annotated bibliography

Cross-Examining_Socrates_A_Defense_of_the_Interloc..._----_Pg_1--171.pdf

Home >English homework help >annotated bibliography

C R O S S - E X A M I N I N G S O C R A T E S

This book is a re-reading of Plato's early dialogues from the point of view of the characters with whom Socrates engages in debate. Socrates' interlocutors are generally acknowledged to play important dialectical and dramatic roles, but no previous book has focused mainly on them. Unlike existing studies, which are thoroughly dismissive of the interlocutors and re- duce them to the status of mere mouthpieces for views which are hopelessly confused or demonstrably false, this book takes them seriously and treats them as genuine intellectual oppo- nents whose views are often more defensible than commenta- tors have standardly thought. The author's purpose is not to summarize their positions or the arguments of the dialogues in which they appear, much less to produce a series of bio- graphical sketches, but to investigate the phenomenology of philosophical disputation as it manifests itself in the early dialogues.

J o h n Be v e r sl u i s is Professor of Philosophy at Butler University, Indianapolis.

Beversluis, John. Cross-Examining Socrates : A Defense of the Interlocutors in Plato's Early Dialogues, Cambridge University Press, 2000. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/apus/detail.action?docID=3004497. Created from apus on 2017-11-30 19:42:10.

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

C R O S S - E X A M I N I N G

S O C R A T E S

A Defense of the Interlocutors in Plato's Early Dialogues

JOHN B E V E R SLUIS

Butler University, Indianapolis

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

published by the pres s syndi cate of the univers ity of cambridge The Pitt Building, Trumpington Street, Cambridge, United Kingdom

cambridge uni versity pres s The Edinburgh Building, Cambridge cb2 2ru, UK http://www.cup.cam.ac.uk

40 West 20th Street, New York ny 10011±4211, USA http://www.cup.org 10 Stamford Road, Oakleigh, Melbourne 3166, Australia

: Cambridge University Press 2000

This book is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the

written permission of Cambridge University Press.

First published 2000

Printed in the United Kingdom at the University Press, Cambridge

Typeset in Baskerville and Greek New Hellenic [a o]

A catalogue record for this book is available from the British Library

Library of Congress cataloguing in publication data

Beversluis, John, 1934± Cross-examining Socrates: a defense of the interlocutors in

Plato's early dialogues/John Beversluis. p. cm.

Includes bibliographical references and indexes. isbn 0 521 55058 0 (hardback)

1. Plato. Dialogues. 2. Socrates. 3. Sophists (Greek philosophy) i. Title. b395.b445 1999

184±dc21 99±11232 cip

isbn 0 521 55058 0 hardback

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

To the Memory of

Gregory Vlastos

(1907±1991)

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

In your present situation what are your philosophers doing, except incessantly demonstrating to you, that almost upon every subject you think wrong, that you are almost continu- ally acting wrong, that in your constitution, police, and way of living, almost every thing ought to be otherwise, than it is? ± This is to convince a sick man of his illness. To recover him again, is the chief point! ± But I will lay a wager, that they are as little serious in re-establishing your health, as you are in desiring it. I could give you a good reason for my opinion; but people must not tell every thing they know.

Christopher Martin Wieland, Socrates out of his Senses: or Dialogues of Diogenes of Sinope

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Contents

Preface page ix

Introduction 1

1 The Socratic interlocutor 18

2 Elenchus and sincere assent 37

3 Crito 59

4 Ion 75

5 Hippias 94

6 Laches and Nicias 111

7 Charmides and Critias 135

8 Euthyphro 160

9 Cephalus 185

10 Polemarchus 203

11 Thrasymachus 221

12 Hippocrates 245

13 Protagoras 257

14 Gorgias 291

15 Polus 315

16 Callicles 339

17 The last days of the Socratic interlocutor 377

vii

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Bibliography 384

Index of passages cited 398

Index of names 407

Index of modern authors 410

General index 414

Contentsviii

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Preface

Many books have been written about Plato's early dialogues, but they all focus mainly on Socrates. Everyone acknowledges that Socrates' interlocutors play important dialectical and dramatic roles, but no one has ever written a book which focuses mainly on them. That is what I propose to do. Unlike existing studies, which are thoroughly dismissive of the interlocutors and reduce them to the status of mere mouthpieces for views which are hopelessly confused or demonstrably false, this book takes them seriously and treats them as genuine intellectual opponents whose views are often more defensible than commentators have standardly thought. It is a re-reading of the early dialogues from the interlocutors' point of view. My purpose is not to summarize their positions or the arguments of the dialogues in which they appear, much less to produce a series of biographical sketches, but to investigate the phenomenology of philosophical disputation as it manifests itself in the early dialogues. The idea for such a book ®rst occurred to me during the sum-

mer of 1983 while I was a participant in Gregory Vlastos's NEH Seminar ``The Philosophy of Socrates'' at the University of Cali- fornia at Berkeley. I had applied in hopes of reviving a waning interest in philosophy which had derailed me professionally and even prompted thoughts of a career change. What the encourage- ment of family and friends had not accomplished in more than half a decade, this seminar accomplished in a few weeks. It was a turning point in my life. Every session was absorbing and illuminating ± a claim one can

seldom honestly make and a tribute to Gregory's inspired leader- ship. Of course, everyone who knew him also knows that he would have scolded me for saying this. For one thing, he disliked the term ``inspired'' ± a ``glib'' term which ``says nothing'' and ``bestows

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

empty praise.'' He also hated to be accused of leadership and always insisted that he was not our teacher but our colleague and fellow student. But it was inspired leadership all the same. His preparation and industry were prodigious, and he expected

the same from us. When it was not forthcoming, he could be very blunt and did not pull his punches. No one ever disappointed him twice. In addition to meting out unsparing (and occasionally scath- ing) criticism, he invited ours. The invitation was matter-of-factly extended, without a trace of condescension or the slightest nuance to convey the unspoken message that we were always to bear in mind that the speaker of these words was Gregory Vlastos. His openness to criticism was indelibly imprinted on my mind by a remark he made during one of our early sessions: ``I am one of those scholars who can never get anything completely right by himself.'' This should not be dismissed as mock modesty or fake self-deprecation. Although it was his writings and seminars which had almost singlehandedly triggered the recent explosion of inter- est in Anglo-American Socratic studies, he refused to take credit for it and saw himself as part of an international community of scholars whose work he studied and whose ``reasoned censure'' he solicited. There are few authors whose published writings are so generously sprinkled with acknowledgments of indebtedness to others. But, of course, he was Gregory Vlastos ± a man who had metic-

ulously worked out and tirelessly re®ned a systematic, original, and often highly controversial interpretation of Socratic philosophy which represented the intellectual fruits of a lifetime. His remark- able willingness to countenance criticism coexisted with an equally remarkable unwillingness to budge an inch unless absolutely com- pelled by textual evidence or analytical rigor. He clung to his views tenaciously. Trying to extract a concession from him was like try- ing to extract a stick from the jaws of a bulldog. But it could oc- casionally be done. Having made the concession, he dropped the adversarial stance in the spirit of the Socrates of the Gorgias who is happy to refute others but no less happy to be refuted himself be- cause then it is he who bene®ts. Although in later years I saw him rarely and corresponded with

him infrequently, I was deeply a¨ected by the news of his death. It was hard to believe that there would be no more bulky airmail envelopes stu¨ed with curious sheets of paper of various sizes and

Prefacex

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

colors ± some typed, some hand-written in his characteristic scrawl ± covered with criticism buttressed with all-but-indecipherable citations from the Greek text but always ending with some word of encouragement and the warmest personal regards. My acute awareness that he would have found the overall thesis of this book shockingly wrongheaded is partially alleviated by the fact that although he welcomed agreement wherever he found it, he did not suppress disagreement and never asked for disciples. It is for these reasons that I have been bold enough to dedicate it to his memory. Although I had been toying with this project since 1983, it was

not until my sabbatical in 1988±89 ± spent in Cambridge, England ± that I began working on it in earnest. I am grateful to the Sab- batical Leaves Committee and the administration of Butler Uni- versity for approving my sabbatical; and to the Academic Grants Committee for funding subsequent research during the summers of 1990, 1993 and 1994. I am also grateful to the Classics Faculty of Cambridge University for allowing me to use its libraries, for permitting me to attend lectures, and for welcoming me into its weekly Ancient Philosophy Colloquium. For criticism and discus- sion, I am indebted to Professors G. E. R. Lloyd, Malcolm Scho- ®eld, David Sedley, and, above all, Myles Burnyeat who gen- erously read large chunks of my early drafts and graciously invited me to read a paper based on this material to the Cambridge B Club ± an honor I accepted with fear and trembling. For further criticism and discussion, I am indebted to Robert Arrington, Linda Bell, Nathaniel Beversluis, Joanne Edmonds, Thomas Fox, George Ho¨mann, and, particularly, C. Grant Luckhardt whose willingness to talk about Plato's early dialogues over the years went far beyond the normal limits of human tolerance and whose enthusiasm for this project never ¯agged. More recently, I simul- taneously bene®ted and su¨ered at the hands of two readers for Cambridge University Press ± initially anonymous but later re- vealed to be Myles Burnyeat and David Sedley ± whose searching criticism required me to re-write the whole book and whose ex- hilarating encouragement provided me with the incentive to do it. My debt to them is incalculable. I cannot delude myself into be- lieving that I have adequately responded to all their objections. I can only hope that they will ®nd the ®nal version open to fewer. More recently still, I bene®ted from the detailed and incisive criticism of Debra Nails. My warmest thanks also to Pauline Hire,

Preface xi

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Classics Editor, for her initial interest, helpful advice, and re- markable patience; and to Linda Woodward, my copy-editor, whose keen eye and scrupulous attention to detail caught many errors, ambiguities, and obscurities, and improved this book in many ways. Finally, I am grateful to my wife, Susan, whose acts of super-

erogation over the years have been so numerous and untrumpeted that they have almost ceased to be surprising. Not only did she urge me to spend my sabbatical in England in the face of my periodic worries about uprooting our children; she also commuted to work between Cambridge and London ®ve days a week for the whole year without ever once complaining and still found the time and energy to plan weekend family excursions throughout the length and breadth of the United Kingdom ± ``dragging us o¨ to yet another castle,'' as our son unwisely observed early one foggy Saturday morning. As for our ``uprooted'' and castle-satiated chil- dren, Nathaniel and Ellen, now young adults, I cannot ®nd words to re-capture my memories of them during that golden year. So I will simply thank them for the aplomb with which they adapted to life in England, a¨ecting an astonishingly authentic British accent almost overnight, and for being so much fun.

Prefacexii

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Introduction

Socrates1 believed that the unexamined life is not worth living (Ap. 38a5±6). He also believed that his interlocutors were living un- examined lives. Generations of commentators have endorsed his assessment. For centuries, they have been berating the inter- locutors and exulting in the e¨ortless brilliance and utter ®nality with which they are refuted in one early dialogue after another. The result is a time-honored reading which may be summed up as follows: Searching for moral truth and disavowing moral knowl- edge, Socrates relentlessly examines his interlocutors, only to dis- cover that they are ignorant about moral matters ± often the very matters about which they profess expertise. Unable to de®ne the terms they so carelessly employ or to defend the assertions they so con®dently advance, they are helpless in the face of Socratic interrogation, reduced to aporia,2 and exposed as theoretically and morally bankrupt. Worse still, they are not committed in- quirers, as is evidenced by their stubborn refusal to acknowledge the cogency of Socrates' criticism and by their perverse ten- dency to return to the workaday world unchanged and ready to go

1 By ``Socrates'' I mean the Socrates of Plato's early dialogues among which I include the Apolog y (Ap.), Charmides (Ch.), Crito (Cr.), Euthyphro (Eu.), Gorgias (G.), Hippias Minor (HMi.), Ion (I.), Laches (La.), Protagoras (Pr.), and Republic i (R.). Passages cited are from J. Burnet, Platonis Opera. Unless otherwise indicated, all English translations are from Plato: Complete Works, ed. John M. Cooper.

2 Although aporia is usually translated as theoretical ``perplexity,'' it has wider practical im- plications which may be seen by noticing its connection to other terms of the same family. A poros is a means of passage ± a way out or through. Hence to be aporos is to be without passage. Uncharted seas induce aporia in sea-farers who lack the techneÅ of navigation which enables them to ®nd their way through unfamiliar territory. Socrates tries to induce the same state in his interlocutors by making the familiar unfamiliar. The victim is not just in intellectual di½culty (``un embarras de la penseÂe,'' Manon, 1986: 24); he is also at a loss as to how to act. See also Kofman, 1983: 51±53.

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

about their business as if these confrontations had never taken place. This account is so widely accepted that it may safely be dubbed

``the standard view.'' I use the term ``view'' advisedly, for what is really at work is not so much an explicit theory as an implicit pic- ture. Those who have fashioned it paint with a wide brush, favor- ing bold outlines and stark contrasts. On the one hand is Socrates, the paradigmatically rational man, resolutely ``following the argu- ment wherever it leads'' and tirelessly exhorting his interlocutors to do the same in hopes of convincing them that they do not know what they think they know, and motivating them to join him in searching for the knowledge which is virtue and leads to that true happiness which all men desire and of which no one is voluntarily ignorant. On the other hand are the interlocutors, paradigmati- cally irrational men, resisting the argument at every turn and the most unlikely assortment of blunderers it was ever a sage's misfor- tune to endure. Unlike Socrates, whose examined life is lived in accordance with Reason and whose genuine wisdom ± or kind of wisdom (sojiÂ an tinaÂ , Ap. 20d7) ± consists in acknowledging that he is not wise (23a5±b4), the interlocutors are impervious to Rea- son and content with a semblance of wisdom. Secure in their false conceit of knowledge, they neglect the care of their souls. In thus turning aside from what is of supreme importance and pursuing trivialities such as power, reputation, and wealth, they serve as permanent witnesses to the folly of the unexamined life ± instruc- tive case-histories of men desperately in need of a gad¯y to sting them into moral awareness. It is a very seductive picture. And as an account of what is going on in the early dialogues, it has never been seriously challenged. In this book I challenge it. The standard picture is not based on

a comprehensive and rigorously empirical study of the early dia- logues; it is a highly selective and unabashedly pro-Socratic ac- count based almost entirely on the Apolog y. The textual evidence that can be adduced in support of it is actually quite meager, and its apparent plausibility depends on assigning undue weight to a handful of super®cially impressive but thoroughly unrepresenta- tive passages scattered throughout the early dialogues in which Socrates' dialectical behavior embodies his announced seriousness and concern for the souls of his fellows, as expressed in the Apolog y,

Cross-Examining Socrates2

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

and on withholding due weight from a host of other, much more representative passages in which these lofty ideals are not much in evidence. Before defending himself against the formal indictment brought

by Meletus, which accuses him of not believing in the gods of the city, of introducing new gods, and of corrupting the youth (24b8± c1), Socrates tries to discredit a cluster of earlier charges which accuse him of being a wise man, a wrongdoer, and a busybody who speculates about cosmology, makes the weaker argument stronger, and teaches these things to others (18b4±c1, 19b4±c1). His defense is prompted by his recognition that these charges have not arisen in a vacuum and by his belief that his conduct has been misconstrued by his accusers. In an attempt to prove that where there is smoke there need not be ®re, he sets out to convince the jury that he is not wise and to explain why he is mistakenly thought to be. In the course of defending himself he o¨ers his only public (and more or less ``o½cial'') explanation of what he has been up to for all those years. He begins by ¯atly denying that he speculates about cosmology

and teaches it to others. Indeed, he teaches nothing to others for the very good reason that he knows nothing and, therefore, has noth- ing to teach. How, then, is his present notoriety and reputation for wisdom to

be explained? The question is a fair one, and Socrates replies by recounting a remarkable incident which had taken place many years ago but which, as is clear from the incredulous outburst from the jurors, had not gained wide currency. It seems that his lifelong but recently deceased friend Chaerephon had once asked the Delphic oracle whether anyone is wiser than Socrates and had been told that there is not.3 Puzzled by this oracular pronounce- ment and unaware of possessing any wisdom, great or small, Soc-

3 The historicity of this incident ± never alluded to again in the Platonic corpus ± has often been impugned on the ground that it is a transparent attempt to absolve Socrates of the charge of impiety by a spurious appeal to divine approval (see Robin, 1910: 30, n. 1; JoeÈl, 1921: 769; DupreÂel, 1922: 46, 412, 418, 419, n. 1; Hackforth, 1933: 101±4; Gigon, 1947: 95; Chroust, 1957: 32±33; and Montuori, 1981: 57±143, and 1988: 17, 50±62). However, most commentators think it unlikely that Plato would have invented the story (see J. Adam, 1894: 57±58; Gomperz 1905, ii: 105; Burnet, 1924: 170±71; A. E. Taylor, 1932: 76±77; Ferguson, 1964: 70±73; Guthrie, 1971b: 85±86; Fontenrose, 1978: 34; Brickhouse and Smith, 1989: 4±9, 88±90; Reeve, 1989: 21; and Stokes, 1992: 52±54).

Introduction 3

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

rates had initially tried to refute the oracle (e� leÂ gxwn toÁ manteiÄ on, 21c1)4 by interrogating numerous people with a reputation for wisdom ± including politicians, poets, and craftsmen ± in hopes of ®nding someone wiser. But he had failed. This disappointing ven- ture had convinced him that the god was right: no one is wiser than Socrates ± albeit only in the modest sense that, unlike these others, he does not claim to know what he does not know. Since anyone can become wise in this sense, he concluded that the god had merely used him as an example (paraÂ deigma, 23b1): anyone is wise who, like Socrates, realizes that he has no wisdom. He also concluded that he had been commanded to live the life of a philos- opher, examining himself and others, and demonstrating that they are in the same deplorable epistemic condition as he (28e4±6). Proponents of the standard picture accept all this at face value.

But it is highly problematic. First, Socrates' account is too general: he claims to have interrogated many people with a reputation for wisdom, but he does not identify speci®c individuals. Second, his assessment is unsubstantiated: he claims to have demonstrated that none of these people knew what they thought they knew, but he o¨ers no evidence in support of this claim. We are, in e¨ect, asked to take Socrates' word for the fact ± if it is a fact ± that, when subjected to interrogation, all these unidenti®ed interlocutors had been exposed as fraudulent claimants to wisdom. Why not take his word for it? Has Plato not immortalized Soc-

rates doing exactly what he claims to have done? Why not say that the Apolog y contains the claim and the early dialogues contain the evidence? This reply on Socrates' behalf is as question-begging as his own

account. There is a serious discrepancy between what Socrates

4 E� leÂ gcw need not mean ``refute,'' as it is rendered by Burnet 1924: 172, according to whom Socrates tried ``to prove the god a liar'' (so too Ryle, 1966: 177; West, 1979: 106±7, 125; Nehemas, 1986: 305±6; Teloh, 1986: 111; Dorion, 1990: 332; and Brisson, 1997: 138, n. 75); it can also mean ``examine'' or ``test.'' Hence eager to rid Socrates of even the slightest hint of impiety, many commentators claim that he was not trying to refute the oracle, but simply trying to unravel the true meaning of this riddling utterance; his strat- egy was ``interpretative only'' (Reeve, 1989: 23; so too Allen, 1970: 48; Navia, 1985: 185± 86; Woodru¨, 1988: 83±84; Brickhouse and Smith, 1989: 96±97; Nagy, 1989: 79; Stokes, 1992: 34±35; and McPherran, 1996: 224). Since e� leÂ gcw has multiple meanings, the dispute cannot be settled on linguistic grounds. Although I think the overall context favors ``re- fute,'' I will not wage a polemical campaign on this already crowded textual battle®eld.

Cross-Examining Socrates4

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

claims to have achieved, as reported by Plato in the Apolog y, and what his interlocutors think he has achieved, as depicted by Plato in the early dialogues. Although these encounters convinced Soc- rates that his interlocutors were devoid of wisdom, they did not have the same e¨ect on the interlocutors. They are not as uni- formly annihilated as Socrates would have us believe. Unlike Xenophon's Socrates, who is so persuasive that, ``whenever he argued, he gained a greater measure of assent . . . than any man I have ever known'' (Mem. 4.6.15), Plato's Socrates is singularly unpersuasive. This is not a novel claim. Gregory Vlastos said the same thing

almost thirty years ago when he declared that Plato's Socrates ``is not persuasive at all. He wins every argument, but never manages to win over an opponent.''5 Like other proponents of the standard picture, Vlastos is fully aware of the discrepancy, and he accounts for it in the same way. If we accept his diagnosis, the discrepancy is easily explained: we simply blame the interlocutors ± ``those misguided, confused, wrongheaded people whose souls [Socrates] seeks to improve.''6 The implication is that Socrates' arguments are always (or almost always)7 cogent and that his interlocutors' resistance is always (or almost always) indefensible and traceable to their impenetrable obstinacy. Proponents of the standard pic- ture say, in e¨ect, ``True, the interlocutors are `seldom, if ever', persuaded by Socrates' arguments; but they should have been.'' I o¨er a di¨erent assessment. Instead of focusing on the inter-

locutors' resistance to Socrates' arguments and accounting for it in psychological terms, I will focus on the quality of the arguments resisted. On my reading, many of them are fallacious or unsound,

5 1971: 2. In 1991: 292, n. 161, ``never'' is replaced by ``seldom, if ever.'' 6 1983a: 56. Cf. Blundell, 1992: 133±34: ``Socrates' interlocutors su¨er from a whole spec- trum of human weaknesses and failings, whether moral, intellectual, or both, which in- terfere with their ability to philosophize . . . [They] rarely seem to progress from elenctic bewilderment to increased philosophical insight.'' Commentators who o¨er similar as- sessments are legion. Here as elsewhere, I cite only representative opinion; detailed doc- umentation would ®ll volumes.

7 Almost everyone acknowledges that Plato is occasionally guilty of minor logical slips and that he sometimes employs elliptical arguments in need of supplementation (see, e.g., C. C. W. Taylor, 1976: 178±81; Santas, 1979: 233±40, 272±80; and Vlastos, 1991: 140±48). Irwin, 1977: 4, goes further: ``Much of what Plato says is false, and much more is con- fused, vague, inconclusive, and badly defended'' ± a remark which draws stinging criti- cism from Allen, 1996: 166.

Introduction 5

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

and the interlocutors' resistance is defensible. Socrates' arguments are not only criticized by his interlocutors; they often warrant criti- cism and are criticized for exactly the right reasons. The black and white contrast between Socrates and his inter-

locutors is evident not only in the standard picture's account of the early dialogues but also in its approach to them. Its propo- nents revere Socrates and treat his every utterance with meticulous care, reconstructing the man and his thought with complete ®del- ity down to the minutest detail. But they do not lavish comparable attention on his interlocutors, skeptically scrutinizing them with the clinical detachment of the laboratory technician and methodi- cally consigning them to neatly labelled categories ± Naive Youth, Complacent Businessman, Religious Fanatic, Slippery Sophist, Arrogant Rhapsode, Cynical Nihilist ± in which, like bottled specimens, they are placed on permanent exhibit as intellectually inconsequential and morally dubious ®gures unworthy of even the perfunctory attention they usually receive. Like the account, the approach is unrelentingly one-sided, humorless, and unfriendly: the interlocutors are never right, seldom likeable, rarely even tol- erable; wrong and wrongheaded, they elicit only criticism and contempt. I want to substitute a di¨erent picture based on a di¨erent

approach. Instead of accepting Socrates' uniformly negative as- sessment of his interlocutors at face value, I will examine them one by one, assessing them independently on the basis of their own merits and individual showings. I will not approach the early dialogues with the presumption that Socrates is always right and that the interlocutors are always wrong and that it is only their logical incompetence and psychological obstinacy which prevent them from being ``won over.'' In short, mine will be a much less tidy picture. The contrasts between Socrates and his interlocutors will not be etched in such sharp relief, the battle between Reason and Unreason will not be so edifyingly obvious, and the outcome will not be so unanimously pro-Socratic. Proponents of the standard picture peruse the texts in search of

arguments, and they are astonishingly good at extracting them. Having done so, however, they often examine them in vacuo ± in isolation from the persons who marshal them, the situations which generate them, and other contextual considerations which inform them and shape their direction. In the process, dialectical inter-

Cross-Examining Socrates6

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

action tends to become super¯uous, and dramatic detail and psy- chological nuance tend to be dismissed as mere ¯u¨.8 In saying this, I am not smuggling in any esoteric hermeneutical

assumptions about ``the dialogue form'' and why Plato employed it. Nor am I minimizing the importance of the arguments. A study of the early dialogues must attend closely to their argumentative content: the de®nitions and theses advanced by the interlocutors, the arguments Socrates deploys against them, their responses to his arguments, the ensuing complications, and the ®nal resolution or lack of resolution. But the early dialogues do not proceed by argumentation alone, and a study of them should not emphasize the reasoning at the expense of the reasoners. We must also attend closely to the psychological dynamics between Socrates and his interlocutors: the character of the interlocutors, the manner in which they advance their de®nitions and theses, the manner in which Socrates argues against them, the strategies they employ in response to his arguments, his counterstrategies, the reasons why some remain active participants while others try to extricate themselves, and, insofar as it is discernible, the degree of their re- spective commitments to pursue the argument seriously and in a spirit of moral earnestness. We must immerse ourselves in a world of subtleties in which no

one simply wins or loses, and in which logic, although powerfully present, seldom fully penetrates and is not the only operative in- strument of persuasion. The single-minded approach, which at- tends only to the arguments, resembles the boxing fan who watches only in anticipation of the knockout punch, oblivious to the incessant in®ghting which gradually undoes an opponent and renders him vulnerable to the knockout punch when it ®nally arrives. The more prolonged and gruelling exchanges between Socrates and interlocutors like Protagoras and Callicles are strik-

8 An instructive example is provided by J. Adam, 1893: x. Commenting on the psychologi- cally climactic moment in the Protagoras when Socrates ®nally locks horns with the great sophist, he declares: ``At this point the true business of the dialogue begins'' ± as if all that had gone before had been ``nothing but a sort of cranking-up, or a big spoonful of jam to cover up the pill that is now to be administered'' (Kitto, 1966: 235) and hence not part of the ``business'' (or, at least, not the ``true'' business) of the dialogue. Of course, not all commentators neglect dramatic context. Notable exceptions include Deschoux, 1980; Stokes, 1986, and 1992: 26±92; Cornford, 1987: 221±38; Coventry, 1990: 174±96; Arieti, 1991: 2±11; Klagge, 1992: 1±12; Frede, 1992: 201±19; Thayer, 1993: 47±59; and Nails, 1995: 36±50, 228±31.

Introduction 7

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

ingly similar to such brawls. In witnessing them, we must lay aside our contemporary academic preconceptions, forgo the unspeak- able niceties of stylized peer debate before impeccably behaved colleagues, and elbow our way into the crowded Athenian market- place where the philosopher is a street ®ghter and philosophy a spectator sport.9 Instead of treating Socrates' interlocutors as disposable dramatic

packaging, we must engage in the joint exercises of imaginative identi®cation and psychological reconstruction, making every e¨ort to grasp their perceptions of these encounters and taking empa- thetic note of their wide-ranging emotions ± their moments of high enthusiasm and buoyant con®dence, their ¯agging interest and mounting despair, their rising gorges and outraged protests, and their desperate attempts to salvage their views ± not to men- tion their egos ± in spite of heavy damages sustained. Plato often depicts their responses in considerable detail; in fact, he often seems as interested in exploring the roots of the interlocutors' re- sistance as he is in providing Socrates with the arguments whose ostensibly rigorous entailment relations they are so determined to resist. Taken seriously, these details invest the early dialogues with a deeper and much less manageable dimension.10 Socrates' interlocutors are not faceless dramatic devices ± mere

stylistic pegs on which to hang an argument. Nor are they ¯at, one- dimensional characters who are of interest only for the grist they provide for the Socratic mill. They are living presences ± vivid and credible dramatic stand-ins for real people whose strengths and weaknesses, formidableness and absurdity, complexity and sim- plicity have been brilliantly captured by Plato's dramatic genius.

9 The popular picture of Socrates accosting people in the marketplace is rejected by Ryle, 1966: 175±77: ®rst, because bedlam is not conducive to philosophical discussion; and, second, because most of the conversations depicted in the early dialogues are not im- promptu ``kerb-side pesterings,'' but semi-formal and (sometimes) previously arranged debates which take place in secluded palaestrae or private homes. Cf. Burnet, 1924: 87. However, this low-pro®le Socrates is hard to reconcile with the Socrates of the Apology who reminds his jurors of his ubiquitous presence: not only in the marketplace, but also at the bankers' tables and many other public places (17c7±d1) where, gad¯y-like, he had lighted on each of them, everywhere, and all day long (30e5±31a1; cf. 19d1±7, 23b4±7, 31c4±7, 33a6±b3, and 36b5±d1).

10 We must be cautious in ascribing inner psychologies to Socrates' interlocutors. Unlike actually existing people, ®ctional characters may legitimately be said to have inner expe- riences only when they are manifested in their behavior or attested to by a narrator. Plato's masterly depiction of character in the dramatic dialogues and his subtle employ- ment of oratio obliqua in the narrated ones insure that these conditions are satis®ed.

Cross-Examining Socrates8

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

These are recognizable human types ± men from di¨erent walks of life, with di¨erent levels of understanding, and with di¨erent reasons for entering into disputation with Socrates. But however they di¨er, in the ®nal analysis they are remarkably similar: they are seldom, if ever, persuaded by Socrates' arguments. Their in- dividual and collective resistance poignantly underscores the dif- ference between refuting a thesis and persuading a person. It also reveals that the human stakes are high and that genuine persua- sion requires more than crushing dialectical skill and periodic as- surances of one's salutary intentions. The early dialogues do not just depict theoretical debates about

matters of perennial philosophical importance, they also depict the intensely personal nature of these debates. They confront us again and again with a phenomenon which is common in philoso- phy (and in intellectual debate generally) but seldom squarely faced: it is possible to silence the opposition without persuading it. These dialogues seldom end on the note of triumph sounded by proponents of the standard picture. It is true that by the time the argument has run its course Socrates' interlocutors have typically been numbed by his stingray-like attack (M. 79e7±80b2)11 and can no longer say a word, but they are just as typically unpersuaded of the falsity (or inadequacy) of their views. However paralyzing momentarily, Socrates' arguments have no lasting e¨ect.12 Nor are the interlocutors noticeably grateful for the indefatigably optimis- tic, destructive-for-constructive-purposes-let's-start-all-over-again manner by which he reduces them to confused silence. On the contrary, they often manifest deep resistance towards his appar- ently devastating but strangely unconvincing arguments and com- parably deep resentment towards him. Proponents of the standard picture pass lightly over these awkward facts too, seldom pausing to ask what it is about Socrates' arguments that his interlocutors ®nd so unconvincing and, on occasion, infuriating.

11 The image of Socrates as a stingray appears for the ®rst time in the Meno, but it is an apt description of his e¨ect on his interlocutors in many of the early dialogues. Although Socrates' preferred image of himself is that of a gad¯y which awakens and arouses (Ap. 30e1±31a1), he accepts Meno's characterization with the proviso that his philosophical discussions leave him even more numbed than his interlocutors (M. 80c6±d1).

12 As is clear from Alcibiades' would-be testimonial to Socrates' extraordinary e¨ect on him: having confessed that Socrates is the only man alive who can make him feel ashamed, he adds that the minute he is out of Socrates' sight, he behaves exactly as he pleases (Sym. 216a8±b5).

Introduction 9

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Socrates' interlocutors have had a very bad press. To be sure, there are a few who resist his arguments for silly or transparently self-serving reasons. Their all-but-complete lack of engagement with the issues; their unwillingness to follow the most elementary argument; their refusal to draw the required inference or to ac- knowledge it when it is drawn for them; their jejune quibbling, frequent red herrings, and predictable face-saving exits ± all this limpidly reveals the lengths to which uncommitted inquirers will go to circumvent criticism. But the interlocutors are not all like that. There are many

others whose encounters with Socrates begin quite promisingly and whose gradually eroding responsiveness, palpably increasing impatience, and eventual recalcitrance cannot be adequately ex- plained by cataloging their shortcomings. Although Socrates peri- odically assures them that he is simply ``following the argument wherever it leads,'' this universally acclaimed profession of philo- sophical malleability should not be taken in any naively literal sense. Socrates does not passively follow the argument wherever it leads. Arguments do not ``lead'' anywhere. Contrary to what Soc- rates sometimes implies,13 the ``direction'' taken by an argument is not vouchsafed to the elenctic practitioner by a presiding Cosmic Logos in relation to which he is merely the logically attuned and morally neutral medium. It is dependent on the systematically op- erative, usually unargued, and often even unstated methodological assumptions and substantive views of the practitioner: his views about the nature and role of de®nition, his implicit ontology and epistemology, his ®rst-order moral judgments, and his analogies between morality and the technai. These are the factors that deter- mine the ``direction'' in which the argument ``leads.'' Socrates' more astute interlocutors resist his arguments not because they are pig-headed and perversely unwilling to follow them to where he says they lead, but because they do not think they lead there. These interlocutors are not the intellectual pushovers we have read so much about. Even when reduced to silence, they usually behave as if it is the argument itself (or the way in which Socrates has conducted it) which is at fault; and the blame for their logical

13 See Louis, 1945: 43±44: ``Le raisonnement auquel se livre le dialecticien, autrement dit le Logos, est freÂquemment preÂsenteÂ comme un eÃtre vivant.'' See also Robinson, 1953: 8; DieÁs, 1972: 111±25; and Dixsaut, 1994: 27±39.

Cross-Examining Socrates10

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

di½culties is unhesitatingly laid at his doorstep. And when he pro- ceeds to announce that it is not he but the logos that is refuting them, they ®nd the situation intolerable. Although their objections are usually rejected out of hand by Socrates and proponents of the standard picture, I think they are often exactly on target. Socrates' interlocutors not only criticize his arguments; they

also complain bitterly about his dialectical tactics. A recurring case in point is his rule against long speeches (I. 530d9±531a1, Eu. 6c8±9, G. 461d6±7). Socrates cannot tolerate people who go on at length, like gongs that continue to resonate long after being struck (Pr. 328e5±329b1); his ``poor memory'' makes it impossible for him to follow them (Pr. 334c8±d5, HMi. 364b5±c2). He wants (and often demands) short answers: a simple ``Yes'' or ``No'' will su½ce. Many interlocutors resent this restriction. The (at least partial) validity of their protest is never acknowledged. No one advocates unbridled verbosity, but laconic brevity bordering on monosyl- labic utterance is not the only alternative. Although Socrates' brevity requirement is routinely endorsed as a necessary safeguard against rhetorical excess and as a precondition of rigor, its less salutary consequences usually go unnoticed. It not only prevents his interlocutors from being long-winded; it also prevents them from explaining what they mean and do not mean, from qualify- ing their assertions, and from making important distinctions to ensure that their views will get a fair hearing. The requirement is also selectively enforced. Socrates himself is inordinately prone to speech-making ± many of them much longer than anyone else's.14 Although proponents of the standard picture ®nd endless excuses for him,15 Socrates himself is mildly sensitive to this charge and occasionally ``apologizes'' for breaking his own rule. But he con-

14 Stone's description, 1988: 146, of Socrates as ``the most talkative man in Athens'' had precedents in classical antiquity. In Plautus' Pseudolus (464) Simo warns Callipho that Pseudolus ``will talk your head o¨ till you feel as if you're arguing with Socrates.'' (I owe this source to C. Grant Luckhardt.) Socrates' propensity for incessant talk prompted Aristophanes to coin the verb ``to Socratize'' (swkrateiÄ n) and to describe those who fell under its spell as having been ``Socrati®ed'' (e� swkraÂ toun, Birds 1281±82).

15 Socrates ``adapts rhetorical techniques to a moral end'' and ``o¨ers his long speeches for questioning and investigation, rather than expecting them to be swallowed whole'' (Rutherford, 1995: 146±47); Socrates employs the ``Phaedrus principle,'' according to which ``di¨erent conditions of psychai educationally require di¨erent types of logoi,'' and sometimes making a speech is the only way that he can ``educate'' his interlocutor (Teloh, 1986: 1); Socratic rhetoric is ``philosophical'' rhetoric which tries to inculcate ``excellence of soul'' (Allen, 1984: 108).

Introduction 11

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

tinues to break it, thereby tacitly sanctioning the double standard which allows him to exploit to the hilt the persuasive possibilities of rhetoric while forbidding his interlocutors to do the same. Socrates' interlocutors accuse him of many other things: of de-

liberately misconstruing their assertions and of deducing absurd and obviously unintended inferences from them, of employing faulty analogies, of leading them into carefully prepared traps, of employing shame tactics, and of arguing for victory. Above all, they accuse him of tendentiously reformulating their theses in the weakest possible form and refuting straw men.16 Although these objections are usually dismissed as the disingenuous subterfuges of soreheads, I think they are also often exactly on target. Having said this, I want to add two disclaimers. First, in endorsing the interlocutors' objections, I am not treat-

ing them as actually existing people who operate independently of Plato's authorial control and have spotted weaknesses in Socrates' arguments. The early dialogues are not stenographic transcripts that preserve the ipsissima verba of actual conversations to which Plato was privy. As Vlastos rightly points out, ``It is Xenophon who professes to be recalling Socratic conversations he had wit- nessed personally. Plato does no such thing.''17 My claims about the interlocutors are put forth in the full realization that they are dramatic characters whose scripts were written by Plato. Although most of them correspond to actual historical ®gures, their por- trayal, like that of Socrates himself, is ®ctionalized. Accordingly,

16 I cannot accept Shorey's contention, 1937, i: xi, that ``[i]t is Plato's method always to restate a satirized and controverted doctrine in its most plausible form before proceeding to a de®nitive refutation.''

17 1991: 49. Cf. Horne¨er, 1904: 22: ``Die Platonische Dialoge sind voÈllig frei Dramen. Diese Art freier Behandlung entspricht der allgemeinen Anschauungsweise des Alter- tums, die in derartigen Schriftwerken keinerlei historische Treue verlangt.'' See also Munk, 1857: 49±53; Schleiermacher, 1858: 34±35; Bruns, 1896: 211±12; Bury, 1909: 243± 44; Ryle, 1966: 161±62; and Kidd, 1992: 83±84. Classical antiquity had no concept of biography in the post-eighteenth-century sense which burdens the modern biographer with the requirement of scrupulous ®delity to fact. For an illuminating discussion of the relation between biography and ®ction in the early dialogues (and in the SoÅkratikoi logoi generally), see Momigliano, 1971: 23±64. Of course, the fact that Plato nowhere professes to be recalling actual conversations does not license us to infer that the early dialogues contain no reliable information about Socrates or Socratic philosophy and that they sim- ply perpetuate a Platonically-contrived ``Socratic legend,'' as has been argued, inter alia, by Maier, 1913; JoeÈl, 1921; DupreÂel, 1922; Robin, 1910; Gigon, 1947; MagalhaÄes-Vilhena, 1952a, and 1952b; Chroust, 1957; Humbert, 1967; and Montuori, 1981, and 1988.

Cross-Examining Socrates12

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

the focal point of this study is not the historical ®gures themselves, but as portrayed by Plato.18 Second, in defending the interlocutors against Socrates, I am

not rejecting the standard picture as a misreading of the early dialogues. The standard picture derives ultimately from Plato, and its proponents read the early dialogues exactly as he intended them to be read. Nor am I just criticizing Socrates. As Myles Burnyeat has said, ``The programme [of the early dialogues] is one there is every reason to accept as no less a genuine mark of the historical Socrates than the recurrent irony which enlivens its exe- cution. The arguments themselves, however, are . . . the creation of Plato.''19 If Burnyeat is right ± and surely he is ± then to criticize Plato's Socrates is ultimately to criticize Plato. Like proponents of the standard picture, Plato greatly overestimates the arguments he puts into the mouth of Socrates, and he greatly underestimates the objections he puts into the mouths of the interlocutors. Since this is not a book about Socrates but a book about his

interlocutors, many issues which would otherwise cry out for sus- tained attention are treated cursorily in the text, consigned to the notes, and sometimes bypassed altogether. For example, in discus- sing the Socratic elenchus, I am much more interested in discov- ering how Socrates actually argues in speci®c contexts than I am in providing yet another general elucidation of the elenchus itself. I am also much more interested in determining how his inter- locutors are actually a¨ected by his arguments than I am in making theory-laden pronouncements about how they ought to be a¨ected by them. Similarly, in discussing the ``What-is-F ?'' ques- tion, I am much more interested in what the interlocutors make (or do not make) of this question than I am in the question itself, how its formulation may vary from one dialogue to another, whether these variations signal a ``development'' in Plato's thought, how

18 The always interesting but often unreliable Diogenes Laertius reports that, upon hearing Plato read the Lysis, Socrates remarked, ``By Heracles, what a number of lies this young man tells about me'' (3.5). Although probably apocryphal, the remark bears witness to an ever-present possibility. Had the historical counterparts of Socrates' interlocutors read the dialogues in which they appear, they might have registered similar complaints. Indeed, if the ancient testimonia are trustworthy, some of them did. See Athenaeus (The Deipnosophists 505e4±8). For full documentation, see Riginos, 1976: 96±100.

19 1971: 214.

Introduction 13

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Socratic eide di¨er from Platonic forms, and in what sense, if any, Socrates has a theory of forms. In short, I discuss problems inter- nal to Socratic philosophy only when they shed light on the theses advanced by particular interlocutors or on the arguments Socrates employs against them. Any other procedure would have led me down interesting but (for my purposes) distracting byways and resulted in a much longer book ± indeed, in two books at cross- purposes with each other and coexisting uneasily between a single cover. In place of the usual dialogue by dialogue format, I have sub-

stituted an interlocutor by interlocutor approach, devoting a sepa- rate and self-contained chapter to each, except for Charmides/ Critias and Laches/Nicias whose dialectical interaction with Soc- rates (and with each other) is so intimately connected as to pre- clude individual treatment. As for the chapter sequence, it is based on philosophical and thematic considerations, not on chronological ones.20 Hence from the fact that I discuss Laches and Charmides before Euthyphro or Protagoras before Gorgias, no inferences should be drawn about my views ± to the extent that I have views ± about the chronology of the dialogues in which these inter- locutors appear. Of course, a book about Socrates' interlocutors cannot com-

pletely avoid Socrates. Eschewing Erasmus' ``St. Socrates, pray for us,'' I will take a hard look at him too, making no attempt to prettify the results. The Socrates of the early dialogues is not the thoroughly whitewashed Socrates of the standard picture. In their determination to defend his often faulty arguments and to vindicate his often unscrupulous dialectical tactics at almost any cost, its proponents have neglected the insatiably polemical and all-too-human Socrates who operates in full view in Plato's pages and replaced him with a more admirable but largely imaginary substitute ± a hitherto unchallenged way of reading Plato against

20 Many commentators think chronology is the key to understanding Plato and for dis- tinguishing Platonic from Socratic thought (see, e.g., Hermann, 1839; Guthrie, 1975: 39± 56; Vlastos, 1991: 45±80; and Irwin, 1977: 291±93, n. 33). According to this ``devel- opmentalist'' methodology, the views espoused by the Socrates of the early dialogues are (as nearly as can be determined) those of the historical Socrates whereas the views espoused by his (in many respects very di¨erent) counterpart of the middle dialogues are those of Plato. For critiques of this hermeneutical approach, see Kahn 1981, 1988b, 1988c, 1991, and 1992; Cooper 1997, xii±xviii; and, more massively, Nails 1993, and 1995: 53±135.

Cross-Examining Socrates14

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

which this book is a long-overdue protest. In it, I will identify and render explicit a disturbing layer of Socratic philosophizing which, although not completely unnoticed by previous commentators, has been relegated to comparative obscurity. Until it has been con- vincingly delineated, Plato's Socrates will continue to enjoy a rep- utation which is out of proportion to his achievement and he will continue to evade responsibility for a lack of achievement which is not yet part of his reputation. There is a dark and even ominous side to this alleged moral

gad¯y who claims to care for the souls of his fellows but does everything in his power (often by whatever means) to deprive them of whatever shred of insight they possess, o¨ering them nothing in return and (more often than not) leaving them in a state of com- plete confusion without so much as a hint about what to do next. It was precisely this deliberate inculcation of a moral void against which P. T. Geach protested when he judged Socratic dialectic morally harmful.21 Although few have gone that far, the fact re- mains that, however true or untrue to the historical Socrates, Plato's Socrates is a complex and enigmatic ®gure. In saying this, I am not just repeating the complaints of other

philosophers. One recalls the impatience of Wittgenstein:

Reading the Socratic dialogues one has the feeling: what a frightful waste of time. What's the point of these arguments that prove nothing and clarify nothing?22

And Ryle's characteristically trenchant quip:

Socrates has, like a gad¯y, to sting Athens into wakefulness, with almost nothing to show what she is to be awakened to, save to the existence of the gad¯y.23

My complaint is di¨erent. I dare say that at one time or another almost everyone has laid

aside a Socratic dialogue with the nagging suspicion that some- thing is amiss. But what? No diagnosis can carry full conviction

21 1966: 372. Presumably not even Geach would go as far as the anonymous author of a 1768 Godwyn Pamphlet, entitled Socrates Diabolicus: or, The Old Man Exploded. A Declamation Against Socrates, according to whom Socrates' ``captious subtleties'' were designed ``to poison the a¨ections of men, and to seduce them from the road of duty and morality'' (4±5, 13).

22 1989: 14e. 23 1966: 177.

