¹ E.g. Kraut, R., Socrates and the State (Princeton, 1984);Google ScholarBrickhouse, T. and N., Smith, Plato's Socrates (Oxford, 1994);Google ScholarIrwin, T., Plato's Ethics (Oxford, 1995).CrossRefGoogle Scholar

² I am not making any claims in this paper about the historical Socrates. In the set of dialogues commonly denominated ‘Socratic’ (on which see next note), we find not only a fairly consistent philosophy and methodology expressed by the character ‘Socrates’, but it is also clear that Plato is sketching a portrait of a realistic Socrates, i.e. of a character who could have existed just as Plato portrays him, living his life in the particular ways Plato describes, with the particular ethical and political views and aims we are familiar with from the early dialogues. I am sceptical, however, about the extent to which we can know that this character actually corresponds to the historical Socrates. The realism and depth of Plato's portrayal of the character ‘Socrates’, together with our knowledge that the historical Socrates really existed and that Plato knew him well, make it tempting to attribute various positions described in the early dialogues to the historical Socrates. But it is quite possible that this assumption may be selling short the literary power and the philosophical originality of Plato. The case for believing that the views expressed in the early dialogues are those of the historical Socrates relies quite heavily on the testimony of Aristotle (see e.g. Irwin [n. 1], ch. 1). The value of this evidence has recently been challenged, however, by Kahn, C., Plato and the Socratic Dialogue (Cambridge, 1996), esp. pp. 79–88.Google Scholar This said, however, I will treat the early dialogues realistically —as though they are about a real person who questions others about their knowledge of virtue and acts the way the character ‘Socrates’ of the early dialogues acts. Thus, I will talk about the effect that Socrate’ irony seems to have on his interlocutors, and on his society, etc., as though Socrates were a real person. I will also speak about the effect the irony of the character ‘Socraters’ has on the readers of the Platonic dialogue, thereby treating Socrates as the character of a literary work. I think this is a reasonable way to proceed, and that the dialogues themselves, with their vividly realistic portrait of Socrates, ask us to understand ‘Socrates’ in this way.

³ By ‘early’ or ‘Socratic’ dialogues I refer to (in alphabetical order): Apology, Crito, Charmides, Gorgias, Hippias Minor, lon, Laches, Protagoras, Republic 1. For an account of the reasons for this division see Vlastos, G., Socrates: Ironist and Moral Philosopher (Cornell, 1992),Google Scholar ch. 2, and Irwin (n. 1), ch. 1. See Kahn (n. 2) for a contrasting view. I treat these dialogues as a group more because of their similarity in style and substance rather than because of any strong belief about when they were actually composed or about their relation to the historical Socrates (see previous note).

⁴ The problem is strikingly similar to the one posed for knowledge by the Argument from Illusion. If one appearance I have that seems correct is in fact false, how can I know whether any appearance I have is true, since all I have to go on is yet another appearance? In this case, the philosopher wants to ‘get outside’ the appearances in the same way as the commentators I have just described want to ‘get outside’ the text.

⁵ Vlastos (n. 3), ch. 1. Gordon, J., ‘Against Vlastos on complex irony’, CQ 46 (1996), 131–7CrossRefGoogle Scholar, calls Vlastos's work on irony ‘the definitive word on Socratic irony for many Plato scholars in recent years’ (p. 131).

⁶ Some recent discussions of various types of irony in Plato, which differ widely in their aims, scope, and accounts of irony, include: Gordon (n. 5); Gottleib, P., ‘The complexity of Socratic irony: a note on Professor Vlastos’ account’, CQ 42 (1992), 278–9CrossRefGoogle Scholar; Griswold, C., Self-Knowledge in Plato's Phaedrus (Yale, 1986)Google Scholar, see esp. Introduction; Miller, M., “The arguments I seem to hear”: argument and irony in the Crito Phronesis 51 (1996), 121–37CrossRefGoogle Scholar; Roochnik, D., ‘Socratic ignorance as complex irony: a critique of Gregory Vlastos’, Arethusa 28 (1995), 39–52Google Scholar; Rowe, C. J., ‘Platonic irony’, Nova Tellus 5 (1987), 83–101Google Scholar; Sayre, K., Plato's Literary Garden (Notre Dame, 1995), see esp. ch. 2.Google Scholar

⁷ See e.g. Irwin, T., ‘Socratic puzzles’, in Oxford Studies in Ancient Philosophy 10 (1992), 241–66Google Scholar; Brickhouse, T. and Smith, N.‘Review of Vlastos’, in Ancient Philosophy 13 (1993), 395–410CrossRefGoogle Scholar; both are discussed below.

⁸ This insight may be most evidenced in Vlastos's choice of title. In the book itself irony is really only discussed in two chapters (1 and 5) and in an additional note (1.1).

⁹ Consider the following recent writing on Socrates (all written after Vlastos's work on Socratic irony) by Scholars who are all either literally or figuratively students of Vlastos: Richard Kraut's ‘Introduction’ and Penner's, Terry ‘Socrates and the early dialogues’ in R., Kraut (ed.), Cambridge Companion to Plato (Cambridge, 1992),Google Scholar Brickhouse and Smith's Plato's Socrates, and Terence Irwin's, Plato's Ethics. In Irwin, Kraut, and Penner I do not believe the word ‘irony’ is mentioned even once. Brickhouse and Smith (p. 32) do refer to irony, but only in order to dismiss it, claiming that it would destroy the possibility of rationally reconstructing Socratic philosophy: ‘unless there is some answer to be given [that is sincere and not ironic about Socrates’ profession of ignorance], we might as well despair of reconstructing a “Socratic philosophy” from the many arguments and assertions we find him making, for these too, might only be expressed ironically’. Kraut argues strongly for the idea that the Platonic dialogues ought to be understood as straightforwardly presenting the doctrines of Socrates and Plato. Kraut writes: ‘Our best chance of understanding Plato is therefore to begin with the assumption that in each dialogue he uses his principal interlocutor to support or oppose certain conclusions by means of certain arguments because he, Plato, supports or opposes certain conclusions for those reasons. … This methodological principle is not an a priori assumption about how Plato must be read, but is rather a successful working hypothesis suggested by an intelligent reading of the text and confirmed by its fruitfulness’ (p. 29). By implied contrast, the unintelligent reading is one that Plato believes as (at least sometimes) problematic, not necessarily presented straightforwardly in the arguments of the primary interlocutor (usually Socrates). Now Kraut is aiming his arguments mainly against an extreme position that maintains that Plato's real views lie in the ‘unwritten doctrines’, and that the content of the dialogues themselves can be no more than an obscure hint at what those doctrines might be. But Vlastos's claims about the importance of irony (at least with respect to the Socratic dialogues) raise the possibility of differing positions that are not nearly so extreme. Therefore it is striking that Kraut in his essay never considers Vlastos's claims about the philosophical relevance of irony, or mentions it as part of Socratic philosophy. Even Kahn (n. 2), who is concerned to focus on the literary form of Plato's work as well as its philosophical significance, does not discuss irony in any detailed or systematic way. 10 As I believe at least some are, including the account of the disavowal of knowledge, and therefore also the account of the disavowal of teaching—but I will not argue this here.

¹¹ Institutio Oratorio 9.22.44, quoted in Vlastos (n. 3), p. 21.

¹² Of course, these should not be forestalled too much; one who never takes things appropriately seriously (laughing at a funeral, etc.) is a person who cannot face reality, while someone who takes every misfortune as though it is the end of the world is overly serious and has no sense of humour in response to life's inevitable misfortunes. This sort of irony then seems to me to fit the model of an Aristotelian virtue, representing a beneficial quality that is susceptible to both an excess—over-seriousness—and a defect—one who can never be serious. But note that this is not how Aristotle uses εíρωυεíα, which means for him a sort of false modesty; see esp. N.E. 4.7.

¹³ Vlastos does not notice this aspect of this example, treating it as a case of pure mockery.

¹⁴ There is some similarity between the effect of irony and the rhetorical figure of litotes in that both are tropes which hold something back: ‘I really love you’ said ironically is still different and more distanced than ‘I hate you’, and similarly, ‘not bad!’ is different form ‘great!’. It is not the case that the ironic form stings less—in fact, it might sting more in that, in this example, the speaker may be showing distance and detachment from a situation by not giving in to the straightforward expression of his or her feelings.

