Philosophy and Phenomenological Research Harry Frankfurt, The Reasons of Love (Princeton, 2004) Review by Richard Moran

The Reasons of Love is based on Harry Frankfurt’s Romanell – Phi Beta Kappa Lectures of 2000. The book, while brief, provides a kind of summa – a late grand synthesis – of Frankfurtiana. In just 100 pages, it takes up and weaves together nearly all the major themes of Frankfurt’s thought over the past few decades, and all the characteristic virtues of his philosophical writing are on full display here. The prose is highly readable, with a kind of austere grace, while at the same time the philosophical work being done is challenging, penetrating, and does not shy from paradox or controversy. And most of all, the questions being pursued here are fundamental ones not only for professional philosophers, but for anyone facing the question of how to live, or what’s genuinely important in life. If only a fraction of the readers who have made Frankfurt a bestselling author in 2005 find their way to this book, it will be a major advance in the public reception of serious philosophy. The book consists of three lectures. The first is called “The Question: ‘How Should We Live?’”, and focuses on the relations and differences between desiring, caring, and importance. The capacity to care about things, and not simply desire them, is given by the reflexive nature of human thought (17), and what we care about is a matter of our attitudes toward our desires and other attitudes, a question of what desires we are committed to (21). For Frankfurt, caring is not so much a response to objective importance or value as it is a matter of how people and other parts of our world become valuable and important to us. The second lecture, ‘On Love, And Its Reasons’, concerns love as a particular mode of caring, one marked by distinterestedness, particularity, and the identification of oneself with the interests or well-being of the object of love. Love and caring in general are not subject to our immediate voluntary control, but rather are themselves the source of what Frankfurt calls “volitional necessities”, constraints on what it is possible to will, which have their source not in some alien intrusion, but in the will itself (46). Yet paradoxically, “the necessities with which love binds the will are themselves liberating” (64), in part through their role in providing us with final ends, which provide our lives with meaning and focus, and bring the ambivalent, restless self somewhat closer to wholeheartedness. The final chapter, called (after Kant) “The Dear Self”, is a sometimes perverse, but always serious, defense of self-love against the charges brought by moralists over the centuries (although Kant comes in for special criticism), and an argument that self-love should be seen as perhaps the purest and most disinterested form of love. In the end, wholeheartedness and selflove are one and the same (95), and there can be no greater philosophical redemption than that in Frankfurtian moral psychology. Given the sweep of this book, and the dense interlacing of themes from the rest of Frankfurt’s writings, a full discussion of the overall argument would require a study of several of the earlier papers where the themes of autonomy, volitional necessity, and wholeheartedness are 1

given a fuller exposition and defense. What I will do here is raise some questions about the nature of the project Frankfurt is engaged in, and express some uncertainties I have about some particular claims he makes concerning ambivalence and the role of final ends in our lives, as well as about the role of the person as agent in caring and wholeheartedness. In one sense, love is just another human emotion, like disgust, anger, delight or fear, and can take the same sorts of objects. The capacity for it is part of the natural human endowment, however historically determined may be the forms in takes in a given society; and like the other emotions on our list it is both a powerful motivation of human action and the expression of special range of attitudes toward its objects. But unlike disgust or delight, love is also the name of a relationship, primarily a relationship between people, but also extending to places, pets, and professions. In this love is much less of a feeling than an orientation of people toward each other which patterns all their dealings with each other, something that provides a kind of norm that they take themselves to be beholden to, and that structures their sense of what is important, what is trifling, and what is unthinkable between them. This is the love that can be put to the test, that one can strive to be true to, and succeed or fail at. Contrasted with love in these ways, delight or disgust remain ‘mere’ emotions, however powerfully felt or powerfully motivating. They are not the names for the structuring principles of a type of human relationship. Still, love is not unique in this regard either, and hence there is a question about what could be meant by seeing it as foundational to our lives in a way that is different from other primary emotions. That is, there is an issue of how love becomes a matter of specifically philosophical concern, something that either raises or promises to answer philosophical questions for us. For among the welter of human emotions we are prey to, down through the ages it is love that stands out, in a way that distinguishes it from other emotions, in being associated with both mystery and enlightenment. And this fact may itself seem mysterious. For, considered as a powerful human emotion, what is any more specifically puzzling about how and what we may love, or what that love can bring us to do, as compared with how and what we may fear or desire or be repelled by? The primal human drives are several; and are specific, imperious, and intractable in their demands just as love can be. They can raise their own questions about their structure and relations to favored philosophical concepts such as perception, value, will, reason and intention. Philosophers will ask, for instance, whether a certain type of emotion bears a relation to justification and reasons, or should count as a form of perception, what its norms or formal objects might be, how it could be said to have norms in the first place, and what importance, if any, should be given to the experiential or felt aspect of the emotion. We can ask these sorts of questions about any candidate emotion, but these are not the questions that moved thinkers like Plato or Freud or St. Paul to the metaphysical claims they make for love as both a mystery and also something like the fundamental principle of Life itself. By contrast, as puzzling and demanding as they also can sometimes be, the secondary characters of fear or repulsion, not to mention thirst or fatigue, don’t find themselves cast in such central dramatic roles in the story of the ordering and binding together of lives and worlds. How can love, considered as a human emotion among others, even so much as lend itself to such specifically philosophical uses? And that is to ask not just what are the elements in it and in the experience of it that call for some kind of philosophical accounting, but also: to what philosophical concerns can the appeal to love present itself as some kind of answer? For that is the role that love takes 2