Introduction 15

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

unless it answers to one's settled intellectual and emotional expe- rience in reading the dialogues. This is hard to achieve ± partly because the texts are often di½cult to interpret and partly because of other more subtle factors which install themselves between reader and text: sympathy with the Socratic philosophical enter- prise, antipathy towards particular interlocutors, worries about one's ignorance of the staggering (and constantly growing) second- ary literature, and so on. But perhaps the most powerfully in¯u- ential factor of all is the standard picture itself ± that venerable interpretive tradition painstakingly constructed by the best minds in the history of classical scholarship, past and present, whose scholarly erudition, linguistic competence, and analytical rigor one can never hope to equal. Hence instead of reporting one's ®ndings and airing one's misgivings, one minimizes them or ex- plains them away. In any event, that is what I did. For years, I suppressed my misgivings in respectful ± albeit begrudged ± deference to the standard picture. Under its spell, I often reported what I did not think and often failed to report what I did think. As a result of this considerably less-than-honest response to what, in my heart of hearts, I thought was going on in the early dialogues, I found myself accumulating more and more skeletons in my Soc- ratic closet. To what extent my experience is shared by others, I have no way of knowing; I speak only for myself. In this book, I open that long-locked door and allow my shadowy closet dwellers to emerge into the light of day. We have never lacked monumental studies of Plato's works, and

the past several decades have produced a new wave of contempo- rary scholarship in the form of books, articles, commentaries, and published colloquia proceedings whose rigor, clarity, and atten- tiveness to detail have revolutionized our understanding of Soc- ratic philosophy and set new standards for studying it. But our understanding of Socrates' interlocutors has lagged far behind. Since embarking on this study, I have spent so much time in their company that by now they seem like old friends. In this book, I undertake to rescue them: not only from Socrates, but also from the mythology which has accumulated around them as the result of generations of unabashedly pro-Socratic scholarship. I do not claim that this study will require a reassessment of all these dia- lectical underdogs, but I do claim that it will require a reassess- ment of many of them. At the very least, by resisting the urge to

Cross-Examining Socrates16

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

berate and by presenting them in a way which makes their resis- tance to Socrates intelligible, it may make us less inclined to stand naively aghast at their imperviousness to Socratic argumentation and disabuse us of the too-easy criticism with which we have abused them.

Introduction 17

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

c h a p t e r 1

The Socratic interlocutor

The term ``interlocutor'' is standardly used in referring to the people with whom Socrates converses in the early dialogues.1 Ac- cording to the O.E.D., an interlocutor is ``one who takes part in a dialogue, conversation, or discussion'' ± an etymological de®nition which slices the term into its Latin derivatives: inter (between) and loquor (to speak). Interlocutors are people between whom there is speech; less cumbersomely, they are people who talk to each other. This does not take us very far. There are all kinds of conversa-

tions and all kinds of interlocutors, though in ordinary language the term is seldom used. Few would refer to the person with whom they chatted on the morning train as their interlocutor. It is a sti¨ and uncolloquial term, a term that elicits raised eyebrows, sug- gesting a¨ectation and alerting those within earshot that they are in the presence of a stu¨ed shirt. In short, it is a term to be avoided ± unless, of course, one is writing about Plato, in which case one can hardly get along without it. Even some students of Plato reject the term ``interlocutor'' in

favor of less pedantic alternatives like ``partner,'' ``respondent,'' or ``answerer.'' But these remedial substitutes are equally prob- lematic. ``Partner'' fosters the illusion of intellectual equality be- tween participants who are, in most cases, spectacularly unequal and, on occasion, mismatched. ``Respondent'' errs in the opposite direction by reducing one participant to a completely passive role. ``Answerer'' is unsatisfactory too; Socrates' interlocutors do much more than answer questions. What is needed is not a new term or even a better de®nition of an old term, but an elucidation of the concept of the Socratic interlocutor which clari®es his dialectical

1 The Greek equivalent is prosdialegoÂ menov ± a term Plato uses sparingly (see Pr. 342e4; Th. 161b3, 167e8; S. 217d1±2, 218a1, 268b4; and Laws 887e1).

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

and philosophical functions. Accordingly, I will retain the term ``interlocutor.'' In spite of its terminological awkwardness, it best captures the announced philosophical goals and methodological principles which underlie the Socratic elenchus2 and are allegedly operative throughout the early dialogues. Although Socrates never systematically formulates these goals

and principles, they can be extracted from what he periodically does say by way of contextual explanations, rebukes, expressions of puzzlement, directives, and asides. They will emerge even more clearly if we approach the early dialogues indirectly and brie¯y highlight some important di¨erences between Socratic interlocu- tors and the very dissimilar non-Socratic interlocutors who appear in dialogues written by philosophers other than Plato. There are not many of them. Most philosophers have opted for the prose treatise as the preferred vehicle for the dissemination of philo- sophical ideas. Comparatively few have written dialogues; of these, fewer still have done so e¨ectively.3 Since philosophers who write dialogues presumably do so for a

reason, some commentators think it is impossible to understand the Platonic dialogues until we have discovered that reason. Hence the notorious question: Why did Plato write dialogues? This ques- tion, which has given rise to a kind of sub-®eld in Platonic studies, implies that Plato's decision to write dialogues cries out for expla- nation. Numerous answers have been given ± many of them based on the controversial (and usually unargued) assumption that ``the dialogue form'' is sui generis and that ``dialogical'' content cannot be communicated ``non-dialogically'' owing to the fact that Plato never speaks in propria persona.4 Everyone agrees that there are important di¨erences between a

piece of reasoning advanced by an author in a prose treatise and a piece of reasoning advanced by a character in a dialogue. However, before trying to coax esoteric doctrine from this, two points should

2 Although Socrates says next to nothing about his philosophical method ± the term ``method'' (meÂ qodov) does not appear until Ph. 79e3 (see Robinson, 1953: 67; and Vlastos, 1983a: 28, n. 5) ± and although he employs e� leÂ gcw more or less interchangeably with a variety of other verbs (e.g., skopeÂ w, e� reunaÂ w, e� rwtaÂ w, e� xetaÂ zw, skeÂ ptomai, and zhteÂ w) and their various compounds to describe his dialectical role vis-aÁ-vis his interlocutors, the term ``elenchus'' has become permanently entrenched in Anglo-American scholarly parlance.

3 Burnyeat, 1987: 24, thinks the only philosopher who even approaches Plato is Hume. 4 See, e.g., Griswold, 1988: 1; Bowen, 1988: 58±63; and Sayre, 1988: 94±95.

The Socratic interlocutor 19

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

be noted. First, the distinction between the prose treatise and the di- alogue is not exhaustive. Philosophers have set forth their views in a wide variety of literary genres: poetry (Parmenides and Xenophanes), confessions (Augustine and Rousseau), the question-and-answer format (Aquinas), meditations (Marcus Aurelius and Descartes), geometrical proofs (Spinoza), diaries and pseudo-autobiographical narratives (Kierkegaard), novels and plays (Camus and Sartre), aphorisms (Nietzsche), ``remarks'' and ``Zettel '' (Wittgenstein), and so on. Second, the distinction between the prose treatise and the dialogue is not sharp. Philosophers who do not write dialogues often employ interlocutor-like ®gures as pedagogical devices: to anticipate objections, to bring ambiguities to light, to forestall mis- understandings, and to show how easy it is to get things wrong. Even philosophers who do write dialogues often insert long speeches during which the interlocutor is mute and all-but-forgotten. Plato himself is a case in point. The middle and late dialogues make copious use of the method of continuous exposition in the form of elaborate (and virtually uncontested) chains of reasoning, myths, historical narratives, quasi-scienti®c discourses, and legal promul- gations. Although dialogues in name, they read more like extended monologues with occasional audience participation. Even the early dialogues contain long speeches during which dialectical interac- tion is temporarily suspended. I conclude that, in spite of weighty pronouncements about ``the dialogue form'' with which some com- mentators a¿ict us, an empirical approach to the Platonic corpus reveals that the Platonic dialogue is not a unitary, sui generis, and consistently employed alternative to the prose treatise, but a carte blanche stylistic format which assigns high priority to sustained dia- lectical interaction but is not restricted to it. In this book, I will proceed on the assumption that Plato did not write dialogues for mysterious reasons, but rather because, given his Socratically in- ¯uenced conception of philosophy as a collaborative enterprise, the dialogue form was the ideal vehicle for celebrating his mentor and conducting his own philosophical investigations. Since philosophers who write dialogues do so in very di¨erent

ways and for very di¨erent reasons, the concept of the interlocutor cannot be elucidated in general. We must examine each interlocutor ± Socratic and otherwise ± on his own terms and in relation to the philosophical goals and methodological principles of the philoso- pher in whose dialogue he appears. Since limitations of space

Cross-Examining Socrates20

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

prohibit a full-scale comparative study, I o¨er the following re- marks as illustrative. I begin with Augustine's De Libero Arbitrio. In the course of re-

futing the errors of his interlocutor, Evodius, Augustine does not employ any single method of argumentation. However, his dia- lectical procedure often bears a striking resemblance to that pecu- liarly Socratic method of argumentation which Aristotle calls ``peirastic'' in which the interlocutor is refuted ``from [his] own beliefs'' (e� k twÄ n dokouÂ ntwn twÄÎ a� pokrinomeÂ nwÎ , S.E. 165b3±4; T. 100a29±30). For example, asked why he judges adultery wrong, Evodius replies that it is because he would not tolerate adultery on the part of his own wife; and whoever does to another what he does not wish done to himself does what is evil (1.3.6). Augustine responds with the counterexample of the aspiring adulterer whose desire is so overpowering that he o¨ers his wife to the husband of his prospective partner in sin ± an action which Evodius also judges wrong. Alas, retorts Augustine, not according to the princi- ple he has just espoused; for in the counterexample the aspiring adulterer is willing to do the very thing Evodius abhors. Scattered examples of ``peirastic'' argumentation aside, De Libero

Arbitrio consists mostly of extended monologues in which Augus- tine is neither refuting Evodius nor (apparently) even conversing with him.5 In these passages, the dialogue form is purely external and the interlocutor's role becomes increasingly perfunctory and, in the end, non-existent. Highly visible and actively involved in the discussion throughout Book i, Evodius gradually recedes from view. The eclipse of Evodius continues in the succeeding books, where his participation is minimal, and he vanishes altogether during the last nineteen pages of the ``dialogue.'' Augustine's pro- pensity for monologue and conspicuous neglect of Evodius suggest that, in his hands, the dialogue form is little more than a peda- gogical device which enables him to expound positive doctrine. Anselm also wrote dialogues for primarily pedagogical purposes.

That this is so is clear from the preface to Cur Deus Homo:

[I]ssues which are examined by the method of question-and-answer are clearer, and so more acceptable, to many minds ± especially to minds

5 A tendency about which he is a tri¯e self-conscious and for which another interlocutor, Adeodatus, profusely thanks him: ``I am specially grateful that latterly you have spoken without the interruption of questions and answers, because you have taken up and re- solved all the di½culties I was prepared to urge against you'' (De Magistro 14.46).

The Socratic interlocutor 21

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

that are slower . . . Therefore . . . I shall take as my fellow disputant the one who has been urging me to this end more insistently than the others, so that Boso may ask and Anselm answer. (3: 49±50)

Although Anselm often replies to Boso's questions with questions of his own, his purpose is not to draw him into philosophical de- bate, but to pave the way for his own forthcoming solution of the di½culty at hand by eliciting Boso's assent to other propositions on which the solution depends. For example, asked whether men would have died had they not sinned, Boso replies, ``As we believe, [they] would not, but I want to hear from you the rationale of this belief '' (3 : 61). He adds that he fears he would be sinning were he to say anything else (3 : 68). In prefacing his reply with the locu- tion, ``As we believe'' ± a locution which occurs frequently throughout Cur Deus Homo ± Boso is not appealing to prevailing orthodox opinion. Although his ``we'' denotes the collective body of believers, it is not the empirical fact that these propositions are believed that recommends them to Boso. They are not theological endoxa ± religious beliefs common to all the faithful ± but unassail- able truths appropriated by faith independently of and prior to philosophical investigation. If they were merely theological endoxa, Boso's dissent would only be atypical and not, as he fears, sinful. As an interlocutor, Boso is a generic stand-in for the religiously

committed but intellectually perplexed believer. What he wants from Anselm is not truth ± he already has that ± but understand- ing: ``We believe it, but I would like to have a reason for it'' (3 : 74). Yet although Boso is already in possession of truth, it is truth imperfectly grasped. Understanding is not the condition of belief, but it is a coveted desideratum; and the intellectually consci- entious believer makes every e¨ort to augment his understanding (1 : 1; 2 : 50). That God became man is beyond dispute; Boso merely wants to know why. Anselm's explanation follows, set forth dialec- tically by means of question-and-answer, so as to render the Doc- trine of the Incarnation intelligible ``to minds that are slower.'' The philosophical function of the Anselmian interlocutor, like

that of his Augustinian counterpart, is largely pedagogical: he asks questions which Anselm answers, he expresses confusions which Anselm dispels, he poses objections which Anselm demolishes. His humble contributions are dialectically uneventful and rarely in¯u- ence the direction of the discussion. His responses are unfailingly

Cross-Examining Socrates22

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

docile and characterized by a studied passivity: ``It is up to you to explain and up to me to pay attention,'' ``You have satisfactorily answered my objection,'' ``I am ashamed for having asked that question.'' And so it continues for hundreds of pages. In De Veritate the interlocutor is so anonymous that he lacks even a name. Iden- ti®ed simply as ``Pupil,'' he earnestly implores ``Teacher'' to im- part truth and promises to be a good listener (2 : 77). Discussion with such interlocutors is almost wholly devoid of philosophical excitement. There is little sense of intellectual struggle, even less dialectical give-and-take, and never the slightest possibility that the interlocutor might remain unpersuaded. We know from the very ®rst page whose view will prevail. Tame and tractable throughout, Evodius, Boso, and ``Pupil'' comport themselves like well-behaved catechumens in the presence of a revered authority, gratefully embracing the truths vouchsafed to them and devoutly resolving never again to be overtaken by doubt or error. If the dialogues of Augustine and Anselm are little more than

thinly-disguised monologues and if their interlocutors are rather too accommodating, Hume's Dialogues Concerning Natural Religion are genuinely confrontational and Cleanthes, Demea, and Philo are not interlocutors at all. They are rather philosophical protag- onists ± ¯esh and blood proponents of divergent points of view which they defend with considerable acumen. Hume is very sensi- tive about this point. Speaking through Pamphilus, he says:

[T]hough the ancient philosophers conveyed most of their instruction in the form of dialogue, this method of composition has been little prac- ticed in later ages, and has seldom succeeded in the hands of those who have attempted it . . . To deliver a System in conversation scarcely appears natural; and while the dialogue-writer desires, by departing from the direct style of composition, to give a freer air to his performance, and avoid the appearance of author and reader, he is apt to run into a worse inconvenience, and convey the image of pedagogue and pupil.6 (1983: 1)

In Hume, the participants are intellectual equals ± ``[r]easonable men [who] may be allowed to di¨er where no one can reasonably be positive'' (1±2).

6 On the di¨erences between the ``pedagogical'' dialogue (as employed, e.g., by Cicero, Augustine, Anselm, Galileo, Malebranche, Schopenhauer, and Shelley), which presup- poses the inherent inequality of the participants, and the ``dialogue of relative equality'' (as employed, e.g., by Descartes, Berkeley, and Hume), see Levi, 1976: 1±20. See also KoyreÂ. 1962: 17±19, esp. 18, n. 4.

The Socratic interlocutor 23

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

The traditional dialectical roles disappear: there is neither a designated questioner nor a designated answerer ± no speci®c participant whose views are singled out for sustained scrutiny and who must bear full dialectical responsibility for them. Sustained scrutiny there is, but of arguments antecedently formulated and delivered for the occasion, not of theses contextually elicited and jointly explored. The method of criticism is also signi®cantly dif- ferent from that of Augustine and Anselm. A representative exam- ple occurs in Part iii where Philo sets out to discredit Cleanthes' formulation of the Argument from Design. His critique does not take the form of a joint exploration in which Cleanthes is required to assent to each step; instead, he proceeds cumulatively with a series of objections which continues uninterrupted for several pages. But he pays a high price for this strategy. Having gone on at considerable length and, as he thinks, brought into play his heaviest artillery, he is more than a little disconcerted when Cleanthes disputes an earlier premise on which the entire chain of reasoning depends but to which he never assented. Having under- cut Philo's argument, Cleanthes disparages it on the ground that it proceeds ``from too luxuriant a fertility which suppresses [his] natural good sense by a profusion of unnecessary scruples and objections'' (26). The rejection of this single premise enables Cleanthes to circumvent the massive critique launched by Philo who is left ``a little embarrassed and confounded'' (26). By contrast, Berkeley's dialogues are genuine dialectical ex-

changes, and Hylas is an authentic interlocutor and the designated answerer to Philonous' questions. Unlike his Augustinian and Anselmian counterparts, the Berkeleyian interlocutor is not a rapt disciple eager to imbibe wisdom from a revered sage. The cham- pion of common sense, Hylas enters the discussion with strongly held opinions, which he defends with considerable acumen, and with clearly formulated objections, which he advances with clarity and force. As for Philonous, he is remarkably attuned to these ob- jections and takes them very seriously. Unwilling merely to silence Hylas or to settle for his grudging acquiescence, he strives for gen- uine persuasion, considering his objections one by one, ignoring his trivial inadvertences, refusing to put words into his mouth, and allowing him time for stocktaking during which he may review the arguments which have been deployed against him and examine them for possible logical ¯aws. Realizing that the hold of custom

Cross-Examining Socrates24

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

is strong, that people often remain wedded to beliefs after ac- knowledging them to be indefensible, and that what is needed is not just a refutation of false beliefs, but an explanation of their apparent plausibility, Philonous leaves no stone unturned and will not rest until he has dispelled Hylas' doubts. Unlike Hume, Berkeley makes extensive use of ``peirastic'' ar-

gumentation. Philonous continually requires Hylas to assent to each step of the argument and continually reminds him of his previous admissions (see, e.g., 227, 231, 234, 239, 240±41, 243, 246±47, 261), so that he will be ``convinced out of [his] own mouth'' (270). However, Berkeley's purpose is not purely negative. Like Augustine and Anselm, he is not just bent on refuting error; he also wants to expound positive doctrine ± in particular, the ``immaterialism'' he had set forth in his previously published but largely ignored Treatise Concerning the Principles of Human Knowledge. Disappointed by its lukewarm reception, he resolved to try again by presenting his views in more accessible form. The result was the Three Dialogues between Hylas and Philonous which ``introduces [his] notions . . . into the mind in the most easy and familiar manner'' (220) in hopes of inducing the reader, presumably sympathetic to Hylas, to abandon his own materialism as he witnesses its cham- pion going down to defeat. In that sense, Berkeley's dialogues are as pedagogically motivated as Augustine's and Anselm's. For these philosophers, the dialogue form is not a methodologi-

cal necessity, but a stylistic option ± a pedagogical device which enables them to set forth their views in a comparatively un- technical and undemanding way. Since, for them, philosophical truth can be presented either dialectically (in dialogues) or non- dialectically (in prose treatises), there is only a contingent connec- tion between the end of expounding positive doctrine and the means by which it is expounded. The choice of means is, in fact, secondary ± an afterthought, a purely strategic matter to be deter- mined by the intellectual capacities of one's audience. It is time to sum up this brief survey of non-Socratic inter-

locutors. First, however di¨erent in other respects, the Augusti- nian, Anselmian, and Berkeleyian interlocutors are alike in that they are all, to varying degrees, spokesmen for, and dramatic em- bodiments of, error; and error must be refuted. Second, in spite of the considerable amount of space allotted to refutation, it is only a preliminary. Simply to have refuted the interlocutor is not

The Socratic interlocutor 25

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

enough; his false beliefs must be replaced with true ones. Finally, since refutation leads the interlocutor into perplexity and uncer- tainty ± states of mind which Augustine, Anselm, and Berkeley re- gard as regrettable and potentially dangerous ± if the interlocutor cannot ®nd his way out of his di½culties, they stand ready to come to his aid. In turning to Plato's early dialogues, one enters a di¨erent

world. Here, too, are interlocutors aplenty and refutation by ``peirastic'' argumentation. But Socratic interlocutors bear little resemblance to their non-Socratic counterparts. For one thing, the philosopher with whom they have to do op-

erates with radically di¨erent motives. Unlike Augustine, Anselm, and Berkeley, Socrates is not interested in expounding positive doctrine ± not because he is indi¨erent to truth, but because he has none to impart. However, although devoid of wisdom, he claims to be a lover of it ± a searcher in search not only of truth, but also of other searchers. The early dialogues re¯ect the Socratic conception of philosophy as a collaborative enterprise ± a joint search for truth. By a ``joint'' search, Socrates does not just mean a discussion between two (or more) participants. The dialogues of Augustine, Anselm, and Berkeley satisfy that criterion; but they are not joint searches in Socrates' sense. In these dialogues only one participant is searching for truth; the other participant already has it. The interlocutor plays no vital role in the discovery; he merely provides the occasion for the author of the dialogue to communi- cate the truth he has already discovered ± ``[t]o deliver a System,'' in Humean phrase. Socrates has no system. Anyone who claims to have one disquali®es himself as a philosopher. Second, unlike Augustine, Anselm, and Berkeley, Socrates does

not refute his interlocutors in hopes of replacing their false beliefs with true ones, but in hopes of convicting them of ignorance and replacing their false beliefs with a desire for true ones. The proxi- mate end of philosophizing is not the discovery of truth, but the realization that one does not have it. The etymological de®nition of ``philosophy'' as the love of wisdom has become so hackneyed through repetition that it is easy to forget that it originally meant something important. As a lover of wisdom, the philosopher dis- sociates himself from all who claim to be wise. But although phi- losophy is, in that sense, a means to an end ± an activity which (one hopes) will culminate in the discovery of truth ± it is also, for Socrates, an end in itself ± an activity which enables one to live

Cross-Examining Socrates26

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

an examined life. It is in living that life, rather than in enjoying the epistemic bene®ts which result from living it, that the highest human happiness is to be found:

[T]he greatest good for a man [is] to discuss virtue every day and those other things about which you hear me conversing and testing myself and others, for the unexamined life is not worth living for men. (Ap. 38a1±6)

In short, the activity of philosophizing is not a means to happiness, understood as an end distinct from philosophizing and contingently connected to it as a causal consequence; it is happiness.7 Finally, unlike Augustine, Anselm and Berkeley, who deplore

perplexity and uncertainty as regrettable and potentially danger- ous states of mind and do everything in their power to uproot them, Socrates prizes perplexity and uncertainty as desirable and potentially salutary states of mind and does everything in his power to inculcate them. If the interlocutor cannot ®nd his way out of his di½culties, Socrates will not bail him out; he is on his own. Unlike Evodius and Boso, the Socratic interlocutor is not a rapt

disciple who has come to sit at the feet of a revered sage.8 Indeed, insofar as Socrates denied having ever taught anyone anything (Ap. 33a5±6, 33b5±8), he had no disciples (maqhthÂ v).9 He did, however, have intimates,10 many of whom were present at his exe-

7 In this life and possibly the next. It is precisely the opportunity of talking endlessly about virtue which makes the prospect of immortality so attractive to Socrates (Ap. 41c2± 4).

8 The interlocutors encountered by Xenophon's Socrates are very di¨erent. Excruciatingly aware of their intellectual inadequacies and embarrassingly susceptible to his instant ``wisdom,'' they have much in common with the Augustinian and Anselmian interlocutors and would have found them kindred spirits.

9 In referring to Socrates' interlocutors, neither Plato nor Xenophon uses the term ``disci- ple'' (maqhthÂ v). Socrates calls Chaerephon his companion (e� taiÄ rov, Ap. 21a1), and Xan- thippe alludes to his friends (oi� e� pithÂ deioi, Ph. 60a6). Xenophon typically employs touÁ v sunoÂ ntav (Mem. 1.2.64, 1.1.4, 1.2.8, 1.2.17, 1.6.3, 4.3.1, 4.4.25), although he occasionally substitutes touÁ v sundiatriÂ bontav (1.2.3, 1.3.15, 1.4.1), touÁ v e� pithdeiÂ ouv (1.1.6), touÁ v a� podexameÂ nouv (1.2.8), and touÁ v suggignomeÂ nouv (1.2.61).

10 In view of their allegedly close proximity to Socrates, one would expect Xenophon and Plato to be authoritative sources about the members of the Socratic inner circle. But their lists are strikingly di¨erent. Interestingly, neither includes the other. Xenophon's writings contain only one oblique allusion to Plato (Mem. 3.6.1), and Plato never men- tions Xenophon. Diogenes Laertius attributes this mutual chilliness to the intense rivalry between them (2.57, 3.34). Athenaeus also reports that Plato and Xenophon were envious of each other (The Deipnosophists 504e4±f6). He adds that Plato was inimical (dusmenhÂ v, 506a6) and ®lled with malice (kakohqeiÂ av, 507a8±10) towards everyone: ``[T]he day would fail me if I should wish to proceed with all who were abused by the philosopher'' (507a1±3).

The Socratic interlocutor 27

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

cution.11 Unlike Xenophon's Memorabilia, which depicts endless conversations between Socrates and these people, Plato's early dialogues depict only one ± the celebrated exchange with Crito.12 But if the Socratic interlocutor is not a rapt disciple like Evodius

and Boso, neither is he an independent thinker like Hylas. The typical Socratic interlocutor is no intellectual and, in spite of Critias' application of the term jiloÂ sojov to the young Char- mides (Ch. 154e8±155a1), none is a philosopher. Unlike the ``Soc- rates'' of the Phaedo, who surrounds himself with philosophers, and the ``Socrates'' of the Republic, for whom philosophy is a spe- cialized discipline reserved for the select few and then only after years of intensive preliminary immersion in mathematics, geome- try, astronomy, and harmonics, the Socrates of the early dialogues thinks it is open to anyone and everyone to philosophize and is willing to debate all comers ± young or old, foreigner or fellow citizen (Ap. 30a2±4). The prerequisites are minimal: one need only speak Greek and possess a modicum of intelligence, though it is arguable that some of his interlocutors lack even that. Socrates' interlocutors fall into three fairly distinct categories:

they are either young men (Charmides is a mere boy), established professionals (Nicias and Laches are generals, Polus is a rhetori- cian, Euthyphro is a theologian, Ion is a rhapsode, and Gorgias,

11 See Ph. 59b6±c6. Diogenes Laertius (3.35±36) reports that many of the Socratics were ill- disposed towards Plato and that some were openly hostile. Antisthenes attacked him in a dialogue entitled Sathon; Aristippus criticized him for not being present at Socrates' exe- cution, ``though he was no farther o¨ than Aegina''; and Aeschines claimed that the ar- guments advanced in the Crito in favor of Socrates' escaping from prison were actually his own and that Plato put them into the mouth of Crito because he despised Aeschines for being poor ± a remark which Burnet, 1924: 173, dismisses as ``a piece of spiteful Epi- curean tittle-tattle.'' If even a fraction of the gossip, rumor, and innuendo reported by Diogenes Laertius is true, the Socratics were a petty and quarrelsome lot who not only disagreed monumentally among themselves about Socrates' philosophical views but also intensely disliked one another.

12 The others are occasionally alluded to ± albeit usually un¯atteringly. This unlikely group included Aristodemus, who worshiped Socrates and went about barefoot in imitation of him (Sym. 173b1±2); Apollodorus, who lamented his pre-Socratic years as wasted and wretched (Sym. 172c2±173a3); and Chaerephon, who behaved like a wild man (manikoÂ v, Ch. 153b2) whenever he was in the presence of Socrates and who considered himself so adept at his method that he would cross-examine people on demand (G. 447c9±448c3). Although Xenophon never tires of recounting how Socrates improved his companions by teaching them to master their passions, Plato's portrayal suggests that he improved few and that his own inner strength and stability of character were conspicuously lacking in his intimates: Chaerephon was impulsive (Ap. 21a3), unpunctual (G. 447a7±8), and overly- susceptible to physical beauty (Ch. 154c8±d5); Aristodemus could not hold his liquor (Sym. 176c1±3); and Apollodorus lacked self-control (Ph. 59a7±b4, 117d3±6).

Cross-Examining Socrates28

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Protagoras, and Hippias are sophists), or prosperous employers of manual laborers (Cephalus and Polemarchus owned what was probably the largest shield factory in the Piraeus, and Crito is an urban dweller who owns several farms).13 Socrates' interlocutors are comprised of a comparatively narrow

sociological group. For one thing, they are all men. Women appear infrequently and ¯eetingly in Plato's dialogues. At Sym. 176e4±10 Socrates unceremoniously enjoins a female ¯utist to vacate the premises so the men can talk philosophy in peace.14 Except for (the probably mythical) Diotima, whose remarks are narrated by Soc- rates (Sym. 201d1±212a8), and Aspasia, whose speech he repeats for the bene®t of Menexenus (Men. 236d4±249c9), the only woman in the whole Platonic corpus who actually says anything is Xanthippe; and she is allotted only one sentence which is narrated by Phaedo: ``Socrates, this is the last time your friends will talk to you and you to them'' (Ph. 60a5±6). Although Phaedo pillories her remark as ``the sort of thing that women usually say'' (60a4±5), I ®nd it quite touching and, in view of her reputation, decidedly unshrewlike.15 But it falls on deaf ears. Absorbed in philosophical re¯ection about the mixed sensations of pleasure and pain in his chained legs, Socrates has no time for irrelevancies. He responds to his wife with a stony silence and instructs Crito to have her escorted from the cell forthwith ± which he does, with Xanthippe wailing and beating her breast. Moments before his execution she makes another appearance accompanied by their three sons and ``the

13 For an illuminating survey of the walks of life from which Socrates' interlocutors are drawn, see Vidal-Naquet, 1984: 273±93.

14 Actually, it is hard to tell whether she is banished because she is a female or because she is a ¯utist. Plato had an inordinate dislike for the instrument because of its wide har- monic range (R. 561c6±d2), and he excluded ¯utes, ¯utists, and even ¯ute-makers from his ideal society (R. 399c7±d6).

15 Xanthippe's reputation as a shrew derives largely from Diogenes Laertius who reports that she regularly scolded Socrates and once tore his coat from his back in the market- place and then proceeded to drench him with water (2.36±37). (One suspects that many interlocutors would have liked to do the same thing.) In response to Alcibiades' criticism of his wife as an intolerable nag, Socrates sagely con®des that he puts up with her for the same reason that riders put up with spirited horses: just as, having mastered these, they can more easily cope with docile creatures, so he, owing to the society of Xanthippe, can more easily cope with humanity at large. Diogenes solemnly adds that it was to such words that the Pythian priestess bore witness when she declared that no one is wiser than Socrates (2.36±37). The only commentator known to me who interprets Xanthippe's re- mark in the Phaedo as that of a shrew is Brun, 1960: 25: ``DeÁs que Xanthippe nous eut apercËus, ce furent des maleÂdictions et des discours tout aÁ fait dans le genre habituel aux femmes.''

The Socratic interlocutor 29

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

women of his household'' (ai� oi� keiÄ ai gunaiÄ kev, Ph. 116b2), but is again quickly dispatched. Not only are there no female interlocu- tors in the early dialogues, Plato never portrays Socrates convers- ing with a woman.16 Although willing to enter into philosophical debate with anyone and everyone ± ``young and old, foreigner and fellow citizen,'' he does not add ``man or woman.''17 Socrates' interlocutors are comprised of a comparatively nar-

row sociological group in a second way. Although the early dia- logues abound with allusions to skilled craftsmen whose expertise, grounded in an understanding of the rational principles underly- ing their technai, serves as a model for the moral expertise for which Socrates is searching, and although Socrates numbers the craftsmen among those he interrogated during his search for someone wiser than himself, Plato never portrays him conversing with a craftsman.18 However his interlocutors may di¨er in age, background, and education, they all move in the higher echelons of society ± Athenian and otherwise. Unlike their non-Socratic counterparts, who participate eagerly

in philosophical discussion and often initiate it, Socrates' inter- locutors typically become embroiled unwittingly and against their better judgment. A casual remark, instantly rued, about being on the way to court (or something equally humdrum) suddenly ac- quires momentous importance, and they quickly ®nd themselves being drawn into a discussion for which they have little relish and less competence. Socrates' method of argumentation is coercive. He tries to force his interlocutors to a particular conclusion ± ``to get [them] to believe something, whether [they] want to believe it

16 Xenophon, on the other hand, portrays him conversing with the beautiful and scantily clad hetaera Theodete (Mem. 3.11.1±18) and reports another conversation with Aspasia (2.6.36).

17 However, he does add that in the hereafter, if there is one, it would be a source of ``ex- traordinary happiness'' to examine the men and women there (Ap. 41c2) ± a remark which prompts Vlastos, 1991: 110, n. 15, to conclude that women were not excluded, in princi- ple, from philosophical debate and that their absence from the early dialogues is trace- able to the fact that they were ``not in the public places where Socrates could reach them'' ± a sociological barrier which is removed in the next life. Vidal-Naquet, 1984: 282, is less apologetic: ``Les femmes sont citoyennes de la ReÂpublique, elles ne sont pas recËues dans la socieÂteÂ platonicienne des dialogues.''

18 Neither does Xenophon. However, he does depict a conversation between Socrates and Euthydemus that takes place in a cobbler's shop where Socrates and his companions often gathered (Mem. 4.2.1).

Cross-Examining Socrates30

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

or not.''19 But although his method of argumentation is coercive, his manner of engaging his interlocutors in discussion is not: par- ticipation is voluntary (Ch. 158e3, La. 188a6±c3, and 189a1±3). At the same time, it is all-but-impossible to avoid. Socrates' unfailing urbanity, combined with his willingness to discuss any subject, however trivial, and his uncanny ability to judge character, enable him to lure people into discussion in spite of their misgivings. But if Socrates' interlocutors enter into philosophical discussion

voluntarily, they also enter into it blindly, advised neither of the constraints which govern the Socratic elenchus nor that its pur- pose is refutation. Lysimachus is one of the few who is forewarned about this ± albeit not by Socrates but by Nicias:

You don't appear . . . to know that whoever comes into close contact with Socrates and associates with him in conversation must necessarily, even if he begins by conversing about something quite di¨erent . . . keep on being led about by the man's arguments until he submits to answering questions about himself concerning both his present manner of life and the life he has lived hitherto. And when he does submit to this question- ing, you don't realize that Socrates will not let him go before he has well and truly tested every last detail. (La. 187e6±188a3)20

The typical Socratic interlocutor is unaware of all this and is told only that he must refrain from making long speeches and say what he really believes. Some are not even told that. Although seemingly spontaneous and even desultory, the Soc-

ratic elenchus is, in fact, highly rigorous. Having been lured into discussion ± often on false pretenses (Euthyphro is ¯attered into believing he can be of genuine assistance) ± the Socratic inter- locutor has no idea of what he is in for. Entering into disputation with Socrates is like inadvertently strolling into a mine®eld. By the time he realizes what is happening, it is too late. The refutation is as swift as it is unexpected. His astonishment is compounded by the fact that the refutation is apparently self-in¯icted. At each step of the argument, he is pointedly asked whether he assents to the propositions proposed for his assent. Dissent is not only possible, it is invited. But he does not dissent. The propositions seem perfectly

19 Nozick, 1981: 4. For a di¨erent view, see Irwin, 1986: 49±74. 20 At Ap. 29e3±30a2 Socrates describes his treatment of his interlocutors in almost identical

terms.

The Socratic interlocutor 31

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

innocuous ± even truisms. Yet once his assent is given, it assumes a life of its own and quickly becomes the instrument of his undoing. Although freely o¨ered, his every utterance binds and fetters him; his every assertion negates something else he has said or wants to say. It is an unpleasant and unnerving experience ± the experience of the interlocutor as dialectical target and of refutation not only as annihilation but, apparently, as self-annihilation. Embellishing the O.E.D.'s de®nition, a Socratic interlocutor is

``one who takes part in a dialogue, conversation, or discussion'' which is conducted for the surreptitious but allegedly salutary pur- pose of refuting him ``from his own beliefs,'' thereby exposing his false conceit of knowledge and infusing him with self-knowledge. From Socrates' point of view, an interlocutor is someone who mistakenly (and often arrogantly) supposes that he knows some- thing which he does not know. Such persons are deluded and in need of having their delusions exposed. In thus engaging his interlocutors, he claims to be caring for their souls and discharg- ing his divine mission. A further insight into the concept of the Socratic interlocutor

may be gained by reviewing how Socrates came to believe he had this mission. As we have seen, when confronted with the astonish- ing ± and, to his mind, dubious ± Delphic pronouncement that no one is wiser than he, Socrates had initially tried to refute it by interrogating various people with a reputation for wisdom in hopes of ®nding someone wiser. Having found such a person, he had intended to appear before the Pythia and confront her with living proof of her error: ``This man is wiser than I, but you said I was [the wisest]'' (Ap. 21c2).21 In short, he had treated the Delphic pronouncement like any other dubious claim. Although not an interlocutor in the strict sense, since he cannot be directly inter-

21 Contrary to what Socrates implies, the oracle did not say that he is the wisest of men, but that no one is wiser ± a claim which is compatible with others being as wise insofar as they, too, acknowledge that they have no wisdom and which seems to entail that these other hypothetical wise men would have the same divine mission. I say ``seems to entail'' because Socrates goes out of his way to explain that his belief in his divine mission was not based solely on the oracle's pronouncement, but also on dreams, commands, and every other way in which the gods make their wishes known to human beings (Ap. 33c4±7). In any event, he does not object to others behaving as if they had the same mission. With- out a trace of disapproval and even with a trace of amusement, he approvingly alludes to certain young men who, with him as their model, go about examining people who think they know something when, in fact, they know little or nothing (Ap. 23c2±7).

Cross-Examining Socrates32

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

rogated, the god is nevertheless treated as a kind of interlocutor ± an interlocutor in absentia ± and his claim is targeted for refutation. The search for someone wiser than Socrates is on. Unlike Socrates' typical allusions to wise men, these remarks

should not be taken ironically. The search was undertaken in complete seriousness and in hopes of refuting the oracle, that is, in hopes of demonstrating that, on any straightforward interpre- tation, the proposition ``No one is wiser than Socrates'' is false. This proposition can be false only if there is someone wiser than (or as wise as) Socrates. And it can be known to be false only if Socrates can ®nd him. Hence his disappointment upon discovering that the very people thought to be the wisest were, in fact, the most de®cient in wisdom (Ap. 22a1±4). This was not a mere cor- roboration of what he had expected all along; it was a genuine empirical discovery. The search for a counterexample with which to refute the oracle had failed. Since the class of persons wiser than Socrates is now known to

be a null class, it would be the height of folly to continue search- ing for its members. And Socrates did not. Unable to demonstrate that the oracle's claim was false, he concluded that it was true, re- interpreted its meaning, and concluded that he had been com- manded to live the life of a philosopher (jilosojouÄ ntaÂ . . . deiÄ n zhÄ n, Ap. 28e5), examining himself and others, thereby helping the god's cause (Ap. 23b4±c1). The hitherto dubious claim that no one is wiser than Socrates has been pronounced irrefutable (a� neÂ legktov, Ap. 22a7±8),22 and the purpose of the divine mission is to vindicate it. The serious search for someone wiser than him- self has been replaced by tongue-in-cheek irony about pretenders. The Socrates of the early dialogues has been born. There is another ``kind of interlocutor'' in the early dialogues ±

an omnipresent, sinister, and undi¨erentiated entity which darkens Plato's pages and hovers over them like a menacing cloud. This entity is ``public opinion'' and it is embodied in the views of ``the Many.'' Like the god at Delphi, it cannot be directly interrogated; but it is sometimes indirectly interrogated with some unfortunate

22 Those who translate e� leÂ gcw as ``examine'' or ``test'' (rather than as ``refute'') are com- mitted to translating a� neÂ legktov as ``unexaminable'' or ``untestable'' ± thereby foisting on Socrates the decidedly odd (if not self-contradictory) claim that his purpose in testing the oracle was to demonstrate that it is untestable.

The Socratic interlocutor 33

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

interlocutor serving as its representative. I say ``unfortunate'' be- cause it is axiomatic in the early dialogues that the opinions of ``the Many'' are, at best, muddled and, at worst, false. Any inter- locutor foolish enough to answer on their behalf has dug his own dialectical grave. I said earlier that I would retain the term ``interlocutor'' be-

cause, in spite of its terminological awkwardness, it best captures the announced philosophical goals and methodological principles which underlie the Socratic elenchus and are allegedly operative throughout the early dialogues. One ®nal principle needs to be mentioned. Socrates sometimes suggests that there is an important and pe-

culiarly Socratic kind of reciprocity between himself and his inter- locutors. This reciprocity is easily overlooked if we attend only to their respective roles as questioner and answerer. Super®cially considered, the questioner seems to enjoy all the dialectical ad- vantages. Everyone knows it is easier to pick apart someone else's position than to set forth a coherent position of one's own. Ac- cording to Socrates, however, there is a deeper dimension to the Socratic elenchus which reveals that this initial impression is false. What it is may be seen by recalling that his announced goal is to care for the souls of his fellows by convicting them of their igno- rance and motivating them to take up the philosophical quest. Hence arises the view of the interlocutor as a person whose life is on the line ± the patient, the defendant, the accused, who is (in Jaeger's phrase) intellectually ``stripped.''23 According to o½cial Socratic elenctic theory, one cannot pursue philosophy with maxi- mum pro®t by oneself in isolation from others. One needs a ques- tioner, an examiner, a critic ± someone to save one from oneself, from one's ignorance, complacency, and sloth by calling into question one's deepest certainties and revealing that one does not know what one thinks one knows. However, if the early dialogues show anything, they show Soc-

rates' monumental failure. The recalcitrant and unpersuaded in- terlocutor is not a phenomenon peculiar to some of the dialogues, but a phenomenon common to most of them ± to all of them if Vlastos is right in claiming that Socrates never manages to ``win over'' an opponent. Hence if Socrates' announced goal is his only

23 1943±45, i: 34. Socrates uses the same metaphor at Ch. 154e5±7.

Cross-Examining Socrates34

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

reason for living the life of a philosopher, why, in view of his uni- versal (or all-but-universal) failure, does he continue to believe it is a worthwhile activity ± not to mention the highest form of human happiness? Why does he not abandon his divine mission as a sin- gularly hopeless and thankless task? What are we to make of his apparently inexplicable willingness to invest such inordinate amounts of time and energy in the company of these unresponsive and seemingly impenetrable interlocutors? A possible explanation is that he persists because he under-

stands that the examined life is di½cult and is not unduly dis- couraged by failure. A second possibility is that he persists be- cause, although he never completely ``wins over'' an opponent, he occasionally makes marginal progress. Yet another possibility is that he persists out of obedience to the god of Delphi. Although each of these explanations has a certain plausibility,

Socrates gives a di¨erent one. It is this explanation which reveals the allegedly deeper dimension of the Socratic elenchus. According to Socrates, he persists in the face of universal (or all-but-universal) failure because caring for the souls of his fellows is not his only reason for living the life of a philosopher; it is not even his most important one. Although Socrates' divinely appointed task is to examine his interlocutors and deprive them of their false conceit of knowledge, the Socratic elenchus is neither wholly adversarial nor wholly altruistic. Socrates seeks out interlocutors: not just for their sakes, but also for his own sake. If his interlocutors need him, as he manifestly believes, he needs them, too. Elenctic examina- tion is always self-examination (Ap. 38a4±5). Rebuked by Critias on the ground that he is just trying to refute

him, Socrates replies:

[H]ow could you possibly think that even if I were to refute everything you say, I would be doing it for any other reasons than the one I would give for a thorough investigation of my own statements ± the fear of un- consciously thinking I know something when I do not. And this is what I claim to be doing now, examining the argument for my own sake pri- marily, but perhaps also for the sake of my friends. (Ch. 166c7±d4)

Similarly, before refuting Gorgias, Socrates expresses the hope that his interlocutor is the same kind of man as he:

And what kind of man am I? One of those who would be pleased to be refuted if I say anything untrue, and who would be pleased to refute

The Socratic interlocutor 35

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

anyone who says anything untrue; one who, however, wouldn't be any less pleased to be refuted than to refute. For I count being refuted a greater good, insofar as it is a greater good for oneself to be delivered from the worst thing there is than to deliver someone else from it. (G. 458a2±7)

In short, the deeper dimension of the Socratic elenchus consists in the fact that, in examining his interlocutors ± and, presumably, improving their souls ± Socrates is simultaneously examining him- self ± and, presumably, improving his soul. To philosophize by oneself in isolation from others is to deny the necessity and to de- cline the risk of scrutiny by others. These passages shed further light on the concept of the Socratic interlocutor and the indis- pensable dialectical and philosophical functions he allegedly per- forms. It is, in fact, no exaggeration to say that the whole Socratic philosophical enterprise is unalterably grounded in the necessity of having interlocutors. Without them, the enterprise collapses, and the possibility of philosophizing in the deepest and potentially most bene®cial sense is lost. Without a questioner, fraudulent claimants to wisdom are deprived of what they need most ± a critic. But without an answerer, the questioner is deprived of what he needs most ± an interlocutor against whom his own views can be tested. If the answerer has access to the logos only when sub- jected to interrogation, the questioner has access to it only when provided with an answerer. In the process, the views of both are tested: the answerer's by the questioner and the questioner's by the logos. This, then, is the concept of the interlocutor which underlies

the Socratic elenchus. At least, it is the ``o½cial'' view. Socrates is hard on his interlocutors, but for excellent reasons ± he is improv- ing their souls ± and he is equally hard on himself. This announced goal is at once noble and puzzling: noble, because it bespeaks a deep moral seriousness; puzzling, because it is (for the most part) a misdescription of his actual goal ± which is not to improve any- one, but simply to win arguments. In the following chapters, we will see that, and how often, this is the case.

Cross-Examining Socrates36

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

c h a p t e r 2

Elenchus and sincere assent

Before turning to the interlocutors and defending my thesis that Socrates' frequent use of faulty arguments and unscrupulous dia- lectical tactics calls into question his announced seriousness and concern for the souls of his fellows, I must address an objection which will have occurred to many readers and which, if sound, would undercut my thesis and render this study super¯uous. The objection is that I am ignoring ``the `say what you believe' con- straint,''1 that is, the Socratic methodological principle that his in- terlocutors may neither advance theses which they do not believe nor concur with theses propounded by Socrates unless they sin- cerely assent to them.2 Socrates' allegedly unwavering commit- ment to this principle is one of the orthodoxies of contemporary Anglo-American Socratic scholarship. I say ``allegedly'' because although Socrates occasionally invokes this sincere assent require- ment, its importance has been greatly exaggerated by many recent

1 The terminology is Vlastos's, 1983a: 35. In 1994: 7, ``constraint'' is replaced by ``requirement.''

2 The requirement is variously interpreted. Some commentators think that it applies not only to the interlocutors but also to Socrates, and that it forbids him to employ premises which he does not believe (see Crombie, 1962, i: 26; C. C. W. Taylor, 1976: 155±61; Irwin, 1977: 39; MeÂron, 1979: 211; Vlastos, 1981b: 223, n. 5; 1991: 132±35, esp. 134, n. 17. Others think that it applies only to the interlocutors on the ground that Socrates always argues ad hominem, i.e., from his interlocutors' beliefs; hence it is not necessary that he be- lieve the premises he employs, it is su½cient that they believe them (see Stokes, 1986: 1±35; Teloh, 1986: 1±2, 16±18; Kahn, 1983: 113, n. 62; Frede, 1992: 212; Brickhouse and Smith, 1994: 14±16. Even Vlastos admits that Socrates occasionally employs premises he does not believe; but he exonerates him on the ground that in these passages he is not engaged in ``standard'' elenchus, which demonstrates that his interlocutors' beliefs are false and re- quires him to employ only premises he thinks are true, but in ``indirect'' elenchus, which demonstrates that their beliefs are inconsistent and does not require this (1983a: 38±39, esp. nn. 29 and 32). Brickhouse and Smith ®nd other excuses, denying that Socrates is ``directly deceitful'' in employing premises he does not believe and asserting ± for reasons unknown to me and undisclosed by them ± that he must divulge his actual views only ``[w]hen asked directly'' (1994: 16, n. 30).

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

commentators.3 In this chapter, I will try to show that it is not nearly so ubiquitously present and systematically operative as they think. Passages in which the requirement is invoked include Cr. 49c11±

d1:

Crito, see that you do not agree to this, contrary to your belief (paraÁ doÂ xan);

R. 349a4±6:

[W]e mustn't shrink from pursuing the argument and looking into this, just as long as I take you to be saying what you really think (a� per dianohÄÎ );

Pr. 331c4±d1:

Don't do that to me! It's not this ``if you want'' or ``if you agree'' busi- ness I want to test. It's you and me I want to put on the line, and I think the argument will be tested best if we take the ``if '' out;

and G. 500b5±7:

[B]y Zeus, the god of friendship, Callicles, please don't think that you should jest with me . . . or answer anything that comes to mind, contrary to what you really think (paraÁ taÁ dokouÄ nta).

Indeed, at G. 495a8±9 Socrates tells Callicles that he cannot be a suitable dialectical partner if he speaks contrary to what he really believes. No one has attached greater weight to these (and other similar)

passages than Vlastos, according to whom Socrates ``will not de- bate unasserted premisses ± only those asserted categorically by his interlocutor, who is not allowed to answer `contrary to his real opinion'.''4 Insofar as the sincere assent requirement disallows ``i¨y'' ± that is, tentative and non-committal assertions ± it invests the Socratic elenchus with an ``existential dimension''5 which en-

3 Among them, Robinson, 1953: 15±16; Gulley, 1968: 33±34, 41; Burnyeat, 1971: 214; Irwin, 1977: 39; 1995: 20, 29±30, 48±49, 121; Stokes, 1986: 29±33; Teloh, 1986: 11; Saunders, 1987: 31, 278, n. 3; Benson, 1989: 591±99; Frede, 1992: 211±12; Penner, 1992b: 144±45; Brickhouse and Smith, 1994: 13±14; and Rutherford, 1995: 79. The requirement is so deeply entrenched in and taken for granted by Anglo-American Socratic scholarship that one recent commentator merely alludes to it in passing as ``a familiar feature'' of the Socratic elenchus (McPherran, 1996: 54, n. 70).

4 1983a: 29. See also 1956: liv, n. 13; 1971: 20; 1991: 14, 111±13, 123, n. 69; 1994: 7±10, 135± 36.

5 The terminology is Vlastos's, 1983a: 37.

Cross-Examining Socrates38

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

ables Socrates to examine his interlocutors' lives as well as their theses6 and radically distinguishes Socratic dialectic, which is con- cerned with truth, from sophistic eristic,7 which is only concerned with victory.8 Although this edifying dichotomy is deeply entrenched in Anglo-

American Socratic studies, it should be taken with a grain of salt. In particular cases it is very hard to distinguish Socratic dialectic from sophistic eristic;9 in fact, they are often indistinguishable.10 The Socrates of the early dialogues is not all of a piece; on the contrary, he is a complex and elusive ®gure whose overall dia- lectical performance cannot be adequately accounted for by either category. Although he habitually a½rms that he is searching for truth and habitually denies that he is arguing for victory, he seems inordinately intent on winning arguments ± often by whatever means. In a word, his actual elenctic practice often bears little re- semblance to his o½cial elenctic theory. This duality of purpose is expressive of a corresponding duality of character which accounts

6 There is no scholarly consensus about how ± or even whether ± the testing of theses and lives are connected. Vlastos, 1983a: 36, denies that there are two kinds of elenchus ± one philosophical, the other therapeutic ± and claims that there is only one which must do both jobs at once ± a claim contested (albeit for very di¨erent reasons) by Woodru¨, 1986: 26, and Penner, 1992b: 144±45.

7 Eristic (from e� riv ± strife, quarrel, contention) was a competitive or ``agonistic'' form of argumentation in which one participant tried to defeat the other by any means ± fair or foul.

8 For standard discussions of the di¨erences ± real or alleged ± between dialectic and eristic, see Moraux, 1968: 277±311; Guthrie, 1971b: 27±54; Kerferd, 1981: 59±67; Rankin, 1983: 13±29; Irwin, 1986: 61±63, 1992b: 63±69; Benson, 1989: 591±99; Poulakos, 1995; and Wardy, 1996. Robinson's magisterial study (1953) is notable for its familiarity with the texts and its non-apologetic tone. With disarming forthrightness, Robinson points out that although Plato always praises dialectic and always denounces eristic, he never ade- quately explains the di¨erence. At each stage of his life, he applied the term ``dialectic'' to ``whatever seemed to him at the moment the most hopeful procedure'' and he applied the terms ``eristic'' and ``sophistry'' to ``whatever seemed to him at the time the danger most to be avoided'' (70).

9 As Robinson, 1953: 85; Kidd, 1967: 481; and Irwin, 1992b: 65, candidly acknowledge. 10 Ryle, 1966: 119, thinks the early dialogues are specimens of eristic and, with characteristic

irreverence, characterizes them as ``elenctic question-answer tussles.'' Although this view has found little favor among recent and contemporary commentators, the fact is that Socrates' dialectical tactics often render him liable to the charge of eristic. His demand for brief answers is a case in point. Requiring short answers was a common sophistic tactic: at Euthyd. 295b6 Dionysodorus prohibits Socrates from requesting clari®cation and requires him to answer questions even if he does not fully understand what he is being asked, at 296a1±2 he faults him for answering more than he was asked, and at 296a8±c2 he tells him to stop qualifying his answers. Socrates often employs the same tactics.

Elenchus and sincere assent 39

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

both for the ``best, wisest, and most just'' Socrates celebrated at Ph. 118a16±17 ± and for the ``not too scrupulous'' Socrates mildly rebuked by Guthrie11 and severely taken to task by many inter- locutors. His dialectical aberrations include numerous violations ± some of them quite blatant ± of the sincere assent requirement. Socrates' unscrupulous dialectical tactics have not gone un-

noticed. Although few endorse Grote's contention that ``[a] person more thoroughly Eristic than Sokrates never lived,''12 many accuse him of deliberate sophistry ± of feeding his interlocutors premises he thinks false, of drawing inferences he thinks fallacious, and of ``every [other] trick of logic and rhetoric.''13 Having done so, how- ever, they vindicate him on the ground that although his tactics are often dubious, his motives are always exemplary: he is expos- ing ignorance and caring for souls. As we shall have many occa- sions to observe during the course of this study, many commenta- tors ± Vlastos is not one of them ± think that the crucial di¨erence between dialectic and eristic is a di¨erence in motivation: the former is serious and aims at moral improvement whereas the latter is frivolous and merely aims at victory. Apparently, the philosophical end justi®es the dialectical means.14 I reject this approach. Throughout this book, I will proceed on

11 1975: 439. 12 1867, iii: 479 ± a criticism rejected by Vlastos, 1983a: 31, n. 14, on the ground that Grote

employs the term ``eristic'' with ``culpable looseness'' to mean ``contentiousness'' which, although admittedly one of Socrates' failings as a philosopher, is not the same thing as sophistry.

13 See, e.g., Woodbridge, 1929: 269±70; Robinson, 1942: 101±2; Jaeger, 1943±45, ii: 64; Dodds, 1959: 249; Sprague, 1962; FriedlaÈnder, 1964, ii: 19, 181, 254; Sinaiko, 1965: 14±15; Ryle, 1966: 119; Gulley, 1968: 24, 45; DieÁs, 1972: 296; Guthrie, 1975: 439; Kahn, 1983: 93, 1992: 252±53, and 1996: 110±11, 112, n. 13, 114±16, 189, n. 11, 237; Kidd, 1967: 481; McTighe, 1984: 226; Stokes, 1986: 34; Teloh, 1986: 1±2, 16±18; ChreÂtien, 1987: 3±6; and Graham, 1992; 152±54. AngeÂlopoulos, 1933: 29, thinks that Socrates was a sophist between 430 and 420. According to his analysis, the sophist depicted by Aristophanes in the Clouds is an accurate description of the Socrates he knew while the moral philosopher depicted by Plato and Xenophon is an equally accurate description of the very di¨erent Socrates they knew many years later. So, too, Mazel, 1987: 27±29.

14 The lengths to which some commentators are prepared to go in vindicating Socrates' di- alectical tactics are vividly illustrated by Nickolas Pappas's assessment of the arguments against Thrasymachus in Republic i. Having faulted Socrates, inter alia, for changing the subject, for failing to de®ne crucial terms, for using faulty analogies, for generalizing from particular cases, and for employing sloppy arguments which depend on ambiguity, wordplay, and legerdemain, Pappas defends him on the ground that these ``¯awed'' ar- guments are merely ``®rst sketches [whose] obvious faults . . . betray the overcompression of deep truths'' (1995: 45±50 passim).

Cross-Examining Socrates40

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

the assumption that allegations of faulty reasoning are a serious matter. Unlike Teloh,15 who thinks that Socratic dialectic should not be judged primarily by ``the excellence of its logic,'' I think that is the sole criterion by which it should be judged. In response to McKim's puzzling contention16 that ``the very features of Soc- rates' argument[s] that are logically objectionable are precisely those that render [them] psychologically telling,'' it is su½cient to point out that a ``logically objectionable'' argument is a worthless argument; and it remains worthless in spite of the fact that some unwary interlocutor is taken in by it or ®nds it ``telling.'' Vlastos acknowledges that Socrates sometimes employs unscru-

pulous dialectical tactics. In chapter ®ve of Socrates, Ironist and Moral Philosopher ± entitled ``Does Socrates Cheat?'' ± he cites several passages (mostly from the Protagoras) in which he is portrayed as ``complicated, devious, cunning, and not averse to playing pranks on his interlocutors upon occasion.''17 However, he categorically denies that Socrates ever cheats and vigorously defends him against all charges of eristic and deliberate sophistry. Although he is some- times guilty of dialectical trickery, he always remains ``free of resort to deceit . . . when arguing seriously.''18 This, of course, presupposes a criterion by which to tell when Socrates is ``arguing seriously'' and when he is not. According to Vlastos, the early dialogues provide one: ``when Socrates is searching for the right way to live, in circumstances in which it is reasonable for him to think of the search as obedience to divine command, his argument cannot involve wilful untruth.''19 Whenever this criterion is satis®ed, Socrates is ``arguing seriously'' and is never guilty of dialectical trickery or deviousness; however, when- ever it is not satis®ed, he is not ``arguing seriously'' and is often guilty of these things. In producing this criterion, Vlastos is trying to tie down ``a loose thread'' in an earlier paper20 in which he had denied that ``for good purposes (pedagogical or polemical) . . . [Socrates] would ever (knowingly, and in a serious vein) assert

15 1986: 1. 16 1988: 37. 17 1991: 133. Criticism of Socrates is not novel in Vlastos's writings (see 1956: xxiv±xxv,

xxxviii: 1971: 15±17; 1983a: 31, n. 14; and 1988: 92). 18 Ibid.: 133±34. 19 Ibid.: 134. Speci®c passages from which Vlastos extracts his criterion include Ap. 28d6±

a1, 29d2±30a7; Ch. 166c7±d2; and G. 500c1±4. 20 1981b: 223, n. 5.

Elenchus and sincere assent 41

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

categorically a false premise or endorse a fallacious argument.'' What was missing there ± a su½ciently precise criterion of Soc- ratic ``seriousness'' ± is ostensibly supplied here. But there is still something missing. Given Vlastos's account of

the Socratic elenchus, ``arguing seriously'' is not a su½cient con- dition for Socrates' never knowingly endorsing a false premise or drawing a fallacious inference; it is only a necessary condition. Another condition must be added, namely, when he is engaged in ``standard'' elenchus, which does not countenance these maneu- vers, rather than in ``indirect'' elenchus, which does.21 Only when both these individually necessary conditions are satis®ed does the required su½ciency obtain. So whenever Socrates knowingly as- serts a false premise or endorses a fallacious argument, Vlastos's analysis provides him with two outs: (i) ``I was not `arguing seri- ously','' and (ii) ``I was `arguing seriously', but I was not engaged in `standard' elenchus.'' Vlastos's criterion of Socratic ``seriousness'' signals a valiant at-

tempt to explain something which those who accuse Socrates of deliberate sophistry never even attempt to explain, namely, how a philosopher who knowingly employs fallacious and unsound argu- ments can be reconciled with a morally earnest truth seeker.22 The di½culty is a real one, but Vlastos has not solved it. His criterion is open to several objections. First, it restricts the range of applicability of ``elenctic argu-

ment'' to those passages in which Socrates is ``searching for the

21 See n. 2. 22 Whether Plato portrays Socrates knowingly employing fallacious or unsound arguments

is a controversial question. According to Robinson, 1942: 97±103, Plato had no word for ``fallacy'' and his ``consciousness'' of ``fallacy-as-such'' was ``very small''; however, he clearly recognized some fallacies ± in particular, those involving ambiguity and illicit conversion ± and often portrayed Socrates employing them to ``deceive'' his inter- locutors. The same thesis is advanced by Sprague, 1962; McKim, 1988: 36±37, 46±48; Vickers, 1988: 84±119; and Rutherford, 1995: 91. It is, of course, emphatically rejected by Vlastos. I myself am as wary of Robinson's (largely unveri®able) speculation about Plato's ``consciousness'' of fallacy as I am of Vlastos's (sometimes excessively apologetic) justi®cations of Socrates' often faulty arguments and unscrupulous dialectical tactics. Both are insu½ciently attentive to Ryle's salutary reminder, 1966: 206±7, that logic was in its infancy in Plato's day ± a fact which makes it unlikely that a clear distinction between a valid and an invalid argument predated the early dialogues and, therefore, ill- advised anachronistically to assume that Plato ``just knew from the start'' the di¨erences between good and bad arguments. See also Lloyd, 1979: 60±61, 100±102, 123, n. 331). Whether Plato ``realized'' that some of the arguments he puts into Socrates' mouth are fallacious, I do not pretend to know. Since my purpose is not historical but philosophical, I avoid all authorial inferences and assess Socrates' arguments as arguments.

Cross-Examining Socrates42

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

right way to live'' in obedience to the god at Delphi and, there- fore, ``arguing seriously.'' This overly strict criterion entails that all non-argumentative or argumentative but ``non-serious'' pas- sages fall outside the purview of the Socratic elenchus and must be described in some other way, for example, as dialectical pre- liminaries, interludes, or postludes, during which Socrates is not searching for the right way to live and may, therefore, presumably behave however he pleases. This way of carving up the dialogues into ``serious'' and ``non-serious'' portions smacks of overinter- pretation the sole purpose of which is to excuse Socrates' dia- lectical lapses and to protect his reputation as ``the honest arguer we know him to be.''23 Second, I suspect that a rigorous application of Vlastos's crite-

rion to the early dialogues would yield an alarming number of pas- sages in which it would be impossible to determine on the basis of any non-arbitrary criterion of ``seriousness'' (as opposed to play- fulness, playful seriousness, serious playfulness, tongue-in-cheek irony, mockery with serious intent, lighthearted bubble-pricking, etc.) when, and to what extent, Socrates is ``arguing seriously'' and hence when, and to what extent, the criterion has been satis®ed. Part of what makes him so endlessly fascinating ± not to mention infuriating ± is that his ``serious'' moments cannot be divorced from his ``non-serious'' ones without doing violence to both. His behavior cannot be consigned to one of two luminously clear and transparently obvious categories in a way that enables us to say, ``Now he is being serious, but now he is not.'' Indeed, if by ``seri- ous'' one means humorless, unrelieved gravity, and the complete absence of the sense that he is thoroughly enjoying his interlocu- tor's befuddlement, Socrates is seldom, if ever, completely serious; on the other hand, if by ``not serious'' one means ¯ippant, cavalier, and wholly unconcerned about the right way to live, he is seldom, if ever, not serious. Third, it is not only impossible to apply Vlastos's criterion in a

way that enables us non-arbitrarily to distinguish passages in which Socrates is ``arguing seriously'' from those in which he is not; it is not even clear how the criterion itself is to be understood. Ac-

23 1991: 279. Klosko, 1983: 363, rightly points out that Vlastos is often ``sorely taxed'' to rid Socrates' arguments of the appearance of sophistry. He adds that ``by unearthing suit- able tacit premises, a determined scholar would be able to absolve any argument of this charge, no matter how sophistical in appearance.''

Elenchus and sincere assent 43

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

cording to Vlastos, Socrates is ``arguing seriously'' whenever he is ``searching for the right way to live,'' that is, whenever he is phi- losophizing, examining himself and others, in obedience to the god at Delphi. But how are we to determine when he is doing that? Here Vlastos provides no help. His criterion purports to tell us what counts as ``seriousness,'' but it does not tell us what counts as philosophizing. Although apparently perspicacious and unproblem- atically applicable, it is, in fact, vacuous: Socrates is never know- ingly guilty of dialectical trickery or deviousness when he is ``ar- guing seriously''; therefore, whenever he is knowingly guilty of these things, he is not ``arguing seriously.'' In short, instead of elu- cidating the criterion of ``seriousness'' independently of dialectical trickery and deviousness in a way that enables us to approach the early dialogues empirically, criterion in hand, and ask whether Socrates is, in fact, ever guilty of these things, Vlastos de®nes ``ar- guing seriously'' in a way that precludes the possibility. So, of course, he never is. That is not exegesis; it is apologetics. Vlastos's refusal to allow anything to count against his assertion that Soc- rates never cheats renders it textually unfalsi®able. Either we need a more satisfactory criterion of Socratic ``seriousness''24 or we need to abandon the distinction between ``serious'' and ``non-serious'' argumentation. I opt for the latter.25 If we do abandon it, we are left with an enigmatic ®gure whose

astonishingly mixed bag of tricks and constantly ¯uctuating attitu- dinal stance towards his interlocutors invest his overall dialectical performance with an ambiguity that resists all neat classi®cations. If those critics who accuse him of being an accomplished sophist are guilty of exaggeration, it is an exaggeration of a truth worth retrieving. What that truth is may be seen by looking more closely at the

sincere assent requirement and its alleged role in (what Vlastos calls) ``standard elenchus'' ± that peculiarly Socratic form of

24 MeÂron, 1979: 41±42, o¨ers one: Socrates is being serious whenever he reveals his true thoughts (``sa vraie penseÂe''). But this criterion is as di½cult to apply as Vlastos's. Indeed, if Michael Stokes (about whom more later) is right, Socrates never (or rarely) reveals his ``true thoughts.''

25 For further criticisms of Vlastos, see Beversluis, 1993. For a di¨erent assessment, see Irwin, 1992a: 242±43, who is ``entirely convinced'' by Vlastos's ``devastating case'' against the charge that Socrates cheats.

Cross-Examining Socrates44

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

``peirastic'' argumentation by which Socrates refutes his inter- locutors ``from their own beliefs'' and which Vlastos has schema- tized as follows:

(i) The interlocutor asserts a thesis, p, which Socrates considers false and targets for refutation.

(ii) Socrates secures agreement to further premises, say q and r (each of which may stand for a conjunct of propositions). The agreement is ad hoc: Socrates argues from fq, rg, not to them.

(iii) Socrates then argues, and the interlocutor agrees, that q and r entail not-p.

(iv) Socrates then claims that he has shown that not-p is true, p false.26

This is a lucid schematization of the formal structure of (some of ) Socrates' arguments. But it purchases its lucidity at the intolerably high price of neglecting the informal complexity of the Socratic elenchus and the contexts in which it is conducted. Socrates' interlocutors resist his arguments for many reasons.

Sometimes they disagree with one (or more) of the premises. Sometimes they dispute the analogy by which he deduces a dam- aging inference from their previous admissions. Sometimes they dispute the inference itself. But the most interesting form of resis- tance occurs when an interlocutor, intimidated by the deftness with which Socrates can deduce unwelcome consequences from whatever he says, is reluctant to say anything at all. In such cases, Socrates' task is to elicit his initial assent. But the phenomenon of giving (or withholding) one's assent is

not as straightforward as it seems. Assent is not an either/or a¨air. There are degrees of assent, ranging from the unquali®ed ``Cer- tainly'' (paÂ nu ge) and ``Yes, by Zeus'' (naiÁ maÁ DiÂ a), to the more circumspect ``It seems so'' (jaiÂ netai or e� oike), to the still less ac- commodating ``I suppose'' (oi� mai), to the downright grudging ``If you say so'' and ``So be it'' (e� stw). Except for ``Certainly'' and its semantic equivalents, all these locutions signal quali®ed assent; and quali®ed assent is not assent simpliciter. Accordingly, in recording the logical commitments that Socrates manages to extract from his interlocutors, it is not su½cient to report that they have assented to

26 1994: 11. For an earlier (and marginally di¨erent) formulation, see 1983a: 39.

Elenchus and sincere assent 45

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

the propositions proposed for their assent ± not even if one adds, as some do,27 that they never renege on their assent once they have given it. Reneging is not an either/or a¨air either. We also need to know the manner in which and the degree to which they have as- sented. Socrates' interlocutors do not typically ``assert categori- cally'' the premises proposed for their assent; their responses are usually more quali®ed and guarded. Nor should we overlook the equally important and rather more

surprising fact that Socrates seldom acknowledges his interlocu- tors' reluctance to give their unquali®ed assent. Indeed, he habit- ually minimizes (or ignores) their misgivings and presses for their assent: by asking leading questions, by resorting to mock ¯attery, by employing shame tactics, by soliciting the aid of other auditors, and even by threatening to leave (Pr. 335c2±6). These psychologi- cally manipulative tactics suggest that, although o½cially insisting on sincere assent, he is often willing to settle for quali®ed, and sometimes heavily quali®ed, assent. In view of the momentous importance that many commentators

have ascribed to the sincere assent requirement, it is surprising that they have paid almost no attention to the question of what counts as sincere assent. Accordingly, before proceeding further, I will try to elucidate this murky ± and almost wholly unanalyzed ± concept. The ®rst thing to notice is that it is not the case that you are

authorized to ascribe a belief that p to me only if I assert p or assent to it when it is proposed as a belief candidate. Beliefs are not (or are not merely) occurrent mental states which are only contingently connected to behavior; they are also partly disposi- tional. Bain28 may have gone too far in saying that belief has no meaning except in reference to actions, but he was surely right in saying that nothing can count as a belief ``that does not directly or indirectly implicate our voluntary exertions.'' The same point is made more clearly by Price,29 according to whom ``A believes that p'' is a dispositional statement which is equivalent to a series of conditional statements describing how A would behave in certain circumstances. Similarly, Braithwaite30 distinguishes ``entertaining

27 See Kraut, 1983: 65±66; and Vlastos, 1983b: 72±73. 28 1880: 112. 29 1969: 275±76. 30 1932±33: 132±33.

Cross-Examining Socrates46

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

p'' ± the ``subjective or phenomenological'' component of belief ± from ``being disposed to act as if p were true'' ± the ``objective or behaviouristic'' component ± and argues that the latter is the dif- ferentia which distinguishes actually believing p from merely enter- taining it in thought. By the ``di¨erentia'' of belief, he does not mean a set of speci®c actions, but rather a set of dispositions to act which are ``actualized in suitable circumstances.''31 Beliefs are often ascribed to (or withheld from) people on the basis of their behavior: not just their overt actions, but also their emotions (or lack of them), their responses (or failures to respond), their facial expressions and tone of voice (or lack of them), and other things they say (or do not say) and do (or do not do). It is not the case that it is only in asserting (or assenting to) p that I incur epistemic responsibility for believing p together with whatever logical or be- havioral commitments believing (or assenting to) it may entail. In short, asserting (or assenting to) p is not a necessary condition for ascribing a belief that p to me. But neither is it a su½cient condition. Just as I can believe (or

assent to) p without asserting it, so I can assert (or assent to) p without believing it. The mere fact that I assert (or assent to) some proposition p is not an adequate criterion for ascribing a belief that p to me ± not even if I preface my assertion with hearty as- surances that I ``mean what I say'' and other displays of sincerity. Sincerity is not (or is not merely) an occurrent inner state either. If it were, there would be no public criteria of sincerity and hence no way of telling who is (and who is not) sincere. But although sincerity is not wholly behavioral and dispositional, the criteria for ascriptions of it are. Like belief ascriptions, sincerity ascriptions are criteriologically connected to behavior. Far from being an inner state which must be inferred, often problematically, from behavior, sincerity (or insincerity) is often manifested in behavior. People radiate sincerity (or insincerity). It is written all over them. It is, of course, reasonable for you to assume that if I believe

that p, I will assert p in apposite circumstances and assent to p when it is proposed for my assent. But the converse is not true.

31 Ibid.: 137. So, too, Cohen, 1992: 115: ``[T]o accept that p is to adopt the policy of taking the proposition that p as a premiss in appropriate circumstances,'' i.e., to include p ``among one's premises for deciding what to do or think in a particular context.''

Elenchus and sincere assent 47

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Taken by itself ± detached from speci®c contexts and disconnected from my past, present, and future behavior ± no utterance (asser- torial or otherwise) is a su½cient condition for a belief ascription. Like ascriptions of inner states generally, belief ascriptions have public criteria. If you are reasonably to ascribe a belief that p to me, there must be some discernible connection between my belief that p and my past, present, and future behavior. If I claim to believe that p but never (or seldom) behave in appropriate ways or behave in ways which are incompatible with believing it, you may reasonably doubt that I believe it, however vigorously I assure you that your doubts are groundless. This behavior includes linguistic behavior. Not only must my

beliefs be compatible with my past, present, and future actions; they must also be compatible with my past, present, and future beliefs and broader set of linguistic practices. Beliefs are seldom analyzable in- dividually in isolation from a person's belief-set. Belief-candidates do not typically present themselves individually in a way that en- ables a person to acquire them one at a time. Nor are they typically abandoned one at a time. To acquire or to abandon a belief is not typically to change one's mind about the truth-value of a single proposition; it is typically to acquire or to abandon a cluster of propositions whose truth-values are systematically interconnected and which stand or fall together. Hence if I assert (or assent to) some proposition p which is not manifested in, or is incompatible with, my past, present, and future behavior ± linguistic or other- wise ± and which stands in no discernible connection to, or is in- compatible with, my overall belief-set, you may reasonably doubt that I really believe what I say I believe. Such utterances do not count as sincerely held beliefs. The importance of all this for determining whether the propo-

sitions to which Socrates' interlocutors assent are expressive of their sincerely held beliefs can hardly be exaggerated. Any even remotely plausible account must take careful note both of the compatibility (or lack of it) between what they say and how they behave and of the continuity (or lack of it) between the proposi- tions to which they assent during their dialectical encounters with Socrates and the propositions they brought with them into the discussion and to which they would have assented (or would have been disposed to assent) before these encounters. In saying this, I am not succumbing to the temptation of turn-

Cross-Examining Socrates48

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

ing the early dialogues ``face-about'' in hopes of ``®nding more on the back.''32 To ask: What would Socrates' interlocutors have said (or have been inclined to say) before their encounters with him? is not like asking: What subjects did Hamlet study in Wittenberg? or: How many children had Lady Macbeth? Unlike the latter, which are pseudo-questions about quasi-factual details to which the text provides no answers, the former are questions about character, belief, and motivation. The fact that Socrates' interlocutors are ®ctional characters does not entail that they did, thought, and be- lieved nothing ``before'' their encounters with him. As in literary works generally, the chronology of a Platonic dialogue is not con- ®ned to the events that take place between its covers. Although ®c- tional characters have no empirical reality (and hence no empirical pasts and futures), they have imaginative reality (and hence imagi- native pasts and futures). Imaginatively considered, they are ``real'' enough ± as ``real'' as Tess of the D'Urbervilles or Hamlet.33 Critics who write books with titles like What Happens in Hamlet?

do not con®ne themselves to events that take place during the play. They also talk about events that took place before it begins. In doing so, they are not guilty of conceptual confusion. Lots of things happened before Hamlet begins: Claudius killed Hamlet's father, Gertrude hastily (albeit unwittingly) married her husband's murderer, Hamlet was overtaken by melancholy and became dis- enchanted with Ophelia and with women generally, and so on. In addition to performing many actions before the play begins, Hamlet held many beliefs ± some of them true (e.g., Ghosts exist), others false (e.g., My father died of natural causes). He was also a very di¨erent person. We know this because the other characters are constantly pointing out that he is not his old self and marvel- ing at the change that has come over him. If it is intelligible to talk about what Shakespeare's characters

did and believed before the play begins, it is equally intelligible to talk about what Socrates' interlocutors did and believed before the dialogues begin. To cite but a single example, before the Euthyphro begins, Euthyphro's father had been guilty of criminal negligence which resulted in the death of one of his slaves ± an act which had

32 The terminology is Kitto's, 1966: 15±16. 33 As Stanley Cavell, 1987: 102±3, rightly observes, to say that the existence of the charac-

ters encountered in literature is ``®ctional'' is not an answer to a question; it is the name of a philosophical problem.

Elenchus and sincere assent 49

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

morally outraged Euthyphro and impelled him to prosecute his father for murder. To deny that these events took place and that propositions about them and about the beliefs of the persons who performed or witnessed them have truth-values ± i.e., are true or false within the world of the work ± is to misunderstand the ``logic'' of ®ctional statements and the ``reality'' of the characters who utter them. It is also to trivialize the early dialogues by depriving them of their dramatic power and their psychological impact. Since it is intelligible to say that Socrates' interlocutors held

beliefs before their dialectical encounters with him, it is important to determine whether the beliefs they bring with them into the discussion are the same as the propositions they assert (or to which they assent) during it. If there are good reasons for thinking that they are not, then there are also good reasons for denying that Socrates elicits their sincere assent. What do Socrates' interlocutors believe? The obvious answer is

that they believe what they say they believe, that is, they believe the propositions which they assert and to which they assent. If Michael Stokes34 is right, Socrates never (or rarely) advances as- sertions himself; he merely asks questions to ascertain what his interlocutors believe, and the subsequent course of the discussion depends entirely on the answers he receives. For example, when he asks Charmides whether temperance is noble, whether temper- ate men are good, and whether only that which is good can be productive of good (Ch. 160e6±13), we should not assume that he himself believes these propositions. And similarly throughout the early dialogues. In a word, Socrates' questions are questions, not disguised assertions to which he himself is committed.35 The inter- locutor is free to answer these questions in any way he wishes. He need not answer them a½rmatively; but if he does, the answers are his and, therefore, the theses examined are also his. In support of these contentions, Stokes produces two pieces of

textual evidence: ®rst, when Socrates' interlocutors realize that their previous admissions entail the negation of their original theses, they often try to avoid logical shipwreck by saying that

34 1986: 6±7. 35 Ibid.: 4±9. The same analysis is given by Teloh, 1986: 1±2; Kidd, 1992: 82±92; and Frede,

1992: 205±6.

Cross-Examining Socrates50

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

their original theses are indefensible if what Socrates has said is true. Second, Socrates never countenances this move. He always insists that he himself has advanced no assertions whatever and that the inconsistent propositions are traceable to his interlocutor. For ex- ample, when Protagoras tries to saddle Socrates with the thesis that the ``parts'' of virtue are related in such a way that none re- sembles any other, Socrates replies: ``I didn't say that. Protagoras here said that in answer to my question'' (Pr. 330e7±331a1).36 Stokes is right on both counts. But the logical situation is more complicated than he seems to think. It is true that Socrates is the questioner. And questioners ask

questions. However, Stokes overlooks the fact that grammatical interrogatives are not always logical interrogatives and, therefore, not always (or not merely) straightforward requests for informa- tion. Every question has its own logic which is a function of the in- tention and purpose of the questioner in asking it. Di¨erences in intention or purpose on the part of the questioner invest the ques- tions asked with di¨erent illocutionary forces (and illocutionary- force potential) which enable the questioner to perform a wide vari- ety of speech-acts under cover of ``just asking questions.'' Of course, some questions are straightforward requests for in-

formation, asked either in ignorance or with incomplete under- standing which the questioner is trying to remedy, for example, ``Is there a faculty meeting this afternoon?'' But others are not so innocent, e.g., ``There is a faculty meeting this afternoon, isn't there?'' Although this could be a straightforward request for in- formation, it need not be. Indeed, it need not (logically) be a ques- tion at all. Such locutions ± which J. L. Austin calls ``utterances in

36 The Alcibiades I contains what is perhaps the clearest example of this common Socratic rejoinder. When Alcibiades tries to saddle Socrates with an equally unwelcome conclu- sion, Socrates asks: ``Who do you think is saying these things ± me, the questioner, or you, the answerer?'' (113a1±2). When Alcibiades admits that it is his views which are under examination, Socrates emphatically asserts that, in that case, he himself cannot be held responsible (e� meÁ . . . ai� tiaÄÎ maÂ thn, 113c4) either for the theses advanced or for the conclusions arrived at; his interlocutor must bear full responsibility for both (112d7± 113c4). (Even if the Alcibiades was not written by Plato, as many believe, few would dis- agree with Shorey's claim, 1933: 415, that it contains nothing which is ``necessarily un- Platonic.'') MeÂron is one of the few commentators who questions the claim that it is always the interlocutor's thesis that is examined. She adds that, in interpreting the early dialogues, we need to be more subtle and examine the overall context to determine which interlocutors respond spontaneously and which respond in ways in¯uenced by Socrates (1979: 13±14).

Elenchus and sincere assent 51

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

a speech situation''37 ± often have a hidden agenda which enables the speaker to ``do things with words'': to seek reassurance, to bemoan his poor memory, to issue a reminder to his hearer, or even to deplore his habitual lack of attendance. Such ``ques- tioners'' advance an assertion ± ``There is a faculty meeting this afternoon'' ± and then ask their interlocutor to endorse it ± ``Isn't there?'' What if he does endorse it? Whose assertion is it? According to

Stokes, it is the interlocutor's. Is Stokes right? If, in response to my ``question,'' ``There is a faculty meeting this

afternoon, isn't there?'', my hearer says, ``Yes,'' it is his answer in the sense that it was he who produced the a½rmative response. However, he was responding; and the proposition to which he was responding was mine. Had I not asked the ``question,'' the matter would probably not have occurred to him. Furthermore, in answer- ing my ``question'' a½rmatively, he did not perform the speech- act of asserting a proposition but the very di¨erent speech-act of con®rming a proposition; and the proposition con®rmed was not his, but mine. Some questions are more loaded still and clearly calcu- lated to secure the desired response, e.g., ``You did promise that I could borrow your car for the weekend, didn't you?'' or ``Jimmy is sorry, aren't you Jimmy?'' Similarly, when queried in the presence of his suspicious wife about whether he approves of adultery, the habitual adulterer can be relied on to issue vehement denials. Although many commentators attach great weight to the sincere

assent requirement and gravely observe that if the answerer fails to abide by it, the discussion will su¨er in serious ways, they seldom add that the questioner has analogous responsibilities and that if he fails to discharge them, the discussion will su¨er in equally serious ways. Socrates often fails. His questions often have a hidden agenda and are calculated to secure the desired re- sponses. Although these responses are (usually) e¨ortlessly elicited and (often) monotonously predictable ± ``Yes,'' ``Necessarily,'' ``To be sure,'' ``Apparently,'' and so on ± they should not be taken at face value and as necessarily expressive of the interlocutors' sin- cerely held beliefs. They often signal nothing more than the philo- sophically uninteresting fact that the interlocutor sees no good reason for not assenting. But it does not follow that he believes the

37 1962: 138.

Cross-Examining Socrates52

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

propositions to which he has assented, i.e., that they are expressive of antecedently formulated and sincerely held beliefs which he brought with him into the discussion and would have a½rmed (or been disposed to a½rm) before it. To cite but a single example, would anyone seriously suggest that, in ``assenting'' to the Socrati- cally proposed thesis that there is a single eidos or idea common to all pious actions ``by which'' they are pious and which constitutes the standard for distinguishing pious actions from impious ones, Euthyphro is giving expression to a belief he held prior to being asked? The suggestion taxes credulity. Although Socrates implies that nothing could be easier, saying

what one really believes is often very di½cult. People hesitate: not because they are unre¯ective or dishonest or hedging, but because they are not sure. Beliefs are not ``cognitive'' states of mind carried about ``in one's head'' like groceries in a bag, so that, upon being asked whether one believes a particular proposition, one need only look within to see whether the belief is there, as one inspects one's grocery bag to see whether one remembered to buy shampoo. Then, too, we are often asked for our opinions on subjects about which we have not made up our minds (``For whom do you intend to vote?'') or on subjects about which we have no opinions and probably never will (``Have you ever heard anyone modulate so e¨ortlessly into such a distant key?''). Nor, pace Vlastos, does the absence of sincere assent necessarily render the interlocutors' assent ``i¨y'' ± tentative, non-committal ± and, therefore, insincere. Here, as elsewhere, radical dichotomies and exclusive disjunctions will only lead us astray. Socrates' interlocutors often assent to prop- ositions routinely, perfunctorily, and uncomprehendingly. They do so for a variety of reasons: to seem more astute than they really are, to give the impression of having followed the argument more closely than they really have, or simply because Socrates expects them to. Judging from the hopeless predicaments in which they often become embroiled, it seems that the answer they should have given ± the only truly sincere one ± is neither ``Yes'' nor ``No,'' but rather ``I have no opinion one way or the other. The proposition has never entered my head.'' Vlastos has a ready reply. The fact that a proposition has never

entered an interlocutor's head is perfectly compatible with saying that he believes it ``in that marginal sense of the word in which we may all be said to `believe' innumerable things that have never

Elenchus and sincere assent 53

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

entered our heads, but are nonetheless entailed by what we believe in the common or garden use of the word.''38 Such beliefs are not ``overt,'' but ``covert.'' Hence ± the example is Vlastos's ± if I overtly believe that Mary is John's sister and that John is Bill's grandfather, then I covertly believe that Mary is Bill's great-aunt ± ``even if I have never thought of that fact ± indeed, even if I do not have a word for `great-aunt' in my vocabulary.'' Formulated in full generality, anyone who overtly believes p and who also overtly believes q and r, which entail not-p, covertly believes not-p. Socrates apparently holds the same view. Why else would he breezily assure Polus that he does not ``really'' believe that it is worse to su¨er injustice than to commit it and that he ``really'' believes the converse, since he believes other propositions which entail it (G. 474b2±5)? Something is wrong here. If I overtly believe p and also overtly

believe q and r, which entail not-p, then I am, of course, logically committed to believing not-p. Hence upon learning the meaning of ``great-aunt,'' I will instantly concur that ``Mary is John's sister'' and ``John is Bill's grandfather'' jointly entail ``Mary is Bill's great-aunt.'' But it does not follow that I covertly believed this all along.39 A proposition which I do not overtly believe (and, in fact, claim to disbelieve) but which is entailed by other propositions which I do overtly believe is not thereby proved to be a proposi- tion which I covertly believe. Polus no more entered the discussion covertly believing that it is worse to commit injustice than to su¨er it than Crito entered Socrates' prison cell covertly believing that he should remain in prison and drink the hemlock. This theory of covert belief is also incompatible with Socrates'

claim that his divine mission consists in demonstrating that his interlocutors are ignorant. On the view before us, Polus is not ignorant; he is just muddled. He thinks he believes that it is worse to su¨er injustice than to commit it, but he ``really'' believes the converse. In that case, however, the required conclusion is not that Polus does not know what he thinks he knows, but rather that he is not aware of what he does know or, at least, truly ± albeit covertly ± believes ± a remarkable claim which entails either that

38 1983a: 51; see also 1994: 23. 39 On the unintelligibility of the notion of covert belief, see Dummett, 1975: 285±86, and

MeÂron, 1979: 14.

Cross-Examining Socrates54

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Polus di¨ers from every other interlocutor in the early dialogues or that all the other interlocutors are in the same epistemic condi- tion and Socrates' description of his divine mission in the Apolog y is inaccurate. It is equally misleading for Vlastos to claim that whenever an

interlocutor advances a thesis which Socrates considers false, he is con®dent that he will always ®nd others in his interlocutor's belief- set which entail the Socratic thesis.40 Since Socrates' ``questions'' are often calculated to secure the desired response, it is not a mat- ter of complete indi¨erence to him how his interlocutor replies, provided that his reply is expressive of what he really believes. It is not as if, owing to incredible good luck, Socrates always just hap- pens to elicit the very answers he needs to refute his interlocutor; on the contrary, he knows exactly what answers he needs and his questions are tailored to secure them. He does not ransack his in- terlocutor's belief-set to ®nd the propositions needed to refute him; he supplies them himself. And if his initial questions do not secure the desired response, he continues asking questions until they do. This policy is often defended on the ground that Socrates

cannot accept just any answers his interlocutor happens to give; he must elicit the right answers. And one elicits the right answers by asking the right questions. Although Socrates has traditionally been celebrated for his ability to do precisely this, the celebration is premature. A questioner can no more claim that he asks the right questions than a clothing retailer can claim that he sells the right sizes. Questions are not right or wrong in themselves, and their rightness or wrongness cannot be assessed in a vacuum. The ``right'' question is the one that accomplishes the questioner's purpose(s); the ``wrong'' question is the one that does not. Ac- cordingly, a question can be ``right'' in one context but ``wrong'' in another. Furthermore, insofar as Socrates ignores what his interlocutor really believes and elicits his assent to premises which he himself supplies, he operates with a surreptitious criterion of what counts as a satisfactory answer which is very di¨erent from his announced criterion: a satisfactory answer is no longer one which is expressive of what his interlocutor really believes, but rather one which enables Socrates to refute him, even if, as is often the case, it is not expressive of what he really believes.

40 1983a: 52±53. See also 1985: 18, n. 44; and 1994: 56, n. 43.

Elenchus and sincere assent 55

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

However, the most crucial question of all is not whether Soc- rates' interlocutors sincerely assent to the propositions proposed for their assent, but the degree of Socrates' own commitment to the sincere assent requirement. And this, I submit, is open to ques- tion on several counts. The ®rst, inconclusive in itself but worth mentioning, is that the requirement is absent from most of the early dialogues. It is, in fact, only invoked in four: the Crito, the Protagoras, the Gorgias, and Republic i.41 Second, even when it is invoked, it is often nothing more than an e¨ective psychological technique for luring the interlocutor into discussion. (Euthyphro is assured that Socrates envies his uncommon theological grasp and wishes to become his disciple.) Third, and more problematic, having invoked the requirement at the outset of a discussion, Soc- rates does not always enforce it and sometimes waives it alto- gether. Indeed, in the Gorgias and Republic i, he waives it the min- ute the exchange becomes heated, and in the Protagoras he waives it almost immediately.42 Finally, and most problematic of all, even when he does enforce it, he does not always take pains to insure that the thesis he refutes is the one his interlocutor originally a½rmed rather than some tendentious reformulation of it. Thra- symachus is a case in point. In response to his de®nition of justice as the advantage of the stronger, Socrates asks whether he means that the person who eats the most beef, and hence is strongest, is the most just. Thrasymachus' outraged reply is exactly on target:

You disgust me, Socrates. Your trick is to take hold of the argument at the point where you can do it the most harm. (R. 338d3±4)

In this passage Socrates is caught in the act of deliberately mis- representing his interlocutor's thesis. His ostensibly humorous but patently lame reply ± ``I was just trying to get you to express yourself more clearly'' (338d5±6) ± is idle. Other interlocutors register similar complaints. Yet even in the face of their protests, Socrates continues to claim that he is refuting them ``from their

41 Vlastos, 1991: 111, n. 21, accounts for its absence by claiming that although the require- ment is ``generally taken for granted'' and ``mentioned only when there is special need to bring it to the interlocutor's notice,'' it is a ``standing rule'' of elenctic debate.

42 According to Vlastos, 1983a, Socrates waives the requirement ``only as a pis aller and under protest, so that the argument may go on'' (38); at R. 340c1±2 his ``apparent will- ingness'' to waive it is ``ironical'' (34, n. 24), and at Pr. 333b8±c9 he suspends it only to allow Protagoras to ``save face'' (38). As we will see, Socrates waives the requirement much more often than Vlastos thinks and for very di¨erent reasons.