¹⁵ The importance of opposites is also apparent in a slightly different use of ‘irony’ where we speak on an event, rather that a statement, as ironic. A person, after spending twenty years protesting and lobbying against the military, ends up becoming a general. We call this ironic because he has ended up as precisely what he worked against his whole life. But in (13) Mae West does not say the opposite of what she really means.

¹⁶ See esp. ‘Socrates’ disavowal of knowledge’ in Vlastos, G., Socratic Studies (Cambridge, 1994), pp. 39–66.Google Scholar

¹⁷ Vlastos, (first in ‘Socratic irony’ in CQ 37 [1987], 79–96CrossRefGoogle Scholar, then in [n. 3], ch. 1 and additional n. 1.1) connects irony to four areas on Socratic thought—the disavowal of knowledge, the disavowal of teaching, the disavowal of politics, and Socratic eroticism—calling them collectively ‘Socrates’ complex philosophical ironies’.

¹⁸ Gordon (n. 5), p. 134, emphasizes the idea of ‘incongruity’ between these different categories.

¹⁹ Here there is some conceptual connection with metaphor. A statement can be called metaphorical in one ordinary sense only by contrast to the literal truth—the metaphor must, as it were, be translatable. Can we call an expression metaphorical while lacking its literal rendition? I suppose something of the sort is intended in a remark like ‘all language is metaphorical’, which suggests that language has no non-metaphorical level. Whatever this might mean, using the concept of metaphor in this way would be quite different from the ‘ordinary’ way—where a metaphorical expression stands in for an idea that can be rendered literally. Something similar is at issue with irony. Someone of a more Deconstructionist stripe might want to claim that all Plato's writing is ironical. Whatever is meant by ‘ironical’ in such a statement, and whatever truth it might contain, it should be clear that this is not the sort of irony I am discussing. For a statement to be ironic in the sense I am interested in, something must be held as the truth against which a statement is then understood as ironic. On this account, then, it makes no sense to say something like ‘all of Plato's writing is ironic’.

²⁰ Complex irony is controversial because it depends, as I discussed as an example above, on agreeing with Vlastos's account of certain versus elenctic knowledge. Recognizing conditional irony (at least in many clear cases) will not be dependent on adopting any particular understanding of Socratic philosophy.

²¹ Of course, from the denial of the antecedent the denial of the consequent does not follow logically. But here we are not dealing simply with the contents of the antecedent and consequent but with the fact that Socrates appears to assert both the entire conditional as well as to hold that the antecedent is false. This suggests at least strong doubts about whether Socrates believes that the consequent is true, especially in the absence of some other antecedent which implies the consequent.

²² See also: Ap. 19el–4, 20b9–cl; Eu. 513–4, 6a9–b2; Hi.Mi. 376b4–6; Pr. 319a8–9; Symp. 218d6.

²³ See Vlastos (n. 3), ch. 1 and esp. ch. 5. As we shall see, Vlastos must explain how interlocutors can partially or totally fail, as they often do, to grasp Socrates’ irony without it being considered deceptive. Both Irwin (n. 7) and Brickhouse and Smith (n. 7) argue that there is a conflict between Vlastos's attribution of irony to Socrates and his contention that Socrates never cheats—both suggest that the attribution of irony is incorrect.

²⁴ Vlastos, focused as he is on the claim that he is reconstructing the philosophy of the historical Socrates, has lost the idea that irony must at least also be a literary device employed by Plato. I make no claims about the historical Socrates.

²⁵ The distinction between the audiences of the inner and outer frame cuts across Gottlieb's ‘in-crowd /out-crowd’ distinction (n. 6, p. 278); see also Gordon (n. 6), esp. p. 136. I agree with Gordon that Vlastos's account needs expanding, but I expand it in a different direction than she does.

²⁶ I speak of Socratic and not Platonic irony because all of the examples I am considering come from the character ‘Socrates’, not because it belongs to the historical Socrates. Of course in one sense the irony is Platonic, in so far as Plato is the author of the entire text, including the irony of the character Socrates. But what would more interestingly be called ‘Platonic’ irony would involve the claim that Plato is being somehow ironic with respect to a dialogue or dialogues as a whole, so the irony is no longer between Plato's character ‘Socrates’ and the reader, but between Plato and the reader. This would be a more elusive form of irony, which I have nothing to say about here. But see Griswold (n. 6) and Rowe (n. 6) for discussions of types of Platonic ironies.

²⁷ Irwin (n. 7), pp. 246–7. Irwin goes on to argue that it is implausible to claim that a careful interlocutor could discern one of Vlastos's paradigm examples of complex irony: the disavowal of knowledge.

²⁸ According to Brandwood, εἰρυεíα,εἰρωυεὐoμαι,εἰρωυικός, and εἰρωυικŵς, and all their variations in form, appear only fourteen times in Plato, only six of those in the Socratic dialogues. None of those instances is identical to the concept of irony as it has been defined above; they all involve the idea of pretending or shamming, common to the meaning of εἰρωυεία, at the time. The same is the case for its uses in Aristotle: see esp. Eudemian Ethics 2.3, 1221a6, a25 and N.E. 2.7 1108a20ff., and 4.7, passim.

²⁹ For a clear example of εἰρωυεία see Pr. 334c8–d1 where Socrates tells Protagoras to make his speeches short because he is so forgetful. Soon (342a–348b) Socrates himself will give the longest speech of the dialogue. We have an example here of a false humility or feigned incapacity, typical of εἰρωυεία. We will see that Socrates often engages inεἰρωυεία as well as conditional irony.

³⁰ Since I am concerned here to critique Vlastos's account of this passage, I have used his translation of φα⋯λoς as ‘stupid’. But it might also be translated as ‘base’, and perhaps this translation might make the conditional irony even clearer than Vlastos's ‘stupid’. If Socrates really has the power to make Alcibiades a better man, Alcibiades would be trying to secure something truly valuable (by Socrates’ lights) and must therefore appreciate this value. Alcibiades, then, would in fact not be base in so far as he is pursuing excellence. Thus the sense in which Socrates literally means the conditional is even more apparent. I owe this point to Jennifer Whiting.

³¹ This example is similar to the Mae West example discussed above.

³² The remaining antecedents are: in (2), knowing how the youth are corrupted and who corrupts them; in (3), knowing how to care correctly for the young so that they may be as good as possible; and in (4), since he has the knowledge in (2) and (3), starting out from the correct beginning of properly caring for the youth.

³³ Some of the clearest examples include: 5a3–4, 6a9–b2, 9b2–3, 15e5–6.

³⁴ See 5a for the avowal of knowledge.

³⁵ Although, as I am about to discuss, Lysimachus’ suggestion of a vote is roundly criticized, it should not be understood as an obviously stupid idea. Lysimachus is looking for someone who knows what he does not, and the idea of finding a majority of knowledgeable people who claim the same thing is a reasonable approach. If two doctors offer conflicting opinions abut the necessity of surgery, it is reasonable to go to a third to seek a ‘majority opinion’. Socrates will criticize this because the aim of the dialogue will be to show that none of them knows what courage is, so that simply amassing a majority opinion of non-knowers will be of no help. But Lysimachus is considering them all as at least more knowledgeable than he himself is.

³⁶ Here as elsewhere, see e.g. Cr 47a-d.

³⁷ Contrast Roochnik, D. (Art and Wisdom [Penn State, 1996, p. 96)Google Scholar who calls Socrates’ irony here ‘manifest’. To us, this is true; but in the context of the Laches alone it is not.

³⁸ As Socrates himself says at Meno 71c. Again, we are faced with a conflict between the manifest fact that no one has successfully withstood an elenchus about virtue, and Socrates says as much in the Meno, and that nevertheless he is here attributing knowledge of virtue to Laches and Nicias. Here is an example of how one can argue that a passage contains irony. We have no reason to think, by contrast, that the claim in the Meno is ironic since it is consistent with all the actual results in the other dialogues.

³⁹ See e.g. G. Vlastos, ‘The Socratic elenchus: method is all’, in Vlastos (n. 16), pp. 1–37.

⁴⁰ Also, as I have mentioned, Socrates takes certain actions as implying a claim to knowledge—such as Euthyphro's prosecution of his father or Laches‘ and Nicias’ speeches on the education of the youth.

⁴¹ I have benefited from the comments of Steve Strange, the Editor, and an anonymous referee. I am especially grateful to James Barrett, Bill Vasiliou, Jennifer Whiting, and Nancy Worman for their criticisms and suggestions.