on in the philosophical reflections of thinkers such as Freud, Plato, Simone Weil, Tolstoy and others for whom love is not only an experience, or an attitude, or a pattern in human life, but also something like an explanatory category. Frankfurt’s book belongs to this tradition of thought about love, in that the concept of love serves as a kind of binding principle to articulate the relations between the wide array of the most central concepts in philosophy, which Frankfurt himself has done so much to place at the center of contemporary philosophical discussion. Love is the concept that ties together the interlocking definitions of the active and the passive; freedom, autonomy, and identification with oneself; the nature of the will and its relation to reason and to care; volitional necessities, wholeheartedness and ambivalence, the nature of final ends and their relation to both intrinsic and instrumental value; the place of morality and meaning in our lives. Frankfurt couldn’t be clearer about the role that is to be played by love in this story: “Love”, we are told, “is the originating source of terminal value. ... the ultimate ground of practical rationality” (55-6). At several points in the book this terminus is presented by Frankfurt as a corrective to what he sees as a characteristic philosophical inflation of the importance or authority of reason or morality in our lives, as well as an exaggerated picture of the degree to which we are self-aware, self-controlling, or agents whose cares are based on reasons or the perception of value. These criticisms can be seen as belonging to a debunking or disillusioning strain in contemporary philosophy, the philosophical critique of philosophy’s compulsive highmindedness, but in Frankfurt’s thought this impulse has little to do with any reductive ambitions. At the same time, however, the invocation of love of all things as the ultimate ground of practical rationality is a philosophical choice not taken lightly, and raises its own questions about its ultimacy in the story being told. For while we are told that, contrary to the hopes of philosophers, reason does not have authority over our actions, and that “morality does not really get down to the bottom of things” (9), there remains room to ask what makes love, considered as a particular human drive, apt for the work Frankfurt wants it to do here. For philosophers of many different stripes will admit (or insist) that in our final story of human motivation Reason or reasons will bottom out somewhere, in the appeal to something not itself Reason. The issue will be over just what “bottoming out” means or is supposed to accomplish, and what considerations or explanatory ambitions lead to the choice of love to fill this role, rather than Habit, Passion, Inertia, Form of Life, Will to Power, or some other ‘original existence’. So one set of questions raised by the book, considered as a guide to Frankfurt’s thought in the rest of his published works, is why there is this appeal to love in particular to play this explanatory role, and why this choice matters, and whether this difference can survive alongside the disillusioning story running parallel with it. Love enters the picture through reflection on what it is to care about something or someone, which is related to but not identical with desiring something, finding it valuable, or finding it important. And we are led to reflection on what we care about through confronting the most basic questions about how we should live (Chapter One), or what Frankfurt calls “authoritative reasoning about what to do”(9). Nothing can answer this question for us without appealing to what we can, or do, or should care about. Caring about something is not the same as wanting it, since we may desire many things that we do not really care about at all. We can 3