Cross-Examining Socrates56

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

own beliefs.'' All the more important, then, that he attend to what they actually say instead of tendentiously reformulating their as- sertions and then refuting his own reformulations. To sum up, Socrates explicitly invokes the sincere assent re-

quirement with only four interlocutors (Crito, Protagoras, Thrasy- machus, and Callicles); he enforces it with only one (Crito); and he does not so much as mention it with twelve (Euthyphro, Hippias, Charmides, Critias, Gorgias, Polus, Ion, Laches, Nicias, Cephalus, Hippocrates, and Polemarchus).43 This meager textual base is too precarious a foundation on which to erect so momentous a meth- odological principle. The requirement is too infrequently invoked, too arbitrarily enforced, and too often suspended to justify elevat- ing it to the status of a ``standing rule'' which, although ``generally taken for granted'' and ``mentioned only when there is a special need to bring it to the interlocutor's notice,'' is nevertheless pre- supposed throughout the early dialogues.44 Given Socrates' frequent violations of the sincere assent re-

quirement, we need to distinguish the arguments to which Vlas- tos's schematization of ``peirastic'' argumentation is applicable (in which Socrates abides by the requirement and refutes the thesis a½rmed by his interlocutor) from those to which it is not (in which he disregards the requirement and refutes a tendentiously refor- mulated thesis). We also need to amend Vlastos's schematization of ``standard'' elenchus as follows:

(i) The interlocutor asserts some thesis, say, p, which Socrates construes, i.e., misconstrues, as p�.

43 I anticipate the following objection. If A lays down a procedural stipulation in the pres- ence of B, C, and D, it applies equally to them all. He need not repeat it to each succes- sively. Once su½ces for a group. Although this objection seems eminently reasonable, it overlooks the crucial question of time. If A lays down the stipulation during the initial discussion with B, it is reasonable to claim that it applies subsequently to C and D. However, if he does not lay it down until the subsequent discussion with D, it is unrea- sonable to claim that it applies retroactively to B and C. That is precisely what happens in the Gorgias and Republic i where the sincere assent requirement is not invoked until the discussion with the third interlocutor ± Callicles in the former and Thrasymachus in the latter.

44 To my knowledge, the only commentators to have noticed this are Kahn, 1992: 255±56, according to whom sincerity is more a ``desideratum'' than a rule or, at best, a rule ``more honored (or at least more often mentioned) in the breach than in the observance'' (see also 1996: 118), and Nails, 1993: 286±88; 1995: 92±95, according to whom Socrates ``gave only lip service'' to the sincere assent requirement and ``merely pretends'' that his interlocutors must answer in accordance with their actual beliefs.

Elenchus and sincere assent 57

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

(ii) Socrates secures his interlocutor's quali®ed (sometimes heavily quali®ed) agreement to further premises, say, q and r.

(iii) Socrates then argues, and the interlocutor agrees, that q and r entail not-p�.

(iv) Thereupon Socrates claims that not-p� has been proved true, p� false.

Vlastos's schematization is inapplicable to all the Socratic argu- ments that fall into the second category. Armed with this distinction, we can provide a less abusive ex-

planation of the interlocutor who, having a½rmed p and to some degree assented to q and r (which entail not-p�), does not abandon p and a½rm not-p�. We need not berate him as stubbornly perverse and unwilling to follow the argument wherever it leads. There is a simpler and kinder explanation. The reason he does not a½rm not-p� is that he never a½rmed p�; he a½rmed p. Even if the thesis refuted is the one originally a½rmed by the

interlocutor, he often still has good reasons for refusing to aban- don it. Every valid deductive argument is either a proof of its conclusion or a disproof of the truth of at least one of its premises. Clearly, the degree of the interlocutor's con®dence in not-p, the conclusion of Socrates' argument, can be no stronger than the degree of his con®dence in q and r, the premises from which it is deduced. Hence having a½rmed p, having given only quali®ed assent to q and r, and having perceived that they entail not-p, he may reasonably begin to suspect that his assent to q and r had been hasty and that he had been ill-advised to yield to the pressure brought to bear on him by Socrates. In such a situation, his reluc- tance to abandon his original thesis would not only be intelligible; it would be justi®ed. Refutation by tendentiously reformulated thesis is not the only

unscrupulous dialectical tactic employed by Socrates. But it is one of the more common ones ± a fact which reveals that the sincere assent requirement is neither as important nor as systematically operative as we have been given to believe.

Cross-Examining Socrates58

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

c h a p t e r 3

Crito

Throughout this study we need to proceed carefully with our eyes on the texts. The reputations of Socrates' interlocutors go before them, and of no one is this more true than Crito. Universally dis- missed as one of the most super®cial and non-morally1 motivated interlocutors in the early dialogues, he is usually described as a wealthy and well-connected businessman who is unduly worried about how his reputation will su¨er if people mistakenly thought that he had been unwilling to provide the funds necessary to save Socrates' life and whose fear of public opinion has prompted him to spring Socrates from prison and thereby e¨ect an eleventh-hour miracle. R. E. Allen's opinion is representative:

Crito's plea begins, ends, and is mainly based on ``how things will look.'' The pivot of his reasoning, to the degree that it has a pivot, turns on the connected concepts of shame and success . . . The good man succeeds and helps his friends . . . The bad man fails, and in his failure he becomes disgraced . . . Crito stands as an advocate, pleading a cause to his friend in behalf of his friend, using, as a pleader will, such terms as he can muster to persuade. The randomness of his persuasion answers to an underlying incoherence in his plea.2

Although this seems to be the received opinion among students of Plato,3 it is a misdescription of Crito and a misdiagnosis of his motivation. Crito is not the well-meaning but misguided simpleton he has

1 Some commentators grudgingly concede that Crito's remarks bear witness to a morality of sorts, but they disparage it as ``average Athenian morality'' (see J. Adam, 1888: xvi; and A. E. Taylor, 1929: 170).

2 1984: 108±9. 3 See, e.g., J. Adam, 1888: xv±xvi; A. E. Taylor, 1924: 258; Burnet, 1924: 178; Shorey, 1933: 84; FriedlaÈnder, 1964, ii: 75; Santas, 1979: 11±12; Tejara, 1984: 17; Teloh, 1986: 118±19; Kraut, 1984: 25; and Brisson, 1997: 180.

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

been made out to be. He has not been belatedly spurred into action by the prospect of social disgrace; he has been involved with Socrates' escalating ± and, to his mind, largely self-in¯icted ± problems with Athens from the very beginning. His caustic capsule summary of recent events betrays his mounting displeasure at the way things have been handled: the whole trial could have been avoided, Socrates' conduct was shameful, and his friends have emerged looking like cowards. Socrates' imminent death is the crowning absurdity (Cr. 45c5±46a8).4 In short, Crito has had enough. He appears in Socrates' prison cell as a loyal but exas- perated critic of his eccentric friend whose imprudent conduct has gotten so out of hand that it threatens death for him, grief for his friends, and abandonment and educational disaster for his sons. His impassioned case ± multi-faceted but uni®ed by an underlying moral principle ± has been largely ignored by commentators. It is also largely ignored by the vaguely attentive Socrates who focuses on one aspect of the case and ignores the rest. It is about time that someone took Crito seriously. The historical Crito was a lifelong friend and benefactor of

Socrates. According to Diogenes Laertius, it was he who liberated Socrates from manual labor, ``being struck by his beauty of soul'' (2.20±21), educating him and providing for his needs. Diogenes also reports that Crito was the author of seventeen philosophical dialogues on a variety of topics ranging from What is knowledge? and That men are not made good by instruction to What is expedient? and On tidiness (2.121). This is more than a little surprising. As portrayed by Plato, Crito is not the sort of man who writes philosophical dialogues. His dramatic counterpart appears in three dialogues: the Crito, the Euthydemus, and the Phaedo. He is also mentioned among those present at Socrates' trial (Ap. 33d9, 38b6). The Crito of the Euthydemus is a prosperous urban dweller and

the owner of several farms. Unlike Cephalus, who is also wealthy but not overly fond of money (R. 330b8±c2), Crito is determined to accumulate as much wealth as possible so his sons will be hand- somely provided for (Euthyd. 304c3±5, 306d2±e3). Although Soc- rates periodically alludes to Crito's obsession with wealth and jok-

4 Grote, 1867, i: 298, n. b calls this a ``remarkable passage'' which reveals that, in the opin- ion of Socrates' closest friends, his trial and death could have been avoided ``without anything which they conceived to be dishonourable to his character.''

Cross-Examining Socrates60

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

ingly urges him to study with Euthydemus and Dionysodorus, whose art is not a hindrance to money-making, he neither rebukes him for his obsession with money nor warns him about its poten- tially harmful e¨ects on the soul of its possessor. Throughout the Euthydemus he seems to accept Crito pretty much for what he is. That is not the case in the Crito. Whereas Socrates is compara-

tively gentle with and even slightly amused by the money obsessed and (apparently) morally neutral Crito of the Euthydemus, he is quite critical of and even somewhat disappointed by the altogether sober and morally earnest Crito of the Crito. Although roughly the same age and from the same deme (Ap.

33d9±e1), Socrates and Crito are strikingly di¨erent: the former is poor,5 shabbily clothed, and devoted to philosophy; the latter is rich, elegantly clothed, and devoted to augmenting his net pro®ts. Unlike some members of the Socratic inner circle, who hang on Socrates' every word and even imitate his manner of dress, Crito is a no-nonsense businessman, and he acts and dresses the part. No going about bare-footed for him. Nor does he completely ap- prove of some of Socrates' other activities: he rebukes him for as- sociating with Euthydemus and Dionysodorus and scolds him for taking lyre lessons at his advanced age in a class made up of young boys who a¿ict him with such constant ridicule that he ®nally re- cruited several other senior citizens to enroll in the class in hopes of di¨using the abuse (Euthyd. 272c3±d1). Crito enjoys eavesdropping on philosophical discussions in his

spare time (Euthyd. 271a1±3) and praises philosophy as ``a charm- ing thing'' (Euthyd. 304e6±7), but it has not taken hold of him. Unlike the other-worldly and metaphysically susceptible Simmias and Cebes, who are deeply a¨ected by Socrates' discourse about the immortality of the soul and ®lled with longing for the bliss awaiting those whose lives have been holy and just, the utterly this-worldly and metaphysically impervious Crito abruptly breaks the spell by urging Socrates to refrain from talking so much on

5 Like many commentators, Burnet, 1924: 98, explains that Socrates was not always poor and attributes his later (and largely self-induced) poverty to his divine mission. For a less awestruck explanation, see Wills, 1993: 37: ``Those who draw a romantic picture of poor Socrates living free of material encumbrance resolutely overlook the fact that a person who could a¨ord to study with the Sophist Prodicus, who lived as an intellectual inquirer without the need to work, and who quali®ed as a property-holding hoplite free to train and campaign in the wars, obviously had an estate of some sort . . .''

Crito 61

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

the ground that over-excitement can counteract the e¨ect of the poison he is soon to ingest (Ph. 63d5±e2). Not normally one to publicize his failures, Socrates candidly admits that he has failed to persuade Crito that the soul is immortal and that, in burying him, he will only be burying his body (Ph. 115c5±d6). Crito lets the remark pass. He is a practical man who prides

himself on living in ``the real world.'' Right now he has more im- portant things to worry about, and there is no time for philosophy; for busy people like Crito, there seldom is. Only when he is in the company of Socrates does it seem like madness (maniÂ an) to have neglected the education of his sons ± especially Critobulus, who is growing up and in need of guidance (Euthyd. 306d6±e3). Like Alcibiades, who is overcome by shame when he is in the presence of Socrates but backslides the minute he is out of his sight (Sym. 216a8±b5), Crito needs Socrates' sobering presence to keep his parental responsibilities in the forefront of his consciousness. Of course it is madness to worry more about his sons' ®nancial security than about their education, and he wonders why it is so hard to remember this. But if Socrates' presence is a powerful reminder of the impor-

tance of philosophy and education, his counsel falls on deaf ears:

[P]ay no attention to the practitioners of philosophy, whether good or bad. Rather give serious consideration to the thing itself: if it seems to you negligible, then turn everyone from it, not just your sons. But if it seems to you to be what I think it is, then take heart, pursue it, practice it, both you and your sons. (Euthyd. 307b6±c4)

Good advice, perhaps; but it is largely wasted on Crito who, in spite of his periodic twinges of guilt and mild attacks of self- knowledge, is an unlikely candidate for philosophy. In turning to the Crito, we encounter a recognizable but, in

many respects, very di¨erent person. A dramatic rather than a narrated dialogue, the Crito depicts a conversation that takes place before dawn in Socrates' prison cell. Asked why he is up and about so early, Crito divulges that he is the bearer of bad news. The ship from Delos, whose sacred voyage had delayed Socrates' execution, has been sighted at Sunium and will soon reach Athens ± which means that Socrates will be executed the next day. Soc- rates begs to di¨er. He has just awakened from a dream in which a

Cross-Examining Socrates62

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

beautiful woman in white informed him that he would not die for three more days. Crito is unimpressed. Like sensible men every- where, he gives little credence to nocturnal revelations; besides, this is no time to be quibbling about the empirical accuracy of prophetic utterances. The situation is desperate. There are deci- sions to be made and things to be done. In the interim between Socrates' trial and execution, Crito has not been idle; he and sev- eral other friends of Socrates have arranged for his escape. In the conversation that follows, Crito tries to persuade him to avail himself of this golden opportunity. He begins by pouring out his heart. Socrates' death will be a

double calamity: not only will he lose an irreplaceable friend; his reputation will also be dealt a shattering blow. People will not believe that he had tried to help and that Socrates had repeatedly resisted his overtures ± a remark which suggests that Crito's pres- ent plea had been preceded by others.6 And what, he rhetorically asks, could be more shameful (ai� sciÂ wn) than to be thought of as a man who cares more about his money than about his friends (44c2±3)?7 Socrates' response is surprising. We expect him to tell Crito to

ignore the unstable and ¯uctuating opinions of ``the Many.'' But although he says this a few Stephanus pages later, he does not say it here. Instead, he tries to soften the blow that Crito thinks his reputation will su¨er by o¨ering him a piece of consolation: the most reasonable people (oi� . . . e� pieikeÂ statoi ), whose opinion alone is worthy of consideration, will understand what has hap- pened and not think less of him (44c6±9). The distinction Socrates makes is not between worrying about what people will think and not worrying about it, but between worrying about what reasonable, as opposed to unreasonable, people will think. In saying this, he seems (temporarily at least) to be operating on Crito's own ground ± not urging him to renounce all concern for his reputation, but directing it to a more worthy object. Crito disagrees. One must worry about public opinion, as is

6 The same thing is implied at Cr. 48e1±3. Socrates tells Crito to stop trying to persuade him with the same arguments.

7 Woozley, 1979: 15±16, thinks Crito's worries are unfounded on the ground that Socrates was not a popular hero and it was highly unlikely that his friends would have been ac- cused of stinginess or cowardice for failing to rescue him.

Crito 63

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

abundantly clear from Socrates' present legal predicament. Get a bad name with ``the Many'' and their capacity for in¯icting the greatest evils has no limit (44d1±5). Socrates thinks otherwise: would that ``the Many'' could in¯ict the greatest evils; for then they would also be capable of the greatest good.8 In fact, they can do neither. Unable to make a man either wise or foolish, they do what they do haphazardly (poiouÄ si deÁ touÄ to o� ti a� n tuÂ cwsi, 44d9±10).9 Crito does not dispute the point, but neither does he pursue it.

Instead, he returns to his previous train of thought and confesses that his impending sense of loss is particularly hard to bear because it is a preventable loss. Socrates need not die. Further attempts at persuasion follow. Crito hopes that Socrates' reluctance to escape is not prompted by worries about his friends' safety; if so, his worries are groundless. The required amount is comparatively small, everyone is willing to contribute and to incur whatever risks are necessary, and potential informers can be bought o¨.10 Finally, Socrates would be welcome in many other cities ± especially in Thessaly where Crito has powerful friends who will give him asylum. In addition to these personal and prudential reasons, Crito o¨ers

several moral reasons ± none of which is based on worries about his reputation. What Socrates is about to do is not right (ou� deÁ diÂ kaioÂ n, 45c5). In opting for death, he is knowingly embarking on a course of action which is harmful to himself and injurious to his friends. He is also derelict in his parental duties, abandoning his

8 Woozley, 1979: 14, thinks this is a non sequitur: from the fact that ``the Many'' can in¯ict the greatest evils, it does not follow that they are also capable of the greatest good. But Woozley has misread the passage. Socrates is not deducing an inference. His remark should be understood in light of his belief that no one errs voluntarily. Men do evil because of ignorance which, if replaced by knowledge, would enable them to do good; hence the greater their present capacity for evil, the greater their future capacity for good.

9 The Greek cannot mean `` `The Many' do whatever occurs to them'' (as many trans- lations have it). See J. Adam, 1888: 32; and Burnet, 1924: 260.

10 Crito is not implying that there are informers lying in wait to testify speci®cally against Socrates. The allusion is to a group of professional informers (sukojaÂ ntav, 45a8) ± public busy-bodies who made their living by threatening to bring charges (often false ones) against the rich who were usually willing to pay them o¨, thereby settling cheaply out of court. Xenophon reports that Crito himself was constantly besieged by these predators, and that Socrates had advised him to employ a counterinformer who could induce them to drop their charges by discovering skeletons in their own closets and threatening them with countersuits (Mem. 1.2.9).

Cross-Examining Socrates64

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

sons and voluntarily consigning them to the status of orphans.11 Instead of acting like a good and courageous (a� gaqoÁ v kaiÁ a� ndreiÄ ov) man, Socrates is taking the easy way out and is guilty of a kind of cowardice (a� nandriÂ aÎ tini, 45d4±8) ± disreputable conduct on the part of someone who professes to care about virtue. I have summarized Crito's case at some length ± partly to con-

vey its urgency and partly to call attention to the variety of argu- ments it contains. Unlike most commentators,12 who think Crito's single (or overriding) reason for urging Socrates to escape is his preoccupation with his reputation, Woozley recognizes that he ad- vances numerous reasons. He also recognizes that, insofar as Crito explicitly invokes the principle that one should never knowingly do what is harmful to others ± much less to one's own children13 ± several of his reasons are moral ones based on considerations about justice.14

11 Socrates' apparent lack of concern for his sons also surfaces at the end of the Apology. At 41e1±42a2 he makes a request in their behalf, but it is a very unusual one. Instead of im- ploring the jurors to look after them and to educate them in his stead, he implores them to avenge him by troubling his sons, as he has troubled Athens, if they care about any- thing more than virtue or if they think they are something when they are nothing. This request, utterly devoid of weak-kneed sentimentality about children and childhood, strikes many modern readers as decidedly cool, if not downright callous. (Woozley, 1979: 121, speaks of Socrates' ``distasteful detachment'' from his children.) It struck Crito the same way. Socrates' views about parental responsibility are highly contra-endoxic ± not only by modern standards, but also by ancient ones. Although the early dialogues pro- vide no reason for thinking that Socrates was unconcerned about his sons, they provide every reason for thinking that he was fundamentally concerned about them not qua sons, but qua potential fraudulent claimants to wisdom. That is, his most fundamental concern was a concern rooted in his divine mission.

12 Grote, 1867, i: 287±98; Teloh, 1986: 119±21; and Arieti, 1991: 137±38, are notable ex- ceptions. Teloh acknowledges Socrates' ``cavalier dismissal'' of most of Crito's reasons, but adds that he does respond to some of them and that the rest have ``clear Socratic answers'' (119).

13 Crito's reasoning is based on the general principle that one should either not have chil- dren at all or one should take their upbringing and education seriously (45d4±5). Soc- rates does not deny the principle, but he does deny that escaping from prison and avoid- ing death is the best way of discharging his parental obligations, thereby relocating Crito's principle within a Socratic hierarchy of value in which one's supreme obligation is not to be a good father, as conventionally understood, but to be a just man (48c2±d5, 54a1±d1).

14 According to Woozley, 1979: 12±13, those who deny that Crito's reasons are moral be- cause they spring solely from worries about his reputation have overlooked an important principle of moral reasoning, namely, ``If it is a good moral reason why A should not act in a certain way that B will su¨er if he does, then it is a good reason, whoever utters it . . . If that Crito and others will su¨er an undeserved reputation in the event of Socrates' death was a good moral reason in favor of Socrates' agreeing to the escape plan, then it was no less a good reason because it was Crito himself who advanced it.''

Crito 65

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Although Woozley penetrates further than most, his analysis is incomplete. That this is so is borne out by his overly harsh assess- ment of (what he takes to be) Crito's ®rst reason, namely, that Socrates' death will deprive him of an irreplaceable friend. Ac- cording to Woozley, Socrates ``does not bother'' to reply to this reason ± a reason which Plato included simply to reveal Crito's ``muddleheadedness in argument.''15 This criticism is unfounded. However deep Crito's personal attachment to Socrates, he does not o¨er this heartfelt sentiment as a reason for escaping from prison. He is not appealing to Socrates' pity (``Your death will make me very sad, therefore you should escape''); he is stating a fact. This is not ``muddleheadedness in argument,'' it is friendship. The reason Socrates ``does not bother'' to reply is that there is nothing to reply to. It is only after giving vent to these personal apprehensions that

Crito o¨ers his ®rst reason ± a reason which must be assessed in light of the declaration of friendship that preceded it. He fears that people who do not know them well will think he refused to save Socrates' life because he cares more about his money than about his friends. This troubles Crito: not because he is a slave to public opinion, but because the allegation is false. He has tried to save Socrates' life ± repeatedly: not simply because Socrates is his friend, but also because he thinks that it is right (diÂ kaioiÂ ), i.e., just, for him to escape and, therefore, that it is also right, i.e., just, for his friends to help him to escape (44e6±45a5). Crito does not want to be morally censured for being unwilling to do the very thing he has been trying to do. He is right. In view of his persistent e¨orts, it would be shameful if people upbraided him with these false alle- gations. It is idle to deny that Crito is worried about his reputa- tion, but his worries are those of a principled moral agent shrink- ing from unjust criticism, not those of a spineless sociopath in quest of peer approval. Again Socrates' reply is puzzling. Although Woozley goes too

far in saying that Socrates treats this reason ``with contempt,''16 he does not take it seriously. He again faults Crito for being unduly concerned about what people will think and again assures him that ``the Many'' cannot in¯ict the greatest evils.

15 Ibid.: 17. 16 Ibid.: 13.

Cross-Examining Socrates66

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Has Socrates contradicted Crito's claim? It certainly looks like it. A few Stephanus pages ago, Crito asserted that ``the Many'' can in¯ict the greatest evils, and Socrates now asserts that they cannot. In fact, however, there is no contradiction here. P and not-p are genuine (as opposed to apparent) contradictories, which signal genuine (as opposed to apparent) disagreement, if and only if both speakers mean the same thing. Crito and Socrates do not. In say- ing that ``the Many'' can in¯ict the greatest evils, Crito means that they can in¯ict irreversible harm on others, for example, they can put innocent people to death. Socrates is not denying that. The disagreement between them is not about whether ``the Many'' are capable of doing such things, but about whether, in doing them, they are in¯icting the greatest evils. Crito thinks they are; Socrates thinks they are not. In saying this, he is not so much denying Crito's claim but rede®ning ± and presumably deepening ± the meaning of ``the greatest evils.'' As he sees it, the greatest evils are not evils in¯icted on the body, but on the soul ± ``that part of us, whatever it is (e� keiÄ no, o� ti pot' e� stiÁ twÄ n h� meteÂ rwn), which is concerned with justice and injustice'' (47e8±48a1). It is true that ``the Many'' can put innocent people to death, but death is not the greatest evil. The greatest evil is not to be killed, but to be made unjust. Hence, appearances to the contrary, Socrates has not contradicted Crito's claim. But he has done something else. In rede®ning ``the greatest

evils,'' he has introduced a new ± and single ± criterion for deter- mining whether it would be just for him to escape. Unlike Crito's multiple criteria, which focus exclusively on bodily harms in¯icted on other people ± in particular, one's friends and children ± Soc- rates' single criterion treats these as secondary considerations and focuses on harms in¯icted on oneself ± in particular, on one's soul. It is true that ``the Many'' cannot in¯ict this kind of evil on a per- son, but a person can in¯ict it on himself by acting unjustly; for, in so doing, he makes himself unjust. According to this criterion, the ultimate reason for calling an action just or unjust is its e¨ect on the soul of the agent who performs it.17 It follows that insofar as Crito

17 As J. Adam rightly observes, this principle, ``far from being altruistic, was dictated by conscious egoism.'' Socrates ``refrained from doing wrong, not out of regard for others, but because of its e¨ect upon his own soul'' (1888: xviii). See also Santas's discussion of the ``prudential paradox'' and Socrates' ``egoistic theory of motivation'' according to which it is always to one's advantage to be just (1979: 185±93).

Crito 67

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

con®nes his arguments to bodily harms and ignores the e¨ects of wrongdoing on the soul of the wrongdoer, he is wasting his time. Socrates is not completely unappreciative of Crito's e¨orts; he

acknowledges that his zeal is estimable ± provided that it is rightly directed (metaÂ tinov o� rqoÂ thtov, 46b1±2). But if it is not, then the greater his zeal, the harder it will be to bear. Accordingly, they must examine his proposal:

I am the kind of man who listens only to the argument that on re¯ection seems best to me. I cannot, now that this fate has come upon me, discard the arguments I used; they seem to me much the same. I value and respect the same principles as before, and if we have no better arguments to bring up at this moment, be sure that I will not agree with you. (46b4±c3)

In short, unless Crito can produce better reasons for escaping from prison, Socrates will not budge. Although unfailingly cour- teous towards his old friend, he treats him as one who ``speaks an in®nite deal of nothing . . . [H]is reasons are as two grains of wheat hid in two bushels of cha¨; you shall seek all day ere you ®nd them, and when you have them they are not worth the search.''18 Before presenting his reasons for remaining in prison and going

to his death, Socrates alludes to several previous discussions be- tween Crito and himself during which they had agreed about a number of important matters; and he wonders whether Crito still holds the same opinions. Crito thereupon dutifully agrees with a series of statements. More accurately, he dutifully gives a½rmative answers to a series of questions, thereby committing himself to their assertorial reformulations: (i) that we should attend only to the useful opinions (taÁ v . . . crhstaÂ v) of the wise (twÄ n jroniÂ mwn) that is, those who know (twÄÎ e� pistaÂ thÎ ) and understand (e� paiË onti ) and ignore the useless (taÁ v . . . ponhraÂ v) opinions of the foolish (twÄ n a� jroÂ nwn); (ii) that we should pay heed to the praise and blame of the wise but ignore the praise and blame of the foolish; (iii) that, in yielding to the opinions of ``the Many,'' we harm that within us which is bene®ted by justice and harmed by injustice and which makes life worth living; and (iv) that we should attend to the opinions of him ``who understands justice and injustice'' (o� ti e� paiË wn periÁ twÄ n dikaiÂ wn kaiÁ a� diÂ kwn) ± ``the one, that is, and the

18 Shakespeare, The Merchant of Venice 1.1.114±18.

Cross-Examining Socrates68

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

truth itself '' (o� ei� v kaiÁ au� thÁ h� a� lhÂ qeia) ``before whom we should feel fear and shame more than before all the others'' ± ``if there is one who has knowledge of these things'' (47a2±48b2).19 Having elicited Crito's assent to these theses, Socrates gently

but ®rmly rebukes him:

So . . . you were wrong to believe that we should care for the opinion of the many about what is just, beautiful, good, and their opposites [even though] the many are able to put us to death . . . [T]he most important thing is not life, but the good life. (48a7±b6)

Here a protest is in order. Crito said nothing of the kind. What he said was that he had decided on moral grounds and independently of the opinions of ``the Many'' that it is just for Socrates to escape. However, given Socrates' refusal to escape, Crito ®nds himself worrying about what ``the Many'' will think. Although this may show that he looks to ``the Many'' for moral approval ± or, at least, shrinks from their disapproval ± it does not show that he looks to them for moral guidance. Not only has Socrates failed to rebut Crito's case; he has also

misrepresented it. The heart of his case is that, insofar as Socrates voluntarily in¯icts avoidable harm on his friends, his family, and himself, he is behaving unjustly. Ignoring the moral dimension of Crito's case and focusing exclusively on his worries about his reputation ± worries which Socrates misunderstands ± he accuses him of basing his moral beliefs on the opinions of ``the Many.'' This is both inaccurate and unfair. Even if Crito is completely wrong about Socrates' duties to his friends, his family, and him- self, as Socrates (and most commentators) ®rmly believe, the claim for which Socrates faults him is a claim he never made. Uncorrected by Crito, Socrates next undertakes to demonstrate

that it would not be just for him to escape. Before doing so, he again emphasizes that he wishes to conduct a joint investigation and again announces that he will withdraw his claim if Crito can refute it. Another series of questions follows, all of which are again answered a½rmatively by Crito who thereby again commits himself to their assertorial reformulations: (i) that we should never

19 The Crito is the only early dialogue in which the existence of a moral expert is treated as a serious possibility. At Pr. 319a10±320b5 Socrates explicitly denies that any such person exists on the ground that virtue cannot be imparted from one person to another. Cf. M. 89e1±96c10.

Crito 69

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

do wrong voluntarily (e� koÂ ntav); (ii) that wrongdoing is always shameful to the wrongdoer; and, above all, (iii) that we should never requite evil with evil ± no matter how unjustly we ourselves have been treated (49a4±e3). Before allowing Crito to answer these questions, Socrates gravely observes that there are not many peo- ple who hold these views, and that those who do and those who do not have nothing in common and even despise one another. Accordingly, before a½rming these theses and allowing them to serve as the starting point (thÄ v a� rchÄ v, 49d9) of the ensuing discus- sion, Crito should consider these matters very carefully and not answer contrary to what he really believes (paÁ ra doÂ xan, 49d1). This prominent invocation of the sincere assent requirement is

one of the star texts cited by commentators intent on installing it as a ``standing rule'' of elenctic debate ± the guarantor of truthful- ness in argumentation and the provider of the ``existential di- mension'' which enables Socrates to examine his interlocutors' lives as well as their theses.20 However, Socrates himself seems to attach much less importance to the requirement and is not as in- tent on securing Crito's sincere assent as these commentators sup- pose. If he were, he would simply ask his questions, invoke the re- quirement, and await Crito's honest answers. But he does not. Behaving as if he half-suspects that Crito might have changed his mind about these previously agreed-on theses, he surreptitiously introduces a new ingredient into the discussion which insures that Crito will rea½rm them. The invocation of the sincere assent requirement is preceded by a powerful psychological appeal to Crito's sense of shame. Socrates sagely observes that he certainly hopes Crito still holds these opinions because they would be no better than children if they were to change their minds about something so fundamental as the injustice of requiting evil with evil (49a4±b1). The strategy works: Crito assures Socrates that he still holds these opinions. There is no reason to doubt him. I am not claiming that the

admission Socrates wrings from Crito is dishonest; I am simply pointing out that he does wring it from him. This is the answer Socrates wants, and it is the answer he makes sure he gets. How- ever, the decisive instrument of persuasion is not the sincere assent

20 Like many Anglo-American commentators, Brisson attaches great weight to the require- ment and describes it as a ``[r]eÁgle essentielle aÁ la discussion'' (1997: 234, n. 59).

Cross-Examining Socrates70

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

requirement but the appeal to Crito's sense of shame with its subtle, irresistible, and patently false suggestion that it is childish to change one's mind about an important issue in one's old age. Having elicited Crito's ``sincere'' assent, Socrates declares that

the next thing he will say ± or, rather, ask (maÄ llon d' e� rwtwÄ , 49e5) ± is whether we should always keep our agreements, provided that they are right (twÎ diÂ kaia o� nta, 49e6), or whether we are free to break them. Predictably, Crito opts for the former. Very well then. In escaping from prison, would Socrates not be breaking his agreement with Athens and thereby trying to destroy the city by disobeying its laws? And would it be a su½cient justi®cation to say that, having been wronged by the city, Socrates may wrong it in return, thereby requiting evil with evil? It looks like the Socratic elenchus is on its way, like a rollercoaster irretrievably set in mo- tion, and that the by-now helpless Crito is just along for the ride. In view of his recent ``rea½rmation'' of and ``sincere assent'' to

the principle that it is always and unconditionally wrong to requite evil with evil, we expect Crito to say that the fact that Athens has wronged Socrates is not a su½cient justi®cation for him to wrong it in return by escaping from prison. But he surprises us by saying ± emphatically and with an oath ± that it is a su½cient justi®cation; indeed, that is exactly what we should say (50c3) ± an astounding reply which reveals that even at this comparatively late stage of the discussion Crito is still in fundamental disagreement with Soc- rates: not only about whether he should escape from prison, but also (and more fundamentally) about whether it is always and un- conditionally wrong to requite evil with evil. Perceiving that Crito is still unpersuaded, Socrates renews his

e¨orts. But with a di¨erence. Again abandoning logic in favor of psychology, he opts for yet another strategy ± neither a ``joint'' investigation in which Crito is required to say what he ``really'' believes nor another appeal to his sense of shame, but rather a long and largely uninterrupted piece of rhetoric ± the famous speech which he puts into the mouth of the personi®ed Laws of Athens (50a6±54d1). This speech ± a Socratic epideixis if there ever was one ± is designed to persuade Crito that were Socrates to escape from prison, he would be breaking his contract with Athens ± a contract which provided him (and his family) with all the ben- e®ts of citizenship in exchange for unconditional obedience to its laws ± and thereby acting unjustly, disobeying the laws, destroying

Crito 71

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

the city, and harming his own soul.21 Again, the psychological strategy works.22 Having concluded his speech, Socrates again addresses Crito in

his own person and informs him that although the words of the Laws resonate so loudly within him that he can hear nothing else, he is still free to dissent. However, unlike his previous invitation, which was prefaced by assurances that he would take Crito's ob- jections seriously and yield if he found them convincing (48d8±e1), this one is prefaced by assurances that further objections would be futile (maÂ thn, 54d6). Crito sensibly replies that he has nothing more to say. His studied resignation, preceded by his renewed insistence that Socrates should escape from prison, reveal that in his heart of hearts he is still unpersuaded. The end of the Crito presents us with an interlocutor who has heard Socrates out and ``assented to'' each of the premises put to him but who continues to disagree with him and ®nally lapses into silence because he realizes that further protests would indeed be futile. Although Socrates ach- ieves a victory of sorts in this dialogue, it is not a dialectical victory. Crito remains loyal: not because he agrees with Socrates, but be- cause he is his friend.23 Henceforth he abandons the role of critic, remaining with Soc-

rates until the very end and attending to the many mundane and often gruesome details. He had been present at the trial where he had o¨ered to pay a ®ne as an alternative to the death penalty. Failing in this, he had tried to prevent Socrates' imprisonment by posting bail. During his imprisonment, he had visited him fre- quently ± perhaps daily. In any event, the jailer had grown accus- tomed to him and had given him special privileges. On the day of execution, he acts as manager of Socrates' estate and personal a¨airs. Unlike Socrates' other friends, Crito is not self-indulgent,

21 See Brisson, 1997: 194: ``Il . . . s'agit du seul dialogue ouÁ un argument moral seÂrieux est mis dans la bouche non de l'un des interlocuteurs, mais dans celle d'une abstraction per- sonni®eÂe, les Lois d'AtheÁnes.''

22 Grote, 1867, i: 307±8, notes that in spite of Socrates' strictures on long speeches, he often makes them himself ``with ability and e¨ect.'' J. Adam, 1888: xx±xxi, goes further: Soc- rates' long speeches reveal ``[t]he limitation of [his] genius'' and the ``half-complete'' nature of his rationalism; had it been complete, ``he would not have shrunk from sub- mitting to the test of his dialectic the whole question of the validity and authority of law, as a condition of the stability of social life.''

23 This is argued with great eloquence by MeÂron, 1979: 212±13. The same point is made in passing by Mazel who observes that Crito remains loyal to Socrates ``[p]ar amitieÂ plus par conviction'' (1987: 207±8).

Cross-Examining Socrates72

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

and his grief has not rendered him useless. At Socrates' request, he directs someone to escort Xanthippe from the cell. He later delivers a message from the executioner and solicits ®nal instruc- tions from Socrates about the children and his burial. He assists him with his bath, is present during his last moments with his family, and ®nally signals the executioner to administer the hem- lock. Temporarily overcome by emotion, he slips from the cell; but soon he is back. In deference to Socrates' last wish, he promises to sacri®ce a cock to Asclepius24 and, moments later, closes the eyes and mouth of the corpse. It would be hard to ®nd a more devoted and steadfast friend. All this seems to go unnoticed by Socrates. Even as Crito e½-

ciently and unobtrusively discharges his thankless task, Socrates seems to remain completely oblivious of him. The Socrates of the Phaedo is a man who seems to have no need of other people and is willing to tolerate them only if they will talk philosophy and con- duct themselves in a way that measures up to his impossibly high standards. He makes little allowance for normal human weakness or even for normal human response.25 When Xanthippe breaks down at the prospect of losing her husband, she is removed from his presence. When his friends are overcome by similar emotions, they are reprimanded and placed on notice that, unless they can control themselves, the same fate awaits them. If his friends are all-too-human, Socrates hardly seems human

at all. He seems more like a disembodied intellect, a rational soul temporarily housed in a body and awaiting release ± if not eagerly, at least without discernible regret. His serenity in the face of death is undeniably impressive, but it is a remote and chilly serenity. As time slips away, his conversation and general demeanor suggest that he would be perfectly content to spend his ®nal hours in soli- tude, pondering the mixed sensations of pleasure and pain in his legs and setting the verses of Aesop to music. While the banished Xanthippe grieves elsewhere and his emotion racked friends do their best to avoid ``womanly'' behavior, Crito takes control of

24 As Chambry astutely notes, ``Socrate mourant a les meÃmes inquieÂtudes que CeÂphale'' (1989: 9, n. 2).

25 J. Adam is one of the few commentators who is sensitive to this. According to him, Plato put ``the fatal argument'' which requires Socrates to place his obligation to the city above his obligation to his family and friends into the mouth of the personi®ed Laws of Athens ``to save Socrates from the charge of sel®shness and lack of feeling'' (1888: 61).

Crito 73

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

the situation and tries to make Socrates' ®nal hours as comfortable and as conducive to rational discussion as the oppressive circum- stances permit. Although Socrates seems completely oblivious of him, Plato most certainly was not. In saying this, I am not denying that Plato's ultimate sympathies

lay with Socrates. He apparently agreed that he should remain in prison and that the objections he had put into Crito's mouth were feeble and beside the point. At the same time, his portrayal sug- gests that the bond between them is not an intellectual one which stands or falls with their theoretical agreement or disagreement. Crito entered Socrates' cell hoping to persuade him to escape. The reasons he advanced seemed cogent to himself and would have persuaded most people. Socrates knows that too, and in his emo- tionally detached way he ``appreciates'' Crito's genuine ± albeit, in his opinion, misguided ± concern. Although he ignores a good deal of what Crito says and misrepresents much of what he does not ignore, Crito seems not to notice; if he does, he does not pro- test. But even if Socrates had examined Crito's case for escaping more carefully, it is unlikely that he would have persuaded him that it would be unjust to escape. The gap between them cannot be bridged by argumentation. Fortunately for Socrates, Crito found another way to bridge it. Crito's loyalty, sel¯essness, and remarkable capacity for uncon-

ditional friendship do not bespeak shallowness and super®ciality. Such a man does not deserve to have his portrait permanently housed in the Gallery of Inept Interlocutors of which proponents of the standard picture are the self-appointed curators. It is not only false but cruel to suggest that his presence in Socrates' cell is traceable to his worries about ``how things will look'' and ``what people will think.'' Nor will it do to assert without quali®cation that he is living an unexamined life. In spite of his lack of philo- sophical acumen, Crito is a man of considerable stature. His less than imposing mind is more than compensated for by his royal heart.

Cross-Examining Socrates74

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

c h a p t e r 4

Ion

``There is an ancient quarrel between poetry and philosophy,'' de- clares the ``Socrates'' of Republic x, and ``if you admit the pleasure- giving Muse . . . pleasure and pain will be kings in your city instead of law or the thing that everyone has always believed to be best, namely, reason'' (R. 607a5±8). It is often said that the quarrel was started by the poets who claimed to be men of great wisdom. But that is to account for it from the point of view of the philosophers. It might equally be said that the quarrel was started by the philos- ophers who denied that the poets were men of great wisdom, thereby accounting for it from the point of view of the poets. And if we are speaking chronologically, that is exactly what we should say. The poets pre-dated the philosophers by centuries. In any event, there was a quarrel; and it was intensi®ed by the fact that the poets were taken much more seriously than the philosophers by the general public. Their works had shaped the beliefs and sensibilities of Greece for generations, and they were looked upon as the de®nitive embodiments and ®nal arbiters of human virtue. What was believed to be true of the poets generally was believed to be preeminently true of Homer ± ``the encyclopaedist'' whose poetry ``as an epic archetype of the orally preserved word was composed as a compendium of matters to be memorized, of a tra- dition to be maintained, of a paideia to be transmitted.''1 All these claims are vigorously contested by the ``Socrates'' of

the Republic according to whom the writings of the poets ± under- standing the term broadly enough to include comic, lyric, and tragic authors ± must be closely scrutinized and methodically purged of their morally objectionable content which, owing to the craft and cunning with which it is presented, inculcates falsehood

1 Havelock, 1963: 49.

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

by misrepresenting the nature of the gods and heroes (R. 377b5± 392c5) and, what is worse, penetrates the soul of the reader/ auditor, causing him to assimilate these images unawares and to become like the objects depicted (R. 395b8±d3). In view of these immanent dangers, it is not enough to instruct the young about what is good (a� gaqoÂ n) and beautiful (kaloÂ n); they must be habi- tuated in goodness and beauty from earliest infancy so that, like carefully selected and pretreated wool, their souls are permanently imbued with the indelible dye of virtue which pleasure, fear, and passion can never rinse out (R. 429d4±430b2). Thus forearmed against the seductive but morally corrupting o¨erings of the ban- ished poets, and immersed in the didactic but morally edifying e¨orts of their approved counterparts, who write according to the prescribed patterns (toiÄ v tuÂ poiv, R. 398b3), they will grow into properly formed adults in whom the love of the Good has taken root. These recurring denunciations of poets and poetry, so charac-

teristic of middle and later Plato, are wholly absent from the early dialogues. Unlike his (in many respects) very dissimilar counter- part in the Republic, the Socrates of the early dialogues promul- gates no theories of education, banishes no poets, and compiles no approved lists of appropriately excised poetry. Misgivings he has, but they are infrequently expressed and, by the standards of the Republic, comparatively mild. Confronted with Euthyphro's claim that, in prosecuting his father, he is following in the footsteps of Zeus, the best and most just of the gods, who punished his father for his crimes (Eu. 5e2±6a5), Socrates does not deplore Euthy- phro's moral corruption at the hands of the poets and advocate curricular reform; he merely marvels at Euthyphro's credulity and confesses that he himself ®nds it hard to believe these Homeric tales (Eu. 6a6±b4). Similarly, when Protagoras recounts how, in distributing the skills necessary for survival, Epimetheus had pro- vided so lavishly for the animal kingdom that he exhausted his supply before he had made provision for human beings ± a blunder which prompted Prometheus to steal ®re and technical expertise from Athena and Hephaestus, thereby infuriating them and incur- ring the wrath of Zeus (Pr. 320d3±322a2) ± Socrates does not rail against theological anthropomorphism and lobby for a morally purged poetic canon; he merely changes the subject. The Homeric tales which elicit severe moral censure from the ``Socrates'' of the

Cross-Examining Socrates76

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Republic, evoke little more than skeptical eyebrow-raising from the Socrates of the early dialogues whose periodic allusions to poetry and poets are made in a mildly amused, tongue-in-cheek manner which suggests that he does not take them seriously enough to war- rant sustained criticism. But although the early dialogues do not bear witness to any

overt quarrel between poetry and philosophy, this ostensibly unru¿ed surface is deceptive. As Grube2 rightly observes, there is no quarrel so long as the poets make no claim to knowledge. But, of course, they do; and so there is.3 However, it focuses on di¨erent issues. Whereas the Platonic

protest is basically a moral one which focuses on objectionable poetic content and its potentially harmful e¨ects, the Socratic protest ± if ``protest'' is not too strong a term ± is basically epis- temological. Socrates' complaint is not that the poets are potential menaces to the moral well-being of society, but that they cannot provide a rational justi®cation for what they say or a rational ex- planation of their ability to say it (Ap. 22b8±c3). In short, the ut- terances of the poets do not count as knowledge (e� pisthÂ mh). And neither, by implication, do those of the rhapsodes ± who are the interpreters of the poets. Rhapsodes ± from r� aÂ ptw (to stitch together) and wÂÎ dhÂ (song) ±

were professional reciters of poetry and a familiar phenomenon throughout Greece and Asia Minor. Poetic recitation was not con®ned to intellectual literary guilds and select co¨ee-houses. Poetry was part and parcel of Greek popular culture, and it was available to the masses from the globe-trotting rhapsodes whose performances at religious festivals and annual games were heavily advertised and well-attended forms of entertainment which at- tracted crowds of 20,000 and upwards (I. 535d4±5). Prominently positioned and clothed in elaborate costumes, they delivered their spell-binding speeches, competing for victory and (usually) large purses ± the Hellenic version of show-biz. But the rhapsodes' importance far transcended their entertain-

ment value. Insofar as their recitations were one of the principal

2 1980: 182. 3 See Flashar, 1958: 1: ``Dadurch wird der `alte Streit zwischen Philosophie und Dichtung', von dem Platon spricht, zu einem unmittelbar philosophischen Anliegen, das sich durch fast alle Dialoge Platons verfolgen laÈsst und das im Ion zum erstenmal Gestalt gewinnt.'' See also Jowett, 1871, i: 496; and Murdoch, 1977: 8.

Ion 77

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

means by which the Homeric tradition was handed down from one generation to the next, they also performed an indispensable educational and cultural function.4 Although books (in the form of papyrus rolls) existed, they were neither readily available nor much in demand.5 Except for the educated few, the Greeks were not readers, but listeners. As a result, their relation to the Homeric texts was analogous to that of most music lovers to music in every era. Either lacking scores or unable to read them, their only access to music is through public performances. Hence the importance of competent performers. The same was true of public access to poetry in classical antiquity. Hence the importance of competent reciters. Only by being transmitted anew to each succeeding gen- eration could this poetic tradition continue to survive as a coher- ent, standardized, and shared set of normative beliefs, attitudes, and practices, and thus continue to articulate the criteria that pro- vided content for the term areteÅ and de®ned what it meant to be human. As the transmitters of this rich but precarious tradition, the rhapsodes answered to an immense and ± except for atypical dissenters like Socrates and Plato ± universally perceived cultural need.6 Like the sophists, who answered to a di¨erent and ± except for

the same dissenters ± equally perceived need,7 the rhapsodes are often assessed in isolation from the historical context that pro- duced them. Usually maligned as opportunistic charlatans whose super®cially impressive but substantively empty recitations pan- dered to public taste and provided them with handsome incomes, they are routinely numbered among the contributors to the decline of Athenian morality. No doubt there were some who answered to this description. But the abuse of a function does not undercut its

4 According to the Hipparchus ± a dialogue traditionally believed to be authentic but now almost universally regarded as spurious ± it was Hipparchus, the oldest son of Peisis- tratus, who brought the Homeric poems to Greece and compelled the rhapsodes to recite them in hopes of educating the citizenry (228b4±c6). Diogenes Laertius claims that the credit belongs to Solon (1.57).

5 For a discussion of Athenian literacy, books, and the minor educational role they played in classical antiquity, see Roberts, 1984: 103±5.

6 In Xenophon's Symposium (3.6) Niceratus reports that he listens to the rhapsodes almost every day.

7 On the unavailability of an education that went beyond the fundamentals, see Kerferd, 1981: 37±38; Roberts, 1984: 94±103; and de Romilly, 1992: 30±56.

Cross-Examining Socrates78

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

need ± a point clearly grasped by the ``Socrates'' of the Republic who, although convinced that existing poetry cannot provide ade- quate moral habituation for the young, never doubted the need for poetry that can. Convinced that neither the sophists nor the rhapsodes were adequate to the educational task, Plato's attitude towards both ¯uctuated between covert distrust and overt hostility. Although he recognized that they were merely symptoms of the lamentable state of Athenian society and not its cause (R. 490e2± 496a9), he seldom had anything good to say about them. For that reason, it is as unwise to rely solely on the Platonic corpus for in- formation about the sophists and rhapsodes as it is to rely solely on the New Testament for information about the Pharisees and the Sadducees. Plato's estimate of the rhapsodes emerges most clearly in his portrayal of the celebrated Ion of Ephesus in the dialogue which bears his name. The Ion depicts a discussion between Ion and Socrates. Although

its dramatic date cannot be precisely determined, Ion's passing re- mark that his native city is presently under Athenian rule (541c3± 4) indicates that it must be before 4168 when the alliance between Athens and Ephesus was dissolved. The preliminary chit-chat sug- gests that this is not the ®rst meeting between the pair, but Ion's pitiful helplessness in the face of Socratic interrogation suggests that it is their ®rst philosophical encounter. In answer to the question: who is Ion? one can only say that Ion

is Ion, and there is none like him. Unknown apart from this dia- logue, he is arguably the most intellectually marginal character in the early dialogues and a strong contender for the title of Soc- rates' Silliest Interlocutor. If Xenophon is right, he was typical of his profession:

[T]he rhapsodes are consummate as reciters, but they themselves are utter fools. (Mem. 4.2.10)

It is hard to imagine a more damning criticism, coming as it does from a writer whose heartfelt but (after a few pages) repetitious and uninteresting ``recollections'' of Socrates leave one with the distinct impression that his hero was little more than an inexhaust- ible conduit of numbingly predictable and eminently forgettable

8 Unless otherwise indicated, all dates are bce.

Ion 79

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

platitudes. If the intellectually inconsequential Xenophon judges you foolish, you must be foolish indeed.9 But if Ion is one of Socrates' most intellectually marginal and

dialectically inept interlocutors, he is also one of the most likeable. Although he labors under an inordinately high opinion of himself ± he thinks he is worthy of being awarded a gold crown by the Homeridae10 ± his vanity is not the o¨-putting vanity of intellectual arrogance, but the endearing vanity of intellectual naivete ± the vanity of the precocious, but untested adolescent.11 Here is a lin- guistic virtuoso whose auditors are putty in his hands whom he can alternately convulse with laughter and move to tears but who, upon being administered a mild dose of Socratic cross-examination, can explain neither the Homeric wisdom he imparts nor how he is able to impart it. Having just returned from Epidaurus where he had won ®rst

prize at the festival of Asclepius, the jubilant Ion is anticipating similar successes and triumphs in Athens ± if the god wills (530b4). Never one to discourage a braggart, Socrates feigns admiration for Ion's erudition which enables him not only to recite the poet's words, but also to elucidate his thought ± the exegetical task which Ion thinks is the most di½cult part of his job and on which Soc- rates focuses most of his critical attention. Oblivious to the irony, Ion immodestly con®des that, in addition to elucidating the poet's ideas, he embellishes them with such ®ne ideas of his own that his epideictic displays excel those of any other rhapsode the world has ever seen. Having good-naturedly endured all this self- congratulatory fanfare, Socrates uncharacteristically requests a demonstration of Ion's rhetorical ability ± the very thing he usu-

9 Strictly, the source of this extremely negative assessment is not Xenophon but Euthyde- mus. Interestingly, the Xenophontic Socrates does not fully endorse it. Although he agrees that no one is ``more stupid'' (h� liqiwÂ teron) than the rhapsodes, he does not attri- bute their stupidity to lack of intelligence, but to the fact they do not know the ``inner meaning'' (taÁ v u� ponoiÂ av) of the poems they recite (Symposium 3.6±7). That the rhapsodes were far from being fools is implied by Grote, 1867, i: 455, and argued with great elo- quence by Flashar, 1958: 21±24.

10 The Homeridae (or Sons of Homer) was a guild of professional rhapsodes located in Chios who claimed to be direct descendants of Homer.

11 A few commentators agree. Stock thinks Ion's vanity is ``harmless'' (1909: vii). So, too, MeÂron, 1979: 162: ``Il est infatueÂ de lui-meÃme . . . non sans cause apparemment.'' But most do not. Woodru¨ judges Ion ``proud'' and ``a ®t target'' for Socratic criticism (1983: 6). Bloom describes him as ``self-satis®ed'' and ``as far from the radical self-doubt of philosophy as a man can be'' (1987: 371).

Cross-Examining Socrates80

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

ally goes out of his way to avoid ± only to beg o¨ the minute Ion agrees. On second thought, he would rather ask a few questions. Having declined the role of audience in favor of that of ques-

tioner, Socrates is surprised to learn that Ion's skill as a rhapsode is con®ned to Homer. Other poets bore him; when their names come up, his attention wanes. He frankly dozes (532b8±c4). This remarkable self-evaluation should not slip by unnoticed. Unlike every other interlocutor in the early dialogues, Ion acknowledges his intellectual and professional limitations at the very outset: when it comes to reciting Homer, he is unsurpassed; but when it comes to reciting other poets, he is useless. Socrates' response is surprising. Instead of rejoicing that he has

®nally found an interlocutor who is aware of his limitations and willing to acknowledge them, he feigns puzzlement and sets out to show that Ion's skill as a rhapsode cannot be con®ned to Homer. Unlike most Socratic interlocutors, who need to realize that they are less competent than they think they are, Ion apparently needs to realize that he is more competent than he thinks he is. To drive this point home, Socrates advances two arguments. The ®rst ± 531a5±d2 ± is as follows:

(i) There are many subjects about which Homer and the other poets say the same things (tau� taÁ leÂ geton), and in those cases Ion can elucidate them all equally well.

(ii) But there are many other subjects about which Homer and the other poets do not say the same things (mhÁ tau� taÁ leÂ gousin), e.g., divination; and in such cases a diviner could adjudicate their disagreements better than Ion.

(iii) However, if Ion were a diviner himself, he could adjudicate these disagreements equally well.

(iv) Therefore since Homer and the other poets speak about many of the same things, e.g., war, dealings between gods and men, and since Ion has agreed that insofar as the poets say the same things, he can expound and elucidate them equally well, his skill as a rhapsode cannot be con®ned to Homer.

Before proceeding to Socrates' second argument, it is worth noticing that the ®rst is fallacious. In saying ± in (i) and (ii) ± that Homer and the other poets sometimes do and sometimes do not ``say the same things,'' Socrates means that they agree about some things, but disagree about others. However in (iv) he means some-

Ion 81

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

thing else. Here ``saying the same things'' does not mean agreeing (as opposed to disagreeing) about the subjects on which they write; it means writing about the same subjects (as opposed to writing about dif- ferent ones). Not noticing this shift of meaning, Ion concurs: ``That's true, Socrates'' (531d3). What he should have said is: ``That's true, Socrates, but it is irrelevant. We were not discussing whether Homer and the other poets write about the same subjects but whether, in writing about them, they agree or disagree. You are equivocating on `say the same things'.''12 Instead of lodging this objection, Ion shifts ground and intro-

duces a new consideration: he admits that Homer and the other poets write about many of the same subjects, but he thinks that Homeric poetry is better (a� meinon, 531d10). Although it is often risky to speculate about what is going on in the mind of a ®ctional character, that is not the case here. In saying that Homeric poetry is ``better'' than non-Homeric, Ion clearly means that it is better qua poetry, i.e., that it is better written and hence superior when judged by aesthetic criteria.13 My use of the term ``aesthetic'' is, of course, anachronistic. As Guthrie14 rightly points out, aesthetic criteria are never mentioned in the Ion. However, they are clearly implied. In saying that Homeric poetry is better than non- Homeric, Ion is making an implicit aesthetic judgment and hence making an implicit distinction between what the poets say and how they say it, that is, an implicit distinction between content and style. As we will see, Socrates does the same thing a few Stephanus pages later. We expect Socrates to challenge Ion by asking on what ground

he claims that Homeric poetry is better than non-Homeric and by demanding a de®nition of beauty which may serve as a standard (paraÂ deigma) for individual and comparative ascriptions of it. But he does not. Ignoring Ion's implicit appeal to aesthetic criteria

12 Woodru¨, 1983: 9±10, acknowledges that at least one of Socrates' arguments trades on an equivocation; however, instead of faulting him for fallacious reasoning, he defends him on the ground that the argument ``plays on words in an illuminating way.'' Guthrie, 1975: 202, n. 1 is less apologetic: ``It is hardly worth pointing out all the fallacies commit- ted by S. in this little work.''

13 For a di¨erent interpretation, see MeÂron, 1979: 164±65, according to whom Ion is not o¨ering an aesthetic evaluation of Homeric poetry, but merely expressing his personal taste: he just likes Homeric poetry better than non-Homeric.

14 1975: 205.

Cross-Examining Socrates82

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

and his implicit distinction between content and style,15 he tries to demolish Ion's claim that his skill as a rhapsode is con®ned to Homer with a second argument (531d12±532b7):

(i) When several people talk about the same subject, e.g., mathe- matics or wholesome food, and one speaks better (a� rista) than the rest, the person who is most competent to judge the good speaker (toÁ n eu� leÂ gonta) is also most competent to judge the bad (touÁ v kakwÄ v leÂ gontav), i.e., these comparative judgments are necessarily made by the same person, namely, the one who possesses the technai of mathematics or nutrition.16

(ii) Formulated in full generality, in any techneÅ, whoever is most competent to judge who speaks better is also (and for that reason) most competent to judge who speaks worse.

(iii) Therefore if Ion is competent to judge that Homer speaks better than the other poets, he must be equally competent to judge that they speak worse. So, again, his skill as a rhapsode cannot be con®ned to Homer.

Exulting in Socrates' apparent victory, Guthrie17 exclaims: ``Who ever heard of an art critic who could assess Polygnotus but no other painter?'' Guthrie is right. It would be patently absurd to say such a thing. But Ion never said it. The apparent cogency of Socrates' second argument depends

on his tendentious interpretation of what Ion did say. The argu- ment tries to demonstrate that it is inconsistent for Ion simulta- neously to assert:

(i) that he is competent to judge which poets speak well and which speak badly,

and:

(ii) that his skill as a rhapsode is con®ned to Homer.

15 See Schaper, 1968: 25: Socrates ``ruthlessly'' overrides Ion's ``rather sensible suggestion.'' 16 The term techneÅ has no exact English equivalent and applies to any activity that involves

expertise. The Greeks made no distinction between what we broadly call ``crafts'' (cob- bling, carpentry, pottery, etc.) and the so-called ``®ne'' arts (poetry, music, painting, etc.). Although often translated as ``art,'' Dodds, 1959: 190, prefers the more neutral ``skill.'' See, too, Murray, 1996: 108.

17 1975: 202.

Ion 83

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

But these propositions are not inconsistent. All Socrates has shown is that to be a competent judge of the superiority of Homeric poetry over non-Homeric, one must be acquainted with both ± hardly an insight to be shouted from the housetops. Furthermore, Ion is acquainted with both. He did not say that he has no knowl- edge of non-Homeric poetry;18 he said only that he has no interest in it. His claim was that his skill as a rhapsode is con®ned to Homer, not that his acquaintance with Greek poetry is con®ned to him. And from the fact that he is acquainted with non-Homeric poetry and judges it inferior to Homeric it does not follow that he must be equally skilled at reciting both. In addition to ignoring Ion's implicit appeal to aesthetic criteria

and his implicit distinction between content and style, Socrates re- de®nes ``speaks better.'' Instead of agreeing that anyone who is competent to judge who speaks well and badly about mathematics (or any other techneÅ ) must be a practitioner of that techneÅ, Ion should have pointed out that this is only true if, by ``speaks better,'' one means ``speaks correctly'' (o� rqwÄ v). He should then have ex- plained that this is not what he meant. By ``speaks better'' (eu� ), he did not mean ``speaks correctly,'' but ``speaks with more stylistic elegance, beauty, and charm.'' Socrates' argument has force only if ``speaks better'' means ``says what is true'' in the sense of cor- rectly assessing the factual accuracy of what the poets say about the subject matter that falls under the techneÅ in question ± in this case, mathematics. Ion never made such a claim. He claimed only that he could assess the stylistic e¨ectiveness with which the poets speak about the subject matter that falls under the techneÅ. An author less favorably disposed towards Socrates (and more

favorably disposed towards rhapsodes) might have provided Ion with the following rejoinder: ``Although I am neither a diviner nor a mathematician nor a doctor, and, therefore, lack the technical expertise necessary to adjudicate substantive disagreements about divination, mathematics, and medicine, I am a competent judge of who speaks better about these subjects from a non-substantive and purely stylistic point of view. You have overlooked the fact that although one must be a surgeon in order to assess the medical content of a lecture on a complicated surgical procedure, one need not be a surgeon in order to judge which of two lecturers is the

18 Pace FriedlaÈnder, 1964, i: 129±30.

Cross-Examining Socrates84

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

better speaker. Hence you have not shown that I am an incompe- tent judge of the aesthetic qualities by virtue of which Homeric poetry is better than non-Homeric.'' It follows that Socrates' second attempt to refute Ion's claim that his skill as a rhapsode is con®ned to Homer is as unsuccessful as the ®rst. None of this is detected by Ion. Thunderstruck by this appar-

ently devastating argument and unaccustomed to tra½cking in such profundities, Ion is reduced to aporia:

Then how in the world do you explain what I do, Socrates? When some- one discusses another poet I pay no attention, and I have no power to contribute anything worthwhile . . . But let someone mention Homer and right away I'm . . . paying attention and I have plenty to say. (532b8± c4)

Puzzled (apparently for the ®rst time) by his inability to recite non-Homeric poetry, he implores Socrates to unravel this mystery by explaining him to himself. The result is the Socratic theory of poetic inspiration. It is not ``by art and knowledge'' (teÂ cnhÎ kaiÁ e� pisthÂ mhÎ ) that Ion

recites Homeric poetry well (532c6); if it were, then he could recite non-Homeric poetry equally well, just as someone who under- stands the technai of painting, sculpture, and music is not con®ned to discoursing about particular painters, sculptors, and musicians, but can discourse equally well about them all. Those who have knowledge of something have knowledge of the whole and not merely knowledge of some of its parts ± much less, of only one part. Since Ion's skill as a rhapsode is con®ned to Homer, it must be explained in some other way. According to Socrates, he does so by divine inspiration (qeiÂ aÎ moiÂ raÎ , 534c1). Before proceeding fur- ther, it should be noted that Socrates' acknowledgment in this passage that Ion's skill as a rhapsode is con®ned to Homer contra- dicts the conclusion of the two foregoing arguments according to which his skill is not ± indeed, cannot be ± con®ned to Homer. Clearly, Ion's skill as a rhapsode either is or is not con®ned to Homer. If it is not, then the Socratic theory of divine inspiration which follows is super¯uous; we need no esoteric explanation of how Ion recites Homer. On the other hand, if it is, then, insofar as the conclusion of the two foregoing arguments is incompatible with this theory, it follows that in this dialogue Socrates simulta- neously argues for both p and not-p.

Ion 85

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Having explained that Ion does what he does by divine inspira- tion, Socrates goes on to claim that the same is true of the poets:

For a poet is an airy thing, winged and holy, and he is not able to make poetry until he becomes inspired (e� nqeov) and goes out of his mind and his intellect is no longer in him . . . [I]t's not by mastery that they make poems or say lovely things (kalaÂ ) about their subjects . . . [but] by a divine gift . . . That's why the god takes their intellect away from them when he uses them as his servants . . . so that we who hear should know that they are not the ones who speak . . . [but] the god himself. (534b3± d4)19

As empirical evidence for these claims, Socrates adduces the sad case of Tynnichus of Chalcis ± an inconsequential and obscure poet who during his whole literary career produced only one poem worth mentioning (a� xÎ iwÂ seien mnhsqhÄ nai ) ± the hymn of praise which everyone is singing (534d6±8). The existence of this poem ± one of the most beautiful (kaÂ lliston, 534d8) ever written ± demonstrates that even the most unaccomplished poet in all Greece20 can become the chosen mouthpiece of the gods. Socrates' use of kaÂ lliston should not go unnoticed. In em-

ploying this term ± an unmistakably evaluative one ± he, too, is making an implicit aesthetic judgment and hence making an im- plicit distinction between what the poets say and how they say it, thereby evaluating Tynnichus' poem qua poem.21 According to the Socratic theory of the creative process, insofar as a poet is unin- spired and composes voluntarily, he is aesthetically impotent. Such

19 It is sometimes said that Plato did not hold this theory of poetic inspiration and merely interjected it as an ironical polemic with no serious philosophical content ± a ``polemi- sche Ironic ohne sachlichen Gehalt'' (Pohlenz, 1913: 186). See also A. E. Taylor, 1929: 40; Tigerstedt, 1969: 28; Tejera, 1984: 320; Levi, 1985: 356; and, more cautiously, Woodru¨, 1983: 5, 8±9). This is vigorously contested by Grote, 1867, i: 459±60; Fried- laÈnder, 1964, i: 132±33; Schaper, 1968: 29±34; and, above all, Flashar, 1958: 20±21, ac- cording to whom Ion's earlier remark that he will be victorious in Athens ``if the god wills'' (e� aÁ n qeoÁ v e� qeÂ lhÎ , 530b4) is not mere pious clap-trap, but a phrase which has ``einen tieferen Sinn'' for Plato and reveals that the theory of poetic inspiration which follows is indeed his. The phrase is not uncommon in Plato (see, e.g., Ph. 69d6, 80d7±8, HMa. 286c3, Laws 632e7, 688e2, 739e5, and 752a8).

20 Woodru¨, 1983: 38, points out that, in spite of Socrates' low opinion of Tynnicus, he was admired by Aeschylus.

21 Cf. Sym. 198b1±199b5 where ``Socrates'' is astounded (e� xeplaÂ gh) by the beauty of Aga- thon's speech about love in spite of the fact that he judges its content false. Both passages count against Allen's contention, 1996: 6, that ``[n]owhere in the Ion is it supposed that poetry possesses an autonomous value'' and Grote's similar contention, 1867, i: 456±57, that Plato ``takes no account'' of the poet ``who touches the chords of strong and di- versi®ed emotion.''

Cross-Examining Socrates86

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

poets can write poems, but none ``worth mentioning.'' It follows that Tynnichus' single aesthetically estimable poem is an involun- tary achievement ± and hence not an achievement at all ± for which he is as little deserving of praise as he is deserving of blame for his previous and subsequent failures. All credit belongs to the Muse.22 Socrates' theory of poetic inspiration di¨ers radically from the

prevailing view.23 Traditionally, the poet invokes the aid of the Muse; but inspiration does not preclude conscious craftsmanship. According to the Socratic theory, on the other hand, it does; the inspired poet is not a skilled craftsman who composes voluntarily in accordance with the rational principles of his techneÅ. Indeed, the inspired poet has no techneÅ ± and hence no skill or craftsmanship at all.24 Insofar as he produces anything of aesthetic merit, he is the passive vehicle of the gods. Out of his mind and devoid of under- standing, he literally does not know what he is talking about. Ac- cordingly, to call a poet ``inspired'' is not a compliment25 but a kind of insult. Little wonder that the poets resented Socrates' claim that they did not compose ``by art or knowledge,'' but by divine inspiration (Ap. 22b8±c2), and that they engaged Meletus to prosecute him on their behalf (23e3±5). In the Ion, then, the term ``poet'' has two distinct senses ± one,

descriptive; the other, normative. Descriptively, a poet is someone who writes verse voluntarily and unaided by divine inspiration. Normatively, a poet is someone who writes verse involuntarily and under the in¯uence of divine inspiration. Although there are many poets in the descriptive sense, the class of poets who are never in- spired is a null class; no one is a poet in this sense during his whole literary career ± not even Tynnichus who, in spite of his otherwise dreary e¨orts, was on one happy occasion the inspired mouthpiece of the gods. In short, there are three classes which serve as refer- ents of the term ``poet'': ®rst, the null class of universally unin-

22 Woodru¨, 1983: 5, describes the Ion as ``one of Plato's riddles'' ± a dialogue in which Socrates tries to convince his interlocutor of the importance of poetry, which is the result of divine inspiration, and of the unimportance of the poet, who is merely the passive medium through which the gods speak.

23 Woodru¨, 1983: 8, describes it as ``quite startlingly new.'' See also Grote, 1867, i: 459± 60; Stock, 1909: viii±ix; Tigerstedt, 1969; Guthrie, 1975: 206±8; Saunders, 1987: 42±43; Canto, 1989: 10±13, 47±50; and Murray, 1996: 7±9, 114.

24 As Woodru¨, 1982: 145, rightly observes, Socrates ``will not allow the poets a techneÅ , even of pure style. All the beauty of a poem comes from the inspiring gods.'' See also Murray, 1996: 10, 102.

25 This is noticed by Stock, 1909: viii±ix; and Allen, 1996: 7.

Ion 87

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

spired poets; second, the ``mixed'' class of poets who are inspired on some occasions, but not on others; and third, the class of poets who are universally inspired ± which is also a null class if even Homer, ``the best and most divine'' of the poets, occasionally had his uninspired moments ± a subject on which Socrates does not express an opinion. What is true of the poets, who interpret (e� rmhneuÂ ete) the utter-

ances of the gods, is equally true of the rhapsodes, who interpret the utterances of the poets and are, therefore, interpreters of in- terpreters (e� rmhneÂ wn e� rmhnhÄ v, 535a9). So Ion's inability to recite non-Homeric poetry has a simple explanation: the necessary in- spiration is lacking. Accordingly, it is not a contingent fact about inspired poets and rhapsodes that they do not know what they are talking about; given the Socratic theory of the creative process, inspiration necessarily precludes understanding. It follows that ``Tynnichus is inspired'' and ``Tynnichus understands the poem he just composed'' are logically incompatible propositions. So are ``Ion is inspired'' and ``Ion understands the poem he just recited.'' Enchanted by this image of poets and rhapsodes as inspired

mouthpieces of the gods, Ion endearingly confesses that Socrates' words have touched his very soul and becomes an instant convert to the Socratic theory of poetic inspiration. But apostasy threatens when Socrates proceeds to deduce an unexpected and unwelcome implication: just as the inspired poet is out of his mind and, there- fore, mad, so is the inspired rhapsode. That is, so is Ion. Ion balks at this conclusion. Although willing to acknowledge

that he is inspired when he recites, he is unwilling to concede that he is mad. Monique Canto26 judges him rather harshly for this, describing him as ``un des personnages les plus antidialectiques des dialogues platoniens'' ± a person who accepts Socrates' con- clusions when they are ¯attering to himself but rejects them when they are not. But surely Ion's reluctance to concede that he is mad admits of a less negative assessment. To concede that would be to concede that the alleged skill for which he is so highly sought after (and so handsomely paid) is not a skill at all and hence not an ability resident in himself and an appropriate object of praise, but a state to which he is involuntarily reduced and for which he de-

26 1989: 21.

Cross-Examining Socrates88

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

serves no credit. The rest of the dialogue is devoted to convincing him that this is, in fact, the case. Asked about which things in Homer he speaks well (eu� ), Ion

con®dently retorts that he speaks well about everything. Asked whether ``everything'' includes things he knows nothing about, he squirms a bit and warily asks which things Socrates has in mind. Socrates thereupon cites a passage from the Iliad which contains instructions about driving a chariot and asks whether a charioteer would not be a better judge of its correctness than a doctor. Ion agrees. But this admission is a mere preliminary to a more general and far more damaging one, namely, that if a charioteer knows more about driving chariots than a doctor, then it would seem that he also knows more about it than a rhapsode. That is, he knows more about it than Ion. Before Socrates can elicit this admission, however, he needs a

further argument. It is this. Since many technai exist and since each has a unique function (e� rgon, 537c6) which is de®ned by its object ± it follows that the practitioner of a particular techneÅ, e.g., car- pentry, cannot, by that techneÅ, know the subject matter which falls under another techneÅ, e.g., medicine, and, therefore, cannot make authoritative judgments about it. Ion concurs, thereby rendering himself a sitting duck for the sequel. Just as charioteers know more than Ion about driving chariots, so also doctors and ®sher- men know more about medicine and ®shing than he. Since Ion possesses none of these technai or any others, it follows, by his own admission (kataÁ toÁ n soÁ n loÂ gon, 540a5±6), that he is not a com- petent judge of who ``speaks well'' about any of them. Now comes the ®nal blow. Since Ion is not a poet either, it follows that he cannot make authoritative judgments about poetry and hence is not competent to judge that Homeric poetry is better than non- Homeric ± unless, of course, he makes this judgment under the in¯uence of divine inspiration, devoid of understanding and in a state of madness. If Ion insists that it is ``by art and knowledge'' that he makes

these judgments, he must prove it. Since so far he has not, he is either refusing to prove it (and is, therefore, behaving unjustly towards Socrates) or he cannot prove it (and is, therefore, inspired and mad when he recites). So which is it? Is Ion unjust or divine (a� dikov . . . h� qeiÄ ov)? Not surprisingly, Ion opts for the latter on the ground that divinity is nobler than madness (542a7). Thus the

Ion 89

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

answer to the question: Why is Ion's skill as a rhapsode con®ned to Homer? is no longer a mystery. It is only when he recites Homeric poetry that he is inspired. Ion concedes the point. But his concession is premature. Although he has been silenced,

he has not been refuted. Socrates' argument depends on three de- monstrably false assumptions: (i) that poetry is reducible without remainder to its factual content about the various technai; (ii) that the factual accuracy of this information is the only criterion by which poetry can be assessed; and (iii) that only the practitioners of the relevant technai are competent to assess it.27 However, before pursuing these objections, a disclaimer is in

order. It is both anachronistic and unfair to criticize an ancient thinker's views about a given subject by appeal to more recent views about that subject. As Guthrie28 rightly observes, although a modern reader may marvel at Socrates' ``total incomprehension'' of the nature of poetry, which reduces it to a handbook of techni- cal instructions, in Plato's day poetry was in fact viewed as ``pri- marily didactic.'' Guthrie's point is well taken. However, it is sig- ni®cant that, in making it, he ®nds it necessary to use the term ``primarily.'' Although Plato (and the Greeks generally) viewed poetry as primarily didactic, its aesthetic dimension did not en- tirely escape them. If it had, the quali®cation ``primarily'' would have been super¯uous. Ion and Socrates do indeed look to poetry for technical instructions about the various technai, but they also look to it for entertainment and delight. That this is so is borne out by the fact that Socrates heaps high praise on Tynnichus' soli- tary poetic achievement and describes it as one of our most beau- tiful (kaÂ lliston) poems and one which everyone is singing. People do not usually sing technical instructions. Although Ion fails to avail himself of it, his earlier implicit dis-

tinction between speaking correctly (o� rqwÄ v) and speaking well (eu� ), provides him with a reply to this argument and the (even by an- cient standards) excessively narrow view of poetry on which it is based. Confronted with Socrates' dilemma ± either you are unjust or you are divine ± Ion is not logically compelled to opt for the latter and to embrace the Socratic theory of poetic inspiration. The dilemma is a false one. Several replies are open to him. First, he could have invoked his earlier implicit distinction

27 See Murray, 1996: 106. 28 1975: 205.

Cross-Examining Socrates90

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

between content and style, and pointed out that although he is neither a diviner nor a mathematician nor a doctor, and hence incapable of making comparative substantive judgments about divi- nation, mathematics, and medicine, that does not prevent him from making comparative aesthetic judgments about passages in which Homer talks about these technai. To be a competent judge of Homeric poetry, which contains technical instructions about the various technai, does not require that one be a practitioner of those technai. Ion's claim that his skill as a rhapsode is con®ned to Homer is problematic only if it is interpreted in light of Socratic doctrine. There is nothing suspect about the ability to discourse authoritatively about a single author. A specialist in Shakespeare at a modern university would not be denied tenure on the ground that his expertise does not extend to other dramatists (Elizabethan and otherwise) who ``talk about the same things.'' Contrary to what Socrates implies, there is no Art-of-Poetry-in-the-Abstract, mastery of which quali®es one as an expert about all poets and ignorance of which disquali®es one as an expert about any.29 Second, Ion could have pointed out that Socrates' contention that only practitioners of a techneÅ are equipped to make authoritative judgments about it is simply false. Possessing a techneÅ is not a nec- essary condition for speaking authoritatively about it. Dramatic and musical expertise are cases in point. Many teachers and critics are uncannily adept at detecting faults in (and improving) the per- formances of actors and instrumentalists far more gifted than themselves. Nor is it a su½cient condition. Many able performers are reduced to incoherent stammering when called upon to ex- plain how they do what they do. Finally, Ion could have pointed out that although he has no idea how he sends his audiences into raptures when (but only when) he recites Homeric poetry, from the mere fact that he cannot explain how he does it (or why he cannot do it when he recites non-Homeric poetry), it does not follow that his ability to do so can only be explained by tracing it to some causal agent other than himself ± namely, the Muse. And it certainly does not follow that he must be mad. Ion denies that he is mad: not because he is intellectually arro-

29 Kahn, 1996: 109±10, is also skeptical about Socrates' ``one-to-one mapping principle,'' according to which anyone who knows the art of poetry can discourse authoritatively about all poets because ``the whole thing is poetry.''

Ion 91

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

gant and impervious to argumentation, but because he knows per- fectly well that he has a techneÅ ± that he possesses a speci®able and highly developed set of rhetorical and dramatic skills which enable him to a¨ect his audiences deeply and by design. Socrates' conten- tion that his ability to produce these e¨ects is not traceable to an ability resident within himself is at odds with his experience as a rhapsode.30 That this is so is borne out by his tell-tale remark that he is carried away by his own recitations and overcome by the same emotions that course through his hearers ± albeit not so overcome that he forgets that the more they weep over his tales of woe, the more he laughs over his soaring pro®ts (535c4±e6). Ion knows exactly what he is doing and how he does it. The fact that he does not a¨ect his audiences in the same way

and to the same degree when he recites non-Homeric poetry is easily explained. Like most people called upon to do things in which they have no interest, Ion's heart is not in it and his perfor- mance is perfunctory and ine¨ectual. Far from being an impene- trable mystery, this is one of the commonplaces of human experi- ence. Only a person determined to quibble and to ®nd fault ± only a Socrates ± could fail to understand it and, having failed, relegate rhapsodes to the status of opportunistic charlatans devoid of all understanding and account for their genuine successes in such a preposterous way. The contention that rhapsodes and poets are devoid of all intelligence and skill and merely passive vehicles of the gods awaiting the necessary ``inspiration'' bespeaks an extra- ordinarily mechanical understanding of a skill and an adolescent, moonstruck view of the creative process. A Divine Comedy or a ``Hammerklavier'' Sonata does not trickle e¨ortlessly from the pen as the involuntary response to an externally applied divine stimulus. As is obvious from the notebooks of countless poets, composers, and artists, what we loosely call ``art'' is hard work which involves meticulous planning, agonizing trial-and-error, and incessant rethinking. And so does its interpretation. It is often said31 that Plato portrays Ion as a simpleton so that

Socrates can make sport of him. But this is not borne out by the philosophical content of the dialogue. Socrates does not take Ion seriously qua Ion, but he takes him very seriously qua rhapsode,

30 As has been pointed out by MeÂron, 1979: 163, n. 20; and Canto, 1989: 142, n. 29. 31 See, e.g., Elias, 1984: 5.

Cross-Examining Socrates92

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

that is, qua mouthpiece of the gods. Plato portrays Ion as a sim- pleton: not to provide Socrates with entertainment, but to enhance the plausibility of the Socratic theory of poetic inspiration. A more intelligent rhapsode would have rendered it less plausible. What is needed is someone exactly like Ion ± a rhapsode indisput- ably capable of stirring his audiences to the depths, but utterly in- capable of explaining how he does it. Socrates never doubts Ion's competence as a rhapsode; he is concerned solely with its nature and source. By accentuating Ion's intellectual ineptness before presenting the theory of poetic inspiration, Plato prepares the reader for it. It is easy to poke fun at Ion, and few commentators have resisted

the temptation.32 However, I want to conclude this chapter on a di¨erent note. Instead of pointlessly (and cruelly) maligning an admittedly incompetent interlocutor, I will pose a series of ques- tions. Has Socrates cared for Ion's soul? Has he really tried to bene®t and to improve him? Has he argued carefully, cogently, and on a level that Ion might reasonably be expected to under- stand? Is Ion really persuaded by Socrates' arguments? Does he really believe that he recites Homeric poetry under the in¯uence of divine inspiration? Does he opt for being divine out of convic- tion or merely out of embarrassment and the desire to save face? Does he sincerely assent to the premises by which Socrates gen- erates (or seems to generate) these conclusions? Does Socrates care whether he sincerely assents? Or whether his critique of poets and rhapsodes has registered? Does Ion depart a better man? Or a more modest rhapsode? Since Ion is not an actually existing person, these are questions

without answers. But insofar as ®ctional characters may be cred- ited with imaginative futures, such questions have a restricted legitimacy and enable us to form a correspondingly imaginative estimate of the e½cacy of the Socratic elenchus. In the Ion it has none.

32 Most recently Murray, 1996: 98: ``Ion . . . is so stupid that he is not worth attacking''; and Kahn, 1996: 113: Socrates has ``clearly won the argument, but his victory cannot be logi- cally consummated because of the stubborn obtuseness of Ion.''

Ion 93

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

c h a p t e r 5

Hippias

With the dubious exception of Alcibiades,1 Hippias is the only Socratic interlocutor with two dialogues named after him: the Hippias Major and the Hippias Minor. Since the former is rejected as inauthentic by some2 and accepted but classi®ed as transitional (rather than early) by others,3 except for a few biographical details, this chapter will be con®ned to the latter. Hippias is a sophist.4 That, of course, is not su½cient to render

him morally suspect and the enemy of virtue. Although traditional commentators tended to view the sophists as an undi¨erentiated entity and to portray them as intellectually inconsequential and even as morally corrupt, this unduly negative assessment is gradu- ally being abandoned in favor of a more just estimate.5 The pos- sessors of a wide range of knowledge and an impressive array of skills, they answered to a perceived educational and cultural need and commanded high fees6 ± a policy much deplored by Plato's

1 The Alcibiades I and II have had a strange history. Included in the third tetralogy of the Thrasyllean canon and regarded as authentic until the late nineteenth century, they were pronounced spurious by Schleiermacher, 1836: 328±36. Subsequent Platonic scholarship has largely endorsed his verdict.

2 E.g., Thesle¨, 1976: 105±17; Leggewie, 1978: 71±72; Kahn, 1985, and 1996: 37, n. 3; and Figal, 1995: 19±20.

3 E.g., Irwin, 1977: 291, n. 33, and (more cautiously) 1995: 12; Vlastos, 1983a: 27, n. 2, 1991: 46±47, and 1994: 71, n. 14; and McPherran, 1996: 18, n. 46.

4 The four so-called ``older'' sophists were all non-Greek: Hippias came from Elis, Prota- goras from Abdera, Prodicus from Ceos, and Gorgias from Leontini. Before acquiring its pejorative connotations ± a trend encouraged (and perhaps initiated) by Plato ± the term ``sophist'' (sojisthÂ v) simply meant ``wise man.''

5 For more sympathetic and appreciative assessments of the sophists, see the pioneering discussions of Sidgwick, 1872: 288±307, and 1874: 66±80; and Grote, 1888, vii: 346±99, according to whom Plato's ``wholesale'' condemnation of the sophists ``betrays itself as the o¨spring . . . of systematic peculiarity of vision, the prejudice of a great and able mind'' (394±95). For more recent studies, see Havelock, 1957; Kerferd, 1981; Rankin, 1983; Vickers, 1988; and de Romilly, 1992.

6 Protagoras' annual income allegedly surpassed that of Phidias and any other ten sculptors combined (M. 91d2±5), and Gorgias and Prodicus reportedly made more money than the practitioners of any other technai (HMa. 282b4±d5).

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Socrates. Lovers of wisdom should not be lovers of gain.7 But while Socrates never tires of faulting them for this,8 he does not condemn the sophists en bloc; on the contrary, his attitude towards and treatment of them varies considerably. Unfailingly deferential towards Prodicus, uncharacteristically gentle with Gorgias, and alternately playful and rough with Protagoras, he is very hard on Hippias.9 In addition to being a sophist, Hippias is Elis' most distin-

guished foreign ambassador; and his earnings on these diplomatic missions exceed those of any other two sophists Socrates may care to name (HMa. 282d6±e8). Although much younger than his col- leagues, he surpassed them in both range and depth. His knowl- edge, bordering on the encyclopaedic, included mathematics, mnemonics, music, rhetoric, grammar, astronomy, genealogy, my- thology, and the history of antiquity. Possessed of a phenomenal memory, he can repeat a list of ®fty words upon hearing it once, and boasts that he can answer any question he is asked. As if that were not enough, he composes tragedies and epics, and even makes his own clothes.10 If Plato's portrayal is accurate, Hippias was not notable for his

modesty. He is constantly ¯aunting his knowledge and boasting about his achievements. Hippias is not only a braggart, he is also something of an exhibitionist ± the sort of person who is forever praising his own abilities and, what is worse, forever threatening to display them. Unlike the less versatile and more modest Ion, who acknowledges that his skill as a rhapsode is con®ned to Homer, Hippias goes out of his way to let everyone know that he can recite any poet (HMi. 363c7±d4). His greatest claim to fame is his linguistic virtuosity which enables him to stage magni®cent

7 The Xenophontic Socrates has additional objections to charging fees: wisdom should be freely shared, especially among friends; to impart it only upon being solicited by a pro- spective client with ready cash is a kind of intellectual prostitution (Mem. 1.2.6±8, 1.5.6, 1.6.5, 1.6.13). Xenophon also reports that Antiphon traced Socrates' self-induced pov- erty, shabby wardrobe, and unappetizing cuisine to his inexplicable refusal to charge a fee (1.6.2±3).

8 Harrison, 1964: 191, n. 44, has collected thirty-one passages from the Platonic corpus. 9 Especially in the Hippias Major where Woodru¨ describes his treatment of Hippias as ``remarkably savage'' (1982: 127).

10 The historical Hippias (circa 470±395) was far from being an empty showman. In addition to his intellectual achievements, he was a doxographer and the author of several trea- tises. Kerferd, 1981: 47±49; and de Romilly, 1992: 114±15, 181±82, describe him as a thinker of some consequence; and Guthrie, 1971a: 284, asserts that he ``has better claims [than most sophists] to be accepted as a serious ethical thinker.'' It is hard to believe that Plato would have agreed.

Hippias 95

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

epideictic displays ± not only in Athens, where he is an annual sellout, but throughout the Greek world. Like many self-impressed people, he is monumentally imperceptive, regally accepting Soc- rates' mock ¯attery and ostentatiously con®ding that he has yet to meet his superior at anything. Hippias' self-congratulatory tendencies have repelled many

readers. A. E. Taylor11 describes him as ``childishly conceited,'' Grote12 thinks his ``silliness and presumption'' both invite and excuse the ``derisory sting'' of Socrates' comments, Woodru¨13 deplores the ``empty core under [his] multifaceted veneer,'' Saunders14 judges him ``stupid and vain,'' and Allen15 disparages him as ``a professor who in his omniscience is unable to follow a Socratic argument.'' These somewhat churlish complaints are balanced by Guthrie's16 characteristically humane remark that, although Hippias is ``bombastic, humourless and thick-skinned,'' he is also a man of ``unsuspecting innocence'' and someone ``with whom it would be di½cult to be angry.'' A dramatic dialogue and, by common consensus, one of Plato's

earliest works,17 the Hippias Minor begins in medias res. Hippias has just completed an epideictic display and Eudicus, the promoter of the event, is eager to know what Socrates thinks of it. Character- istically, Socrates o¨ers no assessment; he would, however, like to ask a few questions. Recalling Eudicus' father's remark that just as the Iliad is a ®ner poem than the Odyssey, so also Achilles is a ®ner man than Odysseus, Socrates would like to know what Hippias thinks: which of the two is better (a� meiÂ nw)? and in what respect (kataÁ tiÂ , 364b4±5)? Hippias con®dently replies that, of the heroes who sailed to

Troy, Achilles is the best (a� riston), Nestor is the wisest (sojwÂ ta- ton), and Odysseus is the wiliest (polutropwÂ taton, 364c3±7).18

11 1929: 29. 12 1867, i: 387. 13 1982: xii. He is more sympathetic later (see 127±29). 14 1987: 215. 15 1996: 25. 16 1971a: 281±82. 17 FriedlaÈnder, 1964, ii: 146, thinks that its composition predates the death of Socrates on

the ground that Plato would not have portrayed his mentor so un¯atteringly after his death. So, too, Ritter, 1933: 39, n. 1.

18 The superlative of poluÂ tropov ± a term with multiple meanings ranging from the com- paratively neutral ``versatile,'' ``resourceful,'' and ``complex,'' to the more pejorative ``cunning'' and ``wily.''

Cross-Examining Socrates96

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

When Socrates suggests that it is not Odysseus but Achilles who is the most wily, Hippias demurs on the ground that Achilles is the most simple (a� plouÂ statov), that is, the most straightforward and honest of men. In support of this contention, he cites a passage from the Iliad in which Achilles expresses strong disapproval of people who think one thing but say another, thereby denouncing liars; in short, Achilles is simple and true whereas Odysseus is wily and false (365b3±5). Socrates ®nds this claim puzzling because it implies that the

true man (o� a� lhqhÂ v) is di¨erent from the false (o� yeudhÂ v), and he wonders whether that was really Homer's view. Hippias curtly assures him that it was; he adds that it is his view too. But Socrates is still unconvinced: Is the true man di¨erent from the false? Since it is impossible to interrogate Homer, he wonders whether Hippias would be willing to answer on his behalf? Hippias agrees.19 Before proceeding further, it is worth savoring the details of

this carefully prepared tableau. Having been prevailed upon by Eudicus, after hours and free of charge, to answer Socrates' ques- tions, the intellectually gifted but dialectically inept Hippias has inadvertently allowed himself to be drawn into a discussion for which his past experience has not prepared him. More accustomed to being ``e¨ective'' than to being precise and more adept at dis- coursing at length and without interruption than at answering a series of speci®c questions which require him to explain exactly what he means, he is about to make his inauspicious debut as a Socratic interlocutor. Socrates palpably relishes the prospect. Un- like most moralists, for whom arrogance and overcon®dence are intellectual vices to be rebuked and uprooted, Socrates welcomes them because they provide him with a polemical advantage which can be further exploited by the lethal combination of mock ¯at- tery and ironic self-deprecation. Like a wave that crests before breaking on the shore, arrogance and overcon®dence provide the psychological momentum that propels the interlocutor into the discussion. Accordingly, they should not be rebuked but encour- aged. The interlocutor must believe that he can provide Socrates with desperately needed instruction and thereby serve as his intel- lectual benefactor. The minute he begins to see through the ironic

19 The shift from the opinion of a poet to that of an interlocutor who is in agreement with him is a common Socratic move. See Pr. 347e1±348a7 and R. 331d4±5.

Hippias 97

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

facade, Socrates' task becomes harder; for he must then constantly ®nd new ways to prevent his interlocutor from trying to extricate himself from the discussion. The psychologically irresistible strat- egy of encouraging him to spread his wisdom abroad is part of Socrates' habitual concealment of purpose; he rarely divulges in advance that his purpose is not to learn from his interlocutor, but to refute him. The shock-value of the elenchus is an essential ingredient. Few Socratic interlocutors are as shocked as Hippias. Having

con®dently asserted that the simple and true Achilles is better than the wily and false Odysseus and that the true man is di¨erent from the false, the walking encyclopaedia who can answer any question he is asked suddenly ®nds himself burdened with two scandalously paradoxical theses: (i) the true man is not di¨erent from the false, they are one and the same (369b3±7); and (ii) the voluntary wrong- doer is better than the involuntary one (372d3±7). Socrates deduces the ®rst thesis as follows. Asked whether the

false, like the sick, are powerless to act or whether they have power (duÂ namiv) to do things, Hippias replies that they have power to do many things ± in particular, to lie and to deceive others (365d7±8). Furthermore, they deceive, not by foolishness, but by a kind of intelligence (jronhÂ sewÂ v tinov). In short, the false know (e� piÂ stantai ) what they are doing and are, therefore, wise in the things about which they are false, that is they are wise at decep- tion (365e6±10). Socrates thereupon de®nes ``the false man'' as ``the man who is intelligent and capable of speaking falsely'' (366b4±5). Hippias accepts the de®nition. He also agrees that anyone who is ignorant and lacks this capacity cannot be false, and that anyone who is intelligent and has it can do whatever he wishes whenever he wishes. For example, being an able mathema- tician, Hippias is best quali®ed to speak truly about mathematics. But for that very reason he is also best quali®ed to speak falsely about it. A less able person might wish to speak falsely, but in- advertently speak truly. But if that is true, as Hippias agrees, then the true man is not better than the false; they are one and the same (367c7±d2). This scandalously paradoxical thesis is then formulated in full

generality: the true and false man are one and the same ± not only in mathematics, but also in geometry, astronomy, and every other science. Socrates thereupon challenges Hippias to ®nd a single ex-

Cross-Examining Socrates98

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

ception to this thesis and, with a rare burst of conviction, assures him in advance that he never will: ``[I]f what I say is true'' (ei� d' e� gwÁ a� lhqhÄ leÂ gw, 369a4±5),20 there are no exceptions. But if the true and false man are one and the same, then Achilles is not better than Odysseus: if Odysseus is false, he is also true; and if Achilles is true, he is also false (369b3±7). Hippias can hardly believe his ears and vehemently protests:

Oh, Socrates! You're always weaving arguments of this kind. You pick out whatever is the most di½cult part of the argument, and fasten on to it in minute detail, and don't dispute about the whole subject under dis- cussion. (369b8±c2)21

His complaint is well founded. This argument has often been judged fallacious and even prompted charges of deliberate sophistry.22 But we need not become embroiled in polemical dispute. Even

if the argument is not fallacious, it is unsound. ``False man'' does

20 Pace Stokes, 1986: 5±7, Socrates here unambiguously advances an assertion himself. 21 Woodru¨, 1982: 85, and Rankin, 1983: 54±55, think that, in registering this protest, Hip-

pias is merely expressing his preference for epideictic discourse which does not fasten upon some triviality (smikroÂ n) but grapples with issues as a whole (o� lwÎ . . . twÄÎ praÂ g- mati, 369c1±2). However, a parallel passage in the Hippias Major suggests that his objec- tion cuts deeper: ``But Socrates, you don't look at the entireties of things, nor do the people you're used to talking with. You people knock away at the ®ne and the other be- ings by taking each separately and cutting it up with words. Because of that you don't realize how great they are ± naturally continuous bodies of being. And now you're so far from realizing it that you think there's some attribute or being (ti h� paÂ qov h� ou� siÂ an) that is true of these both but not of each, or of each but not of both. That's how unreason- ably and unobservantly and foolishly and uncomprehendingly you operate'' (HMa. 301b2±c3). See also Pr. 337c7±d1 where Hippias exhorts everyone to stop their petty bickering and to remember that they are all akin (suggeneiÄ v) by nature, not by law, and that like resembles like by nature (toÁ o� moion . . . twÄ o� moiÂ wÎ juÂ sei suggeneÂ v e� stin). Insofar as this passage suggests that all beings are naturally continuous and interconnected, it bespeaks an explicit and seriously held ontology. However, Socrates refuses to take Hip- pias' position seriously; on the contrary, he parodies it and construes it as a denial of the fact that Hippias and he are distinct individuals (HMa. 301d5±e8).

22 According to Mulhern, 1968: 283±88, poluÂ tropov is a dunamis-term which denotes a person's capacity to behave in a certain way, whereas yeudhÂ v is a tropos-term which de- notes a person's character (troÂ pov) and typical behavior; in saying that Odysseus is false, Hippias does not mean that he is capable of lying, but that he is a liar. Mulhern argues that Socrates starts out with the tropos-sense of yeudhÂ v and de®nes ``the false'' as ``those who speak falsely'' but later shifts to a dunamis-sense and rede®nes ``the false'' as ``anyone who is capable of speaking falsely,'' thereby fallaciously deriving the paradoxical conclu- sion that the true and false man are the same by equivocation. See also Shorey, 1933: 86±87, 89±90; Hoerber, 1962: 121±31; Sprague, 1962: 67±70; and Kahn, 1996: 114. For a sustained critique of Mulhern's analysis, see Weiss, 1981: 289±90, according to whom Socrates argues solely from the dunamis-sense of yeudhÂ v.

Hippias 99

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

not mean ``someone capable of speaking falsely.'' If it did, then the mere capacity to lie would be a su½cient condition for being a false man, and even the most scrupulously honest person, insofar as he has this capacity, would be a false man ± which is absurd. To call someone a false man is not to say that he is capable of lying; it is to say that he is a liar.23 The liar does not just say what is false; he intends to say what is false and does so knowing it to be false. Not only does he fail to speak the truth; he is untruthful. Unlike the capacity to lie, which is a logically necessary condition that must be satis®ed by any prospective liar, being a liar is a contingently true description of those who have the capacity and elect to exer- cise it. The class of those who are false is not coextensive with the class of those who are capable of being false. It is, of course, true that knowledge is a necessary condition for

lying con®dently and successfully. But knowledge of what? Accord- ing to Socrates, the con®dent and successful liar must have knowl- edge of that about which he speaks falsely; otherwise, he might inadvertently speak truly. In short, if one is con®dently and suc- cessfully to say what is false, one must know what is true. But although this is so in some cases, there are many others in which it is not. The fact that I do not know Jill's telephone number does not prevent me from lying con®dently and successfully if asked for it. I need only give out my own number or that of a friend. To lie about Jill's number, I need not know it; I need only know that the number I give out is not it. Provided that I know someone's tele- phone number, and my questioner does not know whose it is, I can lie about Jill's number as con®dently and successfully as someone who does know it. So long as these conditions are sat- is®ed, I need not worry about inadvertently disclosing her number and thereby inadvertently speaking truly. The examples could be multiplied inde®nitely. Even if, intending to lie, I panic and make up a number on the

spot which by an incredible stroke of bad luck turns out to be Jill's, thereby inadvertently saying what is true, I am not thereby transformed into a ``true man.'' Just as the false man does not just say what is false but intends to say what is false and does so knowing

23 Or at least that he ``is not what he pretends to be ± that he is some kind of fake'' (Vlastos, 1991: 277). The mistake was ®rst detected by Aristotle who, criticizing this very argument, observes that ``being false'' does not consist in the capacity (e� n thÄÎ dunaÂ mei ), but in the choice (e� n thÄÎ proaireÂ sei ), to speak falsely (E.N. 1127b12±13; and Meta. 1025a5±13).

Cross-Examining Socrates100

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

it to be false, so also the true man does not just say what is true but intends to say what is true and speaks believing that what he says is true. Although one can inadvertently say what is true, one cannot inadvertently be truthful. Unlike the former, which is an invol- untary act performed in the mistaken belief that one is lying, the latter is a voluntary speech-act performed with the intention of being truthful. Socrates is also mistaken in thinking that having the capacity to

speak falsely entails that one is skilled (e� mpeirov) at doing so. Like all forms of deception, lying is a social activity. It requires a de- ceiver (someone whose intention it is to deceive) and a deceivee (someone who is deceived). The true, i.e., truthful, man has no such intention. Nor would he be a good liar if he did.24 It is not as if the true and false man are equally skilled at speaking falsely ± and hence ``the same'' in that respect ± and di¨er only in that the former elects not to do so. Unless ``being skilled at speaking falsely'' is de®ned in terms of a purely cognitive capacity ± know- ing the truth ± in complete isolation from the situations in which, the persons to which, the intentions with which, and the charac- ters of the persons by which those propositions are uttered, the true and false man cannot be said to have a common skill. Unable to diagnose what is wrong with Socrates' argument but

rightly rejecting its paradoxical conclusion, Hippias rea½rms his original thesis that Achilles is better than Odysseus and defends it by producing textual evidence from the Iliad in which Homer portrays the former as truthful and free of falsehood and the latter as wily and deceptive. Socrates feigns admiration for Hippias' ex- egetical prowess and momentarily debates him on his own turf by producing several counterexamples based on other passages from the Iliad in which it is not Odysseus but Achilles who is portrayed as wily and false. Hippias is unimpressed and accuses Socrates of being a careless reader of texts. What these passages actually show is something very di¨erent: unlike Odysseus, who speaks falsely voluntarily and by design (e� kwÂ n te kaiÁ e� x e� piboulhÄ v), Achilles does so involuntarily (ou� k e� x e� piboulhÄ v . . . a� ll' a� kwn, 370e5±9) and because he is compelled by circumstances.

24 As Vlastos, 1991: 277, n. 134, rightly points out, it is ``surely false that, in general, truthful people are skilled at speaking falsely . . . Considerable talent and much practice would be required to become a skillful liar.''

Hippias 101

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Socrates has a ready reply: in that case, Odysseus is the better man; for, as the foregoing argument has shown, the voluntary liar is better than the involuntary one. At this point, Hippias is reduced to wondering out loud: How can that be true? Do not the laws (and people generally) make allowances for involuntary wrongdoing? Of all the claims advanced by the interlocutors in the early dia-

logues, we expect Socrates to endorse this one. For he appeals to the same principle himself at Ap. 26a1±7 where he assures Meletus that any wrongdoing of which he may have been guilty was invol- untary and thereupon reminds him that, according to both law and popular opinion, involuntary wrongdoers should not be pun- ished, but admonished and instructed. However, he does not endorse it. Unlike Hippias and people generally, who believe that the involuntary wrongdoer is better than the voluntary one, Soc- rates claims to hold the opposite opinion (372d3±7). This unusually ¯agrant violation of the sincere assent requirement25 is all the more culpable because in this passage Socrates not only fails to say what he believes, but says what he does not believe. Having done so, he adds that sometimes the popular view seems more plausible; in fact, he is constantly vacillating between the two and likens his present confusion to a ®t brought on by the foregoing argument. Would Hippias be kind enough to cure his soul of ignorance by answering a few more questions? Suspecting that Hippias' enthusiasm is waning, Socrates urges

Eudicus to remind him of his earlier promise to answer any ques- tion he was asked. Disinclined to continue but perceiving that his reputation is on the line, Hippias reluctantly agrees. To persuade Hippias that the voluntary wrongdoer is better

than the involuntary one, Socrates advances the following argu- ment (373c9±375c3):

(i) There are good and bad runners. (ii) Since rapidity is good and slowness evil, he who runs rapidly

runs well and he who runs slowly runs badly. (iii) He who runs slowly voluntarily is better (a� meiÂ nwn) qua runner

than he who runs slowly involuntarily. (iv) Since running is a kind of doing and since doing is a kind

of performing, he who runs badly performs badly and shamefully.

25 The violation is noticed by Kahn, 1996: 118.

Cross-Examining Socrates102

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

(v) He who runs badly runs slowly. (vi) It follows that the good runner performs these bad actions

voluntarily and the bad one involuntarily. (vii) It also follows that he who runs badly voluntarily is better

than he who does so unintentionally, and the same is true of every other activity.

(viii) Therefore in every activity he who errs voluntarily is better than he who does so involuntarily.

Hippias assents to each premise, but balks at the conclusion: it is monstrous (deinoÂ n) to say that voluntary wrongdoing is the mark of the good man and involuntary wrongdoing the mark of the bad (375d3±4). Socrates is unmoved: if a conclusion follows from an agreed-upon set of premises, it cannot be rejected simply because it is unwelcome. And the conclusion just arrived at certainly seems to follow from what had been said (e� k twÄ n ei� rhmeÂ nwn, 375d5). Since Hippias is still unpersuaded, Socrates must try again. His

®nal argument ± 375d8±376b6 ± is as follows:

(i) Justice is either a kind of capacity (duÂ namiÂ v tiÂ v) or knowledge (e� pisthÄ mh) or both.

(ii) If it is the former, then the more capable soul is more just; and if it is the latter, then the wiser soul is more just and the ignorant soul is more unjust.

(iii) If justice is both a kind of capacity and knowledge, then the soul which has both is more just and the soul which has only one is more unjust.

(iv) The more capable and wiser soul is the better soul and has greater capacity to do what is good or shameful.

(v) When a soul does what is shameful, it does so voluntarily (e� kouÄ sa) by capacity or art (diaÁ duÂ namin kaiÁ teÂ cnhn) ± or both ± which are attributes of justice.

(vi) Since to do injustice is to do evil and since to refrain from injustice is to do good, insofar as the more capable and bet- ter soul does what is just, it does so voluntarily; and insofar as the less capable and worse soul does what is just, it does so involuntarily.

(vii) Since the good man has a good soul and the bad man has a bad one, insofar as the good man does what is unjust, he does so voluntarily whereas the bad man does so involun- tarily.

Hippias 103

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

(viii) Hence he who errs voluntarily and does what is shameful and unjust ± if there is such a man ± is the good man.

Hippias again assents to each premise, but balks at the conclusion. He cannot agree with Socrates that the voluntary wrongdoer is the good man ± a monstrous (deinoÂ n) idea, in his opinion. Nor can Socrates agree with himself (376b8).26 Yet that seems to be the necessary (a� nagkaiÄ on) conclusion. What shall we say about this elusive little dialogue? Leaving

aside the outraged Apelt, who called it ``a kind of apology for sin,''27 many commentators28 think the interpretive key lies in the quali®cation ``if there is such a man'' (376b5±6). According to them, Socrates does not really believe that the voluntary wrong- doer is the good ± much less the better ± man; the argument is a reductio ad absurdum of the assumption that virtue is a techneÅ. The more cautious Irwin and Kahn think the argument is intended to show either that virtue is not a techneÅ or that it is a techneÅ ``unlike every other''29 or ``of quite an unusual sort''30 in that its practi- tioners never voluntarily misuse it.31 Since the end it promotes ± happiness ± is one which everyone necessarily wants, the misuse of the techneÅ of virtue is psychologically impossible.32 In short Soc- rates' psychological eudaemonism ± the doctrine that everyone necessarily desires happiness ± insures that voluntary wrongdoing will never occur. In short, Socrates' argument is a reductio ad ab- surdum intended to show that a denial of the so-called ``Socratic paradox''33 leads to the scandalous conclusion that the voluntary

26 Interestingly, Xenophon's Socrates espouses the same view without batting an eyelash, and his interlocutor (Euthydemus) instantly concurs (Mem. 4.2.19±20).

27 ``[E]ine Art Apologie der SuÈnde'' (1912: 226±27). 28 Among them, Gomperz, 1905, ii: 293±96; Pohlenz, 1913: 65; A. E. Taylor, 1929: 37±38;

Shorey, 1933: 86±87; Gould, 1955: 41±43; Hoerber, 1962: 128; Sprague, 1962: 76; Fried- laÈnder, 1964, ii: 144; O'Brien, 1967: 104±7; Irwin, 1995: 69±70; and Allen, 1996: 25.

29 Irwin, 1995: 69. 30 Kahn, 1996: 118. See also Saunders, 1987: 268±69. 31 See Aristotle (Meta. 1048a7±11), according to whom virtue is not a techneÅ, since every techneÅ

is a ``capacity for contraries'' and, therefore, admits of misuse ± something which is not true of virtue.

32 FriedlaÈnder, 1964, ii: 144, thinks Socrates' argument shows that the proposition ``Volun- tary wrongdoers are better than involuntary ones'' is self-contradictory and, therefore, that the voluntary misuse of the techneÅ of virtue is logically impossible.

33 By the ``Socratic paradox,'' I mean (what Santas, 1979: 190, calls) ``the moral paradox,'' i.e., the Socratic doctrine that ``if a man has knowledge of what is virtuous and also knowledge that it is always better for one to do what is virtuous, then he will always (so long as he has this knowledge and virtuous behavior is in his power) behave virtuously'' ±

Cross-Examining Socrates104

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

wrongdoer is better than the involuntary one ± a conclusion which is avoided by denying that there is such a man.34 Although this interpretation is quite plausible in some respects,

it is quite implausible in others. Its proponents too readily assume that the thesis that no one errs voluntarily ± and that all wrong- doing is, therefore, involuntary ± are already authentic, fully for- mulated, and self-consciously held Socratic doctrines in the Hip- pias Minor ± one of Plato's earliest works. The assumption can (and should) be challenged. For one thing, the so-called ``prudential paradox,'' that is, the thesis that everyone necessarily desires hap- piness, on which the so-called ``moral paradox'' depends, is never mentioned in the Hippias Minor. In fact, it is not until the Euthydemus ± a dialogue variously classi®ed as ``transitional,''35 ``late-early,''36 and ``pre-middle''37 ± that Socrates unambiguously asserts that everyone wishes to prosper (eu� praÂ ttein, 278e3±279a1), that is, to be happy. And it is not until the Meno that he unambiguously asserts that no one desires evil recognized as such, but only because he has mistakenly judged it to be good (77b6±78b2). To be sure, at Ap. 37a5±6 Socrates emphatically declares that he

never voluntarily wronged anyone. But the meaning of this asser- tion should not be divorced from its context. Socrates is on trial for his life. In saying this, he is proclaiming his own innocence, not advancing a philosophical thesis about human conduct univer- sally. Indeed, if we attend closely to what he says when his philo- sophical guard is down, we discover that he has grave doubts about other people. At Ap. 41d7±e1 he asserts that his accusers are morally blameworthy (a� xion meÂ mjesqai ) for trying to harm him ± apparently a blatant case of voluntary wrongdoing, for otherwise

a paradox which is distinct from and presupposes (what Santas calls) ``the prudential paradox,'' i.e., the Socratic doctrine that men desire only good things, the corollary of which is that no one does what is evil knowing it to be evil, i.e., that all wrongdoing is involuntary. Unlike most commentators, who think that, for Socrates, knowledge of what is virtuous is both necessary and su½cient for virtuous conduct, Santas thinks that it is only a necessary condition; what is su½cient for virtuous conduct is not simply knowl- edge of what is virtuous, but knowledge of what is virtuous in conjunction with the univer- sal desire for happiness.

34 Horne¨er, 1904: 6±7, thinks that the argument of the Hippias Minor is a (Platonic) refuta- tion of the (Socratic) doctrine that virtue is knowledge. He ®nds a similar refutation of Socratic intellectualism in the Laches.

35 Vlastos, 1983a: 27, n. 2, 1991: 46±47, and 1994: 71, n. 14. 36 Guthrie, 1975: 266. 37 Kahn, 1992: 38.

Hippias 105

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

they would not be blameworthy. And at G. 480e5±481b1 he im- plicitly allows for the possibility of voluntary wrongdoing by pre- senting Polus with the hypothetical example of a person whose duty it is to harm someone for having wronged someone else and suggests that the most e¨ective way of doing so would be by pre- venting the wrongdoer from coming to trial and being punished, thereby preventing him from being purged of wrongdoing and improved. He adds that should this prove to be impossible, the next best thing would be to arrange for the wrongdoer's escape so that, unpunished and unpurged of his evil, he will live a long and miserable life. One would be hard pressed to think of a more paradigmatic example of voluntary wrongdoing. Interestingly, at Cr. 49a4±5 Socrates asks Crito whether we should ever do wrong voluntarily ± a pointless question if voluntary wrongdoing is psy- chologically (not to mention logically) impossible ± and he pro- ceeds to argue that we should not ± thereby again implying that voluntary wrongdoing is possible. Why prohibit the psychologi- cally (or the logically) impossible and fortify the prohibition with an argument? All this (by no means exhaustive) textual evidence suggests that the thesis that no one errs voluntarily ± and that all wrongdoing is, therefore, involuntary ± are not already authentic, fully formulated, and self-consciously held Socratic doctrines in the Hippias Minor, but rather doctrines at which Plato arrived later, gradually, and in conjunction with a more sophisticated moral psychology as a hard-earned solution to the problem posed in that dialogue. Grote was one of the ®rst to sponsor this view. Speaking of

Hippias' unwillingness to draw the (apparently) required infer- ence, he says:

If there be any argument, the process of which seems indisputable, while its conclusion contradicts, or seems to contradict, what is known upon other evidence ± the full and patient analysis of that argument is indis- pensable, before you can become master of the truth and able to defend it. Until you have gone through such analysis, your mind must remain in that state of confusion which is indicated by Sokrates at the end of the Lesser Hippias.38

Grote thereupon faults Hippias: not only for a½rming the premises of Socrates' argument and denying its conclusion, but also (and

38 1867, i: 397.

Cross-Examining Socrates106

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

more fundamentally) for not being perplexed by Socrates' argu- ments and for not being dissatis®ed with the outcome of the whole discussion. Instead of allowing himself to experience the full im- pact of aporia by taking the argument seriously, Hippias merely disagrees with its conclusion, thereby betraying that he is unwill- ing to acknowledge his own confusion ± one of the ``enemies [with which] the Searcher for truth must contend [and] the stimulus provocative of farther intellectual e¨ort.''39 Grote does not deny that Socrates' arguments are problematic; on the contrary, he bluntly asserts that had Socrates' reasoning been put into the mouth of Hippias, most readers would dismiss the dialogue as ``a tissue of sophistry.''40 However, since everything which in the mouth of Hippias would have been blatant sophistry is advanced by Socrates, with Hippias resisting every step of the way, a di¨er- ent assessment is in order. And Grote provides it: the Hippias Minor does not refute error; it documents confusion and records Socrates' inability to dispel it. In this dialogue, it is Socrates who is confused, and his confusion and resultant vacillation are not feigned, but real.41 It would be pointless to belabor the truism that an unwelcome

conclusion which seems to follow necessarily from premises to which one has assented but which contradicts (or seems to contra- dict) what is known upon other evidence cannot be judged errone- ous by proclamation. But does the unwelcome Socratic conclusion really follow? It does not. The argument by which Socrates derives the para-

doxical conclusion that the voluntary wrongdoer is better than the involuntary one is vitiated by two logical ¯aws. The ®rst is a sys-

39 Ibid.: 397±98. 40 Ibid.: 394±98. So, too, FriedlaÈnder, 1964, ii: 146, who thinks that Socrates' sophistry is so

blatant in the Hippias Minor that, were it not for the testimony of Aristotle, few would endorse it as genuinely Platonic. See also Guthrie, 1975: 195±96, who complains about the ``manifest absurdities'' of this dialogue in which Socrates comes ``perilously near'' the ``logical clowning'' of the two sophists parodied in the Euthydemus.

41 Ibid. The same interpretation is o¨ered by Ovink, 1931: 176±77; and Vlastos, 1991: 275± 79, according to whom the Hippias Minor documents ``honest perplexity'' and portrays Socrates ``in a muddle'' and victimized by a conceptual confusion shared by Plato whose moral insight at the time extended no further. This view is rejected by DupreÂel, 1948: 197, and Annas, 1992: 50±51, according to whom Socrates is merely toying with Hippias. See also Kahn, 1992: 253: ``Since there are serious moral issues here, for Vlastos the fal- lacies must be honest errors and the perplexity sincere. Why is Vlastos so blind . . . to the malicious humor and ¯agrant blu½ng of the Hippias Minor?''

Hippias 107

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

tematic equivocation on the term ``good'' (a� gaqoÂ v) and its com- parative ``better'' (a� meiÂ nwn).42 According to Socrates, since the man best quali®ed to speak truly about mathematics (or any other subject) is also, and for that very reason, best quali®ed to speak falsely about it, it follows that the man who is good at these things (o� a� gaqoÁ v periÁ touÂ twn, 367c4) is the good man (o� a� gaqoÂ v, 367c5± 6). But the inference is fallacious. ``A is good at X-ing'' does not entail ``A is good.'' The proposition ``A is good at X-ing'' does not evaluate character. It does not even primarily evaluate perfor- mance. It evaluates capacity. It is, in short, a dispositional state- ment. The only conclusion that can validly be derived from Soc- rates' argument is a non-moral one, and it is not paradoxical at all. There is a sense in which skilled and voluntary wrongdoers are better than their less skilled and involuntary counterparts. No one would deny that a runner who voluntarily loses a race by running slowly is better qua runner than a runner who would have lost the race no matter how fast he had tried to run. The same applies to wrestlers who voluntarily lose matches, vocalists who voluntarily sing ¯at, and archers who voluntarily miss targets. All are better qua runners, wrestlers, vocalists, or archers. All could have avoided the blunder in question by exercising their respective skills. The fact that they elect not to exercise them does not cast doubt on their expertise or their capacity, but it does cast doubt on their character. The runner who voluntarily loses a race is better qua runner, but not qua participant in the race; the vocalist who vol- untarily sings ¯at is better qua vocalist, but not qua musical per- former; and so on. However, from the fact that runners, wrestlers, musicians, and archers who voluntarily misuse their respective technai are better qua runners, wrestlers, musicians, and archers than their less skilled and involuntary counterparts, it does not follow that they are better qua human beings. It is only by blatant equivocation that anyone could argue that since the former are better at what they do, they are, therefore, better, that is morally better.

42 See Guthrie, 1975: 195; Hoerber, 1962: 127; Mulhern, 1968: 288; and Sprague, 1962: 72. For a vigorous ± but, to my mind, unconvincing ± rebuttal, see Weiss, 1981: 298±304, according to whom aÁ gaqoÂ v, as employed by Socrates throughout the Hippias Minor, does not mean ``morally good''; o� a� gaqoÂ v is ``an incomplete expression'' which becomes ``complete'' by specifying the techneÅ in which he is skilled, e.g., the just man is good at justice. See also Penner, 1973a: 141±43.

Cross-Examining Socrates108

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

This brings us to the second logical ¯aw. Throughout the argu- ment, Socrates implies that insofar as a person voluntarily fails to do what he is capable of doing, he voluntarily misuses his techneÅ. However, in implying this, he overlooks an important distinction. It is true of the doctor who voluntarily poisons a patient, thereby misusing the techneÅ of medicine, but it is not true of the runner who voluntarily loses a race, the wrestler who voluntarily loses a match, the vocalist who voluntarily sings ¯at, or the archer who voluntarily misses the target. Unlike the former, the latter do not misuse their technai; on the contrary, they do not use them at all. I conclude that Socrates is wrong on four counts: (i) in de®ning

the ``false man'' as ``the man who is capable of speaking falsely''; (ii) in claiming that the true and false man are one and the same; (iii) in arguing that ``A is good at X-ing'' entails that ``A is good''; and (iv) in suggesting ± however tentatively or vacillatingly ± that the voluntary wrongdoer is (morally) better than the involuntary one. Hippias, on the other hand, is right on three counts: (i) in denying that the true and false man are one and the same; (ii) in denying ± or, at least, in doubting ± that ``A is good at X-ing'' entails that ``A is good''; and (iii) in denying that the voluntary wrongdoer is (morally) better than the involuntary one. Not only is Socrates ``in a muddle'' about the problem of volun-

tary wrongdoing; the whole dialogue is something of a muddle. Furthermore, it is not Hippias' false (or confused) beliefs that ac- count for the dialectical impasse with which the discussion con- cludes. The blame is wholly traceable to Socrates. It is he who contributes all the false (or dubious) premises.43 It is also he who draws (or is inclined to draw) all the fallacious inferences. Hippias, on the other hand, does not endorse them: and it is to his credit that he does not. Although far from astute and unable to diagnose what is wrong with Socrates' arguments, he perceives that some- thing is wrong with them and ®nally pronounces the conclusion of the dialogue ``monstrous.'' The legitimacy of his protest ± the product of a robust common sense and a Socratically uncon- taminated linguistic sensitivity ± is attested to by Plato's and Aris- totle's subsequent acknowledgments of the possibility of akrasia.

43 This is su½cient to discredit Vlastos's contention, 1983a: 57 ± endorsed by Stokes, 1986: 29±30 ± that the theses investigated by Socrates in the early dialogues are, without ex- ception, advanced by his interlocutors.

Hippias 109

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

On the other hand, there is no textual basis for accusing Soc- rates of deliberate sophistry in the Hippias Minor. The conceptual thicket in which he ®nds himself may very well be symptomatic of ``honest perplexity.'' However, in view of the false premises and fallacious inferences which generate it, the perplexity is self- induced. And its magnitude is intensi®ed by the realization that the solution ®nally arrived at ± the doctrine that all wrongdoing is in- voluntary ± is every bit as counterintuitive and paradoxical as the perplexity it is intended to dispel. Has Socrates cared for the soul of Hippias? He has not. It is

hard to believe that he has even tried. Although we should hesitate before endorsing I. F. Stone's contention that the Hippias Minor is a satire of Socrates who ``outdoes the Sophists in sophistry,''44 it remains true one searches the dialogue in vain for the moral seri- ousness and concern for the souls of his fellows for which Socrates has traditionally been revered.

44 1988: 56±57.

Cross-Examining Socrates110

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

c ha p t e r 6

Laches and Nicias

In the Meno, after uncharacteristically acknowledging that Athe- nian history provides many examples of good politicians who were also good men, Socrates asks:

[B]ut have they been good teachers of their own virtue? That is the point we are discussing, not whether there are good men here or not, but . . . whether virtue (a� rethÂ ) can be taught . . . whether the good men of today and of the past knew how to pass on to another the virtue they themselves possessed, or whether a man cannot pass it on or receive it from another. (M. 93a6±b5)

His doubts are prompted by the fact that good men like The- mistocles and Pericles, who spared no expense to teach their sons everything else, did not teach them to be virtuous. In the Laches, no one ventures the opinion that virtue cannot be

taught; the presumption throughout is that it can. The dialogue depicts a conversation whose dramatic date falls somewhere be- tween the retreat from Delium (424) ± a humiliating Athenian defeat which, according to Laches, could have been avoided had everyone fought as bravely as Socrates (La. 181b1±4) ± and the Battle of Mantinaea (418) in which the historical Laches perished. The scene is an unidenti®ed palaestra where Stesilaus, a teacher of the art of ®ghting in heavy armor, has just completed a dem- onstration. Among those present are Socrates, Lysimachus, and Melesias, the sons of Aristides the Just and Thucycides the Elder; they are accompanied by their sons, also named Aristides1 and Thucydides, for whose bene®t the ensuing discussion is ostensibly conducted. Acknowledged nonentities in Athenian public life, Lysi-

111

1 At Th. 150e1±151a2 ``Socrates'' numbers Aristides among those who had forsaken him too early. Having taken up with evil companions, he su¨ered a miscarriage of the ideas with which he was pregnant and neglected the intellectual o¨spring to which he had given birth through the agency of Socratic midwifery.

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

machus and Melesias blame their fathers for neglecting their edu- cation. Ashamed in their sons' eyes and determined to be better fathers themselves, they have invited Laches and Nicias,2 two cele- brated Athenian generals with sons of their own,3 to solicit their advice about whether the boys should acquire Stesilaus' art as part of an education designed to make them as good as possible. That Lysimachus and Melesias are genuinely concerned about

their sons is beyond dispute. However, their views about what they should learn and why they should learn it leave something to be desired. Even if we reject Stokes's overly harsh judgment that their values are ``of the utmost banality,''4 the fact remains that they are rather super®cial. Although brimming over with good inten- tions, they are chie¯y concerned that their sons receive an educa- tion which will equip them to perform noble deeds (kalaÁ e� rga) and thereby become worthy of their illustrious names. Nicias warmly endorses this heartfelt display of parental con-

cern. So does Laches, but he wonders why Lysimachus and Me- lesias have not solicited the opinion of Socrates who is of the same deme, whose every waking hour is spent discussing the education of the youth, and whose military prowess at Delium attests to the fact that he is eminently quali®ed to give advice about strategic tactics. The somewhat out-of-touch Lysimachus, who had not re- alized that Socrates is the son of his old friend Sophronicus, is elated and immediately turns to Socrates for advice. We expect Socrates to challenge this preoccupation with curric-

ular trivia. But he does not. Although he will later raise issues of greater moment than whether the acquisition of the art of ®ghting in heavy armor should rank high on the diligent parent's list of educational desiderata, for the present he defers to the older and more experienced generals. Nicias enthusiastically recommends the acquisition of this art.

His reasons, unimportant in themselves, are worth noticing for the light they shed on his character and on the petty and increasingly mean-spirited rivalry which quickly develops between him and

2 The historical Nicias was executed in 413 for incompetence which undermined the Sicilian expedition and led to the downfall of Athens.

3 At La. 180c8±d3 Nicias credits Socrates with having secured the services of Damon of Oa as a music teacher for his son Niceratus. Although Damon was a sophist and a pupil of Prodicus, he is Socrates' musical authority throughout the Platonic corpus. In Xenophon's Symposium (3.5), Niceratus reports that his father was so intent on his becoming a good man that he made him memorize the entire Iliad and Odyssey.

4 1986: 42.

Cross-Examining Socrates112

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Laches. According to Nicias, the art of ®ghting in heavy armor is an excellent form of exercise and self-discipline. It is also militarily advantageous. Its acquisition leads to a mastery of the whole art of military science and thus enables its possessor to ascend natu- rally to the rank of general. Above all, it makes him bolder and more courageous (qarralewÂ teron kaiÁ a� ndreioÂ teron, 182c6) in battle, not to mention more impressive to his peers and more in- timidating to his enemies ± a revealing afterthought which shows how much reputation matters to Nicias. Laches disagrees. Prefacing his rebuttal with elaborate assur-

ances that he is loathe to dissuade anyone from acquiring a bona ®de art and even more loathe to contradict a colleague, he immediately proceeds to do both with a gusto which reveals that his apparent civility towards Nicias was actually the preamble to a scathing cri- tique of his views about military tactics and an implicit slur on his professional judgment. In Laches' opinion, Nicias has greatly ex- aggerated the usefulness of the art of ®ghting in heavy armor. For one thing, it is not cultivated in Sparta where nothing militarily advantageous is overlooked. Nor can he recall anyone who ever distinguished himself by means of it. In fact, Stesilaus is living proof of its uselessness. Once during a naval battle, his spear acci- dentally became entangled in the rigging of an enemy ship moving in the opposite direction ± a mishap which required him to run the entire length of his own vessel in a futile e¨ort to extricate it, thereby rendering himself strategically vulnerable and reducing himself to a laughing stock ± an indication of the importance that Laches ascribes to reputation and a tell-tale sign that, in addition to being very competitive and prone to rhetorical overkill, he is something of a gossip. He concludes by noting that Nicias is also wrong in thinking that the possession of this art makes one more courageous in battle. On the contrary, if one is cowardly, it will make him rash; and if one is already courageous, it will prompt jealous colleagues ± of whom Laches appears to be one ± to scru- tinize him all the more closely in search of the slightest slip.5

5 ``Courage'' and ``courageous'' are sometimes infelicitous translations of a� ndreiÂ a and a� ndreiÄ ov (see MeÂron, 1979: 185±203; and Stokes, 1986: 36, 44±45, 55±57). Laches' speech is a case in point. By ``courage'' he does not mean a moral quality which enables its pos- sessor to perform heroic actions from which the faint-hearted shrink; in employing the terms a� ndreiÄ ov and deiloÂ v (184b3±6), he is not so much distinguishing the courageous from the cowardly as the skilled from the unskilled. Stesilaus' behavior is laughable: not because he is a coward, but because he is inept. If we translate a� ndreiÂ a as ``courage,'' the anecdote becomes contextually irrelevant.

Laches and Nicias 113

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Confronted with this disagreement between two experts, Lysi- machus implores Socrates to cast the tie-breaking vote.6 Sardoni- cally asked whether he thinks truth is ascertained by counting heads, Lysimachus confesses that he knows of no other way. Al- though he is universally disparaged for this reply, it is perfectly reasonable given the mildly Kafkaesque situation in which he ®nds himself. Having initially solicited the advice of Nicias and Laches, he had been urged to consult Socrates, who had agreed to shed whatever light he could but only after hearing what the generals had to say. Finding them hopelessly at odds, he repairs once again to Socrates, only to be ridiculed for his Democratic Theory of Truth ± a hectoring tactic which is not only unfair but unkind. This is not mere sentimentalism on my part elicited by the fact that Lysimachus is an old man. In soliciting Socrates' opinion, he is not consulting just any third person, but someone who just minutes ago had been recommended to him as another expert ± indeed, as the foremost expert. In fact, Lysimachus' behavior is eminently sensible. Confronted with a disagreement between two experts, what else can a non-expert do but consult another expert? To prevent Melesias from endorsing Lysimachus' theory of

truth, Socrates provides him with a bit of coaching. If in doubt as to how his son should go about preparing for an athletic contest, would he put the matter to a popular vote or consult a competent trainer? ``The latter,'' replies Lysimachus. Socrates concurs on the ground that such decisions require expertise, and the opinion of an expert should prevail even if he is outvoted by many non- experts (184e11±185a3).7 Furthermore, before consulting an expert, one needs to know what one is consulting him about. The non- plussed Nicias retorts that Lysimachus and Melesias know per- fectly well what they are consulting Laches and him about, namely, the advisability of acquiring the art of ®ghting in heavy armor. Socrates disagrees: what they are chie¯y concerned about is how to make their sons as good as possible. Accordingly, what they really need is not someone skilled in this or that military tac- tic but someone skilled in caring for the soul (tecnikoÁ v periÁ yuchÄ v qerapeiÂ an, 185e4). In thus changing the subject from education as

6 A feeble response, according to Arieti, 1991: 56, which reveals ``the shallowness of [Lysi- machus'] mind'' and ``the popular insipidity'' of entrusting young minds to ``democratic mechanisms.''

7 Cf. Cr. 44c6±9, 47a2±c3, 48a5±7; and G. 471e2±472a1.

Cross-Examining Socrates114

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

preparation for noble deeds to education as care of the soul, Soc- rates leads the discussion in a very di¨erent direction and, in the process, deprives the speci®cally military expertise of Nicias and Laches of whatever authority it may have had. Socrates thereupon lays down an important (and character-

istically Socratic) procedural stipulation: professions of expertise should not be accepted at face value; every claimant must produce his credentials either by identifying the teacher(s) who have im- proved him or by identifying people he has improved (186e3± 187a8). Any candidate who fails to satisfy one (or both) of these conditions disquali®es himself as an expert. Although it is understandable that Socrates should want to

expose the super®ciality of Lysimachus' and Melesias' educational goals, his method of going about it is somewhat devious. Having agreed, as military experts, to serve as consultants about the ad- visability of acquiring the art of ®ghting in heavy armor, Nicias and Laches suddenly ®nd themselves cast in the unsought ± and unprofessed ± role of moral experts and required either to justify their claim to moral expertise or withdraw it. Their already di- minished authority is undermined still further when Socrates ex- presses his ``surprise'' upon ®nding them in disagreement; for there can be no disagreement between real experts. Having laid down his criteria of expertise and having implied that, in view of their disagreement, Nicias and Laches cannot both be experts, he urges Lysimachus to question them in an e¨ort to determine which of them is. Predictably, Lysimachus begs o¨ and implores Socrates to question them in his stead. Before undergoing Socratic cross-examination,8 Nicias reveals

that he is aware of what he is letting himself in for. Addressing Lysimachus, he says:

8 As always, participation in elenctic debate is voluntary (see La. 188a6±c3, 189a1±3; and Ch. 158e4±5).

Laches and Nicias 115

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Viewed philosophically, Nicias' speech is an astute commentary on the ad hominem character of the Socratic elenchus and an ``omi- nous hint''9 that Socrates is about to examine his interlocutor's life as well as his thesis. Viewed dramatically, however, it is a con- descending gesture towards Laches calculated to convey the mes- sage that, so far as Socrates is concerned, Nicias is in-the-know. Having advised his dialectically uninitiated colleague of the rigors of the Socratic elenchus, Nicias announces that he is willing to be tested. He o¨ers several reasons ± all of them self-congratulatory: ®rst, he is accustomed to these discussions and knows what to ex- pect; second, there is no harm in being reminded of one's past misdeeds; and, third, conversations with Socrates always make him think harder.10 He concludes by expansively noting that not every- one is so refreshingly eager for correction and reproof; indeed, many ®nd the experience unpleasant, and he fears that Laches might be one of them. Not about to be upstaged, Laches responds with a rhetorical

¯ourish of his own. He, too, is a lover of discussion (jiloÂ logov) ± albeit with a low tolerance for moral hypocrites whose behavior is at variance with their principles. Accordingly, he has a condition: his conversational partners must be worthy of him. Their deeds must be consistent with their words, that is, they must practice what they preach. If this condition is not satis®ed, he becomes a hater of discussion (misoÂ logov); and the greater the discrepancy, the greater his hatred. Fortunately, Socrates measures up to his lofty standards. He would, therefore, be happy to be examined and, if need be, refuted by this true musician (mousikoÂ v) whose behavior at Delium testi®es to his courage and whose life bespeaks so complete a harmony between his words and deeds that it may without exaggeration be described as the paradigmatic embodi- ment of the authentically Greek mode ± the Dorian (188d2±8).11 This long introduction (178a1±190d2) ± if ``introduction'' is not

a misnomer for what is, in fact, almost half of the dialogue ± ac- complishes several important philosophical purposes. Ostensibly

9 The terminology, admirably apposite, is O'Brien's (1963: 136). 10 Although Xenophon's writings abound with testimonials about Socrates' salutary e¨ects

on his interlocutors, such tributes are rare in Plato. This is one of them. 11 The mode which ``suitably imitate(s) the tone and rhythm of a courageous person who is

. . . facing wounds, death, or some other misfortune, and who, in all these circumstances, is ®ghting o¨ his fate steadily and with self-control'' (R. 399a5±b3).

Cross-Examining Socrates116

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

concerned with the advisability of acquiring the art of ®ghting in heavy armor, it actually provides what is perhaps the clearest statement in the early dialogues of the announced purpose of the Socratic elenchus ± to test the interlocutor ``well and truly'' about ``every last detail.'' It also provides a correspondingly clear de- scription of the ideal interlocutor ± someone who is willing to be tested and, if need be, refuted. Judged solely by their speeches, Nicias and Laches admirably embody this ideal. Unlike most Soc- ratic interlocutors, both know from the start what will be required of them, and both are willing to submit. Judged by their behavior, however, both fall far short and fail to comport themselves like the ideal interlocutors they purport to be. Having inherited control of the discussion, Socrates substitutes

a di¨erent line of questioning. Instead of requiring Nicias and Laches either to justify their claim to moral expertise or withdraw it, he proposes another strategy which will achieve the same result but enable them to proceed at a more fundamental level. He in- troduces it with uncharacteristic laboriousness:

Suppose we know, about anything whatsoever, that if it is added to an- other thing, it makes that thing better, and furthermore, we are able to make the addition, then clearly we know the very thing about which we should be consulting as to how one might obtain it most easily and best. (189e3±7)

Having delivered himself of this obscure utterance, Socrates clari- ®es it as follows:

[S]uppose we know that sight, when added to the eyes, makes better those eyes to which it is added, and furthermore, we are able to add it to the eyes, then clearly we know what this very thing sight is, about which we should be consulting as to how one might obtain it most easily and best. (190a1±5)

Knowing what sight is, ophthalmologists do not tell their patients that it is good; they enable them to see. Analogously, knowing what virtue is, moral experts do not tell their pupils that certain actions are good; they enable them to be good.12 So if Nicias and Laches are to be of any help to the sons of Lysimachus and Melesias, they must know what virtue is and how to inculcate it in

12 As Woodru¨, 1988: 96, aptly puts it: ``An expert on Courage should be like a doctor who knows how to perform a courage-implant and guarantee that the implant will never fail to cause the recipient to act in courageous ways.''

Laches and Nicias 117

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

others; and if they know what it is, they must be able to say what it is (190c3±7). To make the discussion more manageable, Socrates narrows the topic from the ``whole'' of virtue to the speci®c ``part'' with which they are presently concerned, namely, courage (a� ndreiÂ a).13 The question, then, is: What is courage? Laches thinks the question is easily answered: anyone is coura-

geous who stays at his post and faces the enemy (190e4±6). Soc- rates agrees, but complains that Laches has not answered his question: instead of de®ning courage, he has merely produced an example of a courageous action. But many counterexamples come to mind. Can a soldier not display courage in retreating (191a5± 6),14 as well as in many other situations, for example, amid perils at sea, in illness, in poverty, in politics, and so on (191c7±e8)? Laches concurs. Socrates thereupon explains that what he wants is not an example of one or two courageous actions, but a de®nition of courage which explains what is the same in all these cases (tiÂ o� n e� n paÄ si touÂ toiv tau� toÂ n e� stin, 191e10±11) ± the power (duÂ namin) which is denoted by the term ``courage'' and which is common to all actions called by that name (192b5±8).15 Vlastos thinks that, in supplying this list of diverse examples,

Socrates is able to ``extend enormously the range of [the] applica- tion'' of the term ``courage,''16 thereby a¨ecting ``a radical revi- sion'' of the meaning that attaches to it in conventional usage and stretching its application so as ``to break the traditional moral dogma that had kept courage a class-bound, sex-bound, virtue.''17 Laches thinks of courage in primarily military terms ± as para- digmatically the virtue of a soldier. Under Socrates' guidance ± with its characteristic ``tendency towards greater generality and abstraction''18 ± he gradually realizes that its scope is wider. But Socrates' ``guidance'' admits of another interpretation. Al-

13 Albeit without explaining what he means by a ``part'' of virtue ± the very thing for which he faults Protagoras (Pr. 329b5±331a4).

