Plato’s Symposium: Issues in Interpretation and Reception

  Lesher, James, Debra Nails, and Frisbee Sheffield, eds. 2007. Plato's Symposium: Issues in Interpretation and Reception. Hellenic Studies Series 22. Washington, DC: Center for Hellenic Studies.

1. The Symposium as a Socratic Dialogue

Christopher Rowe

This essay will make one very specific claim: that the Symposium is properly to be treated as a Socratic dialogue. In one way it will of course be quite uncontroversial to describe the Symposium as “Socratic”: Socrates is on any account the focus of the whole dialogue, which ends as it begins, with a celebration of particular features of his, and includes a kind of tableau vivant in the middle with the great man as Eros. The Symposium is a kind of repertory of details of the character and life of the Platonic Socrates (whatever the relationship between this Socrates and the real, historical Socrates). [1]

What will concern me in the present essay, however, is rather the philosophical aspect of the dialogue: whether, or to what extent, the philosophy of the Symposium is “Socratic.” In the first place, given the recent tradition of modern Anglo-American scholarship, this will mean asking whether the dialogue is closer, philosophically, to the dialogues we modern Anglophones have come to label as “Socratic,” or “early,” or whether it is, rather, closer to—what are called—the “middle” dialogues. To that question, the answer is apparently simple and straightforward: by the criteria enunciated by that great modern Platonist Gregory Vlastos, [2] whose lead the majority of modern English-speaking readers of Plato follow, and encourage others to follow, the Symposium is a quintessentially “middle” dialogue, above all because it not only contains but ultimately pivots on those special Platonic metaphysical entities, the Forms, of which—allegedly—Socrates never dreamed. Never {9|10} mind that, in stylistic terms (as Charles Kahn has reminded us), [3] the dialogue actually belongs to a very large group of dialogues that includes not only the “Socratic,” metaphysically innocent, dialogues but also Phaedo and Cratylus; few modern readers hesitate to associate Symposium with the Republic and the Phaedrus. (Again, I refer primarily to Anglophone readers: for example, “early/middle/late” as a way of dividing up the dialogues, is—as I have discovered from experience—difficult to translate into French, or into Italian, except by way of reference to Anglo-Saxon habits.)

That the Symposium should be so treated, however, is rather odd. Witness the following passage from one of the most brilliant and illuminating of contemporary writers on the Symposium, Anthony Price:

A remarkable aspect of the Symposium is its loyalty to the Socratic psychology of the Lysis… Agathon throws out the truism that love (erôs) is of beauty (197b5). Socrates elicits the thesis that its object is one’s own happiness by a brisk inference: the lover loves beautiful things to have them for himself; to love beautiful things is to love good things, and to have good things is to be happy; hence the lover desires to be happy (204d5–e7). Happiness is a final end; we need not ask why anyone wishes to be happy (205a2–3). This is not yet decisive, for it might apply to love, but not to desire universally. Even in its broadest sense, a man’s loves might be what we may call his projects (whether these be poetic, chrematistic, gymnastic, philosophic, or erotic, cf. 205a8–d8), but not his natural appetites or incidental inclinations. What I love may be altogether a function of the sort of life I wish to lead and the sort of man I wish to be, whereas what I desire may in part ride free of such central evaluations. It might be that, while all love and desire is for things that one lacks (200a9, e2–9), only all love is ultimately for happiness. However, it serves Socrates’ present purpose, which is to say nothing against erotic desire, that he gives no hint of any divergence or conflict of the kind that serves in the Republic to distinguish rational and irrational desires (IV 436b8–441c2). And there is a sequence of particular indications that he is placing all desires within a eudaimonist perspective. He argues unqualifiedly that personified Love is a pauper: loving and lacking beautiful things, it must also lack all goods, for goods are beautiful (201a9–c5, cf. Lysis 216d2). Yet if its loves were only its projects, it would not have to lack any natural or incidental goods that did not {10|11} fall within those. Further, even involuntary genital responses, male and female, are taken to express love (206d3–e1). Finally, love is taken to be evidenced by the behaviour of brutes (207a6–b6), and human physiological processes (c9–e1); yet if we were to extend the term “project” beyond personal ideals to desires that are fundamental though unthinking, it would draw a line that was never Platonic. Rather, we must take the background assumption to be Socratic: happiness is the ultimate goal of all desire, animal as well as human. Erotic desire has then to be accommodated as a special mode of desiring that which all desire desires; its definition is a theorem derived from a Socratic axiom. [4]

What is particularly interesting (and complex) about Price’s position here is that he holds that the Plato of the Symposium has actually abandoned “the Socratic psychology of the Lysis”; he employs it, keeps it on, as it were (so I take Price to be saying), in this dialogue just because “it serves Socrates’ present purpose, which is to say nothing against erotic desire.” Desire in general, according to that Lysis account, will always be innocent, always aimed at what really is good (for us): if anything goes wrong, whether in our relationships or in our lives and actions in general, the culprits will be our beliefs, which are the only things that can go wrong. Price claims to find confirmation of his proposal, i.e. that Plato had already abandoned the psychology of the Lysis by the time of writing of the Symposium, in the analysis of (some kinds of) human relationships that finally emerges from the larger and allegedly later dialogue (i.e. the Symposium), which he takes to show a singular advance over the more limited, and ultimately disappointing, analysis of the Lysis. (Disappointing, that is, to Price, not to me. I am therefore not particularly impressed by the form of Price’s confirmation of his thesis. However, little will hang on this if, as I claim, Price in fact has no success in showing that the analysis of personal relationships offered in the Symposium is significantly different from what we are offered in the Lysis.) [

Many of the detailed issues involved here I shall leave to one side for reasons of space. I introduce the passage from Price not only because it is so clear and useful a statement of what is, for my purposes, and in general, a central feature of the Symposium (its Socratic-style psychology), but because {11|12} it illustrates the abiding influence of what I may call the “Vlastosian” paradigm of Platonic interpretation. I do not claim that Anthony Price thinks as he does merely because he happens to believe Gregory Vlastos; indeed I have already cited a separate argument of Price’s for associating Symposium with Plato’s “middle” period (i.e. that Plato’s treatment of human relationships is maturer, more developed and successful—though as I have already indicated, I dispute that claim). [6] And elsewhere [7] Price gives a lengthier explanation of his reasons for thinking that Plato has abandoned the psychology he uses to ground Socrates’ eulogy of erôs in the Symposium. Yet, despite the independence of his case, Price may nevertheless still stand as an example—if a usefully complex one—of that clear majority of scholars who continue to treat Symposium as a “middle” dialogue. The difference between Price and others of the same persuasion is that he has recognized the nature of the situation more clearly than they have: namely that, here in the Symposium, we find an allegedly “middle” dialogue that nevertheless contains at its core a psychology that (a) belongs to the “Socratic” dialogues (as normally so-called) and (b) is actually, and deliberately (in its pure form), rejected in other “middle” dialogues, notably Republic and Phaedrus. [8] Socratic psychology, on the one hand; Platonic {12|13} Forms on the other. How are we to explain the mix? This is exactly the question Anthony Price raises. [9] However, the answer I shall give will be radically different from Price’s.

So, let us suppose that the Symposium is “Socratic” in psychology, “Platonic” in metaphysics: according to the standard view, the dialogue will still count as “middle,” because it is Plato’s metaphysical turn, according to the same standard view, that is the decisive moment in the dialogues, the one that marks Plato’s coming of age as a philosopher—although, paradoxically (still by the standard view in question), it is his subsequent abandonment of “middle-period” Forms that brings him to true maturity. I refer here, of course, to that {13|14} other allegedly pivotal dialogue, the Parmenides, in which Plato—allegedly—has the Eleatic philosopher criticize a younger Socrates’ version, or versions, of Form-theory, and find them in need at least of serious overhaul.

This, however, seems to me to be rather more than a rhetorical question; for the cost of such a strategy—as indeed Price himself recognizes—will be to have Socrates praising Love on the basis of a theory he no longer believes in or accepts. For his, and his author’s, real views we must apparently supplement the Symposium with the Phaedrus, which contains an account of erotic love based on a Republic-type treatment of desire that divides it into rational and irrational—with irrational desires capable of successfully opposing the former, and causing the agent to act against what he or she believes to be best. Of course we have no way of ruling out the possibility that Plato should have adopted such a strategy—one of suppressing his real views—for the Symposium, which is by any account the most literary of all the dialogues.

So should we treat the Symposium as a “middle” dialogue, but just as early-middle—an alternative I briefly floated before? On this reading, Plato will not yet have given up on the Socratic psychology, even while he has made the crucial move to Platonic Forms. Quite why scholars have not rushed to adopt this solution, apart from their not—perhaps [15] —having seen the size of the problem, is not immediately clear. But I think that it may well be connected with a general view that the Socratic psychology is simply inadequate; that it obviously fails to meet the facts of human experience. Such is clearly Price’s view:

Terry Irwin, and others, [
18] understand Socrates rather as treating desire as deriving from the agent’s beliefs about the good. Clearly, on this reading, Socrates’ view of desire would be one that by the simple rule of charity one might want to see Plato ditching as soon as possible (much as Price wants Plato to have ditched the rather different account of desire he, Price, attributes to Socrates); and when better to do the ditching than at the very moment of writing the Symposium, when he turns his attention specifically to the subject of erôs? Surely, as soon as he began to think about erotic experience in all its variety, he must have seen the inadequacy of the Socratic treatment?

This is not, however, a particularly compelling argument, given that the Lysis, too, begins and (more or less) ends as a discussion of erôs; and no one, at least in the present context, is suggesting that Plato is not serious about his analysis, there in the Lysis, of erôs in Socratic terms. It is, it seems, just taken as read that “middle” dialogues show us the mature Plato, or at least a maturer Plato: a Plato who has already embarked on his ambitious metaphysical program, and must (surely?) have rid himself of that well-meaning, “innocent,” but ultimately useless treatment of desire. Here is Price again:

In my own view this represents a serious underestimation of Socrates’ theory of desire. The theory does not require that felt desires like hunger, or thirst, or indeed lust themselves be seen as expressions of the agent’s desire for happiness. He or she may feel hungry, thirsty, or lustful, but at the same time be {16|17} mistaken about the truly best (most happy-making) objects for such appetites; in such cases, as in any other in which desire seems to lead us to the wrong (because not happy-making) outcomes, it will make sense to say that the agent “didn’t want that”—even if he or she really did feel hungry, thirsty, lustful at the time. The question is about exactly what desire causes, or helps to cause, actions, and the Socratic claim is that this is the universal desire for happiness, or benefit: “if we act in response to thirst or hunger it will be acting in order to be benefited rather than harmed—that is, the desire in question [the desire that causes the action] is not desire for food, say, but desire for the good in this situation which happens to include eating.” [
20] In short, Socrates need not be interpreted as claiming, in the Symposium or anywhere else, that hunger or thirst have happiness as their object; Price’s complaint here falls—as, equally, does any objection to Socratic psychology that is based on the notion that Socrates thinks beliefs (our beliefs about the good) determine our desires. [21] What Socrates thinks, and claims, is that the desire that causes our actions is always the desire for our (real) good, and that we only ever go wrong, do the wrong things, because our beliefs about what that good is are not up to scratch.

Once again, the issues are too complex to be discussed in full detail here. But I may perhaps be allowed to offer, as a hypothesis, not only that Plato was for a long time interested in what I may broadly term the Socratic theory of action, but that he had good reason to be interested in it (because it is a respectable theory). [22] Given that hypothesis, then we so far seem to have no reason for supposing that Plato disbelieved in the theory when he wrote the Symposium and based the dialogue on it; indeed we have considerably more {17|18} reason to suppose that he believed in it, i.e. because otherwise Socrates will be represented as having won the agôn of logoi about erôs on false pretences. One may still wish to allow the possibility that the Symposium represents a kind of part-historical document, showing how an “innocent,” or alternatively perverse, or merely odd, not to say downright peculiar, Socratic-type defense of erôs could outshine more ordinary defenses, even while being built on half-truths. But that seems hardly an attractive option. Once again, the Symposium admittedly is more literary than other dialogues. However, the expectation must surely be that, in common with—so far as I understand the matter—all the other dialogues that Plato wrote, the Symposium will have been written somehow to persuade; if Socrates/Diotima fail to convince his/her immediate audience, nevertheless, so I assume, the purpose must be to persuade us, and it would be odd, if that were the case, that Plato should have had Socrates expound a theory of erôs based on other ideas that he, Plato, would have rejected as untrue to reality. The only argument I am able to construct for supposing anything of the sort would be (1) that the Symposium is a “middle” dialogue (because it contains Platonic Forms), (2) that the “middle” period sees Plato rejecting the Socratic psychology (i.e. in the Republic), and (3) that therefore (?) he must have rejected it, even if he doesn’t say so, by the time he was writing the Symposium. That is clearly an appalling argument, insofar as not all “middle” dialogues can be expected to have exactly the same spread of features, where these are not connected: if there is no connection between Platonic Forms and a Platonic psychology, then once again the best a supporter of the Symposium-as-“middle-period”-dialogue thesis could do would be treat it as a kind of pre-“middle,” half and half, “transitional” sort of dialogue.

For myself, I prefer a more radical solution, for I have serious doubts about the importance, for the whole Socratic-Platonic project, of Platonic Forms (whatever these may turn out to be). That is: not only is there no necessary connection between the hypothesis of Forms (whatever this hypothesis may amount to) and any particular kind of psychology or theory of action, but most of what Socrates and Plato care about will go through irrespective of whatever metaphysical commitments are involved in believing in (Platonic) Forms. Or, to put it another way, Plato’s thinking about Forms is by way of the articulation of certain proposals about which both Socrates and he were convinced in any case, and on which they rely in any case: that there are such things as goodness and justice, things that we can investigate and have some hope of grasping; and so on. If that claim of mine were to turn out to be true, then the introduction of Platonic Forms would not be a crucial moment in the development of Platonic thought, as it has so often, in recent times and {18|19} in certain parts of the world, been thought to be; or, if it is such a crucial point, it is not exactly a turning-point, except perhaps for metaphysicians. And that is really the crux of the matter. It is Aristotle who first fixes on Forms as the, or a, crucial distinguishing mark between Plato and Socrates; [23] and it is Aristotle for whom Forms are a real issue, just because he thinks them so bad a mistake—he wants, and requires, forms of a different sort. But there seems no particular reason why we moderns should follow Aristotle. (“Separation” is the big mistake, Aristotle says; but Socrates obviously did not set out to deny separation. He just didn’t get that far. Some dialogues contain Forms, some do not; it may be that those that do were written later than those that do not, but that—on the view I am proposing—will have no obvious consequences for our actual interpretation of the dialogues. We might be able to say of a dialogue that refers to Platonic Forms that it is, to that extent, Platonic rather than Socratic, but only in virtue of that single feature; and there may well be other features of them of which the same could be said—if only we knew which these were). [24] But by the same token, the fact that the Symposium contains, and even proposes, a Socratic psychology also makes it Socratic rather than Platonic. And to the degree that the dialogue is about erôs, and about desire, I venture that the dialogue as a whole ought to be treated in the same way—that is, as Socratic rather than Platonic; although at the same time I have no desire to prejudge just how different the Platonic project is from the Socratic, if indeed we can in principle properly distinguish them at all, as opposed to marking off certain specific differences. [25] In terms of the standard ways of talking about the Platonic oeuvre, the Symposium seems to have closer connections with the so-called “Socratic” dialogues like the Lysis, the Charmides, or the Laches than with those other dialogues with which we have become used to associating it, i.e. the other so-called “middle” dialogues; [26] of which, in fact, two others—Phaedo and Cratylus—also belong to the earliest stylistic group. [27] {19|20}

Why does any of this matter? It matters because a Symposium interpreted as a close relation of the Lysis will tend to look very different from a Symposium interpreted as a close relation of the Republic; or so it will, at any rate, given some interpretations of the Republic. I refer here especially to treatments of the Form of the Good in the Republic as if it were something entirely different from the good that functions as the object of all desire in what I have been labeling as the Socratic theory of desire. As an example of this approach, I cite Nicholas White:

