Jump to ContentJump to Main Navigation
Plato's SymposiumThe Ethics of Desire$

Frisbee Sheffield

Print publication date: 2006

Print ISBN-13: 9780199286775

Published to Oxford Scholarship Online: September 2007

DOI: 10.1093/acprof:oso/9780199286775.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 08 December 2016

(p.227) Appendix Socratic Psychology or Tripartition in the Symposium?

(p.227) Appendix Socratic Psychology or Tripartition in the Symposium?

Plato's Symposium
Oxford University Press

(p.227) Appendix

Socratic Psychology or Tripartition in the Symposium?

Appendix Socratic Psychology or Tripartition?

IT is not altogether clear how the psychological theory of the Symposium should be seen in light of other Platonic dialogues such as the Meno and the Republic. Recent scholarship seems divided on the issue of whether and, if so, in what sense, the theory of erōs in the Symposium assumes, or is compatible with, (a) what is referred to as ‘Socratic’ psychology, that appears in the Meno, for example, or (b) the psychological theory of the Republic. The Socratic view is often characterized by the claim that all desires are rational desires, and the psychological theory of the Republic by the inclusion of non‐rational motivation in the soul and its division into parts (leaving aside, for the moment, the issue of how those parts are conceived). Price, Irwin, and Rowe all agree that the Symposium contains what they refer to as the ‘Socratic conception of desire’, or ‘Socratic psychology’.1 Socrates claims that all erōs is for the good (205e7–206a1), and that happiness is the ultimate aim of all erōs (205a1–3). For Price, although Plato may not be (p.228) actually committed to a Socratic theory of desire any more (his metaphysical commitments in the Symposium suggest distance from his Socratic heritage), it serves the purposes of eulogy that Socrates ‘makes no distinction between good and bad love’.2 All desire, on such a view of the Symposium, is rational in the sense that it is based on considerations about the good. There is no evidence of non‐rational motivation of the sort that appears in the discussions of the Republic, and so some reason to take it that the soul is a unitary and exclusively rational entity. For Hobbs and Nehamas, however, the connection between Socrates' account of erōs and the tripartite theory of the soul in the Republic is a strong one. For Socrates divides different lovers into those concerned with physical production and those concerned with the production of fame and philosophy (esp. 208–12).3 It is not quite (p.229) clear to me whether the position is that the Symposium is compatible with tripartition, or whether it is the stronger view that the text assumes it. Nehamas sometimes indicates the latter; for he uses the idea that the soul has parts to argue for an inclusivist reading of the philosopher's ascent.4 In the following I offer no more than a few cautionary remarks designed to bring out the difficulties of deciding the issue one way or another.

There are at least three relevant claims that Socrates makes about the nature of erōs. First, Socrates claims that ‘the whole of desire for good things and for happiness is the supreme and treacherous erōs to be found in everyone’ (205d1–3; trans. Rowe). Second, he claims that everyone always desires to possess good things (205b1, 206a11–12 with 206b1).5 And third, that ‘there is nothing else that people desire except the good’ (205e7–206a1). We need to consider the scope and implications of these claims. The second claim might be taken to mean only that at any given time it will be the case that everyone has a desire for the good, and not that this is the only desire that they have. But a stronger sense of the claim whereby it means that everyone always has a desire for the good and only for the good gains some force from the second claim that ‘there is nothing else that people desire except the good’ (205e7–206a1). If everyone always desires the good, and there are no desires other than those for the good, then this is very suggestive of a ‘Socratic’ psychological picture. All desires (at least those under consideration here) will be rational ones in the sense that they are all and only desires for what the agent perceives to be good (whether or not she is mistaken).6

The idea that all erōs springs from, or consists in, a belief as to what is good was highlighted in the earlier account of the nature of erōs. Eros is, by nature, ‘a schemer after the beautiful and the good’ (203d4). The nature of a (p.230) desiring agent is such that it is moved towards something characterized as kalon or agathon in some respect, and the object in question is desired under that description (201a8–10, b6–7, c4–5, 202d1–3, 203d4). Now, if it is the nature of all erōs to be based on, or to consist in, a belief as to what is good, then we must surely agree with Price, Irwin, and Rowe that this is strikingly similar to ‘Socratic’‐sounding claims made in other dialogues.7

Whether such a reading is enough to rule out the psychological theory of the Republic though, depends on how one interprets the details of that theory, and the scope of this one in the Symposium. According to a prevalent view, part of the novelty of the psychological theory of the Republic is that it introduces desires for simple, unqualified, objects such as for ‘drink itself’ (438a1–5, cf. 439a9–b1) which do not involve considerations about the good. It is the presence of such desires that pull ‘like a beast’ (439b4) away from the commands of reason, that forms part of Socrates' argument for the division of the soul in Republic 4.8 If the Socrates of the Symposium is arguing that all desires are rational ones, then he is ruling out ‘brute’ desires of the Republic's kind. All erōs (contra Pausanias) is ‘heavenly’ by nature and directed towards good things. This is not, of course, to say that we will not have to struggle against error as to the sorts of things that are such as to satisfy desire. But the struggle, if there is to be one, will be between our beliefs about what sorts of things are good and such as to satisfy erōs, and not one between desires. This explains why Socrates' account in the Symposium is heavy on the epistemology (e.g. ‘the ascent’).

Although this reading has some force, there are considerations that make it difficult to rule out a more complex psychological picture. First, the claim that there is nothing else that human beings desire except the good is not incompatible with the Republic. After all, Socrates claims there that ‘This [the good] every soul pursues and does all its actions for its sake’ (505d11). This need not (and, on the standard view, is not) taken to imply that every desire—even those of the lower parts—is for the good. In light of the considerations of Book 4, one might rather take it that even when an agent is driven by a lower desire, it is reason that decides which desire to follow on the basis on an overall judgement about the good of the agent. This kind of overall judgement is the particular provenance of the rational part, though (p.231) in this instance its role will be somewhat stunted (441e4, 442c5). In this way one can see that the good enters into all the soul's pursuits, and that the soul is dominated by a non‐rational desire. This is not uncontroversial. But such a view accommodates the claim that we all desire the good and the possibility of non‐rational motivations in the soul. Hence, if one argued that Socrates' claim in the Symposium is restricted in some such way, then it would be perfectly compatible with the psychological theory of the Republic. If the claim that ‘there is nothing else that human beings desire except the good’ is restricted to saying that when a human being desires it will employ the resources of the good‐directed rational part, then one leaves room for the possibility of non‐rational desires with no concern for the good. One might draw on the fact that it is anthrōpoi rather than epithumiai or some other desire term which is the subject of the relevant claim in the Symposium (205e7–206a1); it is not said that each and every desire is for the good.

The scope of Socrates' claims in the Symposium is not clear. Is Socrates referring to all desires when he claims that there is nothing else that people desire except the good? Or is he referring to a specific kind of desire (e.g. rational desire) which is always, and only, for the good? All that is, in fact, implied by the first claim that ‘the whole of the desire for good things and for happiness’ is erōs is that whenever one has a desire for good things, this is a case of erōs. This leaves it open whether one can also desire things that are not for the good (e.g. drink as such). The point is just that such desires will not count as cases of erōs. If that is the point, then this leaves room for the sort of non‐rational motivation characteristic of the psychology in the Republic. In the claims highlighted above Socrates nowhere states that each and every desire (ἐπιθυμία) is for the good; he says only that erōs is for the good, and that the epithumia for good things is erōs. There is some terminological slippage in his account between different desire terms. Socrates does not stick to erān and cognate words exclusively when describing the operations of erōs. But that just shows that all erōs is a case of desire, and just because all erōs is a case of desire (for good things) that does not imply that all desire is a case of erōs and so after the good. Evidence of the latter claim, broader and stronger than the first, is needed to argue for a clearly ‘Socratic’ account and incompatibility with the Republic. And this we do not have.

Restricting the scope of erōs in this way makes good sense of the distinctive nature of erōs' characteristic activity. Recall that erōs involves pregnancies and beautiful media and so on (206b1–d5). Are we really to suppose that all desires, even those like hunger and thirst, express themselves in such a way? But it makes less sense, perhaps, of two further features of the account. First, Socrates ascribes erōs to animals in the Symposium. If he meant only to discuss rational desire, why did he include creatures explicitly excluded from (p.232) logismos (207b6–c1)?9 Second, he includes sexual desire as a species of erōs for the good (205d6). If even our sexual desires are counted as cases of desire for the good, as well as non‐rational animal behaviour, then why not include thirst, or any other epithumia one cares to consider? Such cases do not sound as if Socrates is talking exclusively about a certain kind of rational desire for the good. A natural response here is to say that even in such cases, those desires we share with animals (e.g. sexual ones) still involve the perception of some kalon or agathon feature of the object in question; they need not be construed as ‘blind drives’. Such cases of erōs will not involve a rational grasp of the kalon (indeed most cases of erōs will not involve that), but there will be some cognitive grasp of the value of the object of pursuit in such cases if they are to count as cases of erōs. Some lower‐grade cognitive endowment will be sufficient for animals to perceive a thing as kalon and to take steps to procure that desired thing. Socrates does not specify just what cognitive endowment animals do have, but some cognition, such as doxa perhaps, might inform their erotic experiences. In the Timaeus animals are said to have doxa (77a–c). If we broaden our conception of what counts as a rational desire in some such way, then one can accommodate such cases within a ‘Socratic’ picture.

But if we do so it would seem that we are also closing the gap with the Republic. For one might broaden the range of desires with a concern for the good in the Republic and so bring the two psychological theories together. One might argue (with Carone, for example) that even the lower parts of the soul in the Republic have a concern with what they take to be their own distinctive good.10 They will not exhibit this concern in the same way as reason: the rational part is the only part with a prioritized and inclusive conception of the good, whereby considerations about the value of the object ground the desire, and take into account the overall good of the agent.11 But if it is the case that they can still, in some sense, be said to concern themselves with (what they take to be) the good, then even lower drives can be seen to exhibit certain characteristic features of ‘Socratic’ psychology. And if so, then the presence of those characteristics in the Symposium does little to decide the issue one way or the other. We need (p.233) more detailed information than we get in the Symposium about the nature and manner of the pursuit of the good in each case to drive a wedge between that account and the Republic.

I turn next to the arguments in favour of tripartite psychology in the Symposium. Those who argue in favour of tripartition in the Symposium do so because of Socrates' later division of different desiring agents. As we have seen, the desire for eudaimonia manifests itself in creative activity in the presence of beauty because this is the distinctively mortal way in which it can achieve a share of divine happiness. It is the account of the creative activities of different desiring agents as they strive to achieve this aim that is apparently suggestive of tripartition. It begins with the following distinctions. First, Socrates claims that there are some human begins who are pregnant in their body and some who are pregnant in their soul more than in their body (206c1–3 with 209a1–2). So, when human beings engage in the kind of productive activity required to achieve a share of eudaimonia, they do so in a way fitting to their nature (i.e. in accordance with whether they are more pregnant in body or soul). Those who are more pregnant in body produce physical offspring (208e3), whilst those more pregnant in soul produce psychic offspring of various kinds (e.g. poems, laws, or philosophical logoi, 209d1–210d5). Second, he divides his account of the productive activities of different desiring agents into the so‐called lower and higher mysteries of erōs. This distinction is one based on differing conceptions of what will secure good things and happiness: those of the lower mysteries are honour lovers (208c3), and those of the higher mysteries are lovers of wisdom (210a–212a). Now, Hobbs and Nehamas take it that those who engage in productive activity of a physical nature manifest the tendencies of the appetitive part of the soul familiar from the Republic. Those who engage in productive activities of a psychic kind for the sake of honour manifest the drives of the thumoeidēs. And those who pursue wisdom manifest the drives of the logistikon. Such aims, argues Hobbs, ‘undeniably overlap with those of appetite, thumos, and reason.’ Nehamas agrees with this view: ‘The two lower kinds of lovers—those who reproduce themselves physically and those who are made famous by their accomplishments—correspond to the two parts of Plato's divided tripartite soul, which makes in that way a clear if indirect appearance in the Symposium.’12

There are numerous difficulties with this view. It is questionable whether we can, in fact, determine three distinct aims of this sort in the Symposium's account. For those who engage in productive activities of a physical kind are included amongst the honour lovers and make their appearance accordingly (p.234) in the lower mysteries of erōs which is characterized by this tendency. The love of honour is definitive of the lower mysteries and dealt with from its start (208c3) and continues until the start of the higher mysteries (209e4). Those who are pregnant in their body and turn towards women are mentioned at 208e1, the clear implication being that they are being considered as members of this group. One should not be surprised at the inclusion of those who are pregnant in their body and beget physical offspring amongst the honour lovers (209a2–3). In the Laws the desire to be remembered by one's descendants is cited as a reason for marriage and the production of offspring (721b–d; cf. Rep. 618b, for the fame which derives from one's ancestors). Also included in the lower mysteries as honour lovers are those who manifest the sorts of concerns more characteristic of the Republic's thumoeidēs: those who desire fame by arranging cities and households and writing poems and laws and so on (209a1–e4).13 So, it seems to be the case that the production of children and the production of poems and laws and such like are both being considered as different ways in which one can secure honour. If that is the case, then there is little evidence for separating out the ‘body‐lovers’ as definitively appetitive in nature.

It is not as if Socrates could not have availed himself of a more Republic‐style tripartite distinction at this point in his account. Earlier it was claimed that examples of the desire for good things and happiness range from the love of money‐making and the love of sports to the love of wisdom (205d4–5). The love of money is characteristic of the appetitive part in the Republic, so why didn't Socrates employ that specification of aims to different desiring agents here, if he was concerned to demarcate three distinct aims and corresponding parts of the soul? Those associated with the body here are concerned not with the gratification of the appetites through the having and spending of money, but with ‘immortality and memory’ through the production of bodily offspring which they believe will provide them with ‘eudaimonia for all time to come’ (208e4). Honour and wisdom, the two goals mentioned in the text, are presented as different ways in which the desiring agents in question conceive of a good central to their conception of a flourishing life. Even the ‘body‐lovers’ exhibit a concern with ‘eudaimonia for all time to come’ (208e4).

The picture that emerges, then, seems rather to be the following. All desiring agents seek good things and happiness (206a11–12). There are those who specify the good central to a flourishing life as honour (208c3), and those who believe wisdom to be central to a good and happy life (p.235) (210a1–212a6). People try to secure this good for themselves by engaging in productive activities of various kinds, and these will be determined not just by a conception of the good in each case, but also by the tendencies predominant in their nature. Those more pregnant in body will try to secure fame through the production of children, providing for themselves ‘immortality, memory and happiness, as they believe, for all time to come’ (208e3–5). Those more pregnant in soul can secure either fame, or wisdom through the production of psychic offspring of various kinds. Socrates' account, in short, functions with two natural tendencies in human beings, those of a physical and those of a psychic kind (206c1–3 with 209a–6), and two specifications of a good central to eudaimonia in each case (honour and wisdom). So, although there are three productive activities mentioned: the production of children, poems and laws etc., and philosophical logoi/true virtue, these do not fit neatly into the three distinct aims characteristic of tripartition in the Republic.

This is a confusing picture for anyone who wants clear evidence of tripartition in the Symposium. But the fact that Socrates' account treats those with predominantly physical tendencies who produce children alongside those who try to secure fame through poetry and law‐making, does not provide evidence against a tripartite psychological undercurrent either. It may be the case simply that Socrates has other distinctions he wishes to foreground in this context such as, for example, an epistemological distinction between those who are concerned with appearances and those who concern themselves with truth and reality. From this perspective both those with appetitive tendencies and those with thumos tendencies can be considered to belong to the same class, in a way that is perfectly compatible with other more complex psychological distinctions. In the Republic those who are dominated by one of the lower parts of the soul (e.g. money lovers, or honour lovers) are sometimes classed together in this way. In Book 9, for example, money lovers and honour lovers are treated alike as those who enslave their reason and are concerned with becoming, as opposed to being; they do not have knowledge of the truth (583b, 585b–587c). The point, I take it, is that those who indulge the lower parts of the soul share some similar traits—from an epistemological point of view. They are both those who fail to develop their reason properly and are concerned with appearances, rather than the truth. There are good reasons for thinking that the epistemological viewpoint is the operative contrast in the Symposium. This is borne out by the contrast made at the end of Socrates' account between productive activities grounded in the truth and productive activities grounded in images (212a4–6). A clear contrast between the desiring agents of the lower and higher mysteries is that the former fail to engage in the right (p.236) kind of method that leads to the form, as we have seen. Consequently their productive activities produce only images of the virtue produced by the philosopher (212a4–6). This contrast is one grounded in different epistemological states, rather than in a psychological structure. But since the Republic provides evidence that this sort of contrast is compatible with other operative contrasts (e.g. the psychological divisions of Book 4), the presence of a twofold classification of desiring agents in the Symposium does not rule out tripartition. The operative epistemological distinctions may be simply a matter of emphasis, rather than indicative of specific psychological commitments. They do, however, dilute the strongest piece of evidence in favour of the tripartite undercurrent.

So far, then, although we cannot rule out the psychological divisions of the Republic, the evidence in favour of tripartition is weak. And even if it were the case that the aims of the desiring agents in the Symposium correspond to the aims of the three parts of the soul in the Republic that would still not be enough to show any more than compatibility with such a theory. Even if it were the case that we could determine three distinct kinds of motive operative in the account, this in itself does not get us to a division of the soul. The claim that there are three kinds of motive in human life is present in the Apology (29d7–e3), and the Phaedo also discusses three kinds of character (82c2–8). Neither of these texts function with a tripartite psychology. So even if we could determine three distinct kinds of motive and corresponding psychological tendencies, we would still need to establish that each individual has all of these psychological tendencies within him.14 It could, of course, be the case that Socrates lays out the forms or kinds of motivation in the Phaedo and the Symposium which receive greater articulation and grounding in a tripartite psychological structure in the Republic. But that would be a different, and certainly weaker, claim.

I turn finally to considering the possibility of psychic conflict in the Symposium. As is well known, in the Republic Plato grounds different forms of motivation in the divided soul, in part, to account for cases of psychic conflict. Socrates assumes that if a desire for something conflicts with a simultaneous aversion to the same object, then the aversion belongs to a different soul part (440a5–6). Although we have no such principle in the Symposium, we can consider whether there is evidence of the sort of psychic conflict that plays a role in the division of the soul in the Republic, and perhaps consider this to be indirect evidence of such a veiw. Is there any evidence of conflict of a kind that can't be dealt with without supposing a (p.237) division of the soul? Of if there are cases of conflicting desires mentioned in the text can they be understood along ‘Socratic’ lines? Although there is no discussion of conflict of desires in the mainstream of Socrates' account, it might be thought that there is, in fact, rich evidence of psychological conflict provided by the case of Alcibiades. Perhaps, it could be argued, his confessional autobiography provides evidence of the very phenomenon that plays such a central role in the division of the soul in the Republic. If so, then perhaps we need to assume a divided soul, after all, in order to understand such conflict. Or, we might suppose just that Plato is signposting a problem that remains to be dealt with when he comes to deal with the nature of the soul in more detail.

Let us examine this case. When Alcibiades recounts the details of his interaction with Socrates he explains how, though he is moved by the power of his words and the goodness of the man himself, he fails to do what he advocates. Instead, he indulges a desire for honour (216a2–c3). He is both attracted to the path of Socratic wisdom and to the honour from the crowd. Here, again, we have the two familiar goals from the lower and higher mysteries, and now they appear to belong to one and the same individual. Let us examine the details. Alcibiades describes himself as follows:

What's more, even now I'm conscious that if I were prepared to listen to him, I wouldn't be able to resist, and the same things would happen to me. For he forces me to admit that although there's much that I lack myself, it's myself I neglect and do the Athenian's business. So I forcibly stop up my ears and I'm off, as if I were running away from the Sirens, to prevent my sitting there and growing old beside him. He's the only person in the world towards whom I have experienced what one wouldn't suppose I had in me—feeling ashamed towards someone, no matter who; it's only towards him that I feel it. For I'm conscious that I'm not capable of arguing against doing what he tells me to do, and that whenever I leave him, I'm giving in to my desire for the honour that comes from ordinary people. In any case I'm off and away from him like a runaway slave, and when I see him I'm ashamed because of what's been agreed between us. Often, I'd happily see him gone from this world; but then, if that were to happen, I'm well aware that my grief would be much greater, and so I just don't know what to do with this man (216a2–c3; trans. Rowe).

Alcibiades is evidently experiencing some kind of conflict. But exactly what sort of conflict is this? On the one hand Alcibiades wants to listen to Socrates and to change his ways, on the other he desires honour from the crowd. Alcibiades has both tendencies within himself, but it is evidently honour that wins out; for Alcibiades ‘stops up his ears’ and runs away from Socrates. (p.238) But when he comes into contact with Socrates he experiences shame at his actions. Now, notice that the conflict here does not seem to be one where Alcibiades experiences a desire for Socrates' wisdom and aversion for it at the same time. It seems rather to be the case that when he is in the presence of Socrates he is enamoured of his words and desires to improve himself, but whenever he leaves him he gives in to a desire to please the crowd. The desire for honour occurs at a later time, when he no longer experiences the desire to improve himself. So, there are two competing motivational tendencies here—for honour and wisdom—and they do reside within one and the same individual—Alcibiades—but they do not occur at the same time in relation to the same object. Alcibiades' conflict seems to be between a desire for one perceived good at one time (wisdom) and a desire for another perceived good at a later time (honour). This, then, does not provide evidence of the kind of psychological conflict that motivates the division of the soul in the Republic.

Alcibiades does not appear to be exhibiting the marks of akrasia. In Davidson's classic formulation: ‘In doing x an agent acts incontinently if and only if (a) the agent does x intentionally; (b) the agent believes there is an alternative action y open to him; (c) the agent judges that, all things considered, it would be better to do y than x.’15 It is the presence of this phenomenon that threatens the ‘Socratic’ claim that we all desire and act in light of some perceived good. But there is little evidence that Alcibiades judges that all things considered it would be better to stay with Socrates (the better course of action) and instead indulges a desire for honour (the worse course of action). Rather, when he is with Socrates he believes at this time that the best course of action is not to do the Athenian's business, but to improve his soul. But whenever he leaves Socrates he is persuaded—by the Athenians, presumably—that the life of honour is the best course of action and so he chooses a different course of action which appears to him, at that time, to be the better one. We do not witness in this case a man who chooses the worse option when at the same time he is persuaded of the fact that it is the worse option. This is borne out further by the fact that it is only when he sees Socrates that he feels shame, the implication being that when he gives in to his desire for honour he is not feeling shame, and so there is no conflict at this time. Contrast the classic case of Leontius in the Republic. Leontius sees some corpses and ‘he had a desire to look at them but at the same time he was revolted and turned away’ (439e–440a). It is this simultaneous desire for and aversion to the same object that motivates the division of the soul in the Republic. And the case of Alcibiades provides no evidence of that. (p.239) Behaviour of this kind need not be explained in terms of psychological divisions, then, but in epistemological terms familiar from the so‐called ‘Socratic’ dialogues. Alcibiades exhibits what one might term ‘weakness of belief’.16 Although he is persuaded by Socrates, his beliefs do not remain long enough to motivate consistent action over time. Without the secure grounding of knowledge one will lead an unstable existence and be subject to the whims of appearances and the clamour of the crowd.

None of the passages considered, then, assume, or require, a division of the soul, and the evidence in favour of this psychological thesis is weak. Although the evidence for the ‘Socratic’ picture is stronger, there is no evidence to rule out tripartition either. The picture that emerges from the Symposium is underdetermined in many ways. It may be the case simply that Socratic desire is the only item on the evening's agenda, and not that it is the only item on the cards.17


(1) Price (1997) 254–5: ‘A remarkable aspect of the Symposium is its loyalty to the Socratic psychology of the Lysis…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…We must take the background assumption to be Socratic: happiness is the ultimate goal of all desire, animal as well as human.’ Irwin (1995) 303: ‘The conversation between Socrates and Diotima in the Symposium begins with the sexual aspect of erōs, as desire for the beautiful (204d). But this description is soon supplemented or replaced by two others: erōs as desire for the good and for happiness (204e), and erōs, as the desire to ‘give birth in beauty’ (206b7). Plato uses ‘erōs’ 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…the Symposium thus eliminates the common conception of erōs in favour of the Socratic conception of desire.’ But Irwin adds later that ‘[the Symposium] neither endorses nor rejects the division of the soul, since he neither affirms nor denies psychological eudaemonism’. Rowe, C. J. ‘The Symposium as a Socratic Dialogue’ (2006) sees ‘Socratic style psychology’ as ‘a central feature of the Symposium’. And he raises a puzzle. 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 the Republic and the Phaedrus. Socratic psychology on the one hand; Platonic forms on the other. How are we to explain the mix?’

(2) Price (1995) 9: ‘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 a 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…. 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.’

(3) Hobbs (2000) 251: ‘I believe that Diotima's speech assumes, if not precisely the tripartite psychology of the Republic, then at least something very close to it. Having claimed that personal immortality is impossible for humankind, she argues that humans can and do pursue three different kinds of substitute immortality; in ascending order of importance these are biological offspring, fame for noble deeds, and the creation of artistic, legislative, educative and philosophical works. Such aims undeniably overlap with those of appetite, thumos, and reason.’ Nehamas (2004): ‘The two lower kinds of lovers—those who reproduce themselves physically and those who are made famous by their accomplishments—correspond to the two parts of Plato's divided tripartite soul, which makes in that way a clear if indirect appearance in the Symposium. The lowest class are ruled by the soul's appetitive part. Plato does in fact say that they are “pregnant in body” and contrasts them with those pregnant in soul. But that does not make the body an independent entity with its own desires. All human desire is, as it also is in the Phaedo…ultimately due to the soul, but some of them occur only when the soul is embodied. These desires are clearly very common to every human being (which is another way of saying that Plato does not leave them behind). Others, though, are not as widely shared—better, they are not as strong in all as they are in some. One such class of desires emerges in the “lesser mysteries” of the Symposium. And Plato's emphasis on the love of honour (philotimia, 208c3), the desire for glory (kleos, 208c5, 209d3) and the craving for fame (doxa, 208d8) and memory (208d5, 209d3) makes the connection between this sort of erōs and the thumoeides part of the soul, which thirsts for time and doxa (cf. Rep. 580d–583a), unmistakeable.’

(4) ‘And since each part of the soul, as we know from the Republic, has its own appropriate pleasure, so each part of the soul, the Symposium implies, has its own appropriate erōs, directed at objects whose beauty each different part of the soul appreciates. For that reason, just as the pleasures of the soul differ in immense degree but are still for all that pleasures, so every object of erōs, however humble in comparison to the beauty of the Form, is still beautiful and, however, dimly, reflects the Form's light.’

(5) Note that ‘always’ is applied both to the possessing of good things, and to the desiring of good things: it is said that we always desire the good, and that we desire [to possess] the good always (205a6–7: πάντας τἀγαθὰ βούλεσθαι αὐτοῖς εἶναι ἀεί, ‘everyone wants to possess good things always’; 205b1: πάντες τῶν αὐτῶν ἐρῶσι καὶ ἀεί, ‘everyone desires good things and always desires them.’ At 206a9 ἀεί seems to qualify εἶναι with τὸ ἀγαθὸν as the subject of εἶναι indicating that everyone desires that the good ‘always’ belongs to them.

(6) On this characterization of what it is to have a rational desire, see Penner (1990) 49.

(7) Compare, for example, Protagoras 358c6–d2, Meno 78b1–2, Gorgias 468b1–2.

(8) See, for example, Kahn (1987) 85 and Irwin (1995) 209 and 214–15 on the grounds for dividing the rational and the appetitive part. Kahn argues that ‘in order to establish the distinction between reason and appetite Plato must here define, for the first time, the notion of desire that is essentially independent of any judgement concerning what is good.’ For a discussion of desires in the Republic see Lorenz (2004).

(9) For other passages where Plato excludes animals from logismos, or logos, cf. Rep. 441a–b, Laws 963e with Ch. 2, sect. 2.

(10) For just such a reading see Carone (2001), who argues (against the standard view of the Republic) that even the desires of the lower parts aim for the good, albeit in some limited sense. In light of the parallel between city and soul she cites the perceived good of the oligarch and the democrat in Book 8: the good for the oligarch is wealth (562b3–4); the good for the democrat, freedom (562b9–10).

(11) On this see Cooper (1999b) 135.

(12) See above, n. 3.

(13) Such types are also said to desire ‘immortal virtue’ (208d7), an ‘immortal memory of their virtue’ (d5–6), and kleos (208c5), all of which, I take it, are various ways of describing their definitive desire for honour (208c3).

(14) On the specific claims that lead to tripartition in the Republic, see Lorenz (2004) 84 n. 3.

(15) Davidson (1980) 21–2.

(16) I borrow the phrase from a paper delivered by Myles Burnyeat at Bristol (2000).

(17) Cf. Kahn (1996) 264: ‘In dialogues before the Republic Plato offers no general account of moral psychology. On the contrary, his discussion of desire is systematically limited to rational desire for the good, since that is the fundamental thought underlying the Socratic paradox.’