Matt Schuler: Love as Knowledge: the Metaphysics behind the Emotion


Love as Knowledge: the Metaphysics behind the Emotion

Matt Schuler

What is the nature of love between human persons? In attempting to answer this question, a particularly fruitful approach is to identify competing paradigms that emphasize various aspects of the concept. Thus, for example, both Martha Nussbaum and Vincent Brümmer ask whether love is fundamentally an emotion, an attitude, or a relation. Although I believe that the emotional, attitudinal, and relational aspects of love are important, in this essay I will argue that love is principally a form of knowledge. It is my belief that, together, the love-as-emotion model and the love-as-knowledge model virtually exhaust the discourse and account for the vast majority of what is necessary to the concept of love; further, an analysis of these two paradigms will show, I hope, that primacy ought to be given to the love-as-knowledge model. In short, I will argue that love is first and foremost a form of knowledge, and that this knowledge is what triggers and informs the emotion of love. But before delving into the positive content of love, I would like to turn briefly to the topic of what love is not.  


This essay is concerned principally with love between human persons. Nonetheless, it will be useful to make some general remarks concerning the appropriateness of our use of the term “love” in certain other circumstances. This will set some preliminary restrictions on our project that will be of particular importance in part III of this essay. 

It is customary to begin by noting—and often marveling at—the extraordinary diversity of the objects of our love: there is romantic love, love of God, one’s children, one’s country, various activities, and so forth. It is often remarked by philosophers confronting this fact that what is needed is an account of love capable of accommodating such disparate forms as those mentioned. This, in turn, often takes the form of identifying the common denominator of every variety of love. Thus Robert Nozick proposes that “what is common to all love is... [that] your own well-being is tied up with that of someone (or something) you love.”[1] In a sense it would be difficult to disagree with this formulation, but our intuition is that love is more complicated than that. Accordingly, the reductionism that gives rise to this sort of solution ought to arouse suspicion. Putting such dissimilar objects on an emotional par and then searching for the lowest common denominator shrinks down the concept we hope to elucidate to a bare and conceptually impoverished axiomatics.

Thus, I suggest that the question, “What do love of football and love for a human person have in common?” should be neither asked nor answered. There simply is no philosophically significant common denominator. I think it can be safely claimed that when we talk about loving football or Chinese food we are only being linguistically lazy. (It would certainly be strange, after all, to insist that one has the relevant emotions concerning a particular cuisine.) Accordingly, we want to be able to dismiss outright as unlikely candidates for the status of “beloved” whatever class of objects includes football and Chinese food. But where is the line drawn? Part of the answer is found in Aristotle, who removes from his consideration “lifeless objects” on account of their inability to reciprocate. But if this were our sole criterion, we might be stuck dismissing such things as love for humanity and love of God (insofar as He is “lifeless” in a corporeal sense) before giving them their due consideration. Although there are a multitude of epistemic and relational questions that arise when a person asserts that she loves God, there appears to be more substance to her statement than there might have been had she instead proclaimed her love for her favorite television show. For present purposes, then, allow the following to suffice. If a person’s affection for a “lifeless object” is prima facie better explained by some other affective category, let that person’s love-claims be evaluated under that rubric rather than the one we are concerned with here. I think we will often find that, rather than love, the more appropriate category is “fondness.” In his Emotions, Robert C. Roberts shows that to be fond of something is to “prefer it, to be disposed to enjoy it, but in a sort of aesthetic way.”[2] Importantly, he points out that although fondness can be the “basis” for emotions, it is not an emotion itself.[3] I think Roberts is right about this. And if it is conceded that whatever one feels about Chinese food, it is decidedly not a feeling with emotional content (but is rather something more appropriately characterized in terms of the appetitive), then it would make good sense to interpret “I love Chinese food” as “I am extremely fond of Chinese food,” and in the process sidestep the problem of fitting exceedingly divergent affections under the same general heading.[4] 

Perhaps this is an uncontroversial suggestion, but its significance in connection with what follows cannot be understated. Let us turn, then, to a somewhat brief discussion of the love-as-emotion model.