also be in no doubt about the intrinsic value of something without caring about it or giving it any importance in our lives. When we care about something, we may well find it valuable, but the caring itself is not a response to its value (38). In this way, caring is not grounded in reasons; but at the same time caring about something is productive of reasons, for caring about something necessarily involves taking its interests as reasons for acting (37). The reasons we have are dependent on what we care about, rather than the other way around. It is in part for this reason that in this story the aim of authoritative reasoning about what to do or care about is confidence, and not truth; wholeheartedness, and not a demonstration of the conclusive reasons that support one’s caring here. But what is confidence grounded in, or what restores confidence when it has been shaken, or when the question arises for the person “whether he has got it right” (23)? As Frankfurt puts it, “Suppose that somehow he becomes concerned about whether he really should care about the things that, as a matter of fact, he does care about. This is a question about reasons.”(23) Frankfurt does not dismiss this question or claim that caring is just the wrong sort of thing to admit of reasons, but he is ambivalent about it in two related ways. He is skeptical about any active role for the person as such in determining what one cares about, and often points out that love and other forms of care are “not under our direct or immediate control” (44). And, secondly, he suspects that the question of what one has reason to care about is “systematically inchoate” (25) because it is inescapably circular: Clarifying the question the inquiry is to explore consists identifying the criteria on the basis of which the exploration is to be pursued. But this comes to the same thing as affirming the judgments concerning what makes one life preferable to another, at which the inquiry aims. One might say, then, that the question is systematically inchoate. It is impossible to identify the question exactly, or to see how to go about inquiring into it, until the answer to the question is known. (25) And he concludes from this that “the most basic and essential question for a person to raise concerning the conduct of his life cannot be the normative question of how he should live. That question can sensibly be asked only on the basis of a prior answer to the factual question of what he actually does care about.” (26). Certainly the inquiry about how to live must begin somewhere, and must start with some initial stock of beliefs, cares, drives, needs, and goals. Otherwise there is nothing to work from, indeed nothing that could motivate the asking of the question in the first place. And it is also true that any such inquiry must begin with some provisional sense of “the criteria on the basis of which the exploration is to be pursued”. But it’s not obvious that this makes the question itself “systematically inchoate”, anymore than ordinary theoretical inquiry is inchoate since it must begin with, and rely on, an initial set of beliefs and standards for making progress. This would amount to begging the question being raised only if the resultant inquiry did not allow for revision or correction of the assumptions with which it began. Similarly, it may be agreed that the normative question of how to live cannot get going without a provisional answer to the factual question of what one does indeed care about, but so far that is just a reason to begin the inquiry there, not to give the factual question any other priority. Obviously large issues in epistemology and the structure of moral theory are at stake here, but it seems that it is a more 4

foundationalist commitment, rather than features of the questions themselves, that is motivating this skepticism about the question, “how should we live?” or what one has reason to care about. Reasons, after all, come in many varieties, and needn’t aspire to the form of demonstrative proof. A person can give his reasons for caring, or caring so much, about his work or his family by articulating its sources of satisfaction for him, and by describing the aspects under which it is expressive of other values and commitments, both the general and the irreducibly particular. The role of reasons in this sort of discourse is not that of a proof, let alone an effort to compel one’s interlocutor to come to care in the same way about the same things. And conversely, another person’s request for reasons in such a case need not be seen as a demand to produce sufficient justification for one’s caring, on pain of being rationally required to abandon it, but rather an invitation to articulate and make (more) intelligible the nature and form of one’s caring. Reasons commonly also play this kind of role, both in dialogue and in solitary reflection about what one cares about, rather than the role of reasons in proof and refutation. Related to Frankfurt’s skepticism about the role of reasons is a recurrent ambivalence about the role of the person as agent in the structure of caring and in achieving wholeheartedness. Frankfurt puts repeated emphasis on the passive aspect of love and care, meaning both in that we cannot simply choose to care about something, and also in the stress on the volitional necessities that loving or caring about something impose on us (44, 45, 46, 49, 63, 80). Love is “not under our direct or immediate voluntary control” (44), and “what we love and what we fail to love is not up to us”(46). However just as the skepticism about the possibility of rational inquiry about what one cares about seems to rely on a restricted notion of reasons and reasoning, the passivity stressed in these passages seems to rely on a restricted notion of what can be relevantly “up to us”. The claim that what we love and care about is “not under our direct or immediate voluntary control” does not tell us much about the role (or lack of one) of the person as agent in this matter when we consider how little in either our thought or action is under our immediate voluntary control. I can manage to get myself to work most days, but this is not something I can make happen simply by deciding to do so. Even the simplest tasks require a vast degree of cooperation from the world, and this goes for ‘mental tasks’ such as recalling a name as much as complex operations like driving to work. Even deciding to go to work may not be directly responsive to the will, or something I can make happen just by my say so. So is there some deeper sense in which love and care are not “up to us”, that is not shared by nearly everything else we manage to accomplish or are otherwise actively involved in? