The Definition of Love in Plato’s Symposium
For anyone who wants to think philosophically about love, the only way to begin is to reflect on the problems first raised in Plato’s Symposium.
by Donald Levy
For anyone who wants to think philosophically about love, the only way to begin is to reflect on the problems first raised in Plato’s Symposium. The dialogue is original in at least two ways—in that it exposes the presuppositions of Greek sexual morality to the sort of critical scrutiny practiced by Socrates, there is simply nothing like it by anyone else before. In addition, the new theory of love and the new ideal of it developed in Diotima’s speech appear to be Plato’s own equally original advance over Socrates’ philosophy.
The dialogue records the brilliant conversation at a dinner party at which Socrates is a guest. Those who speak before Socrates mainly share what Jeffrey Henderson refers to as the typical Greek tendency to glorify the instinct of sex rather than its particular objects (The Maculate Muse, [New Haven, 1975], 205). For them, love (eros) is a god whose beauty and goodness they compete with one another in praising. Even Pausanias, who takes care to distinguish noble from base love, claims that “it is always honorable to comply with a lover to attain excellence” (185b)—even if the lover turns out to be bad, it does the boy credit to have been so deceived! It is this almost universally held belief in the intrinsic value of sexual love against which Socrates sets himself from the start; love, he says, is neither beautiful nor good (though he does not mean it is ugly or bad, either). Love cannot be beautiful because it is the desire to possess what is beautiful, and one cannot desire that which one already possesses, Socrates argues. That love is nothing good in itself, but is merely a means to the attainment of things that are good in themselves is emphasized again at the very end of Socrates’ recital of Diotima’s speech when he says “human nature can find no better helper than love” (212b). Even Socrates’ own love of testing the opinions of others is not exempt from this new test; just as Socrates had surprised Agathon by claiming love is not beautiful, so Diotima bewilders Socrates with the idea that “the object of love is to procreate and give birth in the presence of beauty” (206e). It is not enough, she seems to say, for a philosopher, a lover of wisdom, merely to assist at the birth of ideas in others, playing the midwife, herself barren (to which Socrates often compared himself), examining the new-born ideas for soundness. Such activities have no intrinsic worth; they are of value only if they lead the philosopher to bring forth theories of his own. The genuine lover of wisdom must himself conceive.
The new account of love introduced in the final part of Diotima’s speech is one she is not certain Socrates can understand, she says. This appears to be Plato’s way of signalling the radical shift in what follows from the comparatively simple attempt to define love by finding the element common to all types of love (typical of Socrates’ method) when no distinctions of value among types of love are made (202d–209e), to Plato’s new approach. Now the different types of love are to be ordered hierarchically, one being judged superior to another because its object is inherently better. Further, this hierarchy of love-objects involves another non-Socratic idea—that there is one ultimate object of love to which all the others must be tending in order for them to be objects of love at all. For those who seek to understand love, this absolute beauty, existing apart and alone, is the final goal of all their previous efforts. To achieve the vision of absolute beauty one must first progress from love of physical beauty in an individual to love of all physical beauty; then, love of beauty in the soul leads to awareness of the beauty of activities, institutions, and sciences. Upon surveying all these different kinds of beauty, one will be led to a glimpse of the science whose object is absolute beauty.
This theory of love has appeared defective in at least two ways to Gregory Vlastos, whose “The Individual as Object of Love in Plato” (in Platonic Studies, [Princeton, 1973]) is the most important recent discussion of Plato’s views. According to Vlastos, the defects in Plato’s account of love can be seen by comparing it with the definition of love Vlastos accepts, and which he adopts from Aristotle—“Love is wishing good things for someone for that person’s sake.” Vlastos’s first objection is that since Plato has already defined love as the desire for oneself to possess what is beautiful, his idea of love, however spiritualized it may be, remains essentially ego-centric (ibid., 30). Secondly, Plato does not see that love fundamentally and primarily has persons as its object; for Plato, the love of persons is placed far below the love of an abstract entity, absolute beauty. “What we are to love in persons is the ‘image’ of the Idea in them” (ibid., 31). In a note Vlastos says “This is all love for a person could be, given the status of persons in Plato’s ontology.”
We are to love the persons so far, and only insofar, as they are good and beautiful . . . the individual . . . will never be the object of our love in Plato’s theory . . . [which] does not provide for love of whole persons, but only for love of that abstract version of persons which consists of the complex of their best qualities. (Platonic Studies, 31)
So, for Plato, our affections for concrete human beings are “lesser loves,” as Vlastos paraphrases it (ibid., 32), to be used “as steps” (211c) to the attainment of absolute beauty. Vlastos concludes his criticism by noting the emphatic frequency of this idea (ibid., 32).
