1 In this paper I shall consider Socrates as the Platonic character in the cluster of dialogues that show him testing people for expert knowledge by asking after definitions. Of course this Socrates represents an historical figure, but I shall not try to negotiate the difficult gap between fiction and fact. 2 A question that clearly expects a certain answer, whether positive or negative, is properly taken as evidence for the views of the questioner in most contexts. Michael Stokes has eloquently argued that this principle does not hold for Socrates (1986: p. 9). If not, we would expect either signs of a deceptive irony, or a flexible adaptation of questions to interlocutors. In fact, contexts like 190b are as straight as any in Plato, and Socrates' leading questions there represent a consistent body of views about knowledge and definition, views he uses in conversation with a variety of interlocutors. 3 Cf. Hippias Major 286c-d and 304d-e, Euthyphro 6d-e, Charmides 159a, and Lysis 223b. See also Woodruff, 1982: pp.138-41, and 148n6. 4 I shall not discuss here what it is for a definition to be Socratic, except to say that a Socratic definition states the essence (ov6ia) of its definiendum. For more detail on my general view of this, see Chapter VI in Woodruff (1982).
5 The elenchus is Socrates' method of examining his interlocutors and/or their proposals. Discussion of the elenchus and its use has been sparked by Vlastos (1983), on which see Kraut (1983), Brickhouse & Smith (1984), Polansky (1985), and Woodruff (1986).
6 Vlastos has excluded from the ranks of elenctic dialogues the Euthy- demus, Lysis, and Hippias Major (1983: pp. 57-8, 1985: p. 1). His main grounds for this are "the abandonment of adversary argument as Socrates' method of investigation," and the turn in the Hippias Major to self-criticism. I think it a better method to begin with a central text, the Apology, and to ask what it represents as central to Socrates' method of investigation. Though perhaps illustrated in the passage with Meletus, adversarial argument is no part of the life described in the Apology. Nor is Socratic self-criticism a later development in Pla- to's dialogues (pace Vlastos 1983: p. 58). In the Euthyphro Socrates re- futes definitions he himself has proposed; in the Laches and related dialogues he rejects definition proposals based on teachings that are known to be Socratic. In the Gorgias he represents the elenchus as test- ing his own soul (487a). In the Crito, moreover, as in the Hippias Major and Symposium, he takes pains to represent himself as refuted by the same arguments that refute his partners in discussion, and this could be predicted from the Apology. 7 Charles Kahn (1986: p. 15) argues persuasively for considering the Laches as the introduction to a set of dialogues on moral education and the nature of virtue (Charmides, Euthyphro, Protagoras, Meno, Lysis, Euthydemus), the set as a whole coming after the Apology, Gorgias and other earlier works. Still, the common theme of expert knowledge warrants my study of the Laches and Apology together.
8 Such a method is close to that of the Gorgias, where, by refuting al- ternatives, Socrates shows that his interlocutors actually share his views, whatever they say to the contrary (458a, 509a, etc.). Cf. the Protagorean method of È1tavópOco¡.ux, by which poetry is corrected to what ought (in the philosopher's view) to have been its literal mean- ing. On this, see my forthcoming Plato and Protagoras, p. 8.
9 By "ignorance" I intend "lack of expert knowledge."
10 Below, p. 100 ff. Modern scholars have tended to overlook the negative side of Socrates' teaching. On his disavowal of knowledge, see Woodruff (1982: p. 142) Kraut, (1983 and 1984: pp. 246-9), Vlastos (1985), and especially Lesher (1987). 11 To be evidence for expert knowledge, a definition proposal for Courage would have to identify the essence of Courage. Many sentences say true things about Courage while falling short of that; some
sentences even disclose essential truths about Courage while falling short. The conclusion of the final argument of the Laches plainly avoids denying the Socratic doctrine, while still denying that it identifies Courage: "But what we mentioned earlier turns out not to be a part of virtue. -Apparently not.— Then we have not found what Courage is" (199e9-11). See Appendix I. On the use of surrogates in refutation, compare the use of Ion in place of Homer in the Ion with the use of Nicias for Socrates in the Laches.
12 At 21c7; but see 22e3. The point is that the sort of sophia that the professionals do have is not worth enough to outweigh the attendant ignorance. The distinction Socrates actually uses is not going to be be- tween wisdom and techne, but between one sort of wisdom-techne and another.
13 This generalization is true of most Skeptics, but (surprisingly) not of Aenesidemus and Agrippa, whose arguments follow models that are more Platonic than those of their predecessors and successors. See Woodruff (1986 and forthcoming). 14 ln the Sophist, by contrast, the Visitor describes a purgative sort of elenchus in Socratic terms (230c). 15 Évóehcv'lJI.W.t oit ovx Ëan aocpo5 (Apology 23b7).
16 Ancient Skeptics faced a Stoic opposition that held a developed theory of knowledge; this they could refute dialectically, in its own terms, without committing themselves to any views of the matter. Socrates' opponents have no idea what knowledge is; it is Socrates who insists that expert knowledge requires definitions, and Socrates who
supplies the standards for success in definition. This theory of know- ledge cannot be attributed to any of Socrates' various interlocutors, for he applies the same theory in every case. Far from believing the theory, the interlocutors rarely show signs even of understanding it.
17 On the range of techne in Platonic usage, see Roochnik 1986. 18 Interest in techne in moral and political contexts grew out of the increasing complexity of public affairs in Athens in the later Fifth Century. On this theme see Connor (1971: pp. 125 and 126 n. 68) and Andrewes (1962: p. 83).
19 This view points towards a theory that will be expanded in the Republic: subordinate technai are worthy of the name only if they are under the rule of those who do know the good; you cannot be a good general, for example, unless you are under the command of a good king (cf. Euthydemus 290c).
20 Hoplites were armed in such a way that they were effective only if each soldier maintained his position in the battle line.
21 Nicias' caution and timidity were proverbial long before the disaster to which these qualities led him in Sicily, owing to his failure to take advantage of the situation on Pylos (Plutarch Nicias 8.1, Aristophanes Birds 638-9). Thucydides is kinder to Nicias (vii.86), but represents him as dilatory in Sicily and "overly inclined to divination" (vii. 49-50). Still, Nicias doggedly continued to lead his division in combat after the Athenians under Demosthenes had surrendered on relatively favorable terms. Nicias' courage was singularly lacking in the intel- lectual qualities he recommends in the dialogue; in the end, he displayed a steadfastness that led to unconditional surrender. Laches' recommendation of steadfastness is similarly undercut in the dialogue by his cowardice in the present argument, which he gives up far too easily. See O'Brien 1963: p. 143 n.11.
22 John Beversluis (forthcoming) and J. H. Lesher (1987) argue inde- pendently and forcefully that Socrates is not committed to the view that one must know what F is before knowing that a is F. The main evidence would be Euthyphro 6e and Hippias Major 304d-e. The former does not make knowledge of definitions necessary, but merely useful, for picking out examples. The latter may not apply to knowing that particulars are F.
23 Non-expert knowledge may be of types or tokens, but can never be unrestricted in context, as expert knowledge would have to be. Later
Plato will use "knowledge" for "expert knowledge" only, and turn to various verbs of perception, appearance, and opinion for one's attitude towards truths that are contextually restricted (cf. Republic V, 479a ff., where it is evident that the concern is not with perception narrowly conceived, as "just" is listed as a perceptual predicate). 24 Few scholars have noticed that the examples Socrates considers and rejects are rarely particulars. Nehamas (1975) is a valuable corrective. 25 What actually happened at Plataea, as in most battles, is obscure; but it was evidently thought that some movement of the Spartans saved the day. For a different account of the battle, more flattering to the Spartans, see Herodotus ix. 61-3.
26 You can, however, construct a difficulty, and Socrates may have done so. There is evidence in the Meno 71a and Republic I 354b that Socrates thought that knowing the definition of F is prior to assigning any fea- ture to F. Socrates never makes such a point in the context of the ignorance-proving elenchus, however, where the context precludes disagreement on such issues as whether Courage is noble.
27 A notable exception is Mark McPherran (1985), who argues for a constructivist reading of the second phase of the Euthyphro. 28 "He would be a bold man who insisted that the definition was not Socratic and rescues Socrates' procedure thus" (Stokes, 1986: p. 109). Still, Stokes insists there is "a residue of uncertainty" on the matter. The consensus has been that Nicias' definition expressed a view Socrates held, and that this is evident from the Protagoras and implicit in views Socrates adopts in the Laches. That it is derived from the historical Socrates: Joel (1906: p. 310), Santas (1969: pp. 195-6), and Devereux (1977: pp. 135 n. 15, 136 n. 24), on the basis of: Protagoras 349e- 350c, 358d, 360d; Aristotle Nicomachean Ethics 1116b4-9; EE 1229al4-16; Xenophon Memorabilia iv. 6.11; 111. ix. 1. But Aristotle is likely working from Plato's text, so is not evidence for the historical Socrates. Xenophon alludes to a different doctrine, that those who know what to do in danger are the brave. But that is proba- bly Xenophon's own idea. The best evidence so far is the Protagoras; but this is vulnerable to the objection that Socrates is there engaged dialectically with Protagoras and so would not necessarily say what he believed. But Socrates states his view at 360d etc. as part of his doctrine of the unity of the virtues. We can be sure that Plato's fictional Socrates, at least, held the view in question; and since it is the fictional Socrates who refutes it in the Laches, that is almost enough to raise the question of Plato's intention in having Socrates refute his own doctrine. To clinch the point we will need evidence internal to the Laches. That Socrates endorses Nicias' definition in the Laches: Penner (1973: p. 61 n. 34), as following from these premises: "the brave man is good"; "each man is good in those things in which he is wise." Devereux ob- jects that Nicias' definition does not follow from these premises, but only that the brave man is wise in the respect in which he is good. I support Penner: if the brave man is good only in those things in which he is wise, then he is brave (a fortiori) only in those things in which he is wise. The best Socratic explanation for this would be that it is
wisdom that makes a man brave. Bravery, then, is a sort of wisdom, by inference to the best explanation. Cf. Meno 87d ff. for a version of this argument (pace Devereux 1977: p. 130). The Laches implicitly gives a short form of the Meno argument. 29 See Appendix I on interpretations of the final argument of the Laches.
30 Participants in the Boston Area Colloquium have enriched this pa- per in more ways than I can mention. I am particularly grateful for the comments of Mark McPherran and of a referee for The Proceedings of the Boston Area Colloquium in Ancient Philosophy. As always when I write on Plato, I am enormously endebted to Gregory Vlastos.
31 The issue is, first, whether Socrates disclaims responsibility for the argument at 199c4-dl. Socrates represents 5 there as following from Nicias' logos. Vlastos took this to be the immediately preceding infer- ence, from which (he thought) Socrates hereby distances himself; Devereux and all others take it to be the definition of Courage (1), a more plausible reading for logos. Second is the issue of whether 4 states an identity or a Pauline predication. Vlastos took the latter line, Devereux (rightly, I think) the former. 32 Cf. the Euthyphro's distinction between ousia (essence) and pathos (11a).
Andrewes, A., "The Mytilene Debate: Thucydides 3.36-49" Phoenix 16 (1962), pp. 64-85.
Beversluis, John, "Does Socrates Commit the Socratic Fallacy?" Forthcoming in The American Philosophical Quarterly.