14 ``[A] nice touch'' on Plato's part (Santas, 1971: 185), since Laches had previously praised Socrates' courage during the retreat from Delium. At Sym. 221a5±b1 Alcibiades reports that during that battle Socrates was much cooler under pressure than Laches ± an un- timely truth discreetly suppressed here.

15 Although the Laches and the Charmides are both de®nitional dialogues in which the ``What-is-F ?'' question is prominent, in neither are the objects of Socrates' search de- scribed as eideÅ or ideai, as they are in the Euthyphro. See Vlastos, 1991: 93, n. 46.

16 1956: 1. See also Stokes, 1986: 71±74; and Penner, 1992a: 2. 17 Vlastos, 1981b: 411, n. 3. 18 The terminology is Santas's (1971: 187).

Cross-Examining Socrates118

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

though diverse examples can be employed to ``extend enormously'' the applicational range of a term, they can also be employed for other, less salutary purposes. This passage is a case in point. Soc- rates introduces his diverse examples: not as illuminating supple- ments designed to make it easier for Laches to discover the de®n- ing characteristic common to all those actions and persons to which the term ``courage'' is correctly applied, but as gratuitous complications designed to make it harder for him to discover it. Anticipating trouble, Laches asks whether it is really necessary to state ``what is the same'' in all these examples (192b9±c1). His question is a good one, but Socrates ignores it. Instead of dispel- ling Laches' doubts by explaining why it is necessary, he requires him to accept this doctrine on authority and merely assures him that it is necessary if they are to answer ``their'' question.19 Laches thereupon de®nes courage as ``a certain endurance of

the soul'' (karteriÂ a tiv . . . thÄ v yuchÄ v, 192b9). Socrates is again dissatis®ed. Having rejected Laches' ®rst answer by producing ex- amples of the de®niendum which are not covered by the de®niens, thereby showing that it is too narrow, he rejects his second by producing examples which are covered by the de®niens but which are not examples of the de®niendum, thereby showing that it is too broad. His argument ± 192c4±193d12 ± is as follows:

(i) Courage is noble (kalwÄ n). (ii) When conjoined with wisdom (jronhÂ sewv), endurance is noble

and good; but when conjoined with foolishness (a� jrosuÂ nhv), it is harmful and bad (blaberaÁ kaiÁ kakouÄ rgov).

(iii) What is harmful and bad cannot be noble. (iv) Since courage is noble, harmful and bad endurance cannot be

courage. (v) Hence only wise endurance is courage.

``So it seems'' ( E� oiken), replies Laches.20

19 This is the closest Socrates ever comes to ``explaining'' why, in de®ning a term, it is nec- essary to state the character F-ness common to all Fs. It is, of course, no explanation at all.

20 In inducing Laches to a½rm that courage is wise endurance, Socrates elicits his assent to a thesis which, in view of the counterexamples that follow, he himself rejects. This single passage is su½cient to discredit Vlastos's contention that Socrates would ever ``(know- ingly, and in a serious vein) assert categorically a false premise or endorse a fallacious argument'' (1981b: 223, n. 5).

Laches and Nicias 119

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Here, then, is a new de®nition to examine. Whose de®nition is it? According to Socrates, it is Laches'. Most commentators agree: ®rst, because Laches had agreed to each step of the argument; second, because he had also agreed that the de®nition ± courage is wise endurance ± follows from what he said (kataÁ toÁ n soÁ n loÂ gon, 192d10). Commenting on Socrates' procedure, Vlastos says:

By exploring relations of implication and incompatibility between the proposed statement and all sorts of others . . . Socrates' method makes you see how big in their consequences are matters that seem so picayune and piddling by themselves, and thus how worthy of serious inquiry are things which otherwise would have passed unnoticed . . . And the very fact that you are required to say whether or not you agree to each prop- osition as it is put before you, one at a time, gives you a high incentive to press for clari®cation, for you may soon look like a fool if you agree to something without understanding what exactly you agreed to.21

Vlastos's remarks make Socrates' every move seem perfectly straightforward and eminently fair. At every stage of the argu- ment, the interlocutor is pointedly asked whether he assents to the proposition posed for his assent. He is invited to dissent or to ``press for clari®cation'' so that later on he will not ``look like a fool.'' He need not assent. But if he does, he is expected to abide by his logical commitments and to acknowledge that he has been refuted ``from his own beliefs.'' What could be more reasonable? The only thing Vlastos neglects to mention is that by the end of the discussion the interlocutor will probably ``look like a fool'' anyway. Why? Because in spite of Socrates' ``guidance'' he still manages

to commit some fatal logical blunder? Or because Socrates' ``guid- ance'' is designed to ensure that he will commit it? In Laches' case, the latter is closer to the truth. A general whose understand- ing of courage is con®ned to examples drawn from the battle®eld and who can elucidate it in only the vaguest of terms as ``a certain endurance of the soul,'' Laches has said nothing to suggest that he thinks courage has anything to do with wisdom. Having stated his ``view'' of courage, he is presented with a series of proposi- tions phrased in such a way as to make dissent virtually impossi- ble, informed that they entail that courage is wise endurance, and thereupon identi®ed as the proponent of this de®nition which, al-

21 1956: xlv±xlvii passim.

Cross-Examining Socrates120

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

though it never entered into his head, is said to follow from ``his'' account.22 Returning to the discussion, it has apparently come to light that

courage is not endurance simpliciter but wise endurance. No student of the early dialogues will be surprised by the next question: Wise about what (ei� v tiÂ , 192e1)? About all things ± great and small?23 Surely not. An investor is not courageous because he wisely en- dures in investing his money. Nor is a doctor courageous because he wisely endures in withholding food and water from patients whose condition they would aggravate. Curiously, Socrates does not explain (and Laches does not ask) why these wise endurers are not courageous. But the foregoing argument provides a clue. Since Socrates and Laches have agreed that courage is noble and since the end for the sake of which the wise investor invests ± ®nancial gain ± is not noble, neither is the endurance he displays in pursu- ing it. Hence, however wise his endurance, it is not courageous. But this is not true of the doctor whose end ± the well-being of his patient ± is noble. So why is his wise endurance not courageous? Here the foregoing argument is of no help. Lacking de®nitive tex- tual evidence, we must settle for an answer that is compatible with the text. I suggest the following. Since the obstacles which the doctor

must endure ± the complaints of his disgruntled patients ± are comparatively trivial, they invest his wise endurance with a corre- sponding triviality which prohibits his behavior from being de- scribed as courageous. For an action to be courageous, it is not su½cient that it manifest wise endurance. Two further conditions must be satis®ed: ®rst, the end at which it aims must be noble; second, in pursuing that end, the agent must incur and endure

22 Penner, 1992a: 16±18, acknowledges that Socrates ± ``deliberately'' and even ``wickedly'' ± induces Laches and Nicias to endorse de®cient views of courage. He does the same thing to Euthyphro. If we complain that this is unfair, Penner disagrees: ``Whatever the strict rights and wrongs, fairnesses or unfairnesses, of Socrates' treatment of Euthy- phro,'' it is ``justi®ed'' because he has not ``thought out'' his position (18, n. 30). Presum- ably, he would say the same thing about Socrates' treatment of Laches and Nicias.

23 Commentators di¨er about what Socrates means by ``great'' and ``small.'' Santas, 1971: 192±93, thinks he is making a distinction between being wise in the sense of knowing one's alternatives and their likely consequences, and being wise in the sense of knowing the comparative worth of those consequences. Vlastos, 1994: 110±12, thinks he is making a distinction between moral knowledge and technical expertise. Stokes, 1986: 82±83, dis- agrees with both and argues ± rightly, in my opinion ± that he is making a distinction between the magnitudes of the respective endurances involved.

Laches and Nicias 121

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

some signi®cant risk(s). Although Socrates' counterexamples do not show that Laches' de®nition of courage as wise endurance is false, they do show that it is defectively incomplete. But what is incomplete about it? Here we are in for a surprise.

Since Socrates has argued that courage is neither endurance sim- pliciter nor wise endurance simpliciter, we expect him to pursue his original question: Wise about what? and thereby try to remedy the vagueness of Laches' de®nition.24 But he does not. Instead, he produces a counterexample designed to show that courage does not require wisdom at all. Laches is asked to imagine two soldiers on opposing armies who endure in battle: the ®rst, knowing on the basis of a wise calculation (jroniÂ mwv logizoÂ menon) that his army is superior, that it outnumbers the enemy's, that it is advantageously positioned, and that reinforcements are on the way; the second, knowing that his army is inferior, that it is outnumbered by the enemy's, that it is precariously positioned, and that no reinforce- ments will arrive. Asked which soldier is more courageous, Laches unhesitatingly opts for the latter ± in spite of the fact that his en- durance is more foolish (193b1). More counterexamples follow. What about cavalry-riders, ar-

chers, and divers who incur and endure grave risks without pos- sessing knowledge and/or the requisite skills? Are they also more courageous than their more knowledgeable and skilled counter- parts? Laches thinks so ± in spite of the fact that they are also more foolish (193c8).25 Socrates thereupon concludes that this claim, taken in conjunction with Laches' previous admission that foolish endurance is harmful and bad, is inconsistent with his original as- sertion that courage is noble and entails that courage is not wise, but foolish endurance (193d6±7). Laches reluctantly agrees.

24 Thus Devereux, 1977: 135±36, declares that if Laches had been able to specify the kind of wisdom that is involved in courage, his de®nition would presumably have been accepted by Socrates who all along seems to imply that some important, albeit highly elusive, kind of wisdom is a necessary (and perhaps also a su½cient) condition of courage.

25 At Pr. 349e8±350c5 Socrates employs the same counterexamples to prove the opposite conclusion: those who do such things are not more courageous. Indeed, they are not courageous at all; they are mad (maiÂ nomenoi ). Devereux, 1977: 136±37, thinks this shows that Plato has come to recognize that someone who lacks knowledge and/or the requisite skills can be more courageous than someone who possesses them and that he employs the character ``Socrates'' to criticize the historical Socrates who denied this. So, too, Vlastos, 1994: 117, according to whom Socrates' refutation in the Laches of the very de®nition of courage that he himself had endorsed in the Protagoras suggests that the former post-dates the latter and represents an advance in Plato's understanding of courage.

Cross-Examining Socrates122

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Has Socrates refuted the de®nition of courage as wise endur- ance? He has not. It is true that one cannot consistently a½rm:

(i) Courage is wise endurance,

and:

(ii) The soldier who foolishly endures in the face of overwhelming odds is more courageous than his more fortunate opponent;

or:

(i) Courage is wise endurance,

and:

(iia) Horsemen, archers, divers, etc., who foolishly endure without knowledge and/or the requisite skills are more courageous than their more knowledgeable and skilled counterparts.

It is also true that Laches has a½rmed both pairs of inconsistent conjunctions. Having done so, he is informed that he must with- draw (i). Which he does. But needlessly. ``Courage is wise endur- ance'' is false only if it is possible for a person to be both coura- geous and foolish. That is, (i) is false only if (ii) and (iia) are true. But the truth of each can be disputed: ®rst, by denying that those who endure in the face of overwhelming odds without knowledge and/or the requisite skills are more courageous than those who do not; second, by denying that they are more foolish ± or, for that matter ± that they are foolish at all. Although Laches avails him- self of neither strategy, he should have opted for the second. ``A does not know X'' entails ``A is ignorant of X,'' but it does not entail ``A is foolish.'' No one can know everything, and it is possi- ble to be ignorant of something without being foolish. If I do not know that Lincoln is the capital of Nebraska, I am ignorant of that fact. But I am not necessarily foolish. Whether one is or is not foolish depends on considerations other than what one knows and does not know. A life-long resident of the Arctic Circle would not be judged foolish for failing to know that Lincoln is the capital of Nebraska, but a life-long resident of Lincoln presumably would. Laches could also have invoked the distinction between not

knowing that and not knowing how. Unlike the former, which signals ignorance, the latter signals inability. We do not judge people ignorant because they do not know how to compose sonatas or

Laches and Nicias 123

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

repair air-conditioners. Unlike not knowing that and being ignorant, which are symmetrical and inter-entailing, not knowing how and being unskilled are neither. ``A is skilled at X-ing'' entails ``A knows how to X,'' but ``A knows how to do X'' does not entail ``A is skilled at X-ing.'' Not every case of knowing how is a skill. We do not speak of diners skilled at wielding forks or of pedestrians skilled at walking upright. These abilities are too fundamental to be called skills. To lack them is not to be unskilled, but to be handicapped. Nor does lacking certain skills necessarily render one foolish. If, lacking musical training and incapable of reading a score, I apply for the post of conductor of a major orchestra, I am a fool. How- ever, if, being a poor swimmer or no swimmer at all, I dive into the raging surf to save my drowning daughter, I am not ± even if it costs me my life. On the contrary, I perish performing an action which my obituary might describe as heroic. Since it is possible to be both ignorant and unskilled without being foolish, Socrates' counterexamples do not demonstrate what they are intended to demonstrate. In fact, the counterexamples are importantly di¨erent. In the

®rst, the soldier who is judged more courageous does not lack knowledge; he has exactly the same knowledge as the soldier who is judged less courageous. They di¨er: not because one has knowl- edge which the other lacks, but because the ®rst has knowledge which is reassuring and con®dence-enhancing whereas the second has knowledge which is discouraging and con®dence-diminishing. So their respective courage (or lack of it) cannot be ascribed (or withheld) on the basis of their knowledge (or lack of it); it must be ascribed on the very di¨erent basis of how they are prepared to act and what they are prepared to risk and to endure in light of the knowledge they have ± which is the same in both cases. Since this is so, it is hard to understand why Socrates thinks the endur- ance of the one is more foolish than that of the other or, for that matter, why he thinks it is foolish at all. Unless it is because he does not retreat. But surely Socrates is not suggesting that. It is precisely because he does not retreat but stays at his post in the full knowledge of his precarious circumstances that he is judged more courageous. The epistemic situation is very di¨erent in the second counter-

example. Here, too, unskilled (or less skilled) cavalry-riders, arc- hers, divers, and so on are judged more courageous, albeit also

Cross-Examining Socrates124

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

more foolish, than their more knowledgeable and skilled counter- parts. However, unlike the alleged foolishness of the badly posi- tioned and outnumbered soldier, their alleged foolishness is not traceable to their willingness to act or to endure in the face of con®dence-diminishing knowledge, but rather to their willingness to act and to endure without knowledge and/or the requisite skills. Again it is hard to understand why Socrates thinks their endur- ance is more foolish or, for that matter, why he thinks it is foolish at all. Unless it is because they do not refrain from incurring and enduring these risks. But surely he is not suggesting that either. It is precisely because they do incur and endure them that they are judged more courageous. In short, it is beginning to look like the proposition ``All who endure in the face of con®dence-diminishing knowledge and without the requisite skills are foolish'' is false. There is a more fundamental question. Why does Socrates ask

Laches which soldier is more courageous in the ®rst place? Clearly, because he wants him to say that it is the soldier answering to the description of (ii) or (iia). That is, wishing to refute Laches' de®ni- tion of courage as wise endurance, Socrates feeds him a premise which he ± Socrates ± thinks is false, but which he suspects Laches thinks is true and to which he will assent. And he does. It might even be said that he sincerely assents. But what is such assent worth?26 This raises awkward questions about the whole exchange. If courage is neither endurance simpliciter nor wise endurance sim- pliciter, and if there is no necessary connection between it and wisdom (knowledge, possessing the requisite skills, etc.), then the foregoing search for the speci®c kind of wisdom possessed by the courageous man was a pointless digression, as Socrates implicitly acknowledges a few minutes later when he abruptly drops the whole subject. All of which makes one wonder why he introduced the concept of wisdom in the ®rst place and ascribed such impor- tance to the seemingly crucial, but in fact irrelevant, question: Wise about what? Having jettisoned the apparently super¯uous component of wis-

dom, Socrates suggests that he and Laches allow themselves to be persuaded by the foregoing argument ``to a certain extent'' (toÂ ge

26 This passage provides further textual evidence against Vlastos's contention that ``when Socrates is searching for the right way to live, in circumstances in which it is reasonable for him to think of the search as obedience to divine command, his argument cannot involve wilful untruth'' (1991: 134).

Laches and Nicias 125

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

tosouÄ ton, 193e8), namely, to the extent that courage enjoins en- durance. Some commentators think this is a ``hint'' that Laches is ``on the right track,''27 i.e., that Socrates thinks endurance is an essential component of courage which should be retained as part of its de®nition. This is disputed by others28 and, above all, by Stokes,29 according to whom Socrates' remark is merely ``a piece of urbanity, suggesting the continuance of the discussion . . . in terms which the soldierly Laches can be expected to appreciate and understand.'' At this point, Laches confesses that a certain contentiousness

(tiÂ v . . . jilonikiÂ a, 194a7±8) has taken hold of him. The malady from which he is su¨ering ± highly contagious, as things turn out ± should not be confounded with the ``hatred of discussion'' to which he alluded earlier. Although symptomatically similar, they are distinct. Unlike the latter, which is psychological, contracted through exposure to moral hypocrites, and medically analogous to an allergic reaction, the former is logical, symptomatic of concep- tual impotence, and medically analogous to a constitutional de®- ciency. Laches' contentiousness manifests itself as a con¯ict within himself. Convinced that he knows what courage is, he is vexed be- cause he cannot say what it is ± a rueful allusion to Socrates' ear- lier contention that anyone who knows what virtue (or a virtue) is can say what it is (190c6) ± a contention Laches had con®dently endorsed (190c7).30 Grote31 thinks Laches' vexation is symptomatic of the ``intellec-

tual de®ciency which Sokrates seeks to render conspicuous to the consciousness . . . and which he impugns as the false persuasion of knowledge.'' But this contention is not supported by the text. No ``intellectual de®ciency'' has been ``rendered conspicuous'' to anyone's consciousness ± certainly not to Laches' consciousness. Laches is vexed: not because he has been convicted of ``the false persuasion of knowledge,'' but because he is having trouble put- ting his thoughts into words. The correct de®nition of courage is on the tip of his tongue. In short, being examined by the sublime

27 Devereux, 1977: 136, n. 20. See also Grote, 1867, i: 478±80; Shorey, 1903: 12±13; O'Brien, 1958: 124, 1967: 113; and, by implication, Santas, 1971: 195.

28 Among them, Pohlenz, 1913: 28; and Irwin, 1977: 302, n. 59. 29 1986: 89. 30 See Stokes, 1986: 66, Socrates is unwaveringly committed to the principle that ``If one

knows what something is, one can explain what it is.'' 31 1867, i: 477.

Cross-Examining Socrates126

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

incarnation of the Dorian mode called ``Socrates'' is not proving to be the exhilarating experience he thought it would be. Neither noticeably surprised by nor noticeably sympathetic to

Laches' discom®ture ± aporia is, after all, the intended result of all this ± Socrates adds insult to injury by suggesting that he continue the discussion with Nicias who might be more adept at giving his thoughts the stability of speech (194c4±6). The palpably delighted Nicias jumps at the chance to outshine his colleague ± now turned rival. Ignoring the notion of endurance altogether, he suggests a way out of the impasse: in his opinion, Socrates' and Laches' di½- culties are traceable to the fact that they are overlooking a remark which Socrates has often made in Nicias' presence to the e¨ect that a man is good in the things about which he is wise and bad in the things about which he is foolish ± a remark which implies that if the courageous man is good, he must also be wise (194d1±2). Socrates acknowledges that what Nicias says is true. But his re-

ply is ambiguous. Does he mean that the thesis which Nicias has often heard him assert is true? Or does he merely mean that it is true that Nicias has often heard him assert it? Nicias takes him to mean the former. And with good reason. For Nicias' next de®ni- tion of courage as knowledge of what is to be feared and not feared (194e11±195a1) is identical with Socrates' de®nition at Pr. 360d1±5, except that Socrates' substitutes ``wisdom'' (sojiÂ a) for ``knowl- edge'' (e� pisthÂ mh).32 Socrates neither agrees nor disagrees with Nicias' de®nition; he simply alerts Laches to the fact that Nicias seems to be saying that courage is a kind of wisdom (sojiÂ an tinaÂ , 194d9).33 When Laches vacantly wonders what kind, Socrates urges him to ask Nicias ± which he does, albeit with icy brevity: ``All right'' ( E� gwge, 194e2), that is, Nicias may consider himself asked. However, before Nicias can reply, Socrates intervenes and asks the same question more pointedly. Since the kind of knowledge (or wisdom) Nicias has in mind does not consist in being skilled at playing the ¯ute or the lyre, what does it consist in? What is this knowledge knowledge of?

32 As Vlastos, 1981b: 230, n. 24, rightly points out, ``The allegation one meets frequently in the scholarly literature that no acceptable de®nition of a virtue is ever reached in the Socratic dialogues is false.''

33 As Stokes, 1986: 56, rightly observes, Nicias' de®nition implies that he has changed his mind and no longer thinks that the acquisition of the art of ®ghting in heavy armor is su½cient to induce courage, although he may still think it is su½cient to induce boldness.

Laches and Nicias 127

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Having just failed to answer this question himself and growing more competitive by the minute, Laches applauds this line of ques- tioning and demands that Nicias answer. Gallingly un¯ustered, Nicias replies that courage is knowledge of what is to be feared and not feared, whether in war or in anything else (194e11±195a1) ± a de®nition which Laches judges strange (a� topa) on the ground that wisdom is distinct from courage. But, as Socrates frustratingly points out, that is precisely what Nicias is denying. Laches inter- prets this remark as a call to arms and accuses Nicias of talking nonsense ± an insult which Nicias dismisses as a mere tu quoque traceable to the fact that Laches had just been accused of the same thing. Irked that Socrates is siding so openly with Nicias, who had previously ¯aunted his familiarity with Socrates and his ways, Laches sets out to prove that it is Nicias who is talking non- sense. A piece of amateur elenchus follows. How can Nicias de®ne courage as knowledge of what is to be

feared and not feared? Doctors, farmers, and other experts know what is to be feared and not feared in their respective technai; but that does not make them courageous. Socrates now taunts Nicias by observing that there seems to be something in what Laches is saying. Perhaps su¨ering from a touch of philonikia himself, Nicias abrasively agrees that there may be something in it, but nothing true. Doctors are experts about health and disease, but that is the extent of their expertise. Qua doctors they do not know in particu- lar cases whether health is preferable to disease or whether life is preferable to death.34 A speedy death is sometimes preferable to a lingering illness, and about such matters a doctor's judgment is no better than anyone else's. Competent judgments of this kind re- quire a unique kind of wisdom, namely, the kind possessed by the courageous man. Courage is not just knowledge (or wisdom), but knowledge of a unique kind: knowledge of what is truly to be feared and not feared (195d7±9). Laches trivializes Nicias' distinction by misinterpreting it: he is

obviously talking about prophets who can foresee the future and claiming that they alone are courageous, since they alone know in particular cases when life is preferable to death. Although Laches thinks this is patent nonsense, Socrates disappoints him by saying

34 ``[O]ne of the best moves of any Socratic interlocutor in the whole of the early dia- logues'' (Penner, 1992a: 20).

Cross-Examining Socrates128

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

that there seems to be something in what Nicias is saying. He adds:

Let us ®nd out more clearly what he means (tiÂ pote noeiÄ ), and if he is really saying something, we will agree with him (196c1±4).35

Resuming the role of questioner, Socrates wonders whether Nicias still thinks that courage is knowledge of what is to be feared and not feared, and that such knowledge is a necessary ± and perhaps a su½cient ± condition for being courageous. Assured that he does, Socrates confronts him with a dilemma: in that case, he must either withhold courage from wild animals or credit them with a kind of wisdom possessed by few human beings. Laches explodes with glee and can hardly wait for Nicias to impale himself on one of the horns of this dilemma. But Nicias has a card up his sleeve. He opts for the ®rst horn, but immediately distinguishes courage from boldness, thereby divesting the claim of its absurdity and de- priving Laches of his triumph.36 Animals are not courageous and wise, they are bold and foolish (a� jobon kaiÁ mwÄ ron, 197a8).37 The dialectically-outclassed Laches accuses Nicias of splitting

hairs. But Socrates again takes up Nicias' cause, tracing his deft- ness at drawing distinctions to his association with Prodicus38 and insisting that his view warrants serious consideration. Dripping

35 Although it would be an exaggeration to say that this passage is an anticipation of the concept of Socratic midwifery set forth for the ®rst time at Th. 148e6±151d6, some com- mentators think it is a good example of the sort of thing that makes midwifery an apt image for what Socrates claims to do for his ``intellectually pregnant'' interlocutors in the early dialogues ± i.e., to assist them in giving birth to the ideas that are in their minds. See Burnyeat, 1977b.

36 Another example of Nicias' ``astuteness,'' according to Penner, 1992a: 21. In fact, this distinction was already implicit in Nicias' earlier contention that the art of ®ghting in heavy armor induces boldness and courage (182c6).

37 Although at 184b3±c4 Laches objects to Nicias' distinction between courage and bold- ness and accuses him of ¯outing ordinary language, his criticism of Nicias' claim that the art of ®ghting in heavy armor induces courage is based on (what amounts to) the same distinction.

38 Prodicus of Ceos was famous for his subtle distinctions ± a technique which Socrates ad- mires in spite of its possible abuses (La. 197d1±5, Ch. 163d1±4) and which, as Prodicus' ``disciple'' (Pr. 341a4), he periodically employs with an ingenuity bordering on the per- verse (see, e.g., Pr. 340e5±347a3). The J. L. Austin of classical antiquity, Prodicus' lin- guistic sensitivity bespeaks an attentiveness to and respect for ``our common stock of words [which] embodies all the distinctions men have found worth drawing, and the connexions they have found worth making, in the lifetimes of many generations. . . , [and which] are likely to be more numerous, more sound . . . and more subtle, at least in all ordinary and reasonably practical matters, than any that you and I are likely to think up in our arm-chairs of an afternoon'' (Austin, 1961: 130).

Laches and Nicias 129

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

with acrimony, Laches tries to extricate himself from the discus- sion, but is prevailed upon to stay. Assured that Nicias still thinks courage is a ``part'' of virtue, Socrates argues that his de®nition of courage as knowledge of what is to be feared and not feared is not satisfactory. His argument ± 197e10±199e12 ± is as follows:

(i) Courage is a ``part'' of virtue.39 (ii) What is dreaded is what causes fear and what is not dreaded

is what may be safely dared. (iii) Fear is of expected evils and hence of what is future. (iv) Things are dreaded because they are future evils. (v) Hence courage is knowledge of future evils and future goods. (vi) But knowledge of what is to be feared and not feared is not

just knowledge of the future, but also of the past and present and, therefore, knowledge of all evils and goods.

(vii) Hence since courage is knowledge, it too is knowledge of all evils and goods.

(viii) But then it is not a ``part'' of virtue, but the whole of it. (ix) Hence courage cannot be de®ned as knowledge of what is to

be feared and not feared.

Nicias reluctantly concurs.40 Some commentators41 are very critical of this argument while

others42 ®nd it completely convincing. The crucial step is (vi). Here Socrates argues that, unlike things that induce fear, which are necessarily future, knowledge of things that induce fear is not divisi- ble into temporal categories. Knowledge of fearful things past, present, and future is not three di¨erent kinds of knowledge. Knowledge is universal, i.e., it is knowledge of everything that is to be feared (and hence evil) and not feared (and hence good) ± whether past, present, or future. But the conclusion does not follow. ``Whatever need not be

39 A problematic claim which, according to Santas, 1971: 202, Socrates abandons in the Protagoras in favor of the Doctrine of the Unity of the Virtues. At Eu. 11e7±12e8 piety and justice are also said to be ``parts'' of virtue.

40 Since the de®nition under scrutiny is not Nicias' but Socrates', it seems that the fore- going criticism is self-criticism. Thus Vlastos, 1991: 113, n. 28 declares that in the Laches Plato portrays Socrates attacking a Socratic thesis which is defended by Nicias who is left ``holding the bag.''

41 See, e.g., Santas, 1971: 203±5; Penner, 1973b: 60±62; C. C. W. Taylor, 1976: 107; and Irwin, 1977: 88±89, 302, n. 62.

42 See, e.g., Vlastos, 1981b: 266±69, and 1994: 117±24; and Allen, 1996: 58.

Cross-Examining Socrates130

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

feared (and hence may be safely dared) is a future good'' does not entail ``Whatever is a future good need not be feared (and hence may be safely dared).'' Socrates is guilty of illicit conversion. Many future goods, for example, health, involve no risks whatever and, therefore, do not fall into the category of actions that may be safely dared. Accordingly, a person who has knowledge of what is to be feared and not feared need not have knowledge of all goods and evils, but only of those which are relevant to his legitimate fears. Neither does it follow that insofar as Nicias' de®nition en- compasses only future goods and evils and ignores past and pres- ent ones, it only accounts for a third of courage. Nor is it the case that Nicias' de®nition turns out to be the de®nition of the whole of virtue rather than merely of a ``part'' of it.43 In short, Nicias' de®nition has not been refuted. But the discussion proceeds as if it had been. Nicias' apparent defeat sets his reputation back light years in

the eyes of Laches, miraculously cured of philonikia and back on his feet. The gratifying spectacle of his humbled colleague elicits a torrent of withering irony reminiscent of Socrates in the Euthyphro: ``What? Dropping out of the discussion just when you were about to instruct me about the nature of courage so I could live a better life?'' Having previously advertised himself as a serious inquirer whose voluntary submission to Socratic cross-examination is a tes- timony to his moral earnestness and a source of endless bene®ts, Nicias suddenly shows a di¨erent face. Smarting from his appar- ent defeat and licking his dialectical wounds, he o½ciously chirps that he has said all that anyone could reasonably be expected to say on the di½cult subject of courage. He solemnly adds that he has every intention of remedying whatever de®ciencies he may have exhibited by consulting with Damon ``and others'' ± after which he will return to enlighten Socrates and Laches (200b2±c1). After a terminal exchange of insults by Laches and Nicias, the dialogue concludes with a bit of post-elenctic chit-chat. Laches advises Lysimachus and Melesias to dismiss himself and Nicias, and to secure the educational services of Socrates. Socrates re- a½rms his advisory incompetence and unhelpfully urges everyone

43 In fact, as Kahn, 1996: 167, rightly points out, even if it did, this should be counted a success rather than a failure, since the original project was to de®ne virtue, i.e., the ``whole'' of virtue ± a project which had been shelved on the ground that it would be easier to de®ne one of its ``parts.''

Laches and Nicias 131

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

to seek out the best teachers they can ®nd. Lysimachus thinks this is a splendid idea and suggests that they all reconvene the next day. I said earlier that, when judged solely by their speeches, Nicias

and Laches are admirable embodiments of the ideal Socratic in- terlocutor, but, when judged by their behavior, both fall far short. I will conclude this chapter with a few remarks about how and why. Initially, Laches' de®ciencies seem more glaring and more griev-

ous. Highly competitive and dripping with superiority, he gets o¨ to a bad start. Impatient with logical details and quickly out of his depth, he has soon had his ®ll of argumentation. Unable to de®ne courage, he becomes touchy and inclined to over-statement and over-reaction. When Nicias succeeds him as interlocutor and experiences similar di½culties, he rejoices and eagerly anticipates the demise of his colleague's equally doomed and slowly expiring de®nition of courage over which he circles like a vulture. But we should not be too hard on Laches for his poor showing.

Socrates is deeply implicated. Having dubiously disposed of Laches' attempts to de®ne courage, he deliberately provokes him by trans- forming him into an inverted instantiation of his own metaphor ± his words are not consistent with his deeds ± and by repeatedly and openly siding with Nicias. The escalating rivalry between the two is deliberately exacerbated by an uncommonly divisive Soc- rates who magni®es their slightest disagreement and exploits their increasingly acrimonious one-upmanship to drive a wedge between them while simultaneously issuing periodic reminders that they are not individuals in competition with each other, but a company of truth-seekers embarked on a ``joint'' inquiry. Initially, Nicias makes a better impression. More familiar with

Socrates and his method, he seems less intimidated by philosoph- ical discussion and more willing to pro®t from it. Although Penner may overestimate his logical acumen, he is certainly more astute than many of Socrates' interlocutors; and he is in®nitely more astute than Laches. Eager to assure everyone that he is far from perfect (though not nearly so far as some of his colleagues), he revels in his self-appointed role of media-analyst providing on-the- scene, pre-debate coverage buttressed with evangelical-like per- sonal ``testimonies'' about the Socratic elenchus and its power to produce ``changed lives.'' An experienced participant and grateful

Cross-Examining Socrates132

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

bene®ciary, he has taken time out of his busy schedule to serve as educational consultant and to instruct the dialectically uninitiated about the joys of refutation. However, in spite of his shortcomings, Laches has qualities that

Nicias lacks. Although capable of monumental pettiness and un- commonly eager to see others su¨er as he has su¨ered, he is nei- ther so self-absorbed nor so self-indulgent. He also has a certain psychological resilience. Every time we think we have had about enough of him, he reengages us by bouncing back and resolving to follow the discussion to its conclusion. If his later contributions are negligible and largely negative, he at least remains an active par- ticipant in the discussion. That cannot be said of Nicias. Initially more impressive, he

does not wear nearly so well and, in the end, proves to be little more than a military celebrity doing a Socratic commercial ± someone who would rather chatter about the joys of refutation than experience them. By the end of the dialogue, we have seen through him and realized that his super®cially impressive speech about the bene®ts of submitting to Socratic cross-examination was a bit unctuous and mostly talk. Unlike Laches, who had no idea of what he was getting into but weathered the ordeal surprisingly well, Nicias claims to be an insider, but inadvertently betrays that he is actually impervious to Socratic cross-examination and ineli- gible for its alleged bene®ts. Like his military uniform, his Socratic rhetoric is something to be slipped into whenever he wants to cut an impressive ®gure but discarded as soon as the mood has passed. Ostensibly a living advertisement for Socrates, he is actually a fraud. Pit him in debate with Socrates and he can do little more than quote him and produce clever, ad hoc distinctions. Upon being refuted, he pouts, extricates himself from the discussion, and books an appointment with Damon ± his real but hitherto undisclosed authority who will remedy his de®cient intellectual grasp and pre- sumably make further Socratic conversations super¯uous. What shall we say about this dialogue? Do Socrates' arguments

persuade anyone of anything? Does the aporia to which they lead infuse anyone with self-knowledge? If so, it has eluded me.44 The

44 Even the usually instructive Guthrie makes surprisingly little of the Laches. Initially com- mending it as ``a model of Socratic dialectic,'' he is ®nally reduced to praising its ``con- siderable value as entertaining literature'' (1975: 131).

Laches and Nicias 133

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Laches reveals that although the Socratic elenchus can expose igno- rance, de®ned in Socratic terms, it can neither provide knowledge nor motivate the ignorant, again de®ned in Socratic terms, to search for it. In the end, everyone remains exactly as they were. All they can do is marvel at the time they have wasted and, with their appetites inexplicably whetted for more of the same, resolve to reconvene the next day ± albeit with no speci®c agenda to pur- sue, no experts with whom to pursue it, and nothing better to look forward to than the overnight ``enlightenment'' that Nicias will have absorbed from Damon. Why bother?

Cross-Examining Socrates134

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

c h a p t e r 7

Charmides and Critias

The Charmides is one of three early dialogues in which relatives of Plato appear. The historical Charmides was Plato's maternal uncle, and the historical Critias1 was Charmides' cousin and guardian. Later members of the Thirty, both perished ®ghting against the Athenian democracy.2 The dramatic date of the dia- logue, narrated by Socrates, is 432. Having just returned from the Battle of Potidaea in which he

had displayed great courage and saved the life of Alcibiades (Sym. 220d5±221b1), Socrates repairs to the palaestra of Taureas to ®nd out how philosophy has fared during his absence and whether anyone has distinguished himself in wisdom or beauty or both. His attention is directed to Charmides, a beautiful youth3 full of promise who has just arrived accompanied by his court of ¯ushed and panting lovers. Assured by Critias that he is equally beautiful in soul and already a poet and a philosopher (jiloÂ sojov, Ch. 154e8±155a1), the momentarily smitten Socrates pulls himself together and sets out to examine him. Like many Socratic interlocutors, Charmides is lured into the

discussion on false pretenses. Informed by Critias that there is a doctor on the premises who can cure his morning headaches, Charmides dutifully appears before ``Dr. Socrates'' who author- itatively explains that he is in possession of a leaf the application of which, if accompanied by an incantation, cures a person of his

135

1 According to Xenophon, the historical Critias was an ambitious and violent man who grievously harmed the city (Mem. 1.2.12). He did not pro®t from his association with Soc- rates because people learn nothing from a teacher with whom they are out of sympathy (1.2.39).

2 They are anonymously alluded to in the Seventh Letter (324d1±2) as ``relatives and acquain- tances'' of Plato's.

3 Although most commentators think Charmides is a teenager, MeÂron, 1979: 99, n. 74, points out that both the terminology Socrates uses to describe him ± meiraÂ kion (154b5), neaniÂ skov (154d1), and neaniÂ an (155a4) ± and the examples he uses to refute him ± copying letters (159c3±4), running and jumping (159c13±d2) ± suggest that he is younger.

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

bodily ailments. However, there are two strings attached: the treatment cannot cure one part of the body without curing the whole, and it cannot cure the body without ®rst curing the soul. It a¨ects this double cure by causing temperance (swjrosuÂ nh)4 to be implanted (e� ggiÂ gnesqai ) in the soul which is the ultimate source both of bodily health and bodily illness (155e5±157b1). So if Char- mides wants to be cured of his headaches, he must submit his soul for treatment. The charade ends abruptly when Charmides reveals that he knows perfectly well that the ``doctor'' is Socrates. Critias thereupon declares that Charmides is not only the most

beautiful youth he knows but also the most temperate ± a remark which confronts the blushing Charmides with an awkward di- lemma: he must either deny that he is temperate, thereby contra- dicting Critias and ``many others,'' or agree that he is, thereby praising himself. Socrates comes to the rescue by proposing that they jointly investigate whether he is temperate. Having secured Charmides' agreement,5 he explains what seems to him the best method. If temperance is present in (paÂ restin)6 Charmides (158e7), he must have some opinion about what it is (o� ti e� stiÂ n) and what kind of thing it is (o� poiÄ oÂ n ti, 159a3); for temperance neces- sarily (a� naÂ gkh) gives some intimation of its nature which enables its possessor to form an idea of it (159a1±3). And if Charmides has an idea of what it is, he must be able to say what it is (159a6±7). So: what is temperance?7 Charmides replies that, in his opinion, temperance is doing

4 Temperance was, together with courage (a� ndreiÂ a), justice (dikaiosuÂ nh), and wisdom (sojiÂ a), one of the four so-called ``cardinal'' virtues. A term with no exact English equiv- alent, swjrosuÂ nh is sometimes translated as ``moderation'' or ``self-control.'' Burnyeat, 1971: 216, thinks it is actually untranslatable ``because the phenomena it grouped together for Greek culture do not form a whole to our outlook.'' So, too, Hazebroucq, 1997: 10±12; and Cooper, 1997: 639.

5 As always, participation in the Socratic elenchus is voluntary. See La. 187c1±d5, 188a6± c3, and 189a1±3.

6 Plato often speaks of properties as being ``in'' persons rather than of persons having those properties. Unlike the Euthyphro, in which Socrates treats piety almost exclusively as a property of actions, in the Charmides he treats temperance as a property both of actions and persons and implies that the latter is prior to the former, i.e., that a person can per- form temperate actions if and only if he is temperate. He implies the same thing about courage in the Laches.

7 Although the Charmides concludes with a long discussion of self-knowledge, I cannot accept Saunders's contention, 1987: 165, that what is actually under investigation in this dialogue is not temperance but knowledge ± in particular, ``the strange and problematic concept of `knowledge of knowledge'.'' Even after the discussion takes this quasi-epistemological turn, the fundamental question is whether ``knowledge of knowledge'' is a satisfactory de®nition of temperance.

Cross-Examining Socrates136

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

everything orderly and quietly, for example, walking in the streets, talking, and so on ± in sum (sullhÂ bdhn), that it is a kind of quiet- ness (h� sucioÂ thv tiv, 159b2±6). A. E. Taylor somewhat patron- izingly remarks:

As is natural in a mere lad, Charmides ®xes ®rst of all on an exterior characteristic . . . in the form which would be most familiar to a boy ± the form of decent and modest bearing towards one's elders and ``good behaviour'' generally.8

However, in his haste to disparage Charmides' super®cial under- standing of temperance, Taylor overlooks the fact that he avoids the error ± common in the early dialogues ± of de®ning a moral term by means of examples, thereby confusing de®nition with enumeration of instances. Although Charmides gives several ex- amples of temperate actions, he immediately brings them under a single de®nition9 without having to be coached by Socrates ± a conceptual feat which lends some (but not much) credence to Critias' claim that he is ``already a philosopher.'' Instead of complimenting Charmides for avoiding this blunder,

Socrates ®nds fault with his de®nition. Although some might say that the quiet are temperate, Socrates thinks they are wrong. His argument ± 159b8±160d4 ± is as follows:

(i) Temperance is admirable (kalwÄ n). (ii) It is more admirable to write quickly than quietly, to read

quickly than slowly, to play the lyre quickly and brilliantly than quietly and slowly, etc.; and the same is true of bodily actions generally.

(iii) Agility and quickness are admirable, but slowness and quiet- ness are shameful (ai� scrouÄ ).

(iv) Therefore, since temperance is admirable, it follows that quickness is more temperate than quietness ± at least with re- spect to bodily actions.

The same is true of all ``mental'' actions:

(i) Facility in learning is more admirable than di½culty. (ii) Facility consists in learning quickly whereas di½culty consists

in learning quietly and slowly. (iii) It is better to teach, and to understand, and to remember

8 1929: 50. So, too, Tuckey, 1951: 19. Santas, 1973: 107, is much less condescending. 9 See ch. 6, n. 15.

Charmides and Critias 137

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

quickly and energetically than quietly and slowly; and the same is true of ``mental'' actions generally.

(iv) Therefore, in everything that concerns body and mind, tem- perance is not quietness ± at least not by this argument (e� k ge touÂ tou touÄ loÂ gou).

Although Charmides instantly concurs, both arguments are faulty. For one thing, both depend on bizarre contrasts. Asked whether it is better to write quickly or quietly, Charmides opts for the former. This is a curious response. The opposite of ``quickly'' is not ``quietly'' but ``slowly,'' and the opposite of ``quietly'' is not ``quickly'' but ``noisily.'' Asked whether it is better to read quickly or slowly, he again opts for the former. His answer is equally ill- advised. He should have replied that the question cannot be an- swered in general; it all depends on the book one is reading. Sein und Zeit should be perused slowly whereas Ken Bails Out may be dev- oured at a faster pace. Nor are there any general musical reasons for thinking that it is better to play the lyre quickly and brilliantly than quietly and slowly. It all depends on the piece one is playing. A vivace should be played quickly and brilliantly whereas an adagio should be played quietly and slowly. But even if Socrates' ques- tions had been more carefully formulated and his examples more judiciously chosen, there are far too few of them. Every epagoÅgeÅ is necessarily incomplete, but this tiny Socratic sample does not even begin to warrant the universal conclusion that, in ``everything'' that concerns body and mind, quickness is more temperate than quietness. Both arguments are also fallacious. Each concludes that tem-

perance is not quietness (or a kind of quietness) on the ground that temperance is admirable and quick actions are more admirable than quiet ones. More precisely:

(i) If an action is temperate, it is admirable. (ii) Actions a, b, and c (which are quick) are more admirable than

actions d, e, and f (which are quiet). (iii) Therefore actions a, b, and c are more temperate than actions

d, e, and f.

This argument a½rms the consequent.10 According to Aristotle, fallacies ``connected with the consequent'' (paraÁ toÁ e� poÂ menon) are

10 Of course, the argument a½rms the consequent only if it is formulated hypothetically. But even if it is formulated categorically, as Socrates formulates it here, the argument

Cross-Examining Socrates138

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

``due to the idea that the consequent is convertible. For whenever, if A is, B necessarily is, men also fancy that, if B is, A necessarily is'' (S.E. 167b2±4). These fallacies comprise part of the class of fallacies ``due to accident'' (touÄ sumbebhkoÂ tov):

For the consequent is an accident but di¨ers from the accident because the accident can be secured in the case of a single thing by itself, for ex- ample . . . a white thing and a swan, whereas the consequent always exists in more than one thing; for we claim that things which are the same as one and the same thing are the same as one another. (S.E. 168b28±34)

But, as Aristotle rightly points out, this is not always true in the case of accidental properties: from the fact that snow and swans are both white, it does not follow that snow is more swanlike than rain (S.E. 168b34±35). In arguing that since temperance and quickness are both admirable, and since quick actions are more admirable than quiet ones, it follows that they are also more tem- perate, Socrates fallaciously fastens upon (what Aristotle calls) an ``accidental'' (sumbebhkoÂ v) property common to both. Aristotle calls such arguments ``sophistical'' (sojistikwÄ n) because they ``appear to be refutations but are really fallacies'' (S.E. 164a20± 22), they ``do not a¨ect their object but only appear to do so'' (S.E. 165a4±5), and they succeed only because of their victims' inexpe- rience and lack of argumentative skill (S.E. 164b25±28).11 Socrates further complicates matters by confronting Charmides

with a puzzling dilemma: Either, as he himself is inclined to think, there are no (or very few) quiet actions which are more admirable

remains invalid: ``All temperate actions are admirable'' does not entail ``All admirable actions are temperate.'' This is illicit conversion. Interestingly, when Laches de®nes courage as a certain endurance of the soul (karteriÂ a tiv . . . thÄ v yuchÄ v, La. 192b9), Soc- rates rejects his de®nition for precisely this reason: from the fact that courageous actions exhibit endurance, it does not follow that all actions which exhibit endurance are coura- geous (193a3±d9). Of course, Socrates never talks about valid and invalid inferences, much less about patterns of valid and invalid inference.

11 Commentators say strange things about Socrates' argument. Tuckey, 1951: 19, n. 1, con- cedes that it is ``vitiated by a paralogism,'' but thinks that it ``exposes the inadequacy'' of Charmides' de®nition. Santas, 1973: 113, concedes that it ``seems full of logical faults,'' but thinks that it ``comes very close to doing its job.'' Neither explains how a fallacious argument can ``expose'' anything or ``do its job.'' Lamb, 1925: 7, thinks the argument commits the fallacy of undistributed middle and speculates that Charmides' ``failure to protest . . . was brought out in the discussion at the Academy.'' He adds that Plato ``would perhaps excuse himself by saying that when he wrote the Charmides he was more intent on intellectual drama than on logical accuracy.'' For a recent defense of Socrates against the charge of fallacious reasoning, see Hazebroucq, 1997: 156±66.

Charmides and Critias 139

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

than quick ones or there are many quiet and quick actions which are equally admirable; however, even if the latter were true, it would still be incorrect to de®ne temperance as quietness since, as the foregoing argument has demonstrated, quick actions are no less (ou� c h� tton) admirable, and therefore no less temperate, than quiet ones (160d1±3). He is, of course, wrong about this. The foregoing argument did

not demonstrate that at all. What it purported to demonstrate was not that quick actions are no less admirable, and, therefore, no less temperate, than quiet ones, but rather that they are more admira- ble, and, therefore, more temperate, than quiet ones. In addition to arguing fallaciously, Socrates has misrepresented the conclusion of his own argument. For even if there are many quiet and quick actions which are equally admirable, the proposition:

(i) Actions a, b, and c (which are quick) are more admirable, and therefore more temperate, than actions d, e, and f (which are quiet).

is incompatible with the proposition:

(ii) Actions d, e, and f (which are quiet) are no less admirable, and therefore no less temperate, than actions a, b, and c (which are quick).

Since (i) and (ii) are incompatible, no single argument can estab- lish both.12 In fact, Charmides could have undercut both arguments by

lodging a methodological objection. Asked whether temperance is admirable, he should have replied: ``I have no way of knowing that, Socrates; for, as you yourself constantly imply, one cannot know what properties are predicable of a virtue until one knows what that virtue is. And we have not yet discovered what temper-

12 Santas, 1973: 116±17, defends Socrates on the (to my mind) unconvincing ground that he is not trying to show that quietness is contrary to the nature of temperance, but that it is irrelevant to it; ``[l]ooked at this way, Socrates' argument is . . . convincing, though from the point of view of impeccable logic it still remains faulty.'' Irwin, 1995: 36, and Kahn, 1996: 189, think Socrates' criticisms of Charmides' ®rst two de®nitions of temperance are intended to show that temperance cannot be de®ned in terms of external actions without reference to motivation. ChreÂtien, 1987: 12±13, defends Socrates on the very di¨erent ground that, as a matter of fact, temperance is not the same thing as quietness, thereby apparently endorsing Socrates' argument on the ground that its con- clusion is true.

Cross-Examining Socrates140

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

ance is.''13 I conclude that Socrates has not refuted Charmides' de®nition of temperance.14 But the discussion proceeds as if he had. Again exhorted to say

what temperance is, Charmides asserts that insofar as temperance makes men ashamed or modest, it is the same as modesty (ai� dwÂ v, 160e4). Socrates thereupon lodges another objection. His argu- ment ± 160e6±161b2 ± is as follows:

(i) Temperance is admirable. (ii) Temperate men are good. (iii) Whatever is good makes men good. (iv) Temperance is always good. (v) But modesty is not always good; for, as Homer observes, it is

not good for a beggar to be modest. (vi) Insofar as modesty is both good and not good, it is no more

(ou� deÁ n maÄ llon) good than evil. (vii) Therefore, since temperance is always good and always

makes men good, it cannot be the same as modesty.

Although this is one of the lamest arguments in the early dialogues, Charmides is undone by it.15 But again there is an obvious reply. Charmides should have explained that he does not accept things on Homeric authority. He should have added that, in view of the fact that Socrates always objects when his interlocutors appeal to poetic authority, he is surprised to ®nd him doing the same thing.16 Of course, Charmides says no such thing. Instead, he announces

that he has just recalled another de®nition which seems to him more promising, namely, that temperance is minding one's own business (toÁ taÁ e� autouÄ praÂ ttein, 161b6). When Socrates suggests that Critias is the originator of this de®nition, Critias vigorously

13 At G. 448e2±4 Socrates faults Polus for saying that rhetoric is the most admirable (thÄ v kalliÂ sthv) of all the technai without having de®ned rhetoric ± an error which reveals that he is unequipped for dialectic. See Robinson, 1953: 51: ``[T]he impression vaguely given by the early dialogues as a whole is that Socrates thinks there is no truth whatever about X [the concept or term under discussion] that can be known before we know what X is.''

14 As Chambry, 1967: 266, justly observes: ``L'argumentation de Platon ressemble plus d'une fois aÁ celles des sophistes, plus preÂoccupeÂs d'avoir le dessus que d'avoir raison.'' Cf. MeÂron, 1979: 99±100, who characterizes Socrates' arguments as ``les reÂfutations . . . faibles.''

15 With good reason, according to Saunders, 1987: 168: ``One negative case is enough to dispatch the proposed de®nition.''

16 ``[C]omme si HomeÁre eÂtait un oracle infaillable'' (Chambry, 1967: 267).

Charmides and Critias 141

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

denies the allegation.17 Ignoring his protest, Socrates focuses on the de®nition itself and pronounces it ``a kind of riddle'' (ai� niÂ gmati . . . tini, 161c9): surely its originator did not mean that a temperate person should write only his own name, build only his own house, weave only his own clothes, and so on. No city based on that prin- ciple would be well governed, and surely a temperate city would be a well-governed one. So temperance cannot be ``minding one's own business'' in that sense. Only an idiot (h� liqiÂ ou) would say such a thing (162b1±2). But then what did the originator of the de®ni- tion mean? Charmides does not know and, looking straight at Critias, divulges that he would not be surprised if the originator himself did not know what he meant. Irritated by Charmides' remark, Critias grudgingly acknowledges that the de®nition is his. Before proceeding further, it is worth pausing to notice that

Critias' irritation is misplaced; it was not Charmides but Socrates who parodied his de®nition beyond recognition and made him look like a fool. Although many commentators are put o¨ by Critias' irritation,18 it is perfectly understandable. Having been lured into the discussion and asked to improve on sheer idiocy, he has fallen victim to a typical Socratic tactic which should not be euphemisti- cally described as an attempt to help him ``make his meaning clearer.''19 Determined to reinstate himself as a thinker of conse- quence, he agrees to succeed Charmides as Socrates' interlocutor. What is going on here? Why does Socrates pronounce Critias'

de®nition ``a kind of riddle''? His ostensible reason is that ``mind- ing one's own business'' cannot mean writing only one's own name, building only one's own house, and so on. He is, of course, right. But that is not a su½cient reason for pronouncing the de®- nition a riddle. Its apparent obscurity is not traceable to Critias' actual view but to Socrates' tendentious interpretation of it ± an

17 Horne¨er, 1904: 56±57, thinks the true originator is Socrates. Having noted that Nicias advances a de®nition of courage which he ascribes to Socrates (La. 194d1±2), Horne¨er suggests that in both dialogues Plato is criticizing Socratic doctrine by putting it into the mouth of an interlocutor. See also Vlastos, 1991: 113, n. 28: ``[T]he procedural form of elenctic argument prevents [Socrates] from making any of his own doctrines the target of elenctic refutation by himself ''; hence ``when Plato ``wants to show Socrates attacking a Socratic thesis he hands that thesis over to an interlocutor, making him its proponent pro tem.''

18 FriedlaÈnder, 1964, ii: 67±68, thinks it reveals his intemperance and marks him as a future tyrant.

19 Teloh's gloss is representative: ``Socrates' ironical misinterpretations have an important function: they force the author of the account to clarify it'' (1986: 62).

Cross-Examining Socrates142

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

interpretation which is all the more strange in view of the fact that at R. 369e6¨. ``Socrates'' bases his ideal society on the very same principle, that is, the so-called Principle of the Division of Labor according to which everyone must ``mind his own business'' by doing that particular work (e� rgon) for which he is best suited by nature. From the moment that principle is introduced, its meaning is treated as luminously self-evident; and everyone is expected to give his immediate and unquali®ed assent. And everyone does. It never occurs to Adeimantus or Glaucon to pronounce the princi- ple a ``riddle'' and to ask whether ``Socrates'' means that everyone should write only his own name, build only his own house, and so on. Why, then, does it occur to Socrates? The answer, I think, is this. Socrates' request for ``clari®cation''

enables him, like an expert hunter with a well-trained bird-dog, to ¯ush a new interlocutor out of the protective social shrubbery. Un- able to sit by and allow himself to be reduced to a laughingstock, Critias is manipulated into entering the discussion in self-defense. Having assented to the truism that craftsmen make (poieiÄ n) things

for other people, he is asked how they can be temperate. Critias fails to see the problem. And with good reason ± there is none. However, Socrates begs to di¨er and explains the ``di½culty'' as follows. If, as Critias has suggested, temperance is ``minding one's own business'' and if craftsmen do (praÂ ttein) other people's busi- ness, does it not follow that they are intemperate (163a1±4)? The answer, of course, is that this does not follow at all; Socrates has confounded making with doing.20 Critias spots the mistake at once and rightly points out that making and doing are two very di¨er- ent things. Craftsmen can make things for other people without doing their business. Indeed, in making good and useful things for others, they are ``minding their own business'' (163b3±c8). Although Critias is obviously right, Socrates accuses him of over-

subtlety ± a technique he has learned from Prodicus ± and unhelp- fully asserts that the things we refer to are more important than the terminology we use in referring to them (163d1±e3), thereby diverting attention from his own mistake and falsely implying that

20 A piece of ``shameless sophistry'' (Guthrie, 1975: 159). ``A sophistic trick'' by which Soc- rates ``perverts'' the obvious meaning of Critias' statement and invests it with a ``ludi- crous'' meaning of his own (Saunders, 1987: 169). Similar criticisms are made by Fried- laÈnder, 1964, ii: 72; Santas, 1973: 108; and Hyland, 1981: 73.

Charmides and Critias 143

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Critias is taking refuge in mere verbal quibbling. However, he thereupon proceeds to trivialize his own objection. Having just faulted Critias for his preoccupation with inconsequential linguis- tic niceties, he now demands more precise de®nitions of ``making'' and ``doing'' ± the very niceties just disparaged. Does Critias think that this ``doing'' or ``making'' ± or whatever he wishes to call it ± is temperance? He does. He also thinks that temperance is the doing of good things (163e10±11). Having set forth his view of temperance, Critias asks whether

Socrates agrees with it. Socrates' response is noteworthy: Critias is not to worry about that yet (pw); what matters is not what he ± Socrates ± thinks, but what Critias thinks (163e6±7). That little word ``yet'' ± so tantalizing but so misleading ± implies that al- though it is presently Critias' turn to say what he thinks, it will soon be Socrates' turn. The implication is false, as we discover a few Stephanus pages later, when in response to Critias' question as to whether Socrates agrees that temperance is self-knowledge, Socrates replies:

[Y]ou are talking . . . as though I professed to know the answers to my own questions and as though I could agree with you if I really wished. This is not the case ± rather, because of my own ignorance, I am contin- ually investigating in your company whatever is put forward. (165b5±c1)

This is a surprising response from someone who just minutes ago had said that it was not his turn to say what he thought yet.21 Critias is in for another disappointment. Having just de®ned

temperance, he now ®nds that Socrates has no intention of exam- ining his de®nition. Apparently no longer interested in discover- ing what temperance is, Socrates abruptly changes the subject and confesses that he is surprised to learn that Critias believes that it is possible for a temperate person to be ignorant of his own temper- ance, i.e., that a person can be temperate without knowing it (164a1±3) ± a remark which leads the discussion in a very di¨erent direction. Critias is puzzled by the remark and explains that he does not

believe this at all. Socrates retorts that he is logically committed to believing it by his previous admission that craftsmen can be tem-

21 As usual, commentators rush to Socrates' defense. Hyland's opinion is representative: Socrates refuses to say what he thinks because he wants ``to penetrate to the very core of'' Critias' position (1981: 86).

Cross-Examining Socrates144

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

perate in doing (poiouÄ ntav) other people's business (164a5±7). His argument ± 164a9±c6 ± is as follows:

(i) In curing his patient, a doctor does what is bene®cial (w� jeÂ lima) for himself and for his patient, thereby doing what he ought (taÁ deÂ onta praÂ ttei ).

(ii) He who does what he ought is temperate. (iii) But a doctor cannot know in advance whether his patient will

be cured and, therefore, cannot know in advance whether what he has done is bene®cial.

(iv) Therefore if a doctor can cure his patient (and thereby do what is bene®cial for him) without knowing it in advance, and if he is temperate insofar as he does what is bene®cial, then it follows that he can be temperate without knowing it.

Obviously alarmed, Critias replies that if this conclusion really does follow from his previous admission, he will retract it.22 For he would almost say that temperance is the same thing (au� toÁ touÄ toÂ . . . ei� nai ) as self-knowledge (164d3±4)23 and that the Delphic in- scription ``Know yourself '' is an exhortation to be temperate.24 Socrates has not, of course, shown that Critias is logically com-

mitted to believing that a person can be temperate without know- ing it. The aporetic impasse to which he has been reduced is not traceable to his actual position but to Socrates' misrepresentation of it. However, this time Critias does not realize what is going on and does not protest. Asked whether he remembers saying that craftsmen can be temperate in doing other people's business (164a8), he says he does. But his memory is faulty. He did not say that at all, indeed, that is exactly what he denied:

[H]ave I agreed that those who do other people's business are temperate by admitting that those making other people's things are temperate? (163a10±12)

22 According to Kraut, 1983: 66; and Vlastos, 1983b: 73, Socrates' interlocutors never renege on their previous admissions, i.e., they never retract propositions to which they previously assented. Perhaps not. But the possibility is envisaged here.

23 Irwin, 1995: 39, notes that this de®nition is ``surprising'' because it makes no mention of the non-cognitive component of self-control which ®gures so prominently in the Socratic concept of temperance in other early dialogues and which Socrates himself previously manifested in this one when he resisted the powerful sexual urge elicited by Charmides' beauty.

24 Horne¨er, 1904: 62, thinks this allusion to the Delphic oracle constitutes further textual evidence for ascribing the de®nition of temperance as ``minding one's own business'' to Socrates.

Charmides and Critias 145

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

What Critias actually said was that craftsmen can make things for other people without doing their business; he added that making and doing are two very di¨erent things. However, Socrates ignores what he actually said, puts words into his mouth, and then ``re- futes'' him. Although he claims to be refuting Critias ``from his own beliefs'' (w� v o� soÁ v loÂ gov, 164c2), his refutation depends on his tendentious interpretation of Critias' actual position.25 Even if Critias had said what Socrates claims he said, he would

still not be logically committed to believing that a person can be temperate without knowing it. From the fact a doctor cannot know in advance whether his patient will be cured, it does not follow that if he is cured, the doctor was temperate all along without knowing it. Whether a doctor is temperate is not contingent upon the re- covery of his patients. A temperate doctor is not one who cures his patients but one who does his best to cure them by conscientiously doing what is in their best medical interests. Although he cannot know in advance that he has done the former, he can know that he has done the latter and that he is, therefore, temperate. Socrates' argument is problematic for a second reason. In claim-

ing that it is possible for a person to be temperate without know- ing it, Socrates contradicts his earlier contention that if a person is temperate, he must have some opinion about it, since temperance necessarily gives some intimation of its nature which enables its possessor to form an idea of it (159a1±3). If this is true, as he pre- viously assured Charmides, then it cannot be true, as he now as- sures Critias, that a person can be temperate without knowing it; on the other hand, if a person can be temperate without knowing it, then from the mere fact that Charmides cannot de®ne temper- ance it does not follow that he is not temperate. Socrates cannot have it both ways. Having identi®ed temperance with self-knowledge, Critias again

asks whether Socrates agrees. Socrates coyly replies that before di- vulging his opinion, he needs time to re¯ect. However, the ensuing interlude of Socratic ``re¯ection'' is not the prelude to a momen- tous revelation. Having ``re¯ected,'' Socrates merely invokes the techneÅ-analogy and introduces another gratuitous complication. If

25 A. E. Taylor, 1929: 52, is wrong in claiming that Critias ``has not . . . thought out the im- plications of his own distinction.'' So is Vlastos, 1981b: 207, n. 7, according to whom Critias is ``a pretentious and unclear thinker.''

Cross-Examining Socrates146

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

temperance is knowing something (gignwÂ skein . . . tiÂ ), it must be a kind of knowledge (e� pisthÂ mh tiv); and, like all knowledge, it must be knowledge of something (tinoÂ v, 165c4±6). But of what? ``Of oneself '' (e� autouÄ , 165c7), replies Critias. Socrates neither agrees nor disagrees; instead, he exploits the logic of the techneÅ-analogy. Medicine is knowledge of health, carpentry is knowledge of build- ing, and so on. Indeed, every techneÅ is knowledge of something and invests its possessor with a unique expertise which enables him to produce some ®ne result (kaloÁ n . . . e� rgon, 165d1±2). So if tem- perance is knowledge, it too must produce some ®ne result. What is it? At this point, Critias registers a protest: Socrates is not inquir-

ing rightly (ou� k o� rqwÄ v zhteiÄ v, 165e3). Previously he had asked how temperance is di¨erent from all other knowledge; now he is asking how it is the same,26 thereby overlooking the fact that temperance is di¨erent from medicine and carpentry in that it does not pro- duce some ®ne result. Arithmetic and geometry di¨er from them in the same way. In short, although every techneÅ is knowledge of something and invests its possessor with a unique expertise, it is not the case that every techneÅ necessarily produces some ®ne result (165e3±166a2). Commentators di¨er about how this exchange should be read.

Some think that Critias' objection is symptomatic of Plato's own growing misgivings about the techneÅ-analogy.27 Thus Kahn declares:

Critias succeeds in blocking Socrates' use of epagoÅgeÅ by twice challenging the tacit assumption of similarity on which the argument from analogy must rely . . . [Plato] has now made the reader aware of the weakness of any argument from epagoÅ geÅ, in its implicit assumption of relevant similar- ity between the cases cited and the item under dispute. The reader may now expect greater rigor in the method of argumentation.28

Perhaps. But if ``the reader'' does expect this, the expectation is quickly disappointed. For Socrates continues to make copious use

26 FriedlaÈnder, 1964, ii: 73, defends Socrates on the ground that, in asking both how tem- perance di¨ers from and resembles other kinds of knowledge, he is searching for ``the genus proximum and the di¨erentia speci®ca.'' But this is nothing more than an ad hoc rescue operation conducted under the dubious auspices of Aristotelian logical categories alien to Socratic thought.

27 See, e.g., Guthrie, 1975: 168; Irwin, 1977: 75±76; Saunders, 1987: 170±71; and Kahn, 1996: 191±203.

28 1996: 194.

Charmides and Critias 147

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

of arguments from epagoÅgeÅ and with the same cavalier disregard of dissimilarities. The only discernible di¨erence between his previ- ous and subsequent use of them is that before deducing his next inference ± that, like other kinds of knowledge, temperance has a distinct object ± he acknowledges that this conclusion cannot be known to be true with complete certainty. But he clearly thinks that it is eminently reasonable ± so reasonable, in fact, that he pronounces the opposite view ± Critias' ± ``strange'' (a� topon). The upshot is that Critias' ``challenge'' is largely ignored and Socrates' method of argumentation remains largely the same. Whether Plato actually employs Critias as a mouthpiece for his

own growing misgivings about the techneÅ-analogy is a question which lies beyond the scope of this book. For our purposes, it is su½cient to notice that Socrates does not dispute Critias' com- plaint that he is not inquiring rightly; on the contrary, he concedes the point: ``You are right'' (166a3). However, his concession is hedged with quali®cations. Having just been refuted (or, at least, corrected) about a matter of considerable importance, Socrates does not respond to Critias as he routinely expects his inter- locutors to respond to him. He neither bemoans the fact that he had previously been the unwitting victim of a false belief ± the greatest evil that can befall a human being (G. 458a8±b1) ± nor applauds Critias as his greatest benefactor ± the appropriate response on the part of someone who has just undergone the ben- e®cial experience of refutation (G. 506c2). Instead, he shifts his ground and tries to salvage whatever he can: ``You are right. But . . .'' (166a3). But what? Having acknowledged that he was wrong in thinking

that temperance produces some ®ne result, Socrates insists that he was right about two other things: ®rst, although arithmetic and geometry do not produce a unique result, they do have a unique subject matter; second, their subject matter, like that of every science, is distinct from the science itself. In short, knowledge is always distinct from its object. So if temperance is a kind of knowledge, the same must be true of it. Critias disagrees. Unlike all other knowledge, which is knowl-

edge of something other than itself, temperance is knowledge both of other things and of itself (te a� llwn e� pisthmwÄ n e� pisthÂ mh e� stiÁ kaiÁ au� thÁ e� authÄ v, 166c2±3); furthermore, he thinks Socrates is well

Cross-Examining Socrates148

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

aware of this, but is ignoring it because he is only interested in refuting him. Socrates does not deny that refuting Critias is one of his objectives, but he does deny that it is his only one:

Wrenched from context, as it usually is, this assurance that elenc- tic examination is always self-examination seems to underscore Socrates' philosophical seriousness.29 In fact, it reveals his lack of seriousness. First, it is a mere diversionary maneuver which enables Socrates to conceal his ¯awed methodology by appealing to his (allegedly) salutary intentions. Second, it is patently insincere. That this is so is borne out both by what precedes and by what follows it. Coming, as it does, from someone who had just been re- futed and who had responded to refutation by shifting his ground, the solemn announcement, ``I examine arguments primarily for my own sake,'' becomes singularly unconvincing. And it is even more unconvincing when he proceeds to transfer the argumenta- tive burden from his own shoulders to those of his interlocutor and thereupon exhorts him to attend to the argument instead of worrying about who is refuting whom. The evasion works. Critias drops his protest. Called upon to clarify his contention that temperance is self-

knowledge, Critias repeats that it is a unique kind of knowledge which is both of other things and of itself, that is, it is knowledge of knowledge. Prompted by Socrates, he agrees that it is also knowledge of lack of knowledge; for if temperance is knowledge of what one knows (e� pisthmwÄ n e� pisthÂ mh), it must also be knowl- edge of what one does not know (a� nepisthmosuÂ nhv e� pisthÂ mh, 166e6±7). Socrates thereupon points out that if this is true, then two important epistemic consequences follow: ®rst, only the tem- perate man will know himself (e� autoÂ n) and be able to examine

29 Commenting on this passage, Vlastos writes: ``Moments of self-revelation like these are rare in the early dialogues'' and ``Socrates means to be taken at his word'' (1971: 9±10).

Charmides and Critias 149

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

what he knows and does not know;30 second, only he will be able to examine other people to determine what they know and do not know (167a1±4). Socrates ®nds both claims problematic. His reasoning ± tortuous

in the extreme ± is as follows. If there is a kind of knowledge which is both knowledge of other things and of itself, it must di¨er from every other cognitive or perceptual capacity (duÂ namiv). For if such knowledge exists, it must not only be knowledge of something; it must also be of a nature to be (e� cei tinaÁ toiauÂ thn duÂ namin w� ste tinoÁ v ei� nai, 168b3) knowledge of something, just as that which is greater than something else is not only greater than it, but also of a nature to be greater than it. The question, then, is: Is there anything of which it may be truly said that when a property by virtue of which that thing is related to some other thing is applied to that thing itself, the same re¯exivity invests it with the ``being'' (thÁ n ou� siÂ an) of the thing to which it is related? It does not seem so. For example, if vision were both of other things, i.e., of color, and of itself, then there would be a kind of vision which is not of color, but of vision and lack of vision; if hearing were both of other things, i.e., of sound, and of itself, then there would be a kind of hearing which is not of sound, but of hearing and lack of hearing. The same is true of magnitude and number and similar things: ``[If ] we should discover something greater that is greater than the greater things and than itself, but greater than nothing than which the other greater things are greater,'' then ``if it were actu- ally greater than itself, it would also be less than itself.'' Similarly, ``anything that was the double of other doubles and of itself would . . . be half of itself and of the other doubles.'' Similarly, anything that is more than itself will also be less; anything heavier will also be lighter; anything older will also be younger; and so on. But it seems absurd to say this about vision, hearing, and magni- tude. And it seems equally absurd to say it about knowledge. It would seem that knowledge can no more know itself than vision can see itself, or hearing hear itself, or desire desire itself, or wish wish itself, or love love itself, or fear fear itself, or belief believe itself (167c8±168a4).

30 The shift from temperance as knowledge of oneself to knowledge of one's knowledge is sometimes judged fallacious (see, e.g., Tuckey, 1951: 33±37; and Saunders, 1987: 169±70, 192). For rebuttals, see Brun, 1960: 63±64; Santas, 1973: 119, n. 12; Guthrie, 1975: 169; and ChreÂtien, 1987: 22.

Cross-Examining Socrates150

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

For reasons which I cannot fathom, Critias claims to understand this argument and thinks that its conclusion follows necessarily. Socrates seems pleased with the argument too and thinks it shows that it is highly unlikely that such knowledge exists. However, he is unwilling categorically to deny that it does. It would take ``some great man'' to determine whether ``there are things that apply to themselves,'' and, if so, whether temperance is one of them (169a1). Having suggested that they investigate this di½cult matter on some future occasion, he proposes that they assume for the sake of the discussion that such knowledge does exist and asks Critias to explain how it would be bene®cial (w� jeÂ limon, 169c1), that is, how it would enable its possessor to know what he knows and does not know (169d5±8). Methodologically speaking, this is a very puzzling passage.

Having just acknowledged the limits of analogical argumentation, with ``its implicit assumption of relevant similarity between the cases cited and the item under dispute,'' Socrates proceeds to re- fute Critias' de®nition of temperance with an analogical argu- ment. Although Critias concedes that Socrates' conclusion follows, his previous remarks about relevant dissimilarities provide him with a cogent rejoinder. In response to Socrates' contention that it is ``strange'' (a� topon) to say that temperance is knowledge both of something else and of itself ± a claim which makes temperance di¨erent from every other kind of knowledge ± Critias should have said: ``Of course it is strange. Temperance is di¨erent from every other kind of knowledge. That was precisely my point. You cannot hope to persuade me that I am mistaken in believing that temper- ance is di¨erent from every other kind of knowledge by means of an epagoÅgeÅ designed to show that temperance is di¨erent from every other kind of knowledge!'' Of course, Critias does not lodge this objection. Instead, in

reply to Socrates' query about how knowledge which is both of other things and of itself would be bene®cial, he produces a series of analogies which tell against his repeated assurances that tem- perance is di¨erent from every other kind of knowledge: just as a person who has swiftness and beauty will be swift and beautiful, so also a person who has self-knowledge will know himself. Socrates does not dispute the tautological claim that a person who has self- knowledge knows himself; what he wants to know, however, is how this (alleged) knowledge would be bene®cial in the sense of en-

Charmides and Critias 151

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

abling its possessor to know what he knows and does not know (169e6±8). Critias thinks the answer is obvious: it is because self- knowledge and knowing what one knows and does not know are the same (tau� toÂ n, 170a1). Socrates is inclined to disagree. His reasoning ± 170a6±171c10 ±

is as follows:

(i) If there is such a thing as knowledge of knowledge, its subject matter (or content) is con®ned to what is and what is not knowledge.

(ii) But the knowledge that something is knowledge is not the same as the knowledge that something is healthy or musical; the latter are known by the technai of medicine and music whereas the former is known by knowledge.

(iii) Therefore if a person has no knowledge of medicine or music but only knowledge of knowledge, he will know that he knows something, but not what he knows.

(iv) So if temperance is knowledge of knowledge, the temperate man will not know what he knows and does not know; he will only know that he knows and does not know something.

(v) Nor will he know what other people know and do not know; he will know only that they have some knowledge.

(vi) So the temperate man will not be able to distinguish a com- petent doctor from a quack (or any other genuine practitio- ner of a techneÅ from an imposter). To do that, he must be a doctor himself.

(vii) Therefore, if temperance is knowledge of knowledge and lack of knowledge, then insofar as it enables its possessor to know only that he and other people know something but not what they know and do not know, it is not bene®cial but useless.31

31 This argument has recently been championed by Kahn, 1996: 199, who agrees that ``one cannot test the credentials of others in regard to a subject in which the tester himself lacks expertise.'' Kahn thinks the argument has ``serious consequences'' for the Socratic enterprise of testing the wisdom of others: for if one cannot distinguish a doctor from a quack (or any other genuine practitioner of a techneÅ from an imposter) unless one is a doctor (or a practitioner of the relevant techneÅ ) oneself, how can Socrates, who is a prac- titioner of none of these technai, expose false claimants of knowledge? According to Kahn, he cannot. A similar objection is raised by Tuckey, 1951: 66±67, who acknowl- edges that Socrates' claim that a non-expert cannot distinguish genuine practitioners from imposters seems inconsistent with his claim that the craftsmen knew many ®ne things (Ap. 22d1±2) ± a claim which presupposes that Socrates can recognize genuine practitioners. However, Tuckey thinks the inconsistency is only apparent and easily re-

Cross-Examining Socrates152

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

This argument is intended to show that the second-order knowl- edge (or meta-knowledge) under examination ± knowledge of knowledge ± cannot enable its possessor to know what he and other people know and do not know unless he also has some ®rst- order knowledge, that is, unless, in addition to knowing what is and is not knowledge, he knows some speci®c subject matter. For it is its speci®c subject matter which distinguishes each kind of knowledge (and each techneÅ ) from every other. Before examining this argument, it is worth noticing a prelimi-

nary point. There is no analogy between knowledge of medicine or music, on the one hand, and knowledge of knowledge, on the other, unless we assume that, like medicine and music, virtue ± and, therefore, temperance, which is one of its ``parts'' ± is a techneÅ; and that, like the possessor of the former, the doctor and the musician, the possessor of the latter, the temperate man, is a practitioner of a techneÅ. However, to assume this is to reinstate the very techneÅ-analogy which, according to some commentators,32 had just been criticized (by Plato with Critias serving as his mouth- piece) on the ground that it relies too heavily on the similarities between virtue and the technai and is insu½ciently sensitive to the dissimilarities between them. Of course, this is more a criticism of Critias' de®nition of temperance than of Socrates' criticism of it. But if Plato really is criticizing (or having second thoughts about) the techneÅ-analogy in the Charmides, this is surely one of the criti- cisms he would have made. Instead, he makes a very di¨erent criticism. The foregoing argument is intended to demonstrate that if Critias is right in thinking that temperance is knowledge of knowledge, then the temperate man cannot know what he or any- one else knows and does not know; he can only know that he and they know something. But the argument is unsound. In order to distinguish a compe-

tent doctor from a quack (or any other genuine practitioner of a techneÅ from an imposter), one need not be a doctor oneself. It is, of course, true that anyone who can make such distinctions must

moved. Following Pohlenz, 1913: 54, he removes it by arguing that Socrates' assessments of technical expertise are based on the application of the elenchus and, therefore, do not require that he himself possess the relevant expertise. Kahn rejects this solution on the ground that there is no such thing as elenctic knowledge, there is only knowledge of medicine, carpentry, etc. (199±200).

32 See n. 27.

Charmides and Critias 153

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

know, in some sense of ``know,'' what a doctor knows. But ``know- ing what a doctor knows'' is ambiguous. It could mean either that one knows the same things that doctors know or that one knows the kinds of things that doctors know. Although the ®rst sense of ``knowing what a doctor knows'' requires that one be a doctor oneself, the second does not. All it requires is that one has some broad but nevertheless genuine understanding of the knowledge and abilities which constitute medical expertise. Only by over- looking this distinction could anyone claim that it takes an expert to recognize an expert. In fact, Socrates has greatly overestimated the di½culty of identifying experts and distinguishing them from non-experts. To be sure, there are some technical and highly specialized sit-

uations in which only an expert can assess someone else's claim to expertise: only a legal expert can assess the expertise of a law student taking a bar exam, only an expert historian can assess the expertise of a candidate for an advanced degree in history, and so on. But there are many other situations which do not require this kind of expertise. Many assessments of expertise (or lack of it) are made, not by determining what someone does (and does not) know, but by observing what he can (and cannot) do. Vacationing in the tropics, I develop a fever and break out in a generalized rash. I book an appointment with the ®rst available doctor who examines me, fails to notice the rash, and concludes that I have the ¯u. Skeptical, I call his attention to my rash covered body and humbly suggest that I have the measles. Astonished by this hitherto unnoticed symptom, he concurs. I do not need a medical degree to realize that I am not in the presence of a brilliant diagnostician. In other situations, assessments of expertise (or lack of it) are made by inspecting the alleged expert's product. Unhandy with tools and ignorant of the principles of engineering, I hire a car- penter to build a house. Upon moving in, I notice that the walls are sagging and that part of the roof has collapsed. I do not need an engineering degree to realize that my contractor is not a skilled practitioner of the techneÅ of carpentry. Such judgments need not be withheld on the ground that I am neither a doctor nor a car- penter. Assessments of expertise (or lack of it) can also be made in many other ways. Sometimes a non-expert need only collect a few bits of obscure or highly technical information which only an expert would be likely to know and then query the alleged expert

Cross-Examining Socrates154

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

to determine whether he knows them. Other determinations of expertise (or lack of it) are easier still. To determine whether you know Greek, I need not know the language myself; I need only hand you a Greek text, ask you to translate, and observe your em- barrassment at being unable to make head or tail of it. Socrates' argument can be faulted on another ground. The

argument tries to show that if temperance is knowledge of knowl- edge and lack of knowledge, then the temperate man can only know that he (or someone else) knows something; he cannot know what he (or someone else) knows. To know that, he must be an ex- pert himself. The unanswered question is: How can the temperate man know that he (or someone else) knows something unless he knows what that knowledge is? The answer is, of course, that he cannot. The same fact (if it is a fact) which prevents the temperate man from knowing what someone knows unless he knows it himself also prevents him from knowing that someone else knows some- thing. Knowing only the criteria which must be satis®ed if some- thing is to count as knowledge, how can the temperate man know that someone's purported knowledge-claims satisfy them unless he knows what he claims to know? Second-order knowledge that someone has knowledge is parasitic on ®rst-order knowledge of what it is. Indeed, to say that someone knows something, for ex- ample, carpentry, is to say what he knows. Socrates advances one ®nal argument again Critias' de®nition

of temperance. It has been agreed that if temperance is knowing what one knows and does not know, it is bene®cial for its pos- sessor. Equipped with this knowledge (and the knowledge of what other people know and do not know), it would seem that a person could live his life unerringly, doing only that about which he has knowledge and assigning all else to others with the requisite ex- pertise. It would also seem that a city based on that principle would conduct itself admirably and well (kalwÄ v kaiÁ eu� praÂ ttein), and those who live admirably and well are happy (eu� daiÂ monav, 172a1±3). According to Socrates, however, this is an illusion. Such an arrangement would not be bene®cial in the deepest and most important sense. It is true that in such a state of a¨airs everything would be done in accordance with knowledge, that is, in accor- dance with the relevant technai and under the direction of ac- knowledged experts. But this would not be su½cient for happiness. Critias is astonished by the curious turn the argument has taken

Charmides and Critias 155

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

± apparently with good reason. For the philosopher who believes that virtue is knowledge to suggest that those whose lives are com- pletely governed by knowledge would not necessarily be either virtuous or happy seems bizarre and utterly un-Socratic. But this initial impression is unfounded. Socrates is not rejecting

the rule of knowledge simpliciter, but only of a certain kind of knowledge. That this is so is borne out by his response to Critias' next question: If we reject the rule of knowledge, what will we put in its place? Critias thinks human happiness is contingent on be- having as knowledge directs. Socrates agrees, but he sees a further problem: knowledge of what? Surely not of shoe-making or work- ing with bronze or wood? So it is not knowledge simpliciter that leads to happiness but knowledge of a certain kind, namely, knowledge of good and evil (toÁ a� gaqoÂ n . . . kaiÁ toÁ kakoÂ n, 174b10). Lacking this, medicine would still produce health, weaving would still produce clothing, and so on; however, this is not su½cient to insure the kind of happiness Socrates is talking about which is not contingent on having these things but in using them properly (174c9± d7).33 So even if there is such a thing as knowledge of knowledge and lack of knowledge, it would not necessarily be bene®cial to its possessor. Socrates thereupon concludes that his (and Critias') previous assumption that temperance is bene®cial now seems hasty and ill-advised. For how can something be bene®cial if it produces no bene®ts? So the discussion has been in vain. At the same time, Socrates cannot bring himself to believe that

temperance is not bene®cial; indeed, he thinks it is a great good. It is only because he is such a poor investigator that he has failed to understand its nature and to discover its bene®ts. Not only have Critias and he failed to discover the nature of temperance; they also made many wrong concessions along the way. They did so in order to maintain the thesis that the temperate man knows what he knows and does not know ± a thesis now seen to be absurd. For nothing can be more irrational (a� logwÂ teron) than to believe that one can know in a sort of way (a� mwÄ v geÂ pwv) what one does not know at all. As a result, the inquiry (h� zhÂ thsiv) is mocking them (175c3±d6). What troubles Socrates most of all, however, is that Charmides will derive no bene®t from his temperance.

33 This doctrine is elaborated more fully in the Euthydemus where Socrates speaks of ``the royal art'' (thÁ n basilikhÁ n . . . teÂ cnhn, 291b5) which, by regulating the subordinate arts by directing them to their proper ends, insures happiness.

Cross-Examining Socrates156

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Charmides again confesses that he does not know whether he is temperate; indeed, in view of Socrates' and Critias' failure to dis- cover what temperance is, how could he know? Nevertheless, he is willing to be ``charmed'' by Socrates every day for as long as he thinks necessary. Critias thinks this is su½cient proof of Char- mides' temperance and urges him never to forsake Socrates. Charmides promises to obey his guardian ± even if has to resort to force ± an ominous forecast of his later willingness to employ force to achieve his political ends. And with this the dialogue concludes. In studying the Charmides, it is important to remember that

Charmides and Critias later became members of the Thirty. It is also important to remember that many Athenians attributed their downfall to their association with Socrates. Since the dia- logue provides no reason for thinking that Plato agreed,34 we must assume that he accounted for their downfall in some other way. The opening pages of the dialogue provide several clues. Here is Charmides ± a beautiful youth full of promise. Although

(allegedly) enamored of temperance, he is also highly susceptible to ¯attery and luxuriates in the attention of his admirers. He is also an intimate of Critias ± a disrespecter of morality and a future tyrant. He thus combines in his person the two deadly ingredients against which the ``Socrates'' of the Republic warns: the capacity to be corrupted and the company of a potential corrupter. Enter Socrates. Momentarily stunned by Charmides' physical

beauty, he quickly recovers and asks whether he is equally beauti- ful in soul. Critias thinks so and adds that he is also the most tem- perate youth he knows. But Charmides is not so sure and agrees to be questioned. Surely if temperance is in him, he can say what it is. But he cannot and is quickly reduced to aporia. Having taxed Charmides' meager dialectical ability to the limits, Socrates enlists the services of Critias. Although somewhat better equipped than Charmides, he too soon founders and ®nds himself agreeing that although temperance is universally praised as the most admirable thing in the world, this assessment is wildly premature in view of the foregoing discussion which seems to suggest that it is not ben- e®cial but useless. Socrates expresses concern for Charmides, albeit not for Critias, and Charmides agrees to participate in future dis- cussions, unproductive though they are.

34 In fact, many commentators think Plato wrote the Charmides, at least in part, to defend Socrates against this charge.

Charmides and Critias 157

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Many commentators have been ba¿ed by the Charmides ± a ``curious and di½cult''35 dialogue ``whose point is very hard to see.''36 The more radical question, of course, is: Does it have a point? Anyone who ventures an answer to this question is obliged to

ponder some hard questions: Has Socrates really conducted a serious investigation into the nature of temperance? Or has he merely tried to persuade his interlocutors that they do not know what it is and that every attempt to ®nd out will inevitably end in failure? Has he really tried to persuade them that temperance is bene®cial and worthy of being cultivated? Or has he merely tried to confuse them? Has he made a genuine e¨ort to discover what they really believe? When they advance their opinions, does he take them seriously and examine them honestly and in good faith? Or does he distort what they say and even put words into their mouths so he can refute them more easily? Has he conducted the argument chie¯y (or solely) for his own bene®t? Or has he also conducted it partly for his interlocutors' bene®t? More radically, has he conducted it for anyone's bene®t? The dialogue provides no reason for thinking that Socrates has

conducted a serious investigation into anything. His purpose throughout does not seem to be truth, but victory. In the process, he employs many arguments which are demonstrably fallacious or unsound. Aporia there is, but its e¨ects are the very opposite of those envisaged in the Apolog y. No one is seized by his ignorance and impelled to take up the philosophical quest. Although more than a third of the dialogue is concerned with self-knowledge, no one achieves it; on the contrary, if the dialogue proves anything, it proves that self-knowledge is di½cult to de®ne, even more di½cult to achieve, and, in the end, probably not worth achieving. By the end of the discussion, Charmides is as confused as he was at the beginning. Perhaps more so. Instead of emerging with his soul im- proved, he is reduced to incoherent stammering about ``improve- ment'' and the Socratic ``charm'' in which he does not believe but which allegedly produces it. In the meantime, Critias looks on, like a helpless parent unsure of what his child needs but sure that he needs something, vacantly praising temperance and inexplicably urging Charmides to cultivate it by attaching himself to Socrates.

35 Guthrie, 1975: 163. 36 Crombie, 1962, i: 211.

Cross-Examining Socrates158

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

Even if Plato did write the Charmides partly to absolve Socrates of all responsibility for the crimes of his former companions, con- versations like the one depicted in it might very well have unin- tentionally undermined the morality of (some of ) his interlocutors. On this subject, Edward Meyrick Goulburn has written well:

[Socrates'] design was to point out to men that they had no clear and consistent notions of moral subjects, and to lead them to form such no- tions. The result, it may be feared, was to puzzle them as to right and wrong, and make them question what their moral sense told them . . . Socrates did raise questions . . . without setting them at rest; and though . . . he did this from his thorough honesty of mind, and with a view to make his countrymen honest-minded, one can quite conceive that in many instances the e¨ect was to make them skeptical . . . Where a few minds were touched and improved, a greater number would accept the di½culties as an excuse for throwing overboard altogether all care for virtue.37

The clear implication seems to be that this may very well have happened to Charmides. In any event, before denouncing him, we should ponder Goulburn's sobering words. We should also ponder the middle-period Plato's arresting

claim that the greater the initial promise, the more lamentable its corruption. The very qualities that make up the ``philosophical nature'' (thÄ v jilosoÂ jou juÂ sewv) are the principal causes of its corruption and destruction when the person is poorly educated and badly brought up, and it is such men who do the greatest harm to cities and individuals (R. 495a4±b6). Although unsparing in his criticism of cities and institutions, Plato is often surprisingly gentle towards the very persons whose views and behavior he most deplores. His criticism is often tempered by his perception of the fragility of youth's ®rm but untested resolve. Viewed in the light of once bright but now faded promise, apparently unredeemable character ¯aws elicit criticism tinged with compassion and regret for what might have been. Such understanding, typical of middle- period Plato, is unavailable to those whose criticism springs solely from moral outrage and can only be achieved by those who are willing to look long and hard in the most unlikely places. Char- mides' end is all the more regrettable in light of his promising beginnings.

37 1858: 51, 60±61.

Charmides and Critias 159

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .

c h a p t e r 8

Euthyphro

Of all the interlocutors encountered in the early dialogues, few are more interesting than Euthyphro.1 A diviner (maÂ ntiv) by profession, he is a claimant to esoteric theological knowledge which enables him to unravel even the knottiest moral issues, and remarkable prophetic powers which enable him to predict the future with uni- versal accuracy. Although much younger than Socrates, he is not a youth;2 his father is an old man and he himself is a familiar ®gure in the Assembly. Convinced that his knowledge of things divine uniquely quali®es him as a theological and moral expert, he seems a made-to-order target for the transparently skeptical and irony drenched Socrates with whom he becomes embroiled in debate about piety (toÁ o� sion)3 on the royal porch of the King Archon after announcing that he is prosecuting his father for murder ± a morally controversial and socially scandalous course of action which has outraged his relatives and infuses the discussion with a sense of practical urgency. The ensuing search for a de®nition of piety is no disinterested academic exercise. Euthyphro's dogged attempt to de®ne this familiar but unexpectedly elusive term is an exercise in self-justi®cation on the part of a zealous moralist de- termined to prove that he knows what he is doing. Plato often

160

1 Unknown outside the pages of Plato, Euthyphro is generally believed to have been an historical person. His historicity is implied at Crat. 396d4±397a1 where he is traced to the deme of Prospalte and credited with having inspired Socrates to study the etymology of names. He is also alluded to at Crat. 399a1, 407d8, 409d1±2, and 428c7.

2 Estimates of his age range from fairly young (Allen, 1970: 20) to middle-aged (A. E. Taylor, 1929: 146).

3 The Greek term has no exact English equivalent. ``Holiness'' has Judeo-Christian over- tones which are foreign to it, whereas ``piety'' sounds quaint and conjures up dreary images of Victorian prudery at which modern readers smile. An action is pious (o� sion) if it is sanctioned by divine (as opposed to human) law (Allen, 1970: 25). The term connotes a religious duty ± that part of justice which has to do with the gods rather than with men (Eu. 12e5±8). I will retain the term ``piety'' for lack of a better one.

C o p yr

ig h t ©

2 0 0 0 . C

a m

b ri d g e U

n iv

e rs

ity P

re ss

. A

ll ri g h ts

r e se

rv e d .