Without trying to deal with the entire array of evidence Vlastos presents to support these criticisms, it is enough to point out in reply to the first objection that Vlastos’s definition of love, compared to which he finds Plato’s defective, seems a definition not of what love is, but of what love ought, perhaps, to be. Fairly clear examples of love abound which do not always conform to our moral ideals of love; the love of children and parents for one another—often negligent, selfish, confused, slow to develop—is one. So it is probably wrong, in defining love, to lay down as a necessary condition of one’s loving a person at all that one seeks what is good for the other for the other’s sake. At least some of the time when we love, we may be seeking what is good for others for our own sake, not theirs, as Aristotle recognizes (Nicomachean Ethics, Book VIII, ch. 1); and we must also consider the possibility that we may not even be seeking what is good for the other at all—“smothering” mothers, murderously jealous husbands are clear examples. (This matter is discussed in some detail in Alice Balint’s Love for the Mother and Mother Love in Michael Balint’s Primary Love and Psycho-analytic Technique, [New York, 1965]). If these examples are granted, then Vlastos’s definition of love does not state a necessary condition of love. Accepting his definition would make it impossible to distinguish between a person’s loving well, and that person being a genuine instance of a lover.
Vlastos’s definition seems not to state a sufficient condition of love any more than it states a necessary condition, since there are cases of persons who seek what is good for others for the sake of the others (i.e., because the others need or deserve good) when love for the others is not the motive, and may not even be present. Nurses, firemen, teachers take care of, seek to do what is good for others, even if love for the others is wholly absent. It may be, as Diotima argues, that love motivates us whenever we achieve anything good; the nurse, fireman, teacher might love the science, art, skill to which each is devoted. But granting this point does not at all narrow the distance between Plato’s theory and the requirement laid down by Vlastos.
I have restricted myself to arguing here that Vlastos’s definition of love is defective;¹ but even if it were accepted, his conclusion that Plato’s idea of love is an egocentric one does not directly follow. For his argument to work, Vlastos must show that desiring for oneself to possess what is beautiful never consists in wishing good things for someone for the person’s sake; Vlastos must show that the first cannot consist in the second. But suppose the beautiful thing one desires for oneself to possess is the good (Symposium, 204e). Further, suppose that some of the time the good one desires for oneself to possess is virtue. At least some of the time, desiring to possess virtue for oneself consists in wishing good things for someone for that person’s sake. It would not be correct to say that wishing good things for someone for that person’s sake is merely a means to acquiring virtue for oneself; the good one seeks to possess for oneself is—to be the cause of what is good for another person for that person’s sake.²
Vlastos’s second objection is actually three tied together—(1) Plato ranks love of persons far below love of other things such as absolute beauty. According to Vlastos, Plato does so because (2) Plato takes love of individuals in themselves to be impossible—only their good qualities can be loved; and partly because (3) Plato understands love of persons to consist in nothing more than love of absolute beauty by way of individual persons. The individual person we love is merely of use as an image of beauty, as a means to it.
It should be noted, in reply to (3), that Diotima does in a sense speak of using particular objects of affection, for example, other persons, to gain knowledge of absolute beauty—but the use to which they are to be put is as examples, instances of beauty, as W. Hamilton’s interpretive translation (of 210d and 211c) suggests. If we use a person in this way, it does not follow that that person cannot really be loved by us—any more than our using Thomas Jefferson as an example of a great president implies that we do not really admire him. Of course, to use a person in that way implies that we do really admire or love him, or in Plato’s case, that we do regard the person as a genuine instance of beauty.
Besides, when Diotima speaks of using examples of beauty, she is speaking of those who seek to be initiated into love’s mysteries, who seek to learn what love really is. For that, a person must understand absolute beauty, and to achieve that, one must use the objects of one’s love as examples, images of absolute beauty. In saying these things, Diotima seems to be thinking of a quite distinctive imaginative process—one people might engage in without being obliged to treat the objects of their affections merely as examples of something else. Certainly, a person might engage in such an activity without necessarily believing that all anyone is ever really doing in loving is using the objects of love as examples of something else, or that using the objects of love as examples of something else is all that we ought to do with them. Diotima’s recommendation of this imaginative process does imply that if (and when) we wish to understand the mysteries of love we must go about it by thinking of the beautiful objects to which we have formed attachments as examples of absolute beauty, leading us onward. But not everyone is always engaged in seeking this, and when not so engaged it would be absurd to treat others merely as instances of something else. It does not appear correct to attribute to Plato the view that we cannot love individual persons, or that we can love them only instrumentally, or that we ought to love them only instrumentally. Diotima does say “This above all others . . . is the region where a man’s life should be spent, in the contemplation of absolute beauty” (211b); that region is not the only region in which we can spend our lives—or even the only region in which we ought to spend our lives; of all the regions in which life should be spent, it is the highest, Diotima says—so there are others.
If Vlastos’s objections do not reveal any basic flaws in Plato’s theory, as I have argued, this does not mean there are no problems in it. The real trouble may be, not as Vlastos thought, that Plato ranks the love of persons far below other sorts of love, but rather that love itself, regardless of its object, has no intrinsic value for Plato, and therefore ranks below things that do have it. The value of love is entirely dependent upon the worth of its object, Socrates had emphasized at the beginning of his discussion; love is at best a mighty helper to human nature—but nothing more.
The oddness of this cannot be avoided, though the logic of the argument may seem good; knowledge, virtue, beauty seem to be inherently superior things to the love we have for them. Whereas they are inherently good, our love for them seems to be good only insofar as it helps us to acquire them. As plausible and insightful as this may sound, it is nevertheless natural to protest that life devoid of love would be worthless, and that love itself therefore must have some great inherent worth. Perhaps Pausanias was not so wrong after all to judge the deceived lover as he did.
But it would be a mistake to suppose that the only alternative to Plato’s treatment of love as merely instrumental in value is the typical Greek view Socrates reacted against. That view saw value in love—but merely because it was pleasurable in itself and productive of excellence. Is there no intrinsic value to love higher than mere pleasure? One solution to this problem would be to argue that the intrinsic value of love is to be found in its being constitutive of the soul; that is, to claim that love is the fundamental activity (or one of them) all souls are necessarily always engaged in, whatever else they may be doing. Then, the worth of love is established, if the worth of the soul is. Such a claim is clearly not part of Pausanias’s commendation of love, nor is it part of any of the praises of love pronounced by the speakers in the Symposium before Socrates. Such a view of love as constitutive of soul might seem to be the one Diotima expresses (205a; 205c) when she says:
“Now do you suppose that this desire [for what is good] and this love are characteristics common to all men, and that all perpetually desire to be in possession of the good, or what?” (205a)
But Socrates’ response leaves it unclear whether he accepts the whole of this view: “That is exactly what I mean; they are common to all men.” Diotima made two distinct claims, and Socrates assented only to the weaker of the two, it seems, that is, to the claim that all men love, at some time or other, we might add. This view is associated with the idea that every man has a master passion—love of money, or of physical prowess, or of wisdom—which are all expressions of the desire for good and for happiness, according to Diotima (205d). This view requires that (a) all men love at some time or other, and (b) each man loves some one thing more than any other thing. It is consistent with these conditions that much of what men do is not done out of love at all. We must keep separate the stronger thesis that all human activity is motivated by love, as well as the thesis that love is the essential activity of the soul. These three views are not equivalent, and Plato does not accept the last one. This we know from the Phaedrus, where the essential activity of soul is said to be eternal motion, self-motion (245e), of which love is perhaps a resemblance. To be sure, love is “the greatest benefit that heaven can confer on us” (245b)—but it is not constitutive of soul.³ Indeed, that love is said to be a type of madness conferred on us at all implies that we are able to exist without it.⁴
A complete resolution of Plato’s doubts about the role of love in the soul would have to take up and reply to his view of the emotions (and therefore of love as well) as alien to intellect, “the best part of the soul” (Phaedrus, 248b). Part of the answer might also draw upon features of his theory of knowledge as essentially recollection. That is, if there were something which, to be known at all, must be loved, it would be difficult to deny that loving it was intrinsically good if knowing it was held to be good in that way. Even if it were granted that knowing it was intrinsically better than loving it, the intrinsic value of loving it would not be undermined. Whether there are any such objects of love and knowledge is a question lying outside the scope of this paper, though it is at least plausible to say that God is such a being, since it is hard to make sense of the claim that someone knows God but does not love God.
Perhaps a different type of case of the following sort illustrates the same point; suppose the only way, the only conceivable way, to gain self-knowledge (or any other kind of knowledge) is through loving others. Further, suppose that loving others well is sufficient for self-knowledge. It would then be hard to deny that love was intrinsically valuable, if the knowledge depending on it was assumed to be intrinsically good. These cases suggest that Plato’s worry about love’s inherent worth rests upon a presupposition hard to justify, namely, that any knowledge or other good reached as a result of love necessarily can be obtained or possessed without love. This presupposition must be false if, as I suggest, love is at least sometimes a necessary condition of recollection.
The subsequent history of philosophizing about love reflects some of these concerns; it is not until Plotinus, I believe, that love is conceived to be constitutive of the soul.⁵ (The idea seems wholly absent in Aristotle’s psychology; love is purely an ethical problem for him.) Augustine’s famous remark, “My love is my weight. To whatever place I go, I am drawn to it by love” (Confessions, Book XIII, chapter nine) implies both that love is constitutive of soul (as he takes weight to be constitutive of body), as well as that whatever good, e.g., knowledge, he achieves, it is the result of that love; this essential connection between love and knowledge receives extended examination in On the Trinity. Both of these thinkers can be seen as struggling with the same problem inherited from Plato—that of understanding love in such a way that its intrinsic as well as its instrumental value is made clear.
NOTES
¹ Whether Vlastos is right to suppose that his definition is the same as Aristotle’s is a complex question. Certainly most scholars have agreed with Vlastos’s interpretation of the definition of love in the Nicomachean Ethics. An excellent opposing interpretation can be found in → W. W. Fortenbaugh’s “Aristotle’s Analysis of Friendship: Function and Analogy, Resemblance, and Focal Meaning,” Phronesis, XX, #1, 1975, pp. 51–62.
² One obstacle to seeing that there is nothing essentially egocentric about Plato’s definition of love probably comes from our imagining an incompatibility between it and Paul’s “Love seeketh not its own” (I Corinthians 13). If Paul is interpreted to mean “Love consists in seeking only what is good for others, never for oneself” perhaps there is something to fear here. But, as modern translations make clear, what Paul meant to say was “Love does not insist on its own way” (Revised Standard Version), “Love is… never selfish” (New English Bible). To be selfish means to ignore or neglect the needs and wishes of others in pursuit of one’s own good. Not being selfish then consists in not ignoring or neglecting others; it need not consist in not pursuing any goods of one’s own at all. It would be an error to make it a necessary condition of love that the lover not be seeking what is good for himself. If this point seems trivial or obvious, the reader might consult Anders Nygren’s Agape and Eros to see the crucial role played in the minds of some scholars by the interpretation of Paul’s remark which I have criticized.
³ Does the claim that love is the greatest benefit that heaven can bestow upon us imply that love must be greater than knowledge or justice, which are inherently good? The implication would succeed only if it were possible for knowledge or justice to be conferred. However, if we take seriously Plato’s doctrine that knowledge is recollection, then even heaven cannot confer knowledge upon us—it must be recollected. That justice cannot be conferred either follows from another of Plato’s views—that virtue (and hence justice) is a kind of knowledge.
⁴ J. M. E. Moravcsik considers, but dismisses, the idea that when the soul reaches the higher stages of the ascent in the Symposium “it no longer has passions or aspirations.” (“Reason and Eros in the Ascent Passage of the Symposium,” in Anton, J. P., and Kustas, G. L., eds., Essays in Ancient Greek Philosophy, /Albany, 1971/, 285–302 at 294) He is forced to consider this possibility because there is an evident absence of emotion-steps in the higher stages of the ascent, in contrast to the lower ones. He concludes that though “eros is still at work in the soul in the later stages, it no longer functions as a guide, thus not appearing in the sequence of steps described. No change in over-all aspiration is needed in order to lead the soul from the contemplation of the sciences to the comprehension of the Forms. Like Virgil in the ‘Divine Comedy’ eros helps as a guide only until we reach the final stages; there contemplation becomes self-sufficient” (ibid., 294). But it is unclear how eros can still be at work in the soul when contemplation becomes self-sufficient, since Moravcsik noted earlier
“in general one can say about the causal influence of eros on the mind that eros is what pushes the mind to new investigations” (ibid., 292).
Presumably, when contemplation becomes self-sufficient, eros ceases to be that which “pushes the mind to new investigations.” Incidentally, the Virgil analogy seems unsuited to Moravcsik’s point, since Virgil vanishes when he ceases to serve as a guide (Purgatorio, Canto XXX); he does not continue to accompany Dante in some non-guidance role.
⁵ “This being, Love, has from everlasting come into existence from the soul’s aspiration toward the higher and the good, and he was there always, as long as Soul, too existed.” (Ennead III. 5.9, translated by A. H. Armstrong, Loeb Classical Library, [Cambridge, Mass., 1967], Plotinus, volume III, at 203).
Brooklyn College of CUNY
Source: Journal of the History of Ideas, Vol. 40, No. 2 (April – June, 1979), pp. 285-291

Read More