do an
"ostentatious, gimmicky game" (1rpc:x-yµc:xrn.w017 1rwóuh) wbich he has "to
play through" (1rc:xítHv). 3 "Shall I begin by taking my own hypothesis about
the one itself and discuss what must happen (,í XP1/ ~vµ/Jc:xívHv) if we
hypothesize either that the one exists (fon) or that it does not?"
3. Toe best translation I can think of for 1rprx-yµ.rxrEtwÓE is the Spanish "aparatoso."
Affairs 1rp&-yµ.rxrrx aren't necessarily "laborious" (as sorne humorless translators would
have it), they are rather "complex."
Toe Dialogism of Plato's Parmenides 217
4. The Demonstrotion as also a Sali,re on Sophistic Deductionism
_w~ º?te, ~ Parmenides begins the exercise, that the tone in wbich he has
sa1d 1t 1s "bis own hypothesis" is, appropriately, one of smiling irony. Toe
~ords_ rij¡; eµc:xu7:ou u1ro0foEw¡; occur at the end of paragraph 136e9-137b4,
1? wbich P_arm_emdes' tone (as_ reported ?Y _Pythodorus to Antiphon) is entirely
hght, self-ll'Omc, and appropnately prelimmary to the virtuoso display he has
co?8ented to. Observe further that the display, or antilogistic performance, is
gmng to follow the rehearsal we have just reviewed of the difficulties (ou
eu,í~ ~ &1ropí~, 133bl, 133a6) in young Socrates' theory of ideas. Afte;
rubb1ng them 10 (at 135a: "the ideas necessarily have these difficulties and
many more besides . . . . "), Parmenides then adds, antinomously, that if,
however, one does not allow being to the ideas, he'll find that bis ability to
c~nvers~ has been destroyed because things will have Iost their identities. To
this ant10omy Parmenides adds, as we saw, the ironic qualification that "only
a man of the greatest natural gifts will understand that for each thing there is
a g~nus and an essence or being-in-itself (ovuíc:x c:xu,r¡ Kc:x0 'c:xu~v); and only
a sflll more wonderful man will find [them] out and be able to teach another
how _to _ju~ge them sufficiently and felicitously." This sequence is telling us
that 1t 1s Just the problem of participation that Parmenides has shown to be
unsolved ~nd}nsolu?le-except (of course) by the "wonder-man of very great
natural g1fts requued to keep young Socrates in countenance and the
conversation from ending.
We c~ now see that readers have been misled by the kindly tone in wbich
~e~des has c~nsoled the young Socrates for the knock-down refutation
admims_tere? to. bis theory, .. and by the politeness that covers the irony of
Parmemdes say1ng that an Ubermensch might sorne day solve the problem in
~e s~me breath ~at he promises young Socrates that there is a great future for
bi~ 10 _th~ pursmt of knowledge, if he can understand the ramifications of the
~tllo~1s_t1c method that he is about to illustrate. But-irony of ironies-misled
hteralist1c readers_here take Parmenides's ironic politeness literally as meaning
that the theory of ideas can be salvaged. That it has not been salvaged in more
than two thousand years goes a long way toward showing that it cannot be
salvaged.
There is yet another lesson that the dialogical reader will not fail to draw
from the situation ?f intellectual tension into wbich Plato has maneuvered bis
characters. Parmemdes goes on to apply Zeno's refutational method to clauses
that, ~hen taken .out of the context of his own poems, reappear as both the
assert1on and derual of a doctrinal question "whether the one is or is not one"
218 The Third Way
4
(E'í'TE ev fonv EÍTE µ~ ev, 137b5). Is it not ironical and paradoxical that,
on one hand, Zeno's method literally applied destroys ali asserti~, including
his master's (should asserti~ be what he was doing)? On the other hand, if
the admonitions of Parmenides' Goddess are taken literally, rather than
poetically, as true propositions, not only are sorne of them easily falsifiable by
the auditor, but also, Zeno's method must be wrong. It follows, if we are
reading the dialogue litemlistically, that Parmenides and his disciple Zeno
cannot both be right; either something is wrong with Zeno's eristic method or
else Parmenides, if he was being dogmatic, is wrong. Only by taking the
dialogue dialogically can master and disciple be in agreement. And the
dialogical reader notes, firstly, that the apparent categorical question is in
reality a propositional precipitare abstmcted from an utterance held in the
suspension of Parmenides' poetic medium, a suspension in which the utterance
is not assertive but exhibitive and, secondly, that the dialogue itself that we are
reading holds ali assertions made by speakers in it in a similar state of
exhibitive suspension. Wrong as it is to take the metaphysical poem of the
original Parmenides as a series of assertions, it is just as much of a misreading
to take Plato's dialogue to be a construction in the assertive rather than the
exhibitive mode of judgment.
As we read on, we find that we are not sure of the reference of " the one"
to be antilogized. Is it the original Parmenides' "Ali:" the All in the sense of
whatever-is-was-and-will-be that ex.eludes nothing and that, in being
conceptualized, has to be thought of as One? Or, in the light of what the
conversation has so far been about, is it just the idea of one, or oneness? Or
is it, thirdly and as sorne suggest, the separate oneness of any idea that is
being antilogized?5 What we must, therefore, do is to let the development and
context of the antilogistic demonstration determine for us what the reference
or references of "the one" might be in the rest of the dialogue. But this new
turn that our inquiry must take may not be allowed to leave out of artistic or
4. Toe ambiguity in these words cannot be overlooked: alternative possible translations
are, "either it is one or it is not one;" "either there is one or there is not one." The
former is a Jogical alternative, wbether the reference is to the original Parmenides' Ali
orto unity as such; the Jatter is an ontological alternative. Fowler translates ontologically
that "the one exists or that it does not exist"-but wouldn't this require the Greek to bave
been "EiTE h fonv EiTE µ.~," without the repetition of h? Grote translates "unum
est" and "unum non est" (vol.iii. p.81f). Brumbaugh's translation of the first hypothesis
(137c4) as "ü one is, it certainty will not be many," takes the alternative to be about
unity; and, as in other neoplatonic interpretations, he takes Parmenides' discourse from
here on to be metaphysical.
5. E .g., H . L. Sinaiko, Love, Knowledge, and Discourse in Plato: Dialogue ami
Dialectic in Phaedrus, Republic, Pannenides (Chicago: University of Chicago Press,
1965), 238f.
Toe Dialogism of Plato' s Parmenides 219
organic relation to the rest of the dialogue either Parmenides' new rhetoristic6
tone or his playfully elaborate antilogizing, if it is true-as I am claiming- that
the JtJ,rmenides is a carefully constructed work of art.
In doing this_, finally, we must point out what too many nondialogical
r:aders h~ve rrussed, namely, that-whatever else it does-this part of the
dialogue 1s at the same time an exhibitive satire on the phenomenon of
second~ elaboration, if not also on the excesses of deductivism, the engine
that drives secondary elaborations in philosophy and the practice upon wbich
~no's refutationism depends. In Plato's brilliant hands, Parmenides can be
s~d to have tempered the _negative impact of bis disciple's refutationism by
bemg endlessly and deduct1vely refutative himself: the point is driven home
that, with enough deduction, any and every claim is refutable. But Parmenides'
parod_y of Zen~'s methodic deductivism also shows that you can prove
any~. you hke by being sheerly deductive, including contradictory
propos1t10ns about the same thing, ifyou don't know which starting points are
not to be accepted. This realization will come as no surprise to readers who
concur in the finding that Plato is a quietly skilful satirizer, in the exhibitive
mode, of the phenomena .of intellectual life and the intricacies of
communicative interaction among disputants. 7 The realization also brings this
part of the dialogue into continuity with the satirical tension in the central
situation of the dialogue according to which-if what Parmenides and Zeno are
s ~ is ~en literalistically-they cannot, though they are master and
disc1ple, be ~n agreement. Nor can we possibly omit noticing, given the above,
that Parmemdes never loses bis virtuoso detachment as he tacks and scuds bis
way over the sea of words on which he is launched. 8
And the reader understands that, though the exercise is for the benefit of
young Socrates, the reason he cannot be the respondent in this parody of
6. Actual!y an antirbetoristic imitation meant to show up the ecbolalic rhetoricism of
such Sophists as Gorgias.
7. As I show to be the case, and in dramatistic detail, for twenty-two dialogues in
Plaro's Dialogues One By One (New York: Irvington, 1984). G. Grote's commeot about
the antinomies in the second half of the Pannenides comes to mind here: "ü these same
dem~nstrati~ns, constructed with care ... for the purpose of proving that the same
premisses will conduct to double and contradictory conclusions, had come down to us ...
under the name of the Megaric Eucleides, or Protagoras or Gorgias- many ... critics
would probably have said ... that they were poor productions worthy of sucb Sophists,
wbo -are declared to have made a trade of perverting truth. Certainly the conclusions of
the demonstrations are specimens of that 'Both and Neither,' which Plato (in the
Euthydemus, 300cff.) puts in the mouth of Dionysodorus as ... [sophistic] defiance" (iii.
82f).
~- The inquiry into the identity or reference of "lt" or "the One," as well as other
po10ts about this satire on antilogistics and deductionism must be left to another occasion.
220 Toe Third Way
deductivist argumentation was given by Zeno when he said earlier in the
dialogue that young Socrates follows arguments, "chasing after" them and
"sniffing [them] out," as closely as a Laconian hound (w<11rEp 'YE ai A&Ka,vO'.,
<1KÚAaKE<; Ev µ,emO{ic; TE Ka, ixvEÚEt(;, 128cl-2). Such an ability, obviously,
would ruin the fun and make the exhibition impossible. That young Socrates
tacitly agrees to remain silent also means that he has caught on to the fact that
what Parmenides has undertaken is an anti-Sophistical parody of extant
sophistries about what-there-is.
In confinnation of this we find, going back to Parmenides' earlier rehearsal
of what bis method will be (starting at 136a4 with young Socrates' "what do
you mean" and going on to 136c9, "What an impossible ... affair, I don't
understand very well," éxµ,i¡xavov . . . 1rp0'.-yµ,areíav Ka, ov u<J,óopQ'.
µavO&vw), that (1) Parmenides' tone has become incantatory and ech~lalic in
a manner that competes with the style of Gorgias's recital on Not-be1ng, (2)
that the description of the hypothetical method is so highly generalized as to
be nearly incomprehensible without exemplification at the same time that it
sounds authoritative, and (3) that it does outline a preset, if complex, schema
for the development of antilogies about anything. 9 And such pre-set schemas
according to which to argue were, of course, the common device and
characteristic resource of all the Sophists. So not only is Plato's Parmenides
out-sophisticizing the Sophists at their deductive games, he is also bringing out
the limitations inherent in the deductive process for the benefit of the
mathematicians as well.
Toe lessons that young Socrates can be seen to have been learning in the
Parmenides thus include a new sense of the limitations of both refutationism
and deductivism. This new sense will temper bis keenness in identifying
paralogisms with an awareness of the limitations of logic itself. But just as the
refutationism of Zeno and the antilogizings of Protagoras had now forced the
climate of opinion to accept the fact that any proposition categorically asserted
was refutable or relativizable, so correspondingly, young Socrates is made by
Parmenides to see that claims must be advanced hypothetically. Toe
reader-auditor of the dialogue, in tum, is invited to see for himself that when
hypotheticals are advanced in argument, the speaker's associated premisses
must be monitored from the very beginning, objected to, or qualified, if the
deduction of consequences from the hypothesis is not to go astray.
Nor will a reading of the dialogue that takes it as a set of carefully
constructed communicative interactions that is both high intellectual satire ami
pedagogical take the concluding antilogistic exercise to be a treatise in
systematic metaphysics. And this, even though there is much to be gained by
9. R. Brumbaugh draws attention to these proof-schemas in his Plato on the One: The
Hypotheses in the Parmenides (New Haven: Yale University Press, 1961).
Toe Dialogism of Plato's Parmenides 221
joining in Parmenides' deductive fun-and-games for those who undertake to
think about unity and plurality or about what-there-is and the All as
metaphysical problems. Toe demonstration of this, like the inquiry into' the
reference of "the One," will be left to another occasion. But a single example
will suffice to show that this other inquiry is worthwhile. Does not the point
made by the first hypothesis that if "the One" is a whole, it cannot have
parts-coun~rin~tive as this is-implicitly raise a question, for example,
about how ideas, 1f they are unitary, can be the ideas of things with parts?
Equally a problem for those who might have on their minds the oneness of the
Ali in Pannenides's poem is that this One-which is All there is and
therefore, the Only thing there is-would have to be a complex to b;
exis«:ntial; for things are only simple in sorne respect and no simples exist,
save m the order of discourse; simples are all implicitly postulational. Toe
seeming paradox is that It, this One or AH, which has no parts, is neither
simple nor only discursive (i. e., nonexistential) because it is the Ali that is
everything. Its existence must, therefore, be conceptual: the essence-as
Santayana might say-of a potentiality, "the Being" of all becoming. It is a
real possibility that is not exemplified because never ful/y exemplified, but
nonetheless exemplifiable because there is what-there-is-was-and-will-be. There
is not nothing, as there would have to not-be if It were not exemplifiable. And
if it is in fact to be the Truth of ali that-ever-was-is-and-will-be, it will include
the attributes of unendingness and ungeneratedness and must, therefore, have
objective being, and be independent of our thinking of it. That it can be
thought of, however, entails, in the terms of Parmenides' poem, that it is; for
there, "if it can be thought it is" (ro -yap avro voúv éurív TE Ka, Eivw,
fr.4).
Toe Pa_rmenides shows, in other words, that in composing bis dialogues,
Plato des1gned them to be not only interesting, challenging, and amusing, but
also to dramatize and teach the need for achieving perspective and observing
?roportio~ in intellectual matters. It is a problem for the neoplatonic
mterpretat1on of Plato, for the intellectual historian, and for the sociologist of
knowledge that the antidialogical tradition has persisted for so long in reading
the dialogues literalistically, in abstraction from their allusive irony and wit
and as solemn tractates seeking to develop or enforce doctrine. But it is not a
problem _that dialogical readers of Plato's ever-intelligent, ever-rewarding
construct1ons have to address- ex:cept perhaps to help literalist readers see that
the burden of proof that what looks, sounds, and reads like a dialogue may be
treated as if it was not a dialogue falls on those who make it their practice to
treat them so.
 V

Between Unwritten Doctrines and

W ritten Dialogue
 13

The Choice between the Dialogues

and the "Unwritten Teachings" :
A Scylla and Charybdis for the Interpreter?

Mitchell Miller

Toe general altematives posed by our panel 1 are these: either the dialogues
are, at least primarily, bearers of doctrine or they are pieces of psychagogical
rhetoric2 aimed at eliciting an ethical-existential becoming3 centered in a
decisively practica! knowledge. 4 My particular contribution is concemed with
a different but related either/or. I am interested in "the so-called unwritten
teachiogs," ta legomena agrapha dogmata. 5 In the context of our discussion,

l. This essay was originally presented as the last of four talles in one of the series of
panels on "Plato: Dramatic and Non-Dogmatic" sponsored by the Society for Ancient
Greek Philosophy in New York in October 1992. The three preceding talles- by David
Roochnik, Drew Hyland, and Francisco Gonzalez, respectively- are cited in n.s 2-4.
2. See David Roochnik, "Socrates's Rhetorical Attack on Rhetoric," included in this
volume.
3. See Drew Hyland, "The Presence and Absence of Plato: On Heidegger's Reading
of Plato," scheduled to appear in The Sovereignty of Construction: Studies in the Thought
of David Lachterman (Rodopi, forthcoming).
4. See Francisco Gonzalez, "Self-Knowledge, Practica! Knowledge and lnsight: Plato's
Dialectic and the Dialogue Form," included in this volume.
5. The whole phrase is Aristotle's at Physics 209bl4-15.
226 Toe Third Way
the issue of the "unwritten teachings" would seem to threaten the interpreter
with this unpalatable choice: either one must go outside the dialogues to the
"unwritten teachings," reported as they are especially by Aristotle in
Metaphysics A6, or, in order to stay within the dialogues, one must reject
Aristotle's testimony as, in sorne proportion, misinterpretation and fabrication.
To many hermeneutically inclined readers of Plato, turning to the "unwritten
teachings" seems like abandoning the dialogues. To borrow Homer's glorious
image, it is making one's journey into Plato vulnerable to the sheerly external
power of a Scylla, letting oneself be pulled up out of the stream altogether and
devoured. On the other hand, to the advocates of the "unwritten teachings,"
especially to the best-known champions of esotericism, the Tübingen school,
to remain within the dialogues, searching out the Schleiermacherian interplay
of form and content and exploring the bottomless depth of Socratic and
Platonic irony, is to let oneself be sucked down into a Charybdis of "empty
and endless "-and fundamentally disoriented- "questioning. " 6
What to do? Given this all too familiar sense of the options, I invite you to
share my surprise severa! years ago when, without any predisposition to do so,
I found my own hermeneutically oriented readings of the lbrmenides, on the
one hand, and of the Statesman, on the other, leading me back through these
dialogues to the "unwritten teachings. "7 I have been working on this issue for
a while now, andas Ido, the "unwritten teachings" that are written only grow
deeper and richer for me.
With this discovery, the choice between the Scylla and Charybdis I have
described has laigely dissolved. To find "the unwritten teachings" in the
dialogues is to be freed from the need to attack Aristotle's reports: and to find
them in the dialogues is to be freed from the need to abandon the dialogues
6. This phrase is Gonzalez's accurate paraphrase ("Dialectic as Self-Knowledge,
Practica! Knowledge and lnsight," n. 36) of Kriimer's polemic against Wieland's
"nondoctrinal" reading of Plato. See especially Kriimer's sustained criticism of the
Schleiermacherian approach to the dialogues in Plato and the Foundations of
Metaphysics, trans. John R. Catan (Albany: SUNY Press, 1990). That taking the
"unwritten teachings" seriously can go hand in hand with a hermeneutically sensitive
approach to the drama and irony of the dialogues is exhibited by Thomas Szlezák's
commentaries in Piaron und die Schriftlichkeit der Philosophie (Berlin: Walter de
Gruyter, 1985).
7. K. M. Sayre's Plato 's Late Ontology: A Riddle Resolved (Princeton: Princeton
University Press, 1983) was a great help to me in this process. Though my own reading
ofthe Parmenides (Plato 's Parmenides: The Conversion ofthe Soul [Princeton: Princeton
University Press, 1986; reprinted in paperback by Pennsylvania State University Press,
1991]) differs fundamentally from his, his ground-breaking effort to recognize the
"unwritten teachings" in the Parmenides and the Philebus displays both the viability and
excitement of the project.
Toe Choice becween the Dialogues and the "Unwritten Teachings" 227
and th: task of interpretation. But how does this bear on the general
alternat1ves that our panel is considering? If the "unwritten teachings" are
after all, written, if they form a deep (even if indirectly conveyed) content'
then do the dialogues not become, after all, bearers of doctrine? And to th;
degree that this is so, does the work of discovering that content lead us away
from the psychagogical and existential function of the dialogues?
In reply to these issues, I want to share, in highly compressed fashion
reflections on three questions: '
(1) What are the "unwritten teachings" that show up in the dialogues?
(2) What is their basic force and character, their spiritual function?
(3) What are the implications of the discovery of them in the dialogues for
our conception of the basic character of the dialogue form and for our
understanding of how to read and interpret Plato?
l.
In '!1etaphysics A6 Aristotle reports that Plato held, among others, these five
teachings:
[1] Th~ forms and the dyad, the Great and the Small, are conjointly
resp?ns1bl: for_ ~.e being of sensibles: forms are the "cause of what [a
sensible thing] 1s, and the Great and the Small are "the underlying matter
of which [forms] are predicated."
[2] Toe One (or what is here the same, Unity) and the dyad, the Great and
the Small, are conjointly responsible for the being of the forms; the One is
"cause of what [a form] is," and the Great and the Small are "the
underlying matter of which [the One] is predicated. "
[3] Toe One is cause of "good" (to eu); the Great and the Small, of "ill"
(to kakós).
[4] "Intermediate" (meta.xu) between the timeless, unchanging, unique forms
and the perishable, changing, many sensibles, there are "the mathematicals"
(ta mathematika); these are intermediate because they are eternal and
unchanging like the forms but many like sensibles.
228 Toe Third Way
[5] Toe forms are-but in sorne qualified sense that preserves the distinction
asserted in [4]-nurnbers. 8
These teachings sh0\1/ up, I would argue, as the key mornents of a
cornprehensive account of the participation of things in forms and of the
interplay among the forms that participation implies. This account is the
implicit content of Pannenides's treatment of participation in the third
hypothesis of the R:mnenides, of Socrates's presentation of the god-given
rnethod of dialectic at Philebus 16c-18d, and of two extended nonbifurcatory
diaireses, Socrates' analyses of the kinds of pleasure and knowledge at
Philebus 46b-52c and 55d-58a and the Eleatic Stranger's elaboration of the arts
concemed with the city at Statesman 287b-291a and 303d-305e. 9 To put this
in the most summary fashion, these passages work together as follows. In the
diaireses of the Philebus and the Statesman, Socrates and the Stranger put into
practice and so exhibit concretely-but in each case without acknowledgrnent
or explanation-the new "god-given" mode of dialectic that Socrates first
introduces, then ostensibly abandons, in his highly schematic methodological
reflections in Philebus 16c-18d. Toe type of eidetic order that this new, non-
bifurcatory rnode of dialectic is fit to disclose is implied by the analysis of
participation that Parmenides offers in the third hypothesis of the R:lrmenides.
Thus, the Philebus and the Statesman, in displaying the general principies and
the concrete results of dialectical rnethod, disclose the eidetic order irnplied by
participation.
8. My inclusion of this thesis and, too, its qualified character require immediate
comment before we step back to look at the whole assemblage of teachings. First, by
contrast with [1)-[4], nowhere in A6 does Aristotle expressly assert that forms are
numbers. Only later, e.g., 991b9-20, 1073a13f., 1080allff., do we find him explicitly
(albeit with varying degrees of certainty and clarity) crediting Plato with this claim.
Nonetheless, it is strongly implied in A6 when Aristotle uses virtually identical language
to say first that/onns are "causes of everything else" [aitia ... tois allois, 987bl8-19] and
then, only two sentences later, that numbers are "causes of the being of everything else"
[aitious .. . tois allois tés ousias, 987b24-25]. At the same time, the identity needs to be
qualified, for Aristotle has also just credited Plato, in thesis [4], with the distinction of
forms from mathematicals. Hence my open-ended "in some limited sense." Even as we
wonder if there are any traces of this teaching in the dialogues, we must also hope that
such traces, should we find them, will show us how to understandjust what the teaching
means.
9. I select these texts only because the way they seem to fit together lets the "unwritten
teachings" emerge as a whole account. But there are other texts as well in which one can
find the "teachings" exhibited. Consider, most notably, the accounts of due measure and
of temperaments and virtues in the Statesman (see Krlimer's analysis ofthe latter in Arete
bei Platon und Aristoteles [Heidelberg: Car! Winter Universitiitsverlag, 1959], 148-168).
The Choice between the Dialogues and the "Unwritten Teachings" 229
Needless to say, to work all this out here is impossible. 10 For the
distillation we require at present, I shall restrict rnyself to summary exegeses
of the analysis of participation in the R:lrmenides and the elaboration of the
arts in the Statesman.
In the third hypothesis, Parmenides reasons that "the one," which itself is
simple and unique, is responsible for the whole-part structure that arises in
"the others" that participate in it. These "others," in and of thernselves
~ndeterminate in magnitude and rnultitude and so not actually existent, receive
mtemal an? extemal "boundaries" from "the one" and are thereby constituted
as a plurality of wholes of parts. If it is right, as I have argued elsewhere 11
to take "the one" to stand for any one defining form, and "the others" to stdid
for those_sensible_s in place and time that it defines, and if it is also right to
take the mdetermmate magnitude and rnultitude to be the instantiation of the
dyad, the Great and the Small, then Parmenides' analysis explicates the first
of ~e five. teachings Aristotle reports: since sensibles (that is, "the others")
rec~:ve the1r "."hole-p~ structure from sorne one defining form (that is, "the
one ) and thetr magrutude and multitude frorn the Great and the Srnall, forros
and the_dyad, the Great and the Small, are conjointly responsible for the being
of sensibles. Moreover, we can also glirnpse the first part of the second of the
fiv_e teachings. The defining form, on Parmenides' analysis, is itself simple and
uruqu~; f'.urther, to ~e responsible for whole-part structure is to bestow unity.
~oth 1? 1ts o~n ~ty and in its causal work, therefore, the defining form
!ns~tiat~s _Uruty 1tsel~..1:hls gives us a first intimation of the role Unity plays
m 1ts conJomt respons1b1hty with the dyad for the being of forms.
T~ ~ee the way the dyad collaborates with Unity, however, we must malee
e~phc1t an~ bring together severa! sets of irnplications in Parmenides' account.
(1). Parmemdes argues that each part of a sensible whole is itself, in its own
~ty, a whole of parts. This implies that just as each sensible thing gets its
uruty as a whole b~ ~arti_cip~ting in sorne one form, so each part gets its unity
as a whole by part1c1patmg m sorne one form of its own. N0\1/, just as parts
rnust befit one another and the whole to which they belong, so, in order to
_10. For exe~eses of different parts of this account, one can see (i) the analysis of the
third h~pothes1s (and related passages in the second and sixth hypotheses) of the
Parmemdes in my "'~~written Teachings' in the Parmenides," Review of Metaphysics
48, 3 (March 1995); (u) the treatment of the way the relevant passages in the Philebus
and the State~man_ dove~il, in my "Toe God-Given Way, " Proceedings of the Boston
A rea .Colloqu1um m Anc1ent Philosophy, Vol. VI (1990), eds. John Cleary and Daniel
Shartrn (Lanham_, Maryland: University Press of America, 1992), pp. 323-359; and (iii)
the more expans1ve account of the whole web of teachings in my "Dialectical Education
and Unwritten Teachings in Plato's Statesman," in The Sovereignty of Construction .
11 . See my Plato's Parmenides, 77, 89-99, 124-125, 139-141, 159-162.
230 Toe Third Way
establish tbis fit, must each form of a part match up appropriately with each
of the other forms of parts and, with them, answer to the form of the whole.
And tbis, finally, implies that a thing's defining form, in requiring of it its
whole-part structure, implicates a definite set of fonns of parts. (ii) A key
respect in which the parts of a sensible must befit one another and the whole
to which they belong is their relative size; to combine as parts, each must be
of a size appropriate to each other and to the whole. But this is to say that the
forms of the parts must pick out, on a continuum from smaller to greater (or
vice vasa), a set of ratios or proportions establishing the relative sizes of the
parts. (iii) Now, tbis continuum is an abstract instantiation of the dyad, the
Great and the Small, and it is here that we can recognize the way the dyad
collaborates with Unity. We have already seen that it is in its bestowal of
whole-part structure on a sensible thing that a defining form instantiates Unity
itself. We have also seen that this bestowal requires, according to (i), a
definite set of forms of parts that, according to (ii), implicate a corresponding
set of ratios on the continuum from smaller to greater. Toe instantiation of
Unity therefore requires the continuum. Hence, as Aristotle reports in teaching
[2], Unity and the dyad, the Great and the Small, are conjointly responsible
for the being of the forms as, to quote from teaching [l], the "cause of what
[a sensible thing] is. " 12
Toe remaining three teachings, [3], [4], and [5], are implied by the complex
structure articulated by [1] and [2]. First, the ratios on the continuum of
relative size articulate the ideal proportions of the parts to one another and to
the whole. These proportions constitute a normative order for actual sensibles.
A sensible will be "good" insofar as its parts reproduce these proportions,
"bad" insofar as they exceed or fall short of them. Now, if we are right to
trace back from these proportions to the forms of parts that call for them and
then to the defining form that calls for these forms of parts and, finally, to
Unity itself as what the defining form instantiates, it is Unity itself that is
ultimately responsible for these proportions. On the other hand. it is the dyad
of the Great and the Small that, in its indeterminateness as a continuum of
relative opposites, is reflected in the excess and deficiency that mark the
failure of "bad" parts to reproduce the normative proportions. Hence, as
teaching [3] holds, the One is cause of "good" (to eu); the Great and the
12. This qualifying clause is important. On the reading I am proposing here, the
"unwritten teachings" do not offer an account of the generation or constitution of the
forms as such; rather, they explícate what is necessary for the forms to play their causal
roles in being responsible for the structure of "things." In the analysis I shall offer
shortly of the diairesis of the arts in the Statesman, for instance, the continuum is
necessary not for the being, as such, of "care" or even fo'r the being of each of the kinds
of art but, rather, for the expression of "care" in the various proportions of material to
spiritual that the fifteen kinds of "care" require in a good city.
Toe Choice between the Dialogues and the "Unwritten Teachings" 231
Small, of "ill" (to kakós). Secondly, these proportions and the continuum itself
from which they are selected are mathematical structures and articulate the
possibilities, "good" and "ill," that sensibles actualize. They are, therefore,
just the sort of "mathematicals," standing "intermediate" between forms and
sensibles, that teaching [4] declares. Finally, we can also now recognize the
qualified sense in which, according to teaching [5], forms can be said to be
numbers. As we have seen, the proportions on the continuum are picked out
by the definite set of forms of parts and, so, since it is this that first implicates
these forms, by the defining form. If we make a distinction between the nature
of a form and its causal power, we can say that in its nature a form is not, but
in its causal work a form is, identical with numbers. That is, even while it is
the case that, as teaching [4] affirms, forms and the proportions they call for
are ontologically different in kind, nonetheless, in picking out these
proportions on the continuum, forms express themselves as numbers. Hence,
in this qualified sense of functional equivalence, forms "are" numbers.
Let me now just indicate the way these five teachings, thus interpreted, are
concretely exhibited in the closing distinctions of the Statesman. To do this,
we must first add one significant qualification. In both the R:irmenides and
Aristotle's reports, the counterprinciple to Unity is the dyad of the Great and
the Small. In the Philebus, by contrast, Socrates treats "greater and smaller"
as but one pair of relative contraries among others; he also names "hotter and
colder," "drier and wetter," "more and fewer," "quicker and slower," and
"high and low" (25c, 26a), and he designates all of them as cases of the
Unlimited (to apeiron). Moreover, he goes on to say that the mixture of the
unlimited and limit, that is, of continua and sets of proportions, gives rise not
only to "beauty and strength along with health" but also to "a further host of
beautiful characters in our souls" (26b). Thus, Socrates extends the reach of
the "unwritten teachings" beyond sensibles, the focus of the third bypothesis
in the Rlrmenides, to include the spiritual as well.
With this in mind we can turn to the Eleatic Stranger's closing distinctions
in the Statesman. At 287b, after proceeding by bifurcation in every one of the
many previous diaireses in the Sophist and the Statesman, the Stranger
suddenly declares that "cutting into two" is "difficult," remarking only that
"the reason will become clearer to us as we proceed" (287bl0-cl). He then
works his way through a division of the arts concerned with the city into these
fifteen kinds:
the "contributory arts":
(1) arts which produce raw materials
(2) arts which produce tools
(3) arts which produce containers
(4) arts which produce vehicles
(5) arts which produce defences
232 Toe Third Way
(6) arts which produce amusements
(7) arts which produce nourishments
the "directly responsible arts":
the arts of services ordinary ... :
(8) the art proper to slavery
(9) the arts of merchants and traders
(10) the arts of heralds and clerks
(11) the arts of priests and diviners
. . . and precious:
(12) rhetoric
(13) generalship
(14) the art of justice
the art directing all these:
(15) statesmansbip
How does this division exhibit the teacbings we have discovered in the
Rirmenides? As I confessed at the outset, to give a full and adequate account
in support of this claim would go far beyond the constraints of the present
occasion. But to give a concrete indication of this is possible. Consider, in
turn, the instantiation of Unity as "care," the instantiation of (not the Great
and the Small but the more general principie these exemplify, namely,) the
Unlimited as the continuum that underlies the fifteen kinds, and the status of
the fifteen themselves as constituting a normative order for cities. (As we
proceed, it may be helpful to keep in view the reconstructive diagram that I
have offered at the end of the essay as an appendix.)
(i) The instantiation of Unity as "care.'' At the close of bis initial diairesis
(258b-267c), the Eleatic Stranger points out that the resultant definition of the
statesman as a kind of herdsman concemed with the "collective nurture of
humans" (anthrópón koinotrophiken, 267dll) opens the field to a host of
challengers; merchants, farmers, bakers, and even teachers of gymnastic and
doctors are all concemed with "nurture" (tes trophes, 268a2). His express
purpose in turning next to the great myth of the ages of Kronos and Zeus is
to remove these competitors. lt is therefore surprisi~ when the first revision
he makes on the basis of the myth is to replace the notion of "nurture" by the
still broader notion of "care" (to theropeuein, 275e3). Rather than eliminating
the challengers, the notion of "care" opens the field still more, summoning a
host of arts that "nurture" would exclude. Why does the Stranger do this? We
can respond on two levels. In terms of the specific issue of statesmansbip, the
myth has shown the inappropriateness of the metaphor of the herdsman: such
a figure, relating to other men as if of a bigher species-that is, as if a
Toe Choice between the Dialogues and the "Unwritten Teacbings" 233
god-would usurp the task of caring for themselves that, creating the need for
the many ans and drawing human beings together into a collaborative
community, gives rise to the city. This implies, however, that statesmansbip
is one among these many arts and that ali together share the essence of being
kinds of "care." Hence-now to put the point in terms of the "unwritten
teacbings " - "care" is that one form that, instantiating Unity, calls for the
fifteen kinds of art as the parts that comprise the community as a whole. Toe
Eleatic Stranger alludes to this whole-part structure when, at 287c3, he
introduces the distinctions to come by saying that the arts must be divided kata
me/e . . . hoion hiereion, "limb by limb . . . like a sacrificial animal."
Understanding "care" requires recognizing which are the kinds of art that, like
the "limbs" of an animal, fit together to make an organic whole.
(ii) The instantiation of the Unlimited as the continuum of the material
and the spiritual. But how is one to recognize these "limbs?" Toe "unwritten
teachings," as we have reconstructed them, provide basic orientation for one
attempting this difficult task. Complementing the instantiation of Unity in the
one form-here, "care"-is .the abstract instantiation of the Unlimited as a
continuum between opposites; the forms of parts that the one form requires
themselves pick out proportions of these opposites on the continuum. To
recognize the "limbs," therefore, one must begin by recognizing the opposites
and the continuum they frame. For us, reading back from the results the
Stranger presents to the orienting "teacbings" they exhibit, the task is to
discem, in the list of fifteen, the underlying continuum. Three signal features
of the Stranger's list enable this. [1] The gradual charocter of the movement
toward statesmanship. With each kind that he distinguishes, the Stranger
pauses to ask whether this is the sought-for art of statesmanship and, by the
relative force and character of bis negations, 13 he makes evident how the list
as a whole charts a path that leads gradually, step by step, from kinds of art
that are least to those that are most like statesmanship. Thus, he indicates the
underlying gradient that bis list articulates. [21 The opposites that fra,me the
continuum. Initially, this gradient appears to be one-directional, oriented
entirely by the goal of statesmanship. In truth, it is a bipolar continuum,
stretched between opposites, with statesmanship itself just one locus on it. Toe
Stranger indicates the bipolarity by a telling revision at 289a9. There,
summarizing the initial seven divisions by wbich he has articulated the merely
"contributory" arts, he points out that "it would have been more just to have
put the kind that produces raw materials" -which he first carne to in bis sixth
cut (288d-e)-"at the beginning (kat' archas)," and he reorders the list
accordingly. (Hence the order in wbich I have given the list above.) That the
13. Por a detailing ofthis point, see Miller, "Toe God-Given Way," 348f., n. 29.
234 Toe Third Way
arts most concemed with the material stand at the farthest remove from
statesmanship makes conspicuous by contrast that statesmansbip is the art most
concemed with the spiritual. Toe Stranger will go on to make this explicit in
the final pages of the dialogue, in wbich he assigns to the statesman
responsibility for generating and preserving the civic virtues of temperance and
courage and (since the balance of these is secured by law) justice in the
citizens. Hence the material and the spiritual emerge as the opposites that
frame the underlying continuum. [3] The range ofproportions. That, in turn,
the list of fifteen marks out a range of proportions between these opposites, the
Stranger spurs us to recognize by the provocative reticence with wbich he sets
the bifurcatory mode of diairesis aside at 287b10-cl (quoted above). Why is
it "difficult" (287b10), indeed "impossible" (c4), to "cut into two?" Toe
Stranger cannot mean that the arts cannot be sorted into two overall groups,
for he does just this, invoking the differentiation of "contributory" and
"directly responsible," sunaitioi and aitiai, that he first made in the
paradigmatic diairesis of weaving. Rather, he is forewarning us that the eighth
of bis distinctions-the art of slaves-prevents this bifurcation from separating
the arts into equal halves. To see why, moreover, he positions us to
understand the character of each of the other distinctions as well. Slaves are
at once both "possessions" (ktétous, 289d10) and agents; as instruments, they
are completely heteronomous and thing-like, but their value as instruments
consists in their capacity to perform the actions that each of the other arts
requires. In this sense, they mark the point at wbich the material and the
spiritual, thing and agency, stand in precisely equal balance. Once we see this,
the continuum as a whole emerges as a series of proportions of material to
spiritual, ranging from the preponderance of the material over the spiritual in
the "contributory" arts to the equality of the two in the art of slaves to the
preponderance of the spiritual over the material in the "directly responsible"
arts. To bring this into sharper focus, consider first the "contributory" arts,
(1) through (7) in the Stranger's list. Toe Stranger begins with the arts that
provide unformed stujfs for other arts to shape into things (1); then, within
these latter arts, he sorts out the making of things to produce (2), then to
preserve (3), then to transport (4) other things. Toe things transported, in tum,
may be persons. Hence (4) marks a transition to the arts that make things for
the sake not of other things but of persons, albeit in their physical being, the
arts that make defences and clothes (5), diversions and omaments (6), and
nourishments and nourisbing activities (7). Strikingly, this last series moves
from what is relatively extemal to persons in their physical being to what is
fully intemal- the foods that become part of the body and the exercise that is
itself the body's own activity. Thus, the Stranger leads smoothly into the art
of slaves, in wbich the artisan himself produces bis own body's activity in
service of other artisans. By contrast with the first seven kinds, the final seven
mark out proportions in wbich the spiritual comes increasingly to dominate the
Toe Choice between the Dialogues and the "Unwritten Teachings" 235
material. Of these, the first three are concemed with public practices ranging
from the economic activity of distributing material goods (9) through the more
administrative activities of record-keeping and communication (10) to the
conventional-spiritual activity of directing the variety of public rites (11). Toe
priests' and diviners' concem with conventional religious piety, in tum, leads
into the next three kinds, in wbich the focus sbifts from public practices as
such to the civic virtues-that is, the sorts of good character- that make them
sound and assure the city's flourisbing. In the rhetor who persuades the
citizens to accept the statesman's policies (12), the general who, taking orders
from the statesman, leads the citizens in war (13), and the judge who applies
the law with impartiality (14), the Stranger marks the cultivation of
temperance, courage, and justice. This leaves only the statesman, whose
regulation of education, public honors, and marriages aims at generating and
preserving this temperance, courage, and justice and so itself embodies
political wisdom (15).
(iii) The fifteen kinds as constituting a normative order. We are now in
position to recognize the "unwritten teacbings" in the Statesman-that is, to
recognize how the Stranger's diairesis of the fifteen kinds exhibits the account
of participation we have discovered in the ltJ.rmenides. Unity and the
Unlimited collaborate to enable the normative order for cities. Unity provides,
in its instantiation as "care," the form that implicates the forms of the parts
necessary to the city; the Unlimited provides, in its instantiation as the dyad
of material and spiritual, the continuum on wbich each form of a part picks out
sorne definite proportion (or range of proportions) between material and
spiritual. That the fifteen kinds are these forms of parts, the Eleatic Stranger
indicates by comparing them to the "limbs" (mele) of an animal; that this set
of parts is normative, he indicates by comparing them to the limbs of a
"sacrificial animal" (hiereion), a sacred offering. Toe implication is that a city
which lacks sorne one or several of the fifteen kinds of art would be
incomplete and disfigured; conversely, the city in which ali are present and in
collaborative interplay would be (to develop the way the simile of the hiereion
recalls the Stranger's myth) the model of that human self-responsibility- that
is, of the "care"-that is called for in the absence of a divine shepherd in the
age of Zeus.
2.
Let me now step back and address the second of the three questions I
articulated earlier: what is the basic force and character, the spiritual function,
of these "teacbings?" I want to offer three tentative suggestions.
236 Toe Third Way
(i) Toe notion of dogmata, "teachings" or "doctrines," tends to suggest final
answers, positions that settle questions. This, it seems to me, is one of the
underlying assumptions that motivates many of us at this conference to resist
the view that the dialogues are, as I put it earlier, "primarily bearers of
doctrine." Such a view seems to impose an alien spirit on texts that are ironic
and provocative, open-ended in their inquiry, dramatic rather than systematic
in their unity. What is therefore striking about the "unwritten teachings" (as
we have interpreted them) is that their basic function is not to resolve so much
as to orient inquiry. We have not found in them a deduction of the forms
themselves from prior causes, and we certainly have not found evidence for
a reduction of forms to numbers. On the contrary, it is the "being" of a form
as "cause of what [a sensible thing] isn that Unity and the Unlimited
collaborate to enable; the form itself remains prior to this causal function and
so is itself the basis on which we understand and interpret the interplay of
Unity and the Unlimited, not the other way around. What the "unwritten
teachings" do is to direct us toward the key questions we need to ask of forms
in order to understand their causal power. Thus, first, to understand a form as
an instantiation of Unity is to be turned to the question: what whole-part
structure does it imply for the sensibles or, more generally, the "things" in
place and time that instantiate it? Second, to understand that Unity collaborates
with the Unlimited is to be turned to the question on what continua, or in
terms of what relative opposites, are the parts apportioned to each other? And
third, to understand Unity as the ultimate source of the proportions or ratios
on those continua is to be turned to the question: how is it that those
proportions-and, an ontological level higher, the set of forms of parts that
call for them- express what is normative or "good" and (since, as the
dialogues stress, 14 the good and the proportionate are beautiful) what is
beautiful in the world?
(ii) Thus understood, the "unwritten teachings" seem to articulate the path
of dialectical inquiry on the way down. Whereas the major thrust of much of
Socrates' elenchus in the dialogues is to point to the transcendence of the
forms and to lead thought upwards from the unstable particulars and one-sided
opinions that fall short of the forms, the thrust of the "unwritten teachings" is
to turn attention to the normative order that the forms imply for their sensible
participants. This involves a renewal of interest in features that, to thinking
attempting the way up, first present themselves as obstacles to be left behind.
To bring this psychagogic course more sharply to view, consider two
important texts that exhibit thinking on the way up. In the middle books of the
Republic, Plato has Socrates chart the "conversion" of the soul from becoming
to being (518c) as a turning away from what is material and given to the
14. See, e.g. , Symposium 201c, Philebus 64e.
Toe Choice between the Dialogues and the "Unwritten Teachings" 237
senses towards what is purely intelligible. A major function of the five
ma~ematical disciplines is to teach the thinker to detach bis thought from
sensibles and to bring directly to mind the forms they image; the thinker
prepared by mathematics for dialectic will be capable, in Socrates' emphatic
Janguage, of "1;Ilaking no use at ali of anything sensible but of using forms
~en:iselves,_ go1ng through forms to forms, and ending in forms" (511c). 15
S1milarly, m the first part of the Fbrmenides, Plato presents the youthful
Socra~es mistakenly applying to the forms the sort of whole-part structure and
plurahty that belongs to sensible things. Toe lesson for critical readers is that
the forms must be grasped conceptually as different in kind from their
participants, and in the second part of the dialogue Parmenides provides the
?.cca~!on for this thinking ?Y distinguishing between the unique and simple
one of the first hypothes1s and the many composite "ones" of the second
16
hypo~esis. In contrast to these passages, the "unwritten teachings" turn our
attent10°: back ~ whole-part_ structure, to plurality in a variety of senses, and
to what 1s ~atenal and sens~ble. What makes this turn complementary rather
than contradictory, however, 1s that these are called to mind as the features not
~f forms but, specifically, of _what becomes, that is, of things in place and
time, as these are structured t,y forms. If, on the upward path, the task is to
!et the forms ~emselves emerge in their prescinsion from the features of things
m place and time, the task on the downward path is to disclose the causal
power by which forms are constitutive for things, reconstructing the order they
establish in the world.
. (ii~) ~inally, l w~t to _offer som_e very tentative thoughts 17 on the possible
1mplicat1ons of the unwntten teachings" for the "existential-ethical becoming"
that the.dialo~es a!m to incite. Too often we proceed as though to bring out
the ethical-existential aspect we must downplay the seriousness of the
metaphysical theory in the dialogues. Near the center of the Republic,
however, Plato has Socrates offer an extended reflection that undermines this
either/or:
15. This is Alan Bloom's translation, slightly altered, in The Republic of Plato (New
York: Basic Books, 1968).
16. ~or extended argument for this reading of the Pannenides, see Miller, Plato 's
Parmerudes.
17. A fu~er ~uestion, too complex to take up here, is the relation of the spectacle of
~e Go~d, w1th 1ts. power to form the philosopher's character (for discussion, see my
P!ato_ruc :z-ovocat10ns: Reflections on the Soul and the Good in the Republic," Platonic
Invest~gatwns, ed. Dominic O'Meara [Washington, D.C.: Toe Catholic University of
Amenca Press, 1985], 163-193), to the account of the causal power of the One in the
"~nwritten_ teachings. " That there is a connection to be pursued is implied by the
alignment m teaching [3] of the One with what is good in the condition of things (to eu).
 Toe Third Way
238
_ Por, presumably, Adeimantus, a man who has his understanding _truly tumed
toward the things that are has no leisure to look down toward the affarrs of_h~an
beings and, as a result of fighting with them, to be filled with envy and ill will.
Rather because he sees and contemplates things that are set in a regular
arrang~ment and are always in the same condition-things that neither do injust~ce
to one another nor suffer it at one another's hands but remain all in order accordmg
to reason-he imitates them and, as much as possible, makes himself like them. Or
do you suppose there is any way of keeping someone from imitating that which he
admires and therefore keeps company with?
- It's not possible, [Adeimantus] said.
_ Toen it's the philosopher, keeping company with the divine and the orderly, who
18
becomes orderly and divine, so far as it is possible for a human being. (500c-d)
Socrates's key notion is the "regular arrangement" (taxis, implicit in
tetagmena, 500c2) and "order" (kosmos, 500c4, c9) that disti~shes the
forros from the "affairs of human beings;" it is this that so deeply 1mpresses
the pbilosopher that he is moved to "imi~te" it, becoming _"o:derly ,,a~d
divine" in bis very character. What the spec1fic character of this order is,
however, Socrates leaves open. That the forms do not "do injustice to one
another" suggests only, in the context of bis definition of justice in Book N,
that none impedes the work proper to any other. How, positively, does each
in doing its own work function collaboratively with each oth~r? My
interpretation of the "unwritten teacbings" in section I ab_ove prov1~es the
outline of an an5Wer to this question; it also shows how, as 1s appropnate for
dialectic on the way down, the "teachings" extend the order this cofunctioning
exhibits into the domain of things in place and time. We can bring this back
to view one last time by marking four basic respects in wbich, in the
encompassi~ structure that inquiry oriented by th: "unwritten teacbi°?s"
discloses, there is a distinctive harmonizing of oppos1tes. [l] For any subJect
matter there will be a vertical axis reacbing from the one form to the
unlimi~edly many possible proportions of the opposites on the continuum;
between these extremes stand the limitedly many forms of parts or "limbs"
that, an5Wering to the one form as form of the whole, pick out the appropriate
proportions on the continuum. 19 [2] Toe continuum itself i~ the balanceó
series of possible proportions between the relevant oppos1tes. [3] As a
mathematical structure, the continuum stands between being, that is, the one
form and the set of forros of parts it implicates, and becoming, the many
things in place and time; the set of proportions picked out on the continuum
by the forms of parts, since tlley stand as normative specifications for the parts
18. Again, this is Bloom's translation in The Republic of Plato, slightly altered.
19. Por the triad of one, limited many, unlimited many, see Philebus 16d. Por the
universality Plato has Socrates claim for this schema, see 17d-e.
Toe Choice between the Dialogues and the "Unwritten Teachings" 239
of sensibles, bring being into the very heart of becoming. [4] This integration
itself expresses the collaboration between Unity, instantiated by the one forro's
implication of the forros of parts, and the Unlimited, instantiated concretely
and a~stractly, respectively, by the intrinsically indeterminate play of
contranes that tlle one form first delimits and by the continuum. Toe power
and elegance of this whole structure is striking. Doesn't this make ali the more
compelling Socrates's claim that the pbilosopher's contemplative experience
must have a deeply forroative influence on bis character? Just insofar as it is
the "divinity" of the "order" the pbilosopher "keeps company with" that
exercises this influence, the discovery of the all-encompassi~ order articulated
by the "unwritten teacbings" will only deepen it. Turning back to the domain
of what becomes, he will be opened to the full spectrum of possibilities,
appreciative of the gift of norroative order wbich the forms provide, and
arrested by the fundamental harmonies tllat first constitute the possibility of
this gift. In the grip of this experience, will he not already be moved to live
accordingly, to make bimself, in bis own character, an analogue to the divine
whole he contemplates?
3.
We come, finally, to the third question: what are the implications of the
discovery of the "unwritten teacbings" in the dialogues for our conception of
the basic character of the dialogue forro and for our understanding of how to
read and interpret Plato?
It seems evident, first of ali, that we need to broaden our sense of the
context in wbich we interpret the dialogues (or, at least, those dialogues in
which we have found the "unwritten teacbings" at work)20 to include the
conversations that were going on in tlle Academy. Without Aristotle's report
~n Metap!t>'sics A~. we might never have had occasion to seek the "teacbings"
m the dialogues m tlle first place. But once we do find them, we see in
retrospect tllat not to have looked for them would have been to bypass
something very important. So, it seems, to be adequate to the tllought in the
dialogues requires, paradoxically, tllat we look to the Academy.
20. Should this list include all of the dialogues (as e.g., the Tübingen school, taking
an essentially unitarian approach, would argue) or only a selection of the "later"
dialogues (the approach exemplified by Sayre in Plato 's Lau Ontology)? My preference
~ to leave this general question open and to take each dialogue as much as possible in
1ts own terms, letting its dramatic framing, especially, provide the starting-point for
determining the context in which it should be read and tuming to the "unwritten
teachings" where reading in that context makes •it appropriate.
240 Toe Third Way
This way of putting the matter seems to revive the problem of Scylla and
Charybdis. Am I saying, after ali, that to study the "unwritten teachings" we
must look away from the dialogues?21 If discovering the "teachings" in the
dialogues is possible only if we respond to reports extemal to the dialogues,
doesn't our "discovery" amount to reading into the texts something that they
themselves do not intend-and, as a consequence, obscuring what they do
intend? Haven't we, after ali, been seized by Scylla?
I think the answer is no. To begin with, the dialogues themselves, in their
dramatic dimension, do invite us to look to the Academy. A number of them,
at least, give us characters or settings that surely mime ironically characters
and perspectives in play in the Academy. Think only of the figure of the
youthful Socrates in the Fbrmenides, paradigmatic philosopher-in-the-making.
Is he not, with his seminal distinction between forms and things and his
inexhaustible energy for inquiry, the very embodiment of the philosophical
aspirations that must have characterized many of the young Academicians? In
constructing this persona dromatis, is Plato not mirroring back to them both
their hopes and limitations? Or consider the personae "Theodorus" and
"Young Socrates" in the Eleatic dialogues. Theodorus, "our greatest
mathematician and geometer" (tou peri logismous kai ta geometrika krotistou,
Statesman 257a7-8), represents expertise in those studies that, according to
Republic VII, are propaideutic to dialectic, studies that we know were a major
occupation in the Academy. His presence in the dialogues also reminds us of
the presence in the Academy of the still greater master of geometry, Eudoxus.
Young Socrates, in tum, is Plato's remake of an older member of the
Academy well-known, apparently, for his expertise in legal or constitutional
matters. 22 By recasting him as a young student of Theodorus and an eager
and respectful interlocutor for the Eleatic Stranger and (at least potentially)23
the elder Socrates, Plato puts before the Academicians a figure who combines
the two basic projects that, in ali likelihood, drew many of them to the
Academy in the first place: ascending through mathematics to dialectic and,
with dialectical inquiry as the key resource, rethinking the basis and order of
21. Toe following comments concentrate on the problem of whether interpreting the
dialogues will lead us to the "unwritten teachings." Notice that on my view there is no
question about the converse, whether interpreting the "unwritten teachings" leads us to
the dialogues. It is only in the context of the Parmenides, Philebus, and Statesman that
I have been able to find a rich and determínate Platonic significance for Aristotle's
reports in Metaphysics A6.
22. See the allusion in the Second Letter, 358d-e. Note also Aristotle, Metaphysics
Zll:1036b25.
23. Was Young Socrates to be the elder Socrates' interlocutor in the projected
Philosopher? See Cornford's interesting speculations (building on Dies) in Plato 's Theory
of Knowledge (London: Routledge & Kegan Paul, 1935), 168-169.
Toe Choice between the Dialogues and the "Unwritten Teachings" 241
political life. Clearly, the mimetic functions of these personae are pointed and
deliberate. This implies in tum that Plato targets the young Academicians as
bis readers of "first intention"24 and invites them to interpret the dialogues
against the background of what was going on in the Academy.
Does this put us, standing at such a distance from the Academy and
knowing so little about it, at an insuperable disadvantage? For a complicated
set of reasons, I think not. Close study of the comportment of the personae
just mentioned suggests a surprisingly complex situation in the Academy. 25
On the one hand, the general atmosphere in the conversations in the trilogy of
the Theaetetus, Sophist, and Statesman is cordial and friendly; there is no trace
of the personal hostility and contentiousness that is such a conspicuous feature
in the dramatic action of so many other dialogues. On the other hand, there is
pervasive evidence of what we might call substantive alienation, of a lack of
comprehension that separates the would-be philosophers from philosophy itself.
In the Theaetetus, for instance, Theodorus confesses to an antipathy for
"abstract discussions" (165a) and repeatedly tries to avoid Socratic elenchus,
appealing to Socrates (who is himself seventy) to let him off on grounds of age
(see 146b, 162b, 164eff., 168eff., 177c, 183c). Theaetetus, in tum, never
capitalizes on bis powerful mathematical education to recognize that knowledge
requires intelligible, not sensible, objects. In the Statesman, even more
strikingly, Young Socrates twice leaps to agree that the search for
statesmansbip is complete when in fact the Stranger does not think so at ali
(267a and c, 277a), mistakenly presumes bis own understanding of the
Stranger's method of bifurcatory diairesis (262a), and never asks the Stranger
for an explanation of bis sudden abandonment of bifurcation in the passage we
studied earlier (see 287cff.). In all of these passages he is authoritarian in bis
deference. Recognizing this, the Stranger warns bim that the time may come
someday when (no longer able to defer to a guiding elder) he will grow
impatient with the discipline of dialectical method, and he provides Young
Socrates with a "prophylactic"26 doctrine with wbich to defend bimself
(283bff.); true to forro, Young Socrates accepts the doctrine with no explicit
reflection on bis potential need of it. Ali of this suggests difficulties in the
24. This is R.E. Allen's useful phrase, in his Plato 's Pannenides: Translation and
Analysis (Minneapolis: University of Minnesota Press, 1983), 197.
25. Toe following paragraph sketches the account I have worked out in detail in my
The Philosopher in Plato 's Statesman (Toe Hague: Martinus Nijhoff, 1980).
26. This is J. B. Stemp's translation in Plato: Collected Dialogues, ed. Edith Hamilton
and Huntington Cairns (Princeton University Press, 1961). Toe doctrine is that of due
measure.
242 Toe Third Way Toe Choice between the Dialogues and the "Unwritten Teachings" 243

Academy. 27 Putting the Academicians on stage befare themselves, Plato "guidance" expressed in those "very briefest statements" of his deepest
seems to point both to their unreadiness for dialectical inquiry into the forros roetaphysical insights.
and, what greatly complicates the problem this poses, their failure to recognize lf it is right to take the dialogues to be both ironic mimeses of conversation
it. in the Academy and acts of indirect communication within that conversation
How did Plato respond to such a situation? If it is reasonable to take as an then taking the context of the Academy seriously does not put us at a~
analogue the way he malees his protagonists in these dialogues respond, then insuperable disadvantage, and taking the "unwritten teachings" seriously does
we can answer: with irony and indirect communication, aimed at provoking not put us at the merey of Scylla. On the contrary, like the Academicians
his associates to self-movement. 28 lndeed, the dramatic depiction of their themselves. we have, as one kind of resource, something very much like
unreadiness is itself such provocation. But if this is right, then not only are the Plato's "briefest statements" - tbr this is just what Aristotle's reports in
dialogues good clues to the kind of conversation that was going on in the Metaplrysics A6 amount to. And again like the Academicians themselves we
Academy (our initial point), but also they must themselves have been elements have, as another kind of resource, the indirect and ironic communication of the
within the conversation. On this interpretation of the situation, we can imagine dialogues. What is needed is to bring these together in inquiry.
that Plato was as reticent as the language of the Seventh Letter suggests,
reserving expression of his deepest metaphysical insights for "those few
capable of discovering the truth for themselves with a little guidance"29
(341e)-and even then restricting himself to "the very briefest statements"
(bra.chutatois, 344e2) needed to provide that "guidance." Otherwise and for
the most part, he would have spoken indirectly, either face to face with the
same sort of pointed inexplicitness we have seen in the Rlrmenides and
Statesman passages we studied earlier or by way of the publication of the
dialogues themselves. In such indirection, he would have been sustained by the
hope that those who felt his irony would awaken, finding in the dialogues'
portrayals, in particular, of crucial limitations of character and thought
occasions for self-recognition and self-transcendence. Such self-knowledge
would be the decisive preparation for making good use, in turn, of the

27. There is an analogous argument to be made regarding the Parmenides. By his

refutations in the first part of the dialogue, Parmenides shows Socrates that he does not
yet understand his own distinction between forms and things. He exposes Socrates as still
too much in the grip of "the opinions of men" (130e) and as "attempting to define
beautiful andjust and good and each one ofthe forms too soon, before [he] has had (the]
preliminary training [he needs]" (135c-d). On the other hand, the youthful Socrates of
the Parmenides differs from Young Socrates in the Statesman by becoming aporetic and
asking for help. This makes it ali the more striking, however, that Parmenides, even as
he agrees to Socrates's request, immediately replaces him with the young "Aristotle" as
his interlocutor; young "Aristotle," now excessively compliant (see 137b), will later
become one of the most vicious of the Thirty Tyrants (see 127d). Por interpretation, see
my Plato's Parmenides, 78 with, e.g., 89-91, 144-145.
28. See R. Schaerer, "Le mécanisme de l'ironie dans ses rapports avec la dialectique,"
Revue de metaphysique et de morale 48 (1941): 181-209.
29. This and the following quotes are drawn from L.A. Post's translation in Plato's
Co/lected Dialogues.
244
Appendix: Reconstruction of the List of Fifteen Kinds

(StaJesman 287b-2091a, 303d-305e)

the form of the whole: "care" for the whole human community
fl\ Select Bibliography of Works
/ 1 \
"contributory slave's "directly responsible Pertaining to the "Third Way"
arts" art arts"
/ \
/ \
/ \ . Toe fullowing is a selection of works relevant to the "third way" in
the forms ofparts: the different forros in which it is pursued by the contributors to this volume.
1, 2, 3 , 4, 5, 6, 7, 8,9,10,11,12,13, 14, 15
nn, t, c, v, d, a, n,sl,m, c, p, r, g, j, st
Different Ways of Interpreting Plato

the continuum between opposites and the proportions picked out by the Albert, Karl. "Zuro Philosophiebegriff Platons." Gymnasium 99 (1992) : 17-
forms of parts: 33.
1----1--1--1----1-- 1-- 1-----1-----1--1--1---- 1--1-- 1----1
the material the material and the spiritual the spiritual Annas, Julia. "Platon le sceptique. " Revue de Meta,physique et Mora.le 95
in predominance in equal balance in predominance (1990): 267-291.
over the spiritual over the material
- -. "Plato the Sceptic." In Methods of Interpreting Plato and his
Dialogues, Oxford Studies in Ancient Philosophy, suppl. vol., ed.
particulars in place and time: James C. Klagge and Nicholas D. Smith, 43-72. Oxford: Clarendon
Press, 1992.
the arts in any actual city that, to instantiate well the forro
"care" and so constitute a good city, must embody the Barnes, J. "Toe Hellenistic Platos." Apeiron 24 (1991): 115-28.
various norroative balances picked out by the fifteen kinds
- -. "Imperial Plato. " Apeiron 26, v. 2 (1993): 129-151.

Berti, Enrico. "Strategie di Interpretazione dei Filosofi Antichi: Platone et

Aristotele." Elenchos 10 (1989): 289-315.

Bowen, Alan C. "On Interpreting Plato. " In Platonic Writings, Platonic

1 Readings, ed. Charles L. Griswold Jr., 49-65. New York: Routledge,

246 Toe Third Way
Chapman & Hall, Inc., 1988.
Brumbaugh, Robert S. Plato for the Modem Age. New York: Crowell-Collier,
1962.
Brumbaugh, Robert S. "Four Types of Plato Interpretation." In Plato 's
Dialogues: New Studies and Interpretations, ed. Gerald A. Press,
239-248. Lanham, Maryland: Rowman & Littlefield Publishers, Inc.,
1993.
Coulter, J. A. The Literary Microcosm: Theories of Interpretation of the Later
Neop/atonists. Leiden: Brill, 1979.
Derrida, Jacques. "La Pharmacie de Platon." In La dissémination, 69-198.
Paris: Éditions du Seuil, 1972.
Festugiere, A. J. "L'ordre de lecture des dialogues de Platon aux Vº/VI•
siecles." Museum Helveticum 26 (1969): 281-96.
Gaiser, K. "Platonische Dialektik-damals und heute." Gymnasium, Beiheft
9 (1987): 77-107.
Goldschimdt, Víctor. Les dialogues de Piaron: structure et méthode
dialectique. Paris: Presses Universitaires de France, 1947.
Griwold, Charles L., Jr. "Gadamer and the Interpretation of Plato." Ancient
Philosophy 1 (1981): 171-8.
- -- ., ed. P/atonic Writings, P/atonic Readings. New York: Routledge,
Chapman & Hall Inc., 1988.
Hermann, Karl Friedrich. Geschichte und System der Philosophie. 1839.
Reprint, New York: Amo Press, 1976.
Howland, Jacob. "Re-reading Plato: Toe Problem of Platonic Chronology
Reconsidered." Phoenix 45 (1991): 189-214.
- - . "Philosophy as Dialogue." Reason R:Jpers 17 (fall 1992): 113-134.
Jaeger, W. "Der Wandel des Platonbildes im neunzehnten Jahrhundert." In
Humanistische Reden und \brtrage, pp. 138-52. Berlin and Leipzig:
W. de Gruyter, 1937.
Bibliography 247
Kramer, H. J. "Kritsche Bemerkungen zu den jüngsten Ausse~en von w.
Wieland and G. Patzig über Platons Ungeschriebene Lehre." Revista,
di FilosQ/ia neoscolastica 74/4 (1982): 579-92.
- - - . "Fichte, Schlegel und der Infinitismus in der Platondeutung."
Deutsche Vierteljahrsschrift für die Literaturwissenschaft und
Geistesgeschichte 62 (1988): 583-621.
- - . "Zur Aktuellen Diskussion um den Philosophiebegriff Platons."
Perspektiven der Philosophie 19 (1990): 85-107.
- - . Plato and the Foundations of Metaphysics. Ed. and trans. John R.
Catan. Albany: State University of New York Press, 1990.
- - . "Altes und neues Platonbild." Méthexis 6 (1993): 95-114.
Lafrance, Y. "Autour de Platon: Continentaux et analystes." Dionysius 3
(1979): 17-37.
- - . "L'avenir de la recherche Platonicienne." Revue des Etudes Grecques
99 (1986): 271-292.
Long, A. A. and Sedley, D. N. The Hellenistic Philosophers. Vol. l.
Cambridge: Cambridge University Press, 1987.
Nails, Debra. "Platonic Chronology Reconsidered." Bryn Mawr C/assical
Review 3 (1992): 314-27.
- - -. "Problems with Vlastos' Platonic Developmentalism." Ancient
Philosophy 13 (1993): 273-91.
- - . "Plato's Middle Cluster." Phoenix 48 (1994): 62-67.
Press, Gerald A., ed. P/ato's Dialogues: New Studies and Interpretations.
Lanham, Maryland: Rowman & Littlefield Publishers, Inc., 1993.
Reale, Giovanni. A History of Ancient Philosophy. Vol. 4: The Schools of the
Imperial Age. Trans. John R. Catan. Albany: State University of New
York Press, 1990.
- - . I tre paradigmi storici nell'interpretazione di P/atone e i fondamenti
del nouvo paradigma. Napoli: Instituto Suor Orsola Benincasa, 1991.
248 Toe Third Way
Rowe, C. J. "On Reading Plato." Méthexis 5 (1992): 53-68.
Schaerer, René. La Questionne Platonicienne. 2d. ed. revised and augmented
with a postscript, A la Recherche de Platon. Neuchatel: Secretariat de
l'Université, 1969.
Schleiermacher, Friedrich. Platons W!rke. 3rd ed. Berlin: Georg Reimer,
1857.
- - - . Introductions to the Dialogues of Plato. Trans. William Dobson.
1836. Reprint, New York: Amo Press, 1973.
Stefanini, Luigi. Platone. Padova: CEDAM, 1932-1935. 2nd ed., Padova:
CEDAM, 1949. Reprint, 1992.
Strauss, Leo. "On a New Interpretation of Plato's Political Philosopby."
Social Research 13 (1946): 326-367.
Szlezák, T. A. "Platon und die neuzeitliche Theorie des Platonischen
Dialogs." Elenchos 10 (1989): 337-357.
- - . Come leggere Platone. Milan: Rusconi, 1991.
Tarrant, Harrold. Thrasyllan Platonism. Ithaca: Cornell University Press,
1993.
Tejera, Victorino. "Toe Hellenistic Obliteration of Plato's Dialogism." In
Plato's Dialogues: New Studies and Interpretations, ed. Gerald A.
Press, 129-144. Lanham, Maryland: Rowman & Littlefield
Publishers, Inc. , 1993.
Thesleff, H. Studies in Platonic Chronology. Helsinki: Societas Scientiarum
Fennica, 1982.
- - - . "In Search of Dialogue." In Plato's Dialogues: New Studies and
Interpretations, ed. Gerald A. Press, 259-266. Lanham, Maryland:
Rowman & Littlefield Publishers, lnc., 1993.
Tigerstedt, E. N. The Decline and Fall of the Neoplatonic Interpretation of
Plato. Commentationes Humanum Litterorum 52. Helsinki: Societas
Scientarum Fennica, 1974.
- - - . Interpreting Plato. Stockholm Studies in the History of Literoture 17.
Bibliography 249
Stockholm: Almqvist & Wiksell, 1977.
Westerink, . L. G., ed. & trans. Anonymous Prolegomena to Plato 's
Plulosophy. Amsterdam: North-Holland Publishing Company, 1962.
Young, Charles. "Plato and Computer Dating." Oxford Studies in Ancient
Philosophy 12 (1994): 227-250.
Oral versus Written Discourse and Plato 's Dialogues
Andrieu, J. Le dialogue antique, Structure et présentation. Paris: Les "Belles
Lettres," 1954.
Edelstein, L. "Platonic Anonymity. " American Joumal of Philology 83
(1962): 1-22
Harris, W. V. Ancient Literacy. Cambridge: Cambridge University Press
1989. '
Havelock, E. Preface to Plato. Cambridge: Harvard University Press, 1963.
. "Toe Orality of Socrates and the Literacy of Plato." In New Essays
on Socrotes, ed. Eugene Kelly, 67-93. Lanham, Maryland: University
Press of America, 1984.
Hirzel, Rudo~f. Der Dialog, Ein literorhistorischer lérsuch. 2 vols. Leipzig:
S. Htrzel, 1895.
Jürss, F. "Platon und die Schriftlichkeit." Philologus 135 (1991): 167-76.
Kuhn, Helmut. "Plato und die Grenze philosophischer Mitteilung." In Idee
und Zahl: Studien zur platonischen Philosophie, ed. H. G. Gadamer
and W. Scbadewaldt, 151-173. Heidelberg: C. Winter, 1968.
Robb, Kevin. Literocy and Fbdeia in Ancient Greece. Oxford: Oxford
University Press, 1994.
Ryle, Gilbert. Plato 's Progress. Cambirdge: Cambridge University Press
1966. '
Tecusan, ~- ''.Logos Sympotikos: Patterns of the Irrational in Philosophical
Dnnking: Plato Outside the Symposium." In Sympotica. A Symposium
250 Toe Third Way
on the Symposion, ed. O. Murray, 229-260. Oxford: Clarendon
Press, 1990.
Thomas, R. Oral Tradition and Written Record in Classical Athens.
Cambridge: Cambridge University Press, 1989.
- - . Literacy and Orality in Ancient Greece. Cambridge: Cambridge
University Press, 1992.
Philosophy anti Rhetoric
Allen, R. E. "Irony and Rhetoric in Plato's Apology." füideia 5 (1976): 32-
42.
Benitez, Eugenio E. "Argument, Rhetoric, and Philosophical Method: Plato's
Protagoras." Philosophy and Rhetoric 25 (1992): 222-252.
Benitez, Eugenio E., and Livia Guimares. "Philosophy as Performed m
Plato's Theaetetus." Review of Metaphysics 47 (1993): 297-328.
Brickhouse, T. C., and N. D. Smith. "Irony, Arrogance and Sincerity in
Plato's Apology." In New Essays on Socrates, ed. Eugene Kelly, 29-
46. Lanham, Maryland: University Press of America, 1984.
Cole, A. T. The Origins of Rhetoric in Ancient Greece. Baltimore: Johns
Hopkins University Press, 1991.
Coventry, L. "Philosophy and Rhetoric in the Menexenus." Joumal of
Hellenic Studies 109 (1989): 1-15.
Farness, J. "Missi~ Socrates: Socratic Rhetoric in a Platonic Text."
Philosophy and Rhetoric 20 (1987): 41-59.
Feaver, D. D., and J. E. Hare. "Toe Apology as an Inverted Parody of
Rhetoric." Arethusa 14 (1981): 205-216.
Roochnik, David. "Socrates's Use of the Techne-Analogy." Joumal of the
History of Philosophy 24 (1986): 295-310.
- - . "Terence Irwin's Reading of Plato." In Platonic Writings, Platonic
Readings, ed. Charles L. Griswold Jr., 183-193. New York:
Routledge, Chapmann & Hall, 1988.
Bibliography 251
- - . "Toe Tragic Philosopher: A Critique of Martha Nussbaum." Ancient
Philosophy 8 (1989): 285-99.
. The Tragedy of Reason: Toward a Platonic Conception of Lagos.
New York: Routledge, Chapman & Hall Inc., 1990.
Rossetti, Livio. "Rhétorique des Sophistes - Rhétorique de Socrate." In ¡
Archea Sophistiki I The Sophistic Movement, 137-45. Papers Read at
the First lntemational Symposium on the Sophistic Movement.
Organized by the Greek Philosophical Society, 27-29 Sept. 1982.
Athens: Kardamitsa, 1984.
- - - . "Toe Rhetoric of Socrates." Philosophy and Rhetoric 22 (1989): 225-
238.
- - - . "Sulla dimensione retorica del dialogare socratico." Méthexis 3
(1990): 15-32.
- - - . "La dimensión retórica de los mitos platónicos." Revista de Occidente
(1994): 72-91.
Vickers, Brian. In Defence of Rhetoric. Oxford: Oxford University Press,
1988.
Zyskind, Harold. "Plato's Republic Book I: An Equitable Rhetoric."
Philosophy and Rhetoric 25 (1992): 205-221.
The Relation of Plato's Philosophy
to Myth, Drama, anti "Literature"
General
Arieti, J. A. Interpreting Plato: The Dialogues as Drama. Savage, Maryland:
Rowman & Littlefield, 1991.
Berger, Harry Jr. "Levels of Discoun;e in Plato's Dialogues." In Literature
and the Question of Philosophy, ed. Anthony J. Cascardi, 75-100.
Baltimore: Johns Hopkins University Press, 1987.
Desjardins, Rosemary. "Why Dialogues? Plato's Serious Play." In Platonic
Writings, Platonic Readings, ed. Charles L. Griswold Jr., 110-125.
New York: Routledge, Chapman & Hall, 1988.
252 Toe Third Way
Edelstein, L. "Toe Function of the Myth in Plato's Philosophy." Joumal of
the History of Philosophy 10 (1949): 463-81.
Famess, Jay. Missing Socmtes: Problems of Plato's 1Mitings. University Park:
Pennsylvania State University Press, 1991.
Frede, M. "Plato's Arguments and the Dialogue Form. " In Methods of
lnterpreting Plato and his Dialogues, Oxford Studies in Ancient
Philosophy, suppl. vol., ed. James C. Klagge and Nicholas D. Smith,
201-219. Oxford: Clarendon Press, 1992.
Friedliinder, Paul. Plato: An Introduction. Tr. H. Meyerhoff. Na.v York:
Bollingen Foundation, 1958.
Frutiger, P. Les mythes de Platon: Etude philosophique et littémire. Paris: F.
Alean, 1930.
Griswold, Charles L., Jr. "Style and Philosophy: Toe Case of Plato's
Dialogues." Monist 63 (1980): 530-46.
- - - . "Irony and Aesthetic Language in Plato's Dialogues." In Philosophy
and Litemture, ed. Doug Bolling, 71-99. Na.v York: Haven
Publications, 1987.
Krentz, Arthur A. "Dramatic Form and Philosophical Content in Plato's
Dialogues." Philosophy and Literoture 7 (1983): 32-47.
Laborderie, Jean. Le Dialogue Platonicien de la Maturité. Paris: Société
d'édition 'Les Belles Lettres,' 1978.
Levi, A. W. "Philosophy as Literature: Toe Dialogue." Philosophy and
Rhetoric 9 (1976): 1-20.
Mattéi, Jean-Francois. "Toe Theater of Myth in Plato." Platonic 1Mitings,
Platonic Readings, ed. Charles L. Griswold Jr., 66-83. New York:
Routledge, Chapman & Hall, 1988.
McCabe, Mary Maigret. "Myth, Allegory and Argument in Plato." Apeiron
25 (1992): 47-67.
McDonald, J. M. S. Charocter-lbrtmiture in Epicharmus, Sophron, and
Plato. Sewanee, Tennessee: Toe University Press, 1931.
Bibliography 253
Merlan, P. "Form and Content in Plato's Philosophy." Joumal of the History
of Ideas 8 (1947): 406-30.
Moors, Kent F. Platonic Myth: An Introductory Study. Washington, DC:
University Press of America, 1982.
Plass, P. "Philosophic Anonymity and Irony in the Platonic Dialogues."
American Joumal of Philology 85 (1964) : 254-78.
Press, Gerald A. "Principies of Dramatic and Non-Dogmatic Plato
Interpretation." In Plato's Dialogues: New Studies and
Interpretations, ed. Gerald A. Press, 107-127. Lanham, Maryland:
Rowman & Littlefield Publishers, Inc., 1993.
Randall, J. Plato: Dromatist of the Life of Reason. Na.v York: Columbia
University Press, 1970.
Reinhardt, Karl. Platons Myth(m. Bonn: 1927.
Rossetti, Livio. "Where Philosophy and Literature Meige in the Platonic
Dialogues." Argumentation 6 (1993): 433-443 .
Schaerer, R. "Le mecanisme de I'ironie dans ses rapports avec la dialectique. "
Revue de metaphysique et de momle 48 (1941): 181-209.
Stewart, A. The Myths of Plato. 1905. Reprint, Carbondale: Southem Illinois
University Press, 1960.
Thayer, H. S. "Meaning and Dramatic Interpretation." In Plato's Dialogues:
New Studies and Interpretations, ed. Gerald A. Press, 47-59.
Lanham, Maryland : Rowman & Littlefield Publishers, Inc., 1993.
Thesleff, H. Studies in the Styles of Plato. Helsinki: Societas Philosophica
Fennica, 1982.
. "Looking for Clues. An Interpretation of Sorne Literary Aspects of
Plato's 'Two Leve! Model. "' In Plato 's Dialogues: New Studies and
Interpretations, ed. Gerald A. Press, 17-45. Lanham, Maryland:
Rowman & Littlefield Publishers, Inc., 1993.
De la Vega, J. S. Lasso. "El Diálogo y la Filosofía Platónica." Estudios
Classicos 54 (1968) : 311-74.
254 Toe Third Way
Wolz, H. G. "Philosophy as Drama: An Approach to Plato's Dialogues."
International Philosophical Quarterly 3 (1963): 236-270.
Woodbridge, Frederick J. E. The Son of Apollo. Boston: Houghton Mifflin,
1929. Reprint, New York: Columbia University Press, 1965.
Zaslavsky, R. Platonic Myth and Platonic Writing. Washington, DC:
University Press of America, 1981.
lnterpretations of Specific Dialogues
Brague, R. Le restant: Supplément aux commentaires de Ménon du Platon.
Paris: J. Vrin, 1978.
Brumbaugh, Robert S. "Toe Meno in Secondary Schools." Teaching
Philosophy 1, no. 2 (fa.U 1975): 107-15
Burger, R. Plato's Phaedrus: A Defense of a Philosophic Art of Writing.
University, Alabama: University of Alabama Press, 1980.
Canto, Monique. L'Intrigue Philosophique. Paris: J. Vrin, 1987.
Eckstein, Jerome. The Platonic Method: An lnterpretation of the Meno. New
York: Greenwood, 1968.
Gonzalez, Francisco J. "Plato's Lysis: An Enactment of Philosophical
Kinship." Ancient Philosophy 15 (1995): 69-90.
Griswold, Charles L. , Jr. Self-Knowledge in Plato 's Phaedrus. N ew Haven:
Yale University Press, 1986.
Hyland, Drew. The Virtue of Philosophy: An lnterpretation of Plato's
Charmides. Athens: Ohio University Press, 1981.
Klein, J. A Commentary on Plato's Meno. Chapel Hill: University of North
Carolina Press, 1965.
Kosman, L. A. "Charmides' First Definition: Sophrosyne as Quietness." In
Essays in Ancient Greek Philosophy, vol. 2, ed. John P. Anton and
Anthony Preus, 203-216. Albany: State University of New York
Press, 1983.
Bibliography 255
Narcy, Michel. Le Philosophe et Son Double: Un Commentaire de
l'Euthydeme de Platon. Paris: J. Vrin, 1984.
O'Brien, M. "Toe Unity of the Laches." In Essays in Ancient Greek
Philosophy, ed. J. P. Anton & G. L. Kustas, 303-315. Albany: State
University of New York Press, 1971.
Rosen, Stanley. Plato's Sophist. New Haven: Yale University Press, 1983.
Schmid, Walter T. 0n Man/y Coumge: A study of Plato's Laches. Carbondale
and Edwardsville: Southern Illinois University Press, 1992.
Sternfeld, Ro~e~ and Harold Zyskind. Plato's Meno: A Philosophy of Man as
Acquzsztzve. Carbondale: Southern Illinois University Press, 1978.
Stokes, M. ~ocmtic Conversations: Dmma and Dialectic in Three Dialogues.
Balttmore: Johns Hopkins University Press, 1986.
Weingartner, R. The Unity of the Platonic Dialogue. Indianapolis: Hackett
1973. '
Dialectic and Dialogue
Bakhtin, M. M. The Dialogic Imagina/ion. Ed. Michael Holquist. Trans.
Caryl Emerson and Michael Holquist. Austin: University of Texas
Press, 1981.
Gadamer, H. G. Dialogue and Dialectic. Trans. P. Christopher Smith. New
Haven: Yale University Press, 1980.
. Plato 's D_ialectical Ethics: Phenomenological Interpretations Relating
to the Philebus. Trans. R. M . Wallace. New Haven: Yale University
Press, 1991.
Griswold, Charles L, Jr. "Reflections on 'Dialectic' in Plato and Hegel."
International Philosophical Quarterly 22 (1982): 115-130.
. "Plato's Metaphilosophy: Why Plato Wrote Dialogues." In Platonic
Writings, Platonic Readings, ed. Charles L. Griswold Jr., 143-167.
New York: Rouúedge, Chapman & Hall, 1988.
Gould, John. The Development of Plato's Ethics. New York: Russell &
256 Toe Third Way
Russell, 1955.
Gundert, Hermann. Der Platonische Dialog. Heidelberg: Carl Winter, 1968.
- - - . Dialog und Dialektik. Amsterdam: Verlag B. R. Grüner, 1971.
- - - . Platonische Studien. Amsterdam: Verlag B. R. Grüner, 1977.
Hyland, Drew. "Why Plato Wrote Dialogues." Philosophy and Rhetoric 1
(1968): 38-50.
Mittelstrass, Jürgen. "On Socratic Dialogue." In Platonic Writings, Platonic
Readings, ed. Charles L. Griswold Jr., 126-142. New York:
Routledge, Chapman & Hall, 1988.
Rorty, Richard. Philosophy and the Mirror of Nature. Princeton: Princeton
University Press, 1979.
Rosen, Stanley. The Limits of Analysis. New York: Basic Books, 1980.
Sallis, J. Being and Logos: The Wzy of Platonic Dialogue. Pittsburgh:
Duquesne University Press, 1975.
Sayre, Kenneth M . "Plato's Dialogues in the Light of the Seventh Letter." In
Platonic Writings, Platonic Readings, ed. Charles L. Griswold Jr.,
93-109. New York: Routledge, Chapman & Hall, 1988.
Sinaiko, H. L. Love, Knowledge, and Discourse in Plato: Dialogue and
Dialectic in Phaedrus, Republic, Parmenides. Chicago: University of
Chicago Press, 1965.
Smith, N.D. "Knowledge by Acquaintance and 'Knowing What' in Plato's
Republic." Dialogue 18 (1979): 281-88.
Stemmer, Peter. Platons Dialektik: Die Früheren und Mittleren Dialoge.
Berlín: Walter de Gruyter, 1992.
Tejera, Victorino. "Methodology of a Misreading: A Critica! Note on T.
Irwin's 'Plato's Moral Theory. "' International Studies in Philosophy
10 (1978): 131-136.
- - - . Plato's Dialogues One by One: A Structuml Interpretation. New
York: Irvington Publishers, 1984.
Bibliography 257
Watson, Walter. "The Voices of the God." In New Essays on Socmtes, ed.
Eugene Kelly, 173-179. Lanham, Maryland: University Press of
America, 1984.
- - - . Architectonics of Meaning. 2d ed. Chicago: University of Chicago
Press, 1993.
Wieland, Wolfgang. "Platon und der Nutzen der Idee. Zur Funktion der Idee
des Guten." Allgemeine 'Zeitschrift jür Philosophie 1 (1976): 19-33.
- - - . Platon und die Formen des Wissens. Góttingen: Vandenhoeck &
Ruprecht, 1982.
- - . "La Cática de Platón a la Escritura y los Límites de la
Comunicabilidad." Méthexis 4 (1991): 19-37.
The Relation Between Plato's Dialogues
and "Unwritten 'Jeachings"
Berti, E. "Über das Verhaltnis von literarischem Werk und ungeschriebener
Lehre bei Platon in der Sicht der neueren Forschung. " In Das
Problem der ungeschriebenen Lehre Platons: Beitrage zum
\érstandnis der platonischen Prinizipienphilosophie, ed. J. Wippern,
88-94. Darmstadt: Wissenschaftliche Buchgesellschaft, 1972.
Ferber, Rafael. Die Unwissenheit des Philosophen oder Wzrum hat Plato die
"ungeschriebene Lehre" nicht geschrieben. Sankt Augustin: Akademia
Verlag, 1991.
- - - . "Hat Plato in der 'Ungeschriebene Lehre' eine 'Dogmatische
Metaphysik und Systematik' vertreten? Einige Bemerkungen zuro
Status Quaestionis.' Méthexis 6 (1993): 37-54.
Fritz. K. von. "Toe Philosophical Passage in the Seventh Platonic Letter and
the Problem of Plato's 'Esoteric' Philosophy." In Essays in Ancient
Greek Philosophy, ed. J. P. Anton and G. L. Kustas, 408-47. Albany:
State University of New York Press, 1971.
Gaiser, K. Protreptik und Riranese bei Platon = Tübinger Beitrage 40.
Stuttgart: Kohlhammer, 1959.
- - . Platons Ungeschriebene Lehre. 2d ed. Stuttgart: E. Klett, 1968.
258 Toe Third Way
- - - . Platone come scrittore filosofico: Sa,ggi sull' ermeneutica dei dialoghi
platonici. Naples: Bibiliopolis, 1984.
Kramer, H. J. Arete bei Platon und Aristoteles: zum ~sen una zur Geschichte
der Platonischen Ontologie. Heidelberg: C. Winter, 1959.
- - . Plato ana the Foundations of Metaphysics. Ed. and trans. John R.
Catan. Albany: State University of New York Press, 1990.
- - . "Das neue Platonbild." Zeitschrift für Philosophische Forschung 48
(1994): 1-20.
Name lndex
Miller, Mitchell. The Philosopher in Plato's Statesman. Toe Hague: Martinus
Nijhoff, 1980. Ackrill, J., 146 n.48
Adam, J. L., 179-180
- - - . Plato's Parmenides: The Conversion of the Soul. Princeton: Princeton Aeschines, 33
University Press, 1984. Aeschylus, 45, 27, 77
Alcidamas, 123, 125
Allen, R. E., 241 n.24
- - - . "Platonic Provocations: Reflections on the Soul and the Good in the
Anaxagoras, 33 n.29
Republic." In Platonic lnvestigations, ed. Dominic O'Meara, 163- Anderson, D., 140 n.32, 145 n.45
193. Washington, DC: Toe Catholic University of America Press, Andrieu, J., 27 n.9
1985. Annas, J., 3 n.3, 9 n.22, 11 n.30
Antisthenes, 33
Reale, G. Fer une nuova interpretatione di Platone: Relettum della metafisica Apelt, O., 144 n.43
dei gmndi dialoghi alle luce delle "Dottrine non scritte." 6th ed. Arcesilaus, 3, 9 n.22, 11-12 n.30,
Milan: Universita Cattolica del Sacro Cuore, 1989. 136, 189
Ardley, G., 143 n.39, 144 n.43
Arieti, J., 17, 26 n.5
- - - . "In che cosa consiste il nuovo paradigma storico-ermeneutico nelle Aristides, 82
interpretazione di Platone?" Méthexis 6 (1993): 135-154. Aristippus, 33
Aristophanes, 34, 137
Sayre, Kenneth. "Review of Kramer's Plato and the Foundations of Aristotle, 16, 20, 30, 32, 39, 53, 58
Metaphysics." Ancient Philosophy 13 (1993): 167-184. n.56, 73, 85, 87, 95- 98, 107, 111,
114 n.3, 115, 121, 131-132, 144,
Szlezák, Thomas. "Dialogform und Esoterik: Zur Deutung des platonischen 192- 193, 225 n.5, 226- 227, 228
Dialogs 'Phaidros. '" Museum Helveticum 35 (1978): 18-32. n.8, 229- 231, 239, 240 nn.21, 22,
243
Augustine, 189- 190
- - - . Platon una die Schriftlichkeit der Philosophie. Berlín: Walter de Aurelius, Marcus, 147 n.54
Gruyter, 1985. Bakhtin, M. M., 211 n.1
Bambrough, R., 146 n.48
- - - . "Zur üblichen Abneigung gegen die agropha dogmata." Méthexis 6 Barilli, R., 93 n.21
(1993): 155-174. Barnes, J., 85 n.5, 146 n.48
Benson, H., 52 n.39
Berger, H. , 13 n.33
Berti, E. , ix n.3
Bloom, A., 237 n.15, 238 n.18
Bluck, R. S., 36 n.39
Blundell, M . W. , 143 n.42
Boder, W., 143 n.39
Bowen, A., 56 n.49
Brickhouse, T. C., 142 n.35
Brocchieri, V. B., 136 n.14
Brumbaugh, R. S., 134 n.7, 173
n.32, 201 n.9, 218 n.4, 220 n.9
Buber, M ., 211 n.l
Buchler, J. , 140 n.29, 192, 211 n.l
Bury, J. B., 48 n.25
Bury, R. G., 26 n.4
Buxton, R. G. A., 138 n.24
Campbell, L., 137 n.20
Carneades, 136, 190
Carpenter, R., 62 n.2
Cervantes, M ., 141-142, 144
Chadwick, J., 62 n.2
Cherniss, H. , 114 n.3, 145 n.46
Cícero, 3- 4 , 8, 35, 82, 88 n.14, 93
n.21 , 147 n.54, 189, 191, 197
Cohen, M.S. , 8 n.20
Cole, A. T., 35-36, 46 n.17, 74-75,
151 n.59
Cook, A., 140, 144
Cornford, P. M. , 240 n.23
Coulter, J. A., 3 n.2
Coventry, L., 143 n.42
Crombie, l. M ., 6 n.13, 133 n.2
260 Toe Third Way
Cross, R. C., 6 n.13, 120 n.2
Dante, 141-142, 144
Darwin, C., 53
Davidson, D., 68 n.14, 69 n.16, 74
n.23
Demosthenes, 34 n.33
Derrida, J., 10, 65-66, 75 n.25
Descartes, R., 162 n.11
Dewey, J., 65, 147 n.55
Diderot, D., 26
Diogenes Laertius, 1, 11 n.30, 25,
27, 31, 32 n.28, 33-34
Dixsault, M., 139 n.28
Dodds, E. R., 86 n.10, 87, 91 n.17
Dover, K. J., 26 n.7, 123
Düring, l., 39
Edelstein, L., 51 n.37
Ehrenberg, V., 29 n.18
Einstein, A., 117, 121
Eisner, R., 142 n.38
Euripides, 27, 45, 49-50, 130
Ferber, R., ix n.3, 9 n.21, 186
nn.54, 55
Fine, G., 12 n.31
Fish, S., 92 n.19
Fowler, H. N., 218 n.4
Frede, M., 36, 134 n.4, 159 n.6
Freud, S., 77, 143 n.40
Friedlander, P., 52 n.39, 138 n.22
Frutiger, P.. 138 n.22
Gadamer, H. G., 10. 96 n.l, 101
n .3, 106 n.5
Gaiser, K., 7 n.17, 38, 95-96 n.1
Gentili, B., 32-33 n.29, 43 n.7 , 75
Gibbon, E., 42, 44
Goethe, J. W. von, 38
Gonzalez, F. J., 18-19, 89 n.15 , 171
n.28, 199, 225 nn.1, 4, 226 n .6
Gorgias (historical), 33, 42, 45, 82
n.3, 125-126, 219 n.6, 220
Gosling, J. , 86 n. l O
Gould, J., 183- 184 n.50
Graham, D., 53 n.41
Greene, W., 54 n.45
Grene, M., 161 n.10, 176 n.35
Griswold, C., 11 n.29, 50 n.35, 51
n.37, 55 n.48, 143 n.39, 158 n.6,
170 n.25, 171 n.27, 183 n.47
Grote, G., 218 n.4, 219 n .7
Gulley, N., 52 n.39
Gundert, H., 10
Guthrie, W. K. C., 3 n.2, 25 n. l, 27
nn.11, 12, 30 n .19, 32 n.27, 38,
43 n.5, 52 n.39, 55 n.47, 120 n.5
Hackforth, R., 50 n.34, 113
Halperin, D. , 10 n.28, 45 n.15, 51
n.36, 58 n.54, 140 n.32, 145 n.45,
149 n.58
Harris, W. V., 31, 42, 47 n.22, 50
n.34, 52 n.38, 62-63
Havelock, E. A., 30-33, 39, 43 n.6,
44- 45 n.11, 63-64, 68 n.16,
69- 72, 74, 142 n.38, 146, 151
n.59
Hawking, S., 117
Hegel, G. W. F., 200
Heidegger, M., ix n.3
Heinemann, F., 86 n.9, 93 n.20
Heisenberg, W., 117
Hermano, K. F., 5, 5-6 n.12, 7 n.15
Herodotus, 35, 48, 51
Hershbell, J., 13- 14, 136 n.16
Hesiod, 54
Hintikka, J., 184-185 n.50
Hirzel, R .. 25-27
Hoffman, E., 44 n.9
Homer, 14, 44-46, 51, 54, 57-58,
71-72, 77,226
Howland, J., 6 n.14, 7 n.15, 136
n.20
Hyland, D., 10 n.24, 183 n.48, 225
nn.1. 3
lrwin, T., 6 n.13, 7 n.15, 84 n.4, 85
n.7, 87 n.13, 156 n.3, 169 n .23
lsaeus, 34 n.33
Isocrates, 123, 125
Jaeger, W., 125 n.10, 139, 144, 146,
151 n .59
Jeffrey, L. A., 68 n.16
Jürss, F., 38 nn.49, 50
Kahn, C., 26 n.5, 45 n.12
Kant, l., 4, 115 , 139, 144
Kenyon, F., 39
Keyt, D., 8 n.20
Kierkegaard, S., 138 n.23
Kirk, G. S., 33 n .29
Name lndex
Kitto, H. D. F., 134 n.8
Klagge, J. C., 8- 9 n.20, 53 n.41, 64
n.8, 73 n.22
K.lein, J., 55 n.46, 134 nn.6 , 9, 166
n.16
Knox, B., 49 nn.30, 31
Kosman, L. A., 145 n.45, 149 n.58,
167 n.21, 172 n.30, 182 n.46
Kramer, H. J., ix n.3, 7 nn.16, 17,
176 n.36, 226 n.6, 228 n .9
Kraut, R., 6 n.13, 9 n.21
Krentz, A. A ., 157 n.4, 161 n.10
Lang, B., 135 n .11
Ledger, G., 6- 7 n.14, 53 n.40, 172
n.29, 180 n.44, 186 n.56
Lévy, C., 3 n.3
Livy, 147 n.54
Long, A. A., 3 n.3, 11 n.30
Lutoslawski, W., 137 n.20
Lyons, W., 178 n.39
MacDowell, D.M. , 34
McCabe, M. M., 6 n.13
McDonald, J. M. S., 30-31 n.21
McK.irahan, R. D. , 142 n.35
McNeal, R. A., 137 n.20
Mead, G. H., 211 n.l
Mejer, J., 25 n. l
Miliadis, J., 48 n.25
Miller, C. L., 201
Miller, M., 20-21, 134 n.5, 146
n.51, 226 n.7, 229 nn. 10, 11, 233
n.13, 237 nn.16, 17,241 n.25, 242
n.27
Mittelstrass, J., 161 n.10
Moors, K., 138 n.22
More, T., 141-142, 144
Morrow, G., 135 n.11
Muecke, D. C., 143 n.39
Müller, C. W. , 30 n.19
Murray, O., 31
Nails, D., 6 n.14, 53 n.40, 137 n .20,
145, 172 n.29
Navia, L., 142 n.35
Nehamas, A ., 139 n.28, 146 n .51,
151 n.59
Newton, l., 115
Nietzsche, F., 77
Notopoulos, 45 n.15
261
Nussbaum, M., 90 n.16, 91 n.18
O'Brien, M., 57 n.52, 170 n.26
Ong, W., 44 n.8
Parry, A., 47 n.24
Parry, M., 43 n.7, 44, 47 n .24, 62
n .2
Patzer, A., 142 n .35
Peirce, C. S., 147 n.55, 211 - 212 n. l
Picard-Cambridge, A., 27 n. 10
Pieri, S. N., 138 n.21
Philodemus, 82
Plass, P., 143 n.39
Plotinus, 97
Polansky, R. , 171-172 n . 28
Polanyi, M. , 160, 161 n.10
Post, L. A. , 242 n.29
Press, G. A., xi, 17- 18, 145 n.46,
201
Prodicus, 42, 123
Protagoras (historical), 33, 42, 125
Quintilian, 82, 88 n.14, 90 n .16, 93
n.21
Randall, J., 10 n.25, 53 n.42, 56
n.50, 111
Rankin, H. D., 144 n.43
Reale, G. , viii n.2, 3 n.2, 7 nn.16,
17
Reinhardt, K., 138 n.22
Rist, J., 183- 184 n.50
Robb, K., 31, 44 n.11, 68 n.16, 136
n.16, 146 n.50
Robinson, R.. 34 n .33, 58 n.55, 67
n.13
Robinson, T. M., 16-17, 7 n. 14
Roochnik, D .. 7 n .15, 15-16, 91
n.18, 156 n.3, 225 nn.1, 2
Rorty, A., 135 n.11
Rorty, R., 65, 67 n.12, 76- 77, 161
Rosen, S., 9, 10 n.28, 166-167 n.19
Rossetti, L., 81 n. l
Rowe, C., 143 n .39
Ryle, G., 26- 28, 31, 33, 43 n.7, 47
n.23, 61, 63-64, 72- 74, 76, 134
n .5, 178 n.39
Sayre, K. , 8 n .18, 145 n.45, 149
n.58, 176 n.35, 185 n .51, 186
n.57, 226 n .7, 239 n.20
Schaerer, R. , 10, 242 n.28
262 Toe Third Way
Schleiermacher, F., 4, 5 nn.11, 12,
37, 140 n.30, 155, 226
Searle, J. R., 133 n. 3
Segal, C., 46 nn.16, 17, 18, 58 n.54
Sextus Empiricus, 84 n.5
Shakespeare, W., 120, 127, 130
Shorey, P., 5, 137 n.19, 147 n.54
Sinaiko, H. L., 218 n.5
Skinner, Q., 67 n.13
Smith, N. D., 142 n.35, 184 n.50
Smith, P. C., 16, 96 n.l, 107 n.6
Sontag, S., 138 n.24, 150
Sophocles, 27, 130
Sophron, 30-31 n.21
Speusippus, 2 n.2, 3 n.3, 16, 114,
117
Stefanini, L., 5 nn.9, 11, 136 n.14,
190
Stemmer, P., 9 n .21
Stemp, J. B. , 241 n. 26
Sternfeld, R., 199, 201 n.9
Stewart, J. A., 138 n.22
Stokes, M. , 56 n.51, 57 n.53, 135
n.11
Strauss, L., 9, 13 n.33
Szlezák, T. A., 32 n.27, 36- 38, 226
n.6
Tuplin, O., 49 n.32, 50 n.33
Thrán, L., 114 n.3
Tuylor, A. E., 26 n.5, 30 n.20, 48
n.26, 114 n.3
Tecusan, M ., 28- 29
Tejera, V., 7 n.15, 20, 140 n.29,
145, 191-192, 219 n.7
Teloh, H., 166 n .17
Tennemann, W. G. , 2 n.l, 4
Thayer, H. S., 135 n.10
Theophrastus, 114 n.3
Thesleff, H., 6 n.14, 53 n.40, 136
n.20, 138 n.41, 146
Thomas, R., 42, 43 n.7, 46 n.17, 47
n.23, 48- 49, 53 n.43
Thompson, J. A. K., 143 n.39
Thrasylus, 27-28
Thucydides, 35, 47
Tigerstedt, E. N., 2 n.l, 4 nn.3, 4,
41, 51 n.37, 56 n.49, 136 n.14,
176 n.36, 190
Ventris, M., 62 n.2
Verdenius, W. J., 142 n.38
Vickers, B., 81 n.1 , 82, 87 n.13
Vlastos, G., 6 n.13, 52 n.39, 142
n.35, 183- 184 n.50
Vries, G. J. de, 143 n.39
Waldenfels, B., 167 n.20
Walsdorff, F., 143 n.41
Watson, W., 19- 20, 139 n.25, 148
n.57, 191 n.5, 203 n.11
Waugh, J., xi, 15, 136 n.16
West, J. M., 14, 59 n.58, 136 n.16
Westerink, L. G., vii n.l, 4 n.3, 155
n.2
White, N. P., 6 n.13
Whitehead, A. N., 112
Wieland, W., 10, 12 n.31, 160 n.8,
176 n.36, 178 nn.37, 38 , 180
Williams, B. , 146 n.48
Wmnigton-lngram, R. P., 45 n.14
Wittgenstein, L., 37
Woodbridge, F. J. E., 10 n.25
Woozley, A. D., 6 n.13, 120 n.2
Xenocrates, 2 n.2, 3 n.3, 16, 114,
117
Xenophon, 28, 35, 43 n.5, 137
Young, C. M., 7 n.14
Zeller, E., 3 n.2
Zimmerman, B., 26 n.7
Zyskind, H., 199, 201 n.9
Subject Index
Academy: New, viii n.2, 2-3, 18,
136, 157, 189; Old, 2, 21, 157,
189, 191; Plato's, 29-30, 123-125,
129, 239-243
allegorical interpretation, 3 n.2, 12,
16- 17
anamnésis. See recollection
anonymity, Platonic, 8-9 n.20, 14, 50
n.35 , 51- 53, 55, 59, 93, 134 n.4,
138, 143-144, 149, 182, 192
Apology, 28, 34, 55, 61, 165 n.14,
191, 197-198, 201- 202, 205-206
aporia in Plato's dialogues,
significance of, 3 n .2, 4 n.6, 14,
59, 136, 148-149, 163-164, 170,
186, 199
arguments: ad hominem, 11-12;
dialectical-rhetorical versus
demonstrative, 16, 96-97;
fallacious, 120, 151; likely (eikos)
versus cenain, 97-99, 100, 102,
114
audiences of Plato's dialogues. See
under dialogues , Plato's
authenticity: of dialogues, 5 n.9; of
Seventh Letter, 186 n.56
authority, Plato 's dialogues as
undermining. See under dialogues,
Plato's
books, cost of in Ancient Greece, 33
n. 29
Channides, 145 n.47, 163-167,
172-174
chronology of Plato's dialogues, 6,
6-7 n. 14, 17, 52, 75 n .26,
120-121, 137, 159, 172, 180 n.44,
202-203; versus dramatic ordering,
137-138, 214
contradictions in dialogues. See under
dialogues, Plato 's
courage, nature of, 168- 171
Cratylus, 35, 131, 179-180
Critias, 4 n.6, 121
Crito, 17, 27-28, 121-123, 130, 142
n.38, 194, 201
deconstructionism, ix-x, 10 n.28, 11,
212 n.2
developmentalist interpretation, viii
n.2, 6 n.13, 8, 52, 202; criticisms
of, 6-7, 7 n.15 , 52-53, 75 n.26,
120- 121, 137, 157, 172, 202-203;
history of, viii, 5- 6. See also
doctrinal interpretation
dialectic, nature of, 34- 35, 96-97,
157, 167, 180-181, 196-200, 228.
See also eristic: distinct from
dialectic; rhetoric: and dialectic
dialogical interpretation, 19- 20, 191,
200, 204-205, 210- 212, 218;
versus monological, 211 - 212
dialogue form, 4- 5, 7 n.15, 9; Plato's
reasons for choosing, 11- 12, 14,
15, 29, 35, 38, 93, 155- 157, 159,

264 Toe Third Way
162- 163, 176- 177, 182-183, 187,
227, 239 (see also dialogues,
Plato's: as undenníning authority);
versus treatise, 6 n.13, 93, 111,
144, 150, 152, 158, 172, 177, 182
dialogues, Plato's: audiences of,
28-30, 33, 51 - 56, 61 n.l, 62-64,
73, 241; authority, as undermining,
35, 50 n.35 , 56, 60, 159 n.6;
consistency in, 111-112, 144-145,
193, 202; contradictions,
inconsistencies in, 5, 36, 56, 119,
140 n.32, 202- 203; dramas,
similar to or differing from, 17,
25-28, 77, 95, 107, 121, 123 n.7,
130, 134- 135, 140, 144, 192-193,
195, 205; as enactments, 139-152;
and historiography in Ancient
Greece, 35; interdependence or
independence of, 121, 200-204; as
parainetic and protreptic, 37-38; as
performances or mimes, 134-135;
not systematic, 119 (see also
doctrinal interpretation: criticisms
of); systematic in a sense, 203-204
(see also system: Plato's
philosophy as); as transition from
orality to literacy, 30- 33, 60. See
also aporía; authenticity;
chronology of Plato's dialogues;
dialogue form; fiction and fact;
form and content; myth; orality
versus written discourse: in Plato's
dialogues; recollection: Plato's
dialogues as occasioning
doctrinal interpretation: characteristics
of, 1, 9 n.20, 133, 135, 193- 195,
204, 221; criticisms of, 144- 152,
175- 176, 190, 194, 196-197,
199-200, 202- 203, 205-206, 221,
236; history of, 2- 8, 189-190. See
also developmentalist
interpretation; esotericist
interpretation; unitarian
interpretation
dramas, Plato's dialogues similar to
or differing from. See under
dialogues, Plato's
dramatic ordering of the dialogues.
See under chronology of Plato's
dialogues
eikos logos. See under arguments
elenchus, 14, 75 n.26, 165 n.15, 236,
241; as legal concept, 33-34
enactments, Plato's dialogues as. See
under dialogues, Plato's
eristic: distinct from dialectic, 34-35,
180, 196-199; in early dialogues,
33, 61 n.l, 197
esotericist interpretation, viii, 7-8, 8
n.19, 20-21, 39, 136; criticisms
of, ix n.3, 8, 149, 183, 185, 190.
See also doctrinal interpretation;
unwritten teachings, Plato's
ethics: incapable of being
demonstrative (apodictic) ora
techne, 90-91, 104-107. See also
rhetoric: and ethics
Euthydemus, 179-181
Euthyphro, 27-28, 58- 59, 125,
127-130, 145 n.47, 171 n.28
fallacies in Plato's dialogues. See
under arguments
fiction and fact, relation between in
Plato's dialogues, 33, 53- 55,
141-142, 144
form and content, relation between in
Plato's dialogues, 13, 138,
149- 150, 156-158, 176-177,
182- 183, 187
Forms, theory of, 12, 20, 119, 145
n.47, 195, 212-221
Toe Good: knowledge of, 173- 175;
Plato's lecture on, 32
Gorgias, 15, 27, 81-93, 138 n.21,
187 n.58, 212
Hippias Minor, 28
Hippolytus, Euripides', 49-50
hypothesis, method of, 16, 103-106,
181
ineffability. See philosophical
knowledge: inexpressible in
Subject Index
writing, nonpropositional or
"vision"
interpretation of Plato 's dialogues.
See allegorical interpretation,
developmentalist interpretation,
dialogical interpretation, doctrinal
interpretation, esotericist
interpretation, nondoctrinal
interpretation, unitarian
interpretation
Ion, 145 n.47
irony, 76- 77, 143 , 183, 197, 199,
217- 221, 226, 242-243
judgment, assertoric versus exhibitive,
140 n.29, 162, 192- 193, 218
knowledge how. See under knowledge
that; philosophical knowledge
knowledge that, 64; versus knowledge
how, 72-73, 77, 177, 178 n.39.
See also under philosophical .
knowledge
Laches, 121, 145 n. 47, 168- 174, 187
n.58
Laws, 4 n.6, 19, 28- 29, 36, 61 n.1,
111, 193- 194, 197,199,201,
205-210
literacy. See orality versus written
discourse; dialogues, Plato's: as
transition from orality to literacy
Lysis, 29, 145 n.47, 171 n.28 , 181
n.45, 201
Menexenus, 125, 131
Meno, 37- 38, 125-127, 145 n.47,
187 n.58, 194, 201
metaphilosophy, meaning of, 158 n.6
method, Plato's. See arguments;
dialectic, nature of; hypothesis,
method of
mimes, dialogues as. See under
dialogues, Plato's
myth (mythos) in Plato's dialogues,
100, 106, 111- 114, 138, 140, 149,
207
Neoplatonism, viii, 2, 12, 114, 157,
187, 221
nondoctrinal (sceptical) interpretation:
characteristics of, 8- 13, 136,
265
191-192, 195, 197, 200-202, 204;
criticisms of, 12- 13, 95, 107,
111- 112, 157, 176 n.36, 187, 190,
192- 193, 196-199, 205; history of,
2-4, 189-190
nonpropositional knowledge. See
under philosophical knowledge
orality versus written discourse,
14-15, 30- 33, 37-38 , 50-51, 75;
in Ancient Greek culture, 31,
38- 39, 42- 51 , 62-63, 214; in
Plato's dialogues , 136. See also
dialogues, Plato 's: as transition
from orality to literacy
Parmenides , 4 n.6, 5 n.12, 20, 61
n.1, 70, 119 n. 1, 121, 142 n.36,
201, 211- 221, 226, 228, 231-232,
235, 237, 240, 242 n.27
Phaedo, 4 n.6, 5 n.12, 16-17, 20,
28, 36, 43 n.5, 95- 107, 122, 130,
136 n.15, 142 n.36, 145 n.47, 146
n.53 , 151, 193, 196, 201, 214
Phaedrus, 4 n.6, 5 n.12 , 27, 31-32,
37, 43 n.5, 50- 51, 75, 120,
123-125, 131, 136, 140 n.32, 145
n.47, 147, 149, 152, 186, 190,
193, 194-196
Philebus, 4 n.6, 35, 86, 136 n.15,
196, 212, 228, 231 , 238 n.19
philosophia , meaning of, 139 n. 28
philosophical knowledge:
inexpressible in writing, 31- 32,
147, 184- 187, 200; "knowledge
how" rather than "knowledge
that," 75-77, 150, 157, 177-182,
183-184 n.50; nonpropositional or
"vision," 14, 39, 146-152, 157,
182- 187; nontransferable,
175-178; as personal orientation or
way of life, 152, 161 , 238- 239;
reflexive, 157, 159- 162; not
techne or technical knowledge,
90-91, 162, 169. See also
knowledge that; rhetoric: versus
philosophy; system, Plato's
philosophy as
Politicus . See Statesman
266 Toe Third Way
Protagoras, 4 n. 6, 29, 50 n.35, 142
n.38, 146, 151, 169 n.22, 175, 176
n.35, 187 n.58, 201, 212
recollection (anamnesis), 99- 100,
127, 145 n.47; Plato's dialogues as
occasioning, 38
Republic, 4 n.6, 6 n.12, 19, 28,
35-36, 38, 51 n.36 , 61 n.l, 70,
77, 92, 113- 114, 116, 119 n.l,
121, 136 nn.14 , 15, 145 n.47, 147
n.54, 169 n.22, 172-175, 178-181,
183 n.49, 187 n .58, 194- 197, 198
n.7 , 199, 201- 202, 205- 209, 214,
236- 238, 240
reflexive knowledge. See under
philosophical knowledge
rhetoric: and dialectic, 96- 97, 99,
100, 104, 106, 218; and ethics, 81,
85- 86; versus philosophy, 91-94,
97; Plato's attitude towards, 81- 82,
84, 90-94; relation to techné, 81,
83- 90, 92- 93
sceptical interpretation. See
nondoctrinal interpretation
Seventh Letter, 32, 35, 37, 130 n.11 ,
135 n .9, 140 n.32, 147, 183-187,
190, 197- 199, 242. See also
authenticity: of Seventh Letter
silence, Plato's. See anonymity,
Plato's
Socrates: not committed to view that
virtue can be defined, 165 n .15;
historical versus Platonic, 52,
56- 60, 137, 142, 149, 151- 152,
197
Sophist, 4 n.6, 6 n.12, 27- 28, 35, 70,
75 n.26, 121 , 201, 208, 212, 231,
241
sophism, sophists, 42, 96- 97,
101-105, 107, 139 n .28, 146, 165,
168, 170 n.25, 216, 219- 220
Statesman (Politicus), 4 n.6, 6 n.12,
16, 19- 20, 27- 28, 70, 75 n.26,
111, 114, 116-117, 197, 199,
201-202, 205-209, 212, 226, 228,
230 n.12, 230- 235, 240- 242
symposia, 14, 28- 29
Symposium, 26, 28- 29, 76, 123-124,
129, 131, 136 n.15, 142 n.36, 147
n.54, 148 n.56, 175, 194, 201
system, Plato's philosophy as, ix n.3,
2, 2-3 n. 2, 3 n.3, 4- 5, 5-6 n.12,
8 n.19, 176 n.36, 200-204. See
also under philosophical
knowledge; dialogues, Plato's
techné (technical knowledge), 15-16,
81-94. See also ethics: incapable
of being demonstrative (apodictic)
or a techné; philosophical
knowledge: not techné or technical Notes on Contributors
knowledge; rhetoric: relation to
techné
temperance, nature of, 163-167
Theaetetus, 4 n.6, 28, 70, 73, 75
n.26, 102, 121, 142 n.36, 145
James A. Arieti is the Graves H. Thompson Professor of Classics at
n.47, 147 n.54, 192, 198, 201, 241 Hampden-Sydney College. His books include Discourses on the First Book of
Timaeus, 4 n.6, 16, 29, 61 n.l, Herodotus (1995), Interpn!ting Plato: The Dialogues as Drama (1991),
111-117, 121,194,201 Longinus on the Sublime: Translation and Commentary (1985), Lave Can Be
treatise versus dialogue fonn. See Found (1977), and The Dating of Longinus (1975). He is the author of
under dialogue form. numerous articles on subjects that include Empedocles, Greek athletics,
unitarian interpretation: characteristics Herodotus, Homer, Horace, Livy, Machiavelli, Philo, Plato, and the
of, viii n.2, 5, 8, 137, 149; Septuagint. He has edited Hamartia: The Concept of Error in the l\éstem
criticisms of, 5, 157. See also
Tmdition. Essays in Honor of John M. Crosset (1983) and The Modem
doctrinal interpretation
unwritten teachings, Plato's: content
Language Association lntemational Bibliogrophy for 1974. lblume lll:
of, 227-228; as orienting inquiry Linguistics (1975). He has delivered over fifty papers at colleges, universities,
rather than providing answers, 236, and professional conferences in North America and Europe. He is currently
242; in Parmenides, 226- 231; in pursuing various projects in philosophy, history, and literature.
Statesman, 231-235. See also
esotericist interpretation Francisco J. Gonzalez is Assistant Professor of Philosophy at Skidmore
virtues: knowledge of is not College. He has published articles on Plato and Aristotle and has recently
definitional, 164- 168, 173-174 completed a book on Plato's dialectical method in the "early" and "middle"
n.33, 186; unity of, 169-170 n.23.
dialogues entitled Dialectic and Dialogue: Plato's Practice of Philosophical
See also Socrates: not committed to
view that virtue can be defined
Inquiry. Future projects include a book on Plato's Lysis, a book on dialectic
writing. See orality versus written in the "late" dialogues, and articles on nonpropositional knowledge in Plato
discourse. and on Heidegger as an interpreter of Plato.
Jackson P. Hershbell is Professor of Classics at the University of
Minnesota. In addition to numerous articles on a wide range of topics, he has
edíted, translated and annotated Pseudo-Plato's Axiochus (Scholars Press,
1981) and lamblichus' On the Pythagorean Wiy of Lije (with John Dillon;
Scholars Press, 1991). He is also editor (with Eric A. Havelock) of
Communication Arts in the Ancient Vtbrld (1978).
268 Toe Third Way
Mitchell Miller is Professor of Philosophy at Vassar College. Author of
P/ato's Pannenides: The Conversion of the Soul (Princeton, 1986, and Penn
State [pb], 1991) and The Philosopher in P/ato's Statesman (Martinus Nijhoff,
1980), he is now at work on a set of essays on the problem of Plato's so-called
"unwritten teachings."
Gerald A. Press is Associate Professor of Philosophy at Hunter College
(City University of New York). He is the editor of P/ato's Dialogues: New
Studies and Interpretations, author of The Development of the Idea of History
in Antiquity, and has contributed articles on Plato and ancient philosophy in a
number of journals and collections. He is Book Review Editor of the Journal
of the History of Phi/osophy and Ancient Philosophy Editor for the Columbia
History of ~stem Philosophy. He is currently completing a monograph on
Plato's Charmenides.
T. M. Robinson is Professor of Philosophy at University of Toronto. His
major publications include Plato's Psychology (1970, 199S2), Contrasting
Arguments: An F.dition of the Dissoi Logoi (1979), The Greek Legacy (1979),
Heroclitus: Frogments (1987). He has written on various aspects of Plato and
the Presocratic philosophers. He is currently completing a monograph
tentatively entitled The Vtbrld as Art-Object: &says on Plato's Timaeus.
David Roochnik is Associate Professor of Philosophy at Boston University.
He has published widely on Greek Philosophy, literature, and rhetoric. His
1991 book, The Trogedy of Reason, was published by Routledge. A second
book, Non-Technical Wisdom: P/ato's Understanding of Techne, is
forthcoming from Penn State Press.
P. Christopher Smith is Professor of Philosophy at the University of
Massachuset1S, Lowell. He received his Ph.D. from Heidelberg University in
1966 after five years of study under H . G. Gadamer and has published
annotated translations of three of Gadamer's books. His own Hermeneutics and
Finitude appeared at Fordham University Press in 1991. He is presently
completing a book entitled Logic, Dialectic, Rhetoric: a Hermeneutics of
Original Argument.
V. Tejera is Stony Brook Professor of Humanities Emeritus. His books
include Art and Human Intelligence (1965), Modes of Greek Thought (1971),
The City-State Foundations of l\éstem Political Thought (1984 & 1993),
Plato's Dialogues One by One (1984), History as a Human Science: The
Conception of History in sorne Classic American Philosophers (1984),
Nietzsche and Greek Thought (1987), Semioticsfrom Peirce to Barthes (1988),
History and Anti-History in Philosophy (with T. Z. Lavine, 1989), Literoture,
Notes on Contributors 269
Criticism, and the Theory ofSigns (1995), Aristotle's Organon in Epitome: The
Poetics, the Rhetoric, the Analytics (1995), The Retum of the King: the
Intellectual V,b,rfare over Democrotic Athens (in press), and American Modem,
The Path Not Taken: Aesthetics, Metaphysics, and Intellectual History in Sorne
C/assic American Philosophers. He is co-editor, with Richard Hart, of Plato 's
Dialogues: the Dialogical Approach (in press); and he is now completing a
monograph Understanding Parmenides: The Poem, the Dialogue, the Ali. His
teaching activity has been mainly in the areas of ancient Greek philosophy,
history of philosophy and intellectual history, American philosophy, aesthetics
and semiotics, comparative literature and humanities. His current research
interest is in the standardized mistranslations and popularizations of key texts
that have biased Western intellectual development in favor of certain
orthodoxies, and misrepresented the thought of key thinkers from Parmenides,
Plato, and Aristotle to Machiavelli, Peirce, and Dewey.
Walter Watson is Professor Emeritus of Philosophy at the State University
of New York at Stony Brook. He is author of The Architectonics of Meaning:
Foundations of the New Pluralism and many articles examining the intrinsic
reasons for irreducible philosophic differences and the consequences of these
differences for the special arts and sciences. He is currently working on the
manifestation of philosophic differences in art.
Joanne Waugh is Associate Professor of Philosophy at the University of
South Florida in Tampa. She writes on early Greek philosophy and aesthetics.
She was a contributor to the special issue of the Monist on Heraclitus, and she
is currently working on a project that discusses Republic V in its historical
context.
Elinor J. M. West was Professor of Philosophy at Long Island University.
Retired, she writes on Plato's philosophizing in the context of the oral culture
of Ancient Greece. Among her publications are articles on the structure of
Plato's dialogues, on Arcesilaus and the skeptical Socrates, and on the contrast
between the oral inspiration of the muses and the Chines e love of visual script.
In progress is a book on Plato and Aristophanes as witnesses for Socrates.