This is the kind of case that I referred to, very briefly, early on in the present essay, according to which there would after all turn out to be a real conceptual link between Platonic Forms and the abandonment of the Socratic psychology; for now (at least so far as White’s own argument goes) the claim, fundamental to the Socratic outlook, that “all desire is for the good” will simply evoke the question “Which good?” “The non-perspectival good (the Form), or what is good for oneself?” And the two may very well be in conflict. The outcome, White says, is not unlike “the modern [concept] that creates the distinction between self-regarding and broader aims”; [
29] and this already seems to allow in the very kind of mental conflict whose existence Socrates’ schema denies, or, better, [30] explains in different terms. However, of this good, or Good, there is rather little sign in the Symposium, which is framed throughout in terms of the “perspectival” good, the good of the agent. [31] More generally, it is tempting to align true “ascent” passages of the Symposium with the Cave simile in the Republic (if only because this too involves an ascent); and {20|21} this comparison may then take one’s eye away from that very preoccupation of the Symposium with the good of the agent, and lead us to suppose that has a better, or different, story to tell about human relationships than the Lysis. In fact, I suggest, it does not: it has essentially the same story to tell. But then, as Penner and Rowe assert (i.e. in their 2005), that is after all not a bad story.

The outcome of my argument, then, is that the Symposium is Socratic in the sense that it shows Socrates interpreting erôs according to the kind of psychology which Plato allows him develop in the alleged “Socratic” dialogues (especially the Lysis)—and to all appearances doing so with absolute seriousness. No philosophical analysis of the Symposium’s main argument that fails to acknowledge this—so I claim—has any real future. And such an outcome surely makes it much more difficult to treat the psychology around which its central argument revolves as something that Plato came to despise, as he emerged blinking into the light of maturity (and developed his allegedly defining, showcase, Theory of Forms).

However, my argument also has a crucial consequence for our reading of the Symposium as a whole, and in particular for our reading of the accounts of erôs given by others present at the party. It is often claimed that the logos of a Phaedrus, or a Pausanias, even—especially—that of an Aristophanes, is somehow meant to contribute positively towards the picture Socrates puts together from what Diotima, the fictional priestess, told him; or else some other way is found of giving at least partial validation—from Plato’s perspective—to other speakers’ treatments of erôs in the dialogue. (Why else, the argument tends to go, would Plato have allocated them so much space?) But if it is the case, as I have urged, that Plato wishes to recommend not merely Socrates’ treatment of erôs, but the theory that it is built on—if it is the case, as indeed I propose, that it is the theory that Plato cares most about, along with its particular application to the analysis of erôs, then that reading of the earlier speeches in the dialogue, as somehow cumulatively building, or building towards, the Socratic perspective, must be wrong. For none of the earlier speakers has the slightest inclination to understand things in the way that Socrates understands them. Socrates’ perspective is just different, and strange. What is more, it comes as a package: earlier speakers may say things that sound like things that Socrates says; but they are not the same things, because the speakers lack the perspective, the whole system, to which they belong, and without which they are mere disiecta membra. What Plato is doing is to contrast the peculiar Socratic view with more ordinary views, not derive it from them—for the fact is that it cannot be derived from them. In order to appreciate fully what Socrates is saying, one has to throw away what one thought one knew, and {21|22} start again. That itself, it seems to me, is the chief point of the whole dialogue: to celebrate the distance that separates Socrates from his audience, and from the rest of us. [32] {22|23}


[ back ] 1. As it happens, I believe that there is a close connection between the two Socrateses, but nothing much hangs on this in the present context. [I wish to thank the participants at the Symposium conference at the Center for Hellenic Studies in August of 2005, especially Frisbee Sheffield and the anonymous readers, for their comments on the earlier version of this paper.]

[ back ] 2. Vlastos 1991:47–49.

[ back ] 3. Kahn 1996:42–48.

[ back ] 4. Price 1997:254–255 (my emphasis); Price admits in a footnote that “[t]his view was barely advanced [in the first edition, of 1989].”

[ back ] 5. For what is, inter alia, a defence of the Lysis’s treatment of relationships between philoi, “friends” of all kinds, and the theory of desire and action on which it is based, see Penner and Rowe 2005, especially part II.

[ back ] 6. In particular, the claim seems to be based at least for the most part on speculation; Price finds gaps in the account of love given by Diotima and Socrates in the Symposium, and proposes a way of plugging them. But, as Penner and I spell out in Penner and Rowe 2005:300–307, Price has offered no compelling reasons why we should accept his diagnosis, let alone his cure.

[ back ] 7. See Price 1995:8–14 (though I am not sure that these pages add substantially to his case).

[ back ] 8. Compare and contrast Irwin (1995:303): “The conversation between Socrates and Diotima in the Symposium begins with the sexual aspect of erôs, as desire for the beautiful (204d). But this description is soon supplemented or replaced by two others: erôs as desire for the good and for happiness (204e), and erôs as the desire to “give birth in beauty” (206b7). Plato uses erôs not in its usual restricted sense, but to refer to the generalized desire for the good from which more specific desires are to be derived (205a–d). In doing this, Plato implies that he can explain a more specific love of persons, and in particular a more specific love of beauty, by appeal to this more general desire”—and Irwin goes on to suggest that “the Symposium [thus] eliminates the common conception of erôs in favour of the Socratic conception of desire”—which he thinks of as abolishing the distinction between different kinds of erôs, and its connection with the common notion of erôs. (I think it right to say that Irwin, at least, has a different view of “the Socratic conception of desire” from my own view of it; I shall return to that point in a moment.) ¶ So far so good with Irwin’s account, in a way. But he adds in a footnote (chap. 18n12): “If [Plato in the Symposium] accepts the division of the soul defended in the [Republic] and the [Phaedrus], his account of erôs in the [Symposium] will not apply to spirited and appetitive desires. In the [Symposium] he neither endorses nor rejects this division of the soul, since he neither affirms nor denies psychological eudaemonism.” Price seems to me to have shown sufficiently clearly that the latter claim is false: the Socrates of the Symposium unmistakably shows his allegiance to (what Irwin calls) “psychological eudaemonism”—even if that allegiance, according to Price himself, is only temporary, and adopted for ulterior reasons. Irwin’s case perhaps depends on the fact that it is mainly Diotima, not Socrates, who advances the Socratic concept of desire. But insofar as Socrates (a) is made to claim that Diotima is his teacher in matters of love (Symposium 201d), (b) represents her as endorsing “psychological eudaemonism,” i.e. in his report of what she taught him, and (c) ends by asserting that he “is persuaded” of what she said (212b2), then (pace e.g. Price—see following note) it is hard to agree that he “neither affirms nor denies psychological eudaemonism.”

[ back ] 9. See Price 1995:9: “… there is no sound inference from the Platonism of the ensuing metaphysics [sc. in Diotima’s account] against the Socraticism of the antecedent Socratic psychology [though Price thinks that Plato does have Socrates increasingly distance himself from aspects of what Diotima proposes]. The combination is still unexpected—which may be why it is overlooked. What could explain it?”

[ back ] 10. There is a category of “transitional” dialogues, according to the Vlastos paradigm; the Symposium is not among them.

[ back ] 11. As a matter of fact it is quite unclear what a Platonic Form is, and what believing in the existence of such things actually entails; fortunately that does not touch the immediate issues (but see further below, on Nicholas White’s reading of the intersection of metaphysics and ethics in Plato).

[ back ] 12. “By retaining a Socratic psychology Plato can combine what Socrates contrasts: Socrates will tell the truth as he sees it, but in Plato’s eyes that will be half-truth too approving of love by half. It is striking that Socrates makes no distinction between good and bad love such as was drawn by Pausanias … and will be recurrent in Plato [Price gives references to Republic, Phaedrus, Laws]. As we shall see, Socrates remains free of moral error in Plato, for his vision of love is blind to those aspects that are not proper objects of eulogy. A Socratic conception of love is an expression of innocence” (Price 1995:9).

[ back ] 13. See preceding note. (Whether in fact Plato’s notions of good and bad love are quite the same as Pausanias’ must surely be in doubt; but still the general point holds.)

[ back ] 14. The reception of Socrates’ logos by the immediate audience is noticeably somewhat lukewarm (general applause/praise [epainos, 212c4], as against universal, loud/uproarious applause [pantas … anathorubêsai, 198a1–2) for Agathon’s contribution; but then that is hardly surprising, given the strangeness of what Socrates has said, and of his transformation of erôs.

[ back ] 15. So Price suggests (see n9 above [“which may be why it is overlooked”]).

[ back ] 16. So too Kahn, with respectable stylometric opinion (cf. n3 above), except that this places both in the earliest group of dialogues, along with the so-called “Socratic” ones. Price, of course, is saying rather that the Phaedo, like the Symposium, is a “middle” dialogue.

[ back ] 17. Price 1995:13–14. (I add: but that—“a redescription of the phenomena …”—is exactly what Socrates is offering us; the redescription is just not in terms of “mental conflict.” But then Socrates specifically denies that the phenomena “constitute mental conflict,” or at least that model of “mental conflict” that Price represents as an incontrovertible fact of human experience.)

[ back ] 18. Among whom is John Cooper: see his review of Irwin’s Gorgias in the Clarendon Plato series: Cooper 1982; also Cooper 1999a.

[ back ] 19. Price 1995:14n7 (note to the passage last cited in the text above).

[ back ] 20. Penner and Rowe 2005:154, commenting immediately on Lysis 220e6–221b3—a passage which, we think, suggests that appetites would exist even if there were no bad things that resulted from them: our appetites too, that is, in principle contribute to our overall drive towards the (real) good, and bad things result only from the misalignment of our beliefs. See in the index under “desire: supposed irrational desires” for references to a succession of passages in Penner and Rowe that progressively illuminate our view of (Socrates’ view of) appetites. (Penner, it should be added, is not committed to any further interpretation of or gloss on the ideas originally expressed in Penner and Rowe 2005 that may be offered in the present essay; such interpretations or glosses, if they in any way get the book wrong, will be trumped by what is said in the book—which in any case offers the longer and more worked-out kind of reply to Price that his elegantly stated case deserves.)

[ back ] 21. I refer here, once more, to the kind of interpretation of Socratic psychology proposed by (e.g.) Irwin and Cooper.

[ back ] 22. In any case, even if it were a bad theory, that would hardly be a good argument for supposing that Plato abandoned it at any particular point—since he evidently entertained it, or at least wrote as if he took it seriously, for a considerable period, i.e. for at least most of that period in which he wrote so-called “Socratic” dialogues.

[ back ] 23. See Aristotle, Metaphysics 1078b30–1079b11.

[ back ] 24. For a longer treatment of the issues here, see Rowe 2005.

[ back ] 25. Indeed, it is important to stress how close the Platonic psychology actually remains to its Socratic counterpart—something that emerges as soon as one compares Plato with Aristotle, whose position is fundamentally different from Plato’s (e.g. and importantly, in that Aristotle thinks of actions done from anger or appetite as voluntary or willing, while Plato—extraordinarily—treats them as involuntary or unwilling, because contrary to the desires of reason).

[ back ] 26. This is not to deny the obvious fact that the old “middle” dialogues have at least one other thing in common, over against the old “Socratic” ones: sheer scale and size. But even that is not a decisive factor; we should have to decide first what to say about Euthydemus and Gorgias, both of which are fairly large-scale productions.

[ back ] 27. Kahn again (n3 above). I advert to this more or less by the way; we need not suppose that stylistic differences go hand in hand with developments in Plato’s philosophical thinking.

[ back ] 28. White 2002:199–200.

[ back ] 29. White 2002:211.

[ back ] 30. See n17 above.

[ back ] 31. I further, and more radically, propose that even the Republic itself should be seen as building on the outcomes of the Lysis and the Symposium, rather than overturning them: see my “The Form of the Good and the good in Plato’s Republic,” first offered in very preliminary form to the Leventis Conference held in Edinburgh in 2005, and to be published in the proceedings of that conference (ed. Penner and others).

[ back ] 32. The introduction of parts of the soul, in Republic IV, then marks the moment when Plato distances himself from the Socratic psychology—even while, as I have suggested, he manages to preserve some central features of it. The Phaedrus is, in part, his new account of erôs, written from a Republic or post-Republic viewpoint, and acknowledging the possibility of bad as well as good erôs (for which—as Anthony Price and I agree, on different grounds—the theory of the Symposium leaves no room). But that is another story (one told, at least in summary form, at Penner and Rowe 2005:307–312.)