Here the important question to ask is, of course, whether love generally adheres to the requisite features of emotions. Whereas Solomon asserts, without argument, that “love is an emotion, nothing else,”[5] others, such as Roberts, recognize the possibility that love differs in an important sense from other emotions. It may even differ so much that it makes sense to call love an emotion only insofar as it predisposes us to a host of (other) emotions (proper).[6] The view that love is a form of attachment that engenders a wide variety of emotions (rather than an emotion in its own right) is supported by the following observations: unlike an emotion, love is (1) dispositional, and (2) involves a perception not of the “local and episodic circumstances of the object” of the emotion, but rather of the “enduring predicates of the object herself [himself, itself].”[7] This is an attractive move, as it addresses the fact that, peculiar among the emotions, love can make us feel joy, anger, sadness, et cetera. At least in terms of degree, the same cannot be said for joy, anger, and sadness themselves. Should love, then, be considered an emotion? Perhaps a more systematic examination will allow us to determine the extent to which it can be properly understood as such. 

Certain features seem necessary to the concept of an emotion. It should be possible, then, to determine which affective phenomena ought to be considered emotional in nature and which ought not to be. Roberts provides an inventory of such features, some of which are listed here as minimal criteria for inclusion in the catalogue of emotions: (1) emotions are “paradigmatically felt” but may occur “independently of the corresponding feeling”; (2) emotions are often accompanied by physiological changes which are sometimes felt; (3) paradigm cases of emotions “take objects”; (4) the objects emotions take are “typically situational”; (5) ordinarily the subject of an emotion “believes the propositional content of his emotion”; (6) it is possible to have emotions “without being able to articulate (all of) their content,” some of which may be nonpropositional; (7)  it is sometimes possible to “exercise voluntary control” over a given emotion, and sometimes not.[8]  

Love has little difficulty satisfying these conditions. The first three are met with ease. Concerning the fourth, it should be pointed out that, in describing an object as typically “situational,” Roberts intends to capture that an emotion is often not directed only at, say, a particular person, but also for a given reason, and against the backdrop of a particular setting. With respect to the three remaining criteria, I think it would be difficult to develop a theory of love that would have trouble meeting them.

But although love fits comfortably within this rubric, the question I will address in part III below is whether any set of requirements exclusively in the domain of the emotions adequately captures the essence of love. It may be that love satisfies a given set of requirements without being appropriately defined by them.[9] Since it is clear, however, that there is at least an emotional component involved in love, we will want to account for love’s possession of features that distinguish it in a significant sense from other emotions. How might we flesh out the specifically emotional content of love? We might begin, as others often do, by observing that as a feeling or an emotion, love is typically contrasted with hate. This approach is promising, since identifying its opposite might help us come closer to isolating love’s essential features. But the fact of the matter is that, although love involves a bestowal of positive value on its object and hate involves a bestowal of negative value, these two emotions are at bottom more similar than they are dissimilar. Importantly, both involve passionate interestedness. One cannot be indifferent toward someone and also hate him; neither can he be indifferent toward someone and also love him. Indifference is therefore more destructive to both concepts than either is to the other. For this reason, I think it is appropriate to regard love as opposed to indifference, as empathy is opposed to apathy. 

It does not seem, however, that indifference is an emotion and so it might be thought that, insofar as love is an emotion, there must be some significant conceptual asymmetry between the two. This difficulty provides the opportunity to bring to light a feature of the love-as-emotion model that we have neglected so far. Indifference, it seems, is an attitude. It is a way of addressing the world and particular people. If love is to be contrasted with indifference, then, it must have some stock in the attitudinal sphere. And indeed it does. Just as a person is capable of regarding others and the world either empathetically or apathetically, so too is she capable of regarding them either lovingly or indifferently. Love, as an attitude, is an individual’s psychological or behavioral projection of certain emotions onto the world and others with the effect that the individual regards those things lovingly. An attitude of love, therefore, requires the emotion of love (or, more precisely, the variety of emotions it disposes us toward) for its formation; it is a turning outward of one’s internal states, a regarding or projecting. 

On the model of love I am advocating, what has been said thus far should be an adequate sketch of the sense in which love is an emotion and an attitude. Perhaps it has occurred to the reader that although it is appropriate to regard love as something that disposes us toward a range of emotions, there are ways in which the emotion model inevitably fails to capture the essence of love. This essence—as I will endeavor to show presently—is constituted primarily by epistemic considerations that fill in the gaps left by the emotion model.


In this section, I hope to provide a foundation for the paradigm discussed above. The view I want to advance is that regarding love as an emotion is appropriate, provided that we consider the emotion to be ultimately a response. What it is a response to is, of course, the subject of this section. My claim is that love, the emotion, issues from a unique kind of knowledge of the beloved. This knowledge is a seeing or perceiving, a privileged insight into the essential truth of the beloved.   

It is commonly assumed that the etiology of love is best explained as the recognition of value in some object. (Call this the value hypothesis.) Something like this has to be right or else we risk leaving unexplained what motivates us to love one person rather than another. The value hypothesis is not, however, immune to difficulties. Emphasizing the role that specific valuable traits or qualities play in our love of persons puts in jeopardy our ability to explain why we are committed to loving certain people and not others when those others possess precisely the same valuable properties as those we do love. In short, it undermines our notion of the beloved as an irreplaceable individual. This criticism is often leveled against the cold, inhuman landscape of Platonic eros. A clear implication of that theory is that as soon as the beloved ceases to exhibit the valuable properties that warrant my love, she ceases to be a suitable object of it. She is an object of my love in the first place only because, in loving those properties, I can ascend to the more noble and lofty love of Goodness and Beauty. As Vlastos, Brümmer, and many others have pointed out, Plato’s view requires that we not “love persons for their own sakes but only to the extent that they instantiate or realize ideals or contribute to the realization of ideals.”[10] But in fact,   

Plato got the matter love grows you love not general aspects or traits but more and more particular ones, not intelligence in general but that particular mind, not kindness in general but those particular ways of being kind.[11]  

For this reason we ought to agree not with Socrates but with the Symposium’s Alcibiades. In her sympathetic portrayal of him, Nussbaum speaks of Alcibiades’s “deep desire to know Socrates,” as well as his “desire to be known”; Alcibiades wants to “open up the other,” an “epistemic” desire to “know everything that he [Socrates] was.”[12] According to Nussbaum, Alcibiades suggests in his speech that “the lover’s knowledge of the particular other, gained through an intimacy both bodily and intellectual, is itself a unique and uniquely valuable kind of [knowledge].”[13] This form of knowledge, which stands in contrast to the form of knowledge Socrates supports (viz., the discovery of universal truths), closely resembles the form I will develop here:

The lover’s understanding, attained through the supple interaction of sense, emotion, and intellect...yields particular truths and particular judgments. [Alcibiades’s] knowledge sees more [than Socrates’s], and differently; it is an integrated response to the person as a unique whole.[14]

This statement, which references (1) the uniqueness or irreplaceability of the beloved and (2) our love for the wholeness of the beloved, embodies in concise form a robust resistance to the logic that takes the value hypothesis too far. As mentioned above, the value hypothesis threatens the doctrine of irreplaceability. The second and related difficulty is that the Platonic emphasis on value stands in direct conflict with the view that, as Tolstoy put it, “when you love someone, you love the whole person.” To be loved for one’s entirety is to be loved for oneself, rather than for any peripheral or unessential features of one’s identity. I would suggest that, in turn, to be loved “for oneself” almost requires that the lover not be able to enumerate various valuable properties of the beloved responsible for his love; something with permanence at the beloved’s core is what sustains the lover’s love throughout myriad changes in physical appearance, personality, and identity more broadly.

What is the nature of this “something with permanence at the beloved’s core?” I think this question can be answered by examining an interpretation of Plato offered by Norton and Kille that renders his theory of love compatible with both love of “the whole person,” and with the view that the beloved is an irreplaceable individual rather than merely a “dispensable placeholder for ideals.”[15] Their account centers on a discussion of the concept of a daimon. (It is important to note that in the present use of this term, we are not referring to Diotima’s suggestion in the Symposium that Love is a daimon, an intermediary between human beings and the gods.[16]  Nor are we referring to the daimon—as Socrates does in the Phaedrus[17] and the Apology[18]as an inner voice that plays the role of conscience. Norton and Kille’s use of the term suggests that in this context we ought to understand a daimon as one’s true self, unique essence, or as Brümmer puts it, “hidden potentiality.”)

According to Norton and Kille, eros is a love that “sees something”—an “ideal possibility”—within the beloved.[19] That ideal possibility is one’s daimon. Human beings are like the busts of Silenus, hollow clay on the outside with a golden figurine hidden inside; eros is the “power which discerns the golden figure within the clay.”[20] The clay is a person’s outward appearance; always flawed in some respect, it constitutes his actuality. But “within his actuality is the golden figurine which love discerns”—his daimon.[21] Each person’s daimon is unique and constitutes his true self; as such, attached to a person’s daimon is a unique value. 

Norton and Kille acknowledge the popular criticisms we raised above (for example, that love of ideals is not love of persons), but they think that their interpretation of Plato—centered on the notion of an “indwelling ideal”—defeats these objections. Such criticisms, they claim, confuse the distinction between actuality and possibility and fail to acknowledge that eros focuses on the latter rather than the former.[22] Eros effectively removes the emphasis from a person’s actuality and focuses instead on his daimon, which he “most essentially is.”[23]

Although this may be a satisfactory response to the charges leveled against eros, I find it unlikely that this interpretation accurately represents Plato’s own view, and it seems even more unlikely that it captures the intention of Diotima’s speech. But this framework accords well with the theory of love I endorse: a viable theory should regard as indispensable the claim that love is “the knowledge and love of someone’s individual daimon,”[24] and is that which “holds before [the beloved] a mirror in which to rediscover his daimon” and therefore enables the beloved to “becom[e] himself.”[25]

In order to illustrate the appeal of this concept, consider once again Tolstoy’s claim that when you love a person, you love the whole person. Against this notion we might cite, for the sake of argument, the following two objections: (1) most if not all of the time, there are things about our beloved that we know about and don’t love (e.g., his failure to put the toilet seat down). In any case, (2) it’s impossible to know everything about a person; there’s always the chance of discovering that one thing about him or her that we don’t love. In light of these remarks, how does Tolstoy’s maxim still seem to ring true? Norton and Kille’s insight nudges us in the right direction: at the most fundamental level, I do not only love the imperfect ways in which a person instantiates ideals, but also (and especially) his daimon. Some of his imperfect actualities are things that, in themselves, I cannot bring myself to love, but his daimon is loveable in itself; further, it constitutes a good so unique and incomparable that the actualities accompanying it—loveable or otherwise—are in some sense overshadowed and excluded from my consideration. 

Couched in slightly different language, we might say that when we truly love a person, the deep knowledge of whom he is at his core renders insignificant our disapproval of the incidental features of his identity that we do not love; we know the core, and that is what we most fundamentally love.[26] As a first approximation, then, we should say that in a significant sense we do love the “whole person,” “all of him”: we love the very features that sustain his unique identity and make him who he is rather than someone else. On this view, what we love—the whole person—is identical with an indivisible core that spills over and saturates every other aspect of the beloved’s identity and infuses those aspects with deep meaning, with the result that we often regard with fondness certain traits or behaviors that would appall us if exemplified by some other person.

This formulation has some plausibility, but it is important to emphasize that a person’s core or daimon is not the whole person; rather, the daimon is a person’s true self, as Norton and Kille suggest. The relevant connection is that a person’s core or daimon manifests itself in the whole person, and manifests itself totally.[27] This emanation of value—the manifestation of the beloved’s core in even his problematic traits—is what allows us to accept those traits. It may be objected that acceptance is not the same thing as love, and so perhaps we have accomplished little in the way of showing that it is possible to love the whole person. The appropriate response to such an objection is that insofar as love of the whole person requires a pure love of each of the component parts of the whole, rather than their mere acceptance, such love is impossible; the problem posed by objections (1) and (2) above remains insoluble.

We can move beyond the impasse generated by the intersection of that problem with Tolstoy’s dictum, however, if only we recognize that we can retain the basic principle that underlies that dictum without necessarily making any mention of the “whole person” as such. If the beloved’s true self, which we love, is manifested in every facet of his being, and if I accept those facets that I do not love precisely on account of their participation in the true self, then effectively I have allowed my love to guide my acceptance of those features, and have established a direct point of contact between not only any undesirable feature and the true (loved) self, but also between my love and a given undesirable feature. 

Perhaps a better way of moving beyond this impasse is to refer back to the preliminary remarks of part I and argue that objections (1) and (2) above wrongly presuppose that undesirable characteristics are within the domain of love. Why should we believe that it is even a matter of loving or not loving uninstantiated universals (e.g., the property of being willing to put the toilet seat down)?[28] In fact, it seems that there is considerable similarity between our affective attitudes concerning the often trivial characteristics we wish our beloved had and the kinds of objects of our “love” we safely dismissed in part I. I do not love bathroom etiquette and my beloved’s conscientious observation of it any more than I love Chinese food; these are both things that I like. I submit that for any unloved characteristic of the beloved, if its opposite were miraculously instantiated, it would not be appropriate to say that I love that new characteristic. I would be pleased (perhaps even exultant), but that would only be because the person that I love now has yet another characteristic that I like. It is not the addition or subtraction of likeable qualities that determines the scope and depth of my love; that idea presupposes that “like” and “love” are commensurable when in fact they are not. (Recall our discussion of “fondness” in What Love is Not, above.) In summing up, it will be useful to quote Brümmer at some length.

I could have admiration for James because he has some admirable characteristics. This implies...that I would admire William as well if he should have the same characteristics... [and that] my admiration for James would stop if he should lose his admirable other words, my admiration is directed primarily at James’ characteristics and secondarily at James as the instantiation of these characteristics. In this respect love is different, since it is directed toward a person as a particular and not as an instantiation. Since I do not love you because of your characteristics, my love for you does not entail that I everybody else who has the same characteristics, nor that I should stop loving you [if you lost them]...My love for you is a love for you and not for your characteristics apart from you. I could also love you in spite of disapproving of your characteristics. The only thing which my love for you excludes is that I should be indifferent to your characteristics...both love and admiration require an objective appraisal of the characteristics of their...objects, but admiration is based on the outcome of this appraisal, whereas love is not.[29]    

This shows that when we talk of loving persons, it is axiomatic that we love something essential, unique, immutable, and irreplaceable. The point is that in order to uphold Tolstoy’s maxim, we need not actually love everything about the beloved precisely because none of the non-core characteristics are even capable of being loved in principle. Rather, it is a matter of liking or not liking these things. This suggests that the entire discourse is misguided: the claim that “in love, we love the whole person,” if tenable, must concern something quite apart from the ordinary discussion of surface-level characteristics, which is what I have attempted to address by instead discussing the beloved’s daimon.   

Such a discussion is necessary because it is important from a theoretical perspective to be able to secure the possibility of loving and being loved as a whole.[30] Only then have we achieved the “full intimacy of love,” in which “the full person is known...”[31] Nozick thinks that to be known fully is for the lover to know about the traits that put one at risk of being unlovable—insecurities, character defects, and “areas of incompetence.”[32] Although such disclosure, issuing from trust and openness, is essential to being loved for the person one truly is, I think that in order for the beloved to be “fully known,” the lover must also have knowledge of what makes the beloved truly irreplaceable and indispensable to him; it will be the knowledge of a unique essence at the beloved’s core, as I’ve suggested. To be known “fully” means that our lover “knows us as we are,”[33] but to be known as we are, I think, means for our true self to be known. And what constitutes that is our daimon.

Two closely related objections that might be raised against what has been said about a person’s core or daimon are that (1) there is no reason to suppose that individuals have cores as such, or that in loving a person we love such a core and that (2) even if persons have such cores, it is not plausible that they are immutable or unchanging (i.e., in loving one’s core, the objection goes, we only love a cluster of contingent, incidental features).The first objection, I think, can be defeated while addressing the second. On the assumption that such a thing as a core exists within human persons, if it were the case that a person’s core could change, we would not be loving the same person over a long period of time and over the course of countless changes (both positive and negative) to that person’s interests, personality, and identity.

Take, for example, the above-mentioned problem of undesirable (unloved) characteristics of the beloved. We hope for the improvement of the beloved, that is, for the elimination of these characteristics. But we don’t hope for this improvement so that we can (finally) love them—we already do, and so what we really desire is that the object of our love be someone we also like through and through. By hypothesis, both before and after the desired improvement we love the person in question. But this shows that our love cannot depend upon those specific characteristics being improved. It also suggests that our love does not depend upon the other set of superficial characteristics (that is, the positive ones not in need of improvement). We can see, then, that it is both necessary and sufficient to posit a core that underlies all characteristics for the purpose of explaining the endurance of love throughout myriad changes to interests, personality, and identity. (It may be useful to tie this concept to inquiries concerning personal identity. From what has been said here, it would be correct to infer that I equate personhood with the possession of such a core. A person’s true or essential self is her daimon. And the suggestion that such a core is unchanging should be accepted lest we succumb to the pitfalls associated with being unable to explain the persistence of persons through time. Thus, the mystery of the endurance of love throughout changes to the beloved runs parallel to the mystery of the persistence of persons through time—both of which are solved by positing a metaphysical core.)  Further, we can see that this core must be immutable and unchanging, or else it would itself be simply a cluster of more characteristics, themselves prone to change without effecting a change in our love. In short, if everything were changing, we would not love the same person over time.[34] Instead, we would repeatedly be pursuing relationships with people who happen at a given time to be the best exemplars of the characteristics we like or “love.”[35]   

From what has been said, it is clear that the sort of knowledge that is required for love is not concerned with the contingencies of the beloved's identity; as a lover, I must be in possession of some knowledge more profound than the mere aggregate of incidental facts (e.g., the beloved's favorite movies or music, or personal history more broadly). Why? Knowledge of the beloved is under discussion precisely because it would be useful to have at our disposal a causal explanation of the formation of love, viz. how and why we begin to love a particular person in the first place. And it seems unlikely that knowledge of any particular (incomplete and contingent) set of personal characteristics—the stuff of mundane knowledge of the beloved—could serve as an adequate explanation of why we begin to love someone. In a trivial sense, loving someone ordinarily requires first knowing (and being attracted to) those personal characteristics, but this has less to do with a direct causal explanation of love than an explanation of why we devote the time to the person in question that’s required for the acquisition of knowledge of their essence or core. And I believe it is rather this latter acquisition that stands in a causal relation with love.

Perhaps the reader will agree at this point that love requires knowledge of the beloved’s core or daimon. Of course, she may still be curious about the precise content of such core—or what it is, exactly, of which we (as lovers) must have knowledge. Unfortunately I do not have an answer that will hold true in every love relation. Part of the reason for this, however, is simply that the contents of daimons are, by definition, unique to specific persons. The further inability to concretely specify even clear formal features—while certainly problematic—is, I think, a problem that any theorist of personal identity faces.

Although I have characterized the emotional response as dependent upon this uncommon knowledge of the beloved’s own profound truth, in some sense it is nonetheless possible to experience emotions ordinarily called “love” without the deep knowledge of the beloved described here. This phenomenon—“love” as simply an emotion without knowledge—is precisely what we find in the courtly tradition, which thrives on distance and suffering. How might we bring this into agreement with the more general claim that love, the emotion, occurs only as a response to knowledge? As mentioned in part II, Roberts entertains the idea that love is properly understood as an emotion only to the extent that it readily disposes us toward a host of other emotions. I think that this is right, and it is especially useful for present purposes. Adapting Roberts’s principle to our model, we might say that what disposes us toward the relevant emotions is indeed love, but that this is simply (or at least initially) the deep knowledge discussed above. If Roberts’s claim (or my modification of it) is correct, then there is nothing to prevent an individual from experiencing those emotions when unprovoked by love (knowledge); indeed, the emotions that love is capable of producing probably occur at a greater frequency in an individual’s emotional life outside of the context of love. Thus one feels joy, sadness, anger, and empathy even when one does not love. Likewise, in the love-without-knowledge phenomenon, the courtly lover experiences some of the emotions characteristically produced by love itself (e.g., passionate desire and suffering, anxiety and frustration) but not love proper. In this sense, it seems more reasonable to call such an individual’s experience not love, but rather something superficially resembling it, that is, a more or less groundless experience of the range of emotions that love is capable of eliciting. It is a mistake to think, as the courtly lover does, that such emotions could be love; at best, they are the sensations and feelings that love produces in us. But we can see that in the context at hand even that is too much, since love’s production of emotions hinges on the prior acquisition of knowledge of an incomparable value of the beloved, the utter absence of which is the defining feature of courtly love.

One final remark: the case of courtly love can, I think, also be used to corroborate the basic presupposition of this project, namely that love ought not to be construed as a mere emotion. The courtly lover has all of the relevant emotions, but other than that they all happen to take the same intentional object, there is nothing to distinguish such a lover from an individual who feels exactly the same emotions without loving anyone. It is both undesirable and inadequate to define love merely as what happens when the feelings of joy, anger, and sadness all happen to concern some individual, and so we must further posit their grounding, knowledge.


Love between persons is a relation which is embedded in an attitude and evidenced by characteristic emotions. But, most importantly, it has as its occasion and its source a deep and intimate knowledge. This knowledge is not unlike the knowledge that Norton and Kille believe is posited by Platonic eros. If their interpretation of Plato is correct, then the theory put forth here would have as its basis metaphysical postulates of a roughly Platonic nature. I have attempted, however, to outline the ways in which this theory escapes the popular criticisms issued against Platonic eros. I have also provided a working sketch of what a “metaphysical core” might look like in form, and I have argued that the possibility of the existence of such cores should be taken seriously. This has been accomplished primarily through a discussion of the terms intrinsic to the philosophy of love. But I have also drawn on resources outside of this field—namely, those in metaphysics itself—and have invoked the subject of personal identity as one that provides prima facie support for the claims I have made. That is, whereas the denial of a metaphysical core leaves us defenseless in the face of the problem of the persistence of human persons, an affirmation of such a core not only bodes well with a general theory of the nature of love, but also serves as a good starting point for any theory of personal identity.     


New College of Florida

Sarasota, Florida

About the Author



Works Cited

Badhwar, N. K. (1989). Friends as Ends in Themselves. In A. Soble (Ed.), Eros, Agape, and Philia: Readings in the Philosophy of Love (pp. 165-187). St. Paul: Paragon House.

Brümmer, V. (1993).  The Model of Love. New York: Cambridge University Press.

Buber, M. (1970). I and Thou. New York: Simon and Schuster

Johann, R. (1966). The Meaning of Love. Mahwah: Paulist Press.

Maritain, J. (1973). Person and the Common Good. Notre Dame: University of Notre Dame Press. 

Norton, D.L., & Kille, M.F. (1989). Philosophies of Love. Lanham: Rowman and Littlefield Publishers.

Nozick, R. (1989). The Examined Life: Philosophical Meditations. New York: Simon and Schuster.

Nussbaum, M. C. (2001). The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy. Cambridge: Cambridge University Press. 

Roberts, R.C. (2003). Emotions: An Essay in Aid of Moral Psychology. Cambridge: Cambridge University Press. 

Soble, A. (1989). Eros, Agape, and Philia: Readings in the Philosophy of Love. St. Paul: Paragon House.

Solomon, R.C. (1981). Love: Emotion, Myth, and Metaphor. New York: Anchor Press/Doubleday.

[1] The Examined Life, p. 68.

[2] Emotions: An Essay in Aid of Moral Psychology, p. 296.

[3] Ibid., p. 297.

[4] Of course, it would not be reasonable—nor would it be desirable—to suggest that in ordinary discourse we restrict our use of the word “love” in this manner. Such a move would do violence to ordinary language. Rather, the point is that certain attachments we are inclined to call love are better explained otherwise; if they are not explained otherwise, they threaten to confound any understanding afforded by a conceptual analysis of love.

[5] Love: Emotion, Myth, and Metaphor, p. 34.

[6] Emotions: An Essay in Aid of Moral Psychology, p. 288. 

[7] Ibid. 

[8] Emotions: An Essay in Aid of Moral Psychology, pp. 60-64. 

[9] It would not be unlike defining man as “merely” a mammal: it is true that he is a mammal, but man has certain features (e.g., reason, the capacity for moral deliberation, etc.) that allow him to fall squarely within that classification without it actually exhausting the meaning of “man.” And if the question were, “What is man?” it would be telling a half-truth to allow any conceivable “mammal criteria” to serve as a complete answer. Likewise it is necessary, although not sufficient, to define love as an emotion—or rather, as having an emotional component. But in order to exhaust the meaning of the term, we need a more robust taxonomy, one that rests comfortably not at class or order but rather at genus or species.          

[10] The Model of Love, p. 113.

[11] The Examined Life, p. 81.

[12] The Fragility of Goodness, p. 189.

[13] Ibid., p. 190.

[14] Ibid., pp. 190-191, italics mine. 

[15] The Model of Love, p. 114.

[16] See lines 202e-203b.

[17] See line 242c. 

[18] See line 31d. 

[19] Philosophies of Love, p. 97. 

[20] Ibid. 

[21] Ibid. This appears to be the same thing Robert Johann is referring to when he speaks of a “deep and profound center which is [a person’s self].” (The Meaning of Love, p. 35)

[22] Philosophies of Love, p. 98.

[23] Ibid.   

[24] The Model of Love, p. 117. The view that knowledge is crucial to love is challenged (rather unsuccessfully, I think) by Stendhal’s theory of crystallization, which holds that the lover interprets everything as proof of the perfection of the beloved and that love therefore thrives on epistemic distance and untruth. Although it is true that we believe untruths about the ones we love, and that sometimes we are resistant to the idea of revising those beliefs (even when their falsehood is staring us in the face), this by no means establishes that illusion is an essential feature of the love relationship. To the extent that his theory has accurately described some interpersonal phenomenon, it must not have been love. If it were, we would require a new term for the experience of those that corresponds to the concept expounded here, and it simply accords better with common sense to retain the positive connotation of the term “love” and regard Stendhal’s theory as descriptive of an immature or incomplete love.

[25] Philosophies of Love, p. 99.

[26] This suggestion accords well with Jacques Maritain’s view that, in love, “we love the deepest, the most substantial and hidden, the most existing reality of the beloved being. This is a metaphysical center deeper than all the qualities and essences which we can find and enumerate in the beloved.” (The Person and the Common Good, p. 39) Similarly, Victor Frankl wrote that “love is the only way to grasp another human being in the innermost core of his personality." (Man’s Search for Meaning, p. 134.)

[27] I am encouraged by having found a similar statement in Martin Buber: “[Whereas before the lover had experienced] things as aggregates of qualities... [in the beloved he sees] the core that reveal[s] itself powerfully in the You, embracing all qualities: the substance.” (I and Thou, p. 81, italics mine.)

[28] In her essay “Friends as Ends in Themselves,” Badhwar agrees that incidental characteristics are not the objects of our love. According to her, the beloved is “seen as loveable on account of what she essentially is, and not just on account of incidental features...” (Cited in Eros, Agape, and Philia, p. 166.)

[29] The Model of Love, p. 152. 

[30] For one thing, we typically feel that our beloved’s love for us is not “genuine” or “complete” if we come to find that he or she loves us not for “who we are” as a whole, but rather for some quality or set of qualities we possess. 

[31] The Examined Life, p. 75.

[32] Ibid. 

[33] Ibid.

[34] It might be objected that in fact we don’t always love the same person over time, and that “falling out of love” would be impossible if the beloved’s core were unchanging. It is true that the reality of falling out of love poses a prima facie difficulty for the view that we most fundamentally love an unchanging core; if personal characteristics are unessential in the formation and therefore also in the decay of love, and if an incomparable and irreplaceable value always inheres in the immutable core of the beloved, what could explain the lover’s withdrawal? I can think of two possible answers. First, and most importantly, recall that knowledge of another’s daimon is a seeing or perceiving. When we stop loving a person, the person’s daimon hasn’t changed; rather, it’s that we’ve stopped perceiving the value of the person’s daimon—we’ve lost the knowledge of it, so to speak. The beloved’s core has remained the same, but the lover’s perception of it has become obscured or perhaps overshadowed by negative non-core features of the beloved. Second, it may be that what we sometimes call “falling out of love” is not the loss of knowledge of the core (or a change in the core itself), but rather the knowledge’s sudden failure to produce in the lover the emotions it once did.    

[35] This state of affairs is precisely what we would expect to be the case if Plato were correct. As Badhwar reminds us, “in the Platonic view...the object of love is not the person, not that individual with those qualities, but rather those qualities in any individual.” (From “Friends as Ends in Themselves,” p. 169. Cited in Eros, Agape, and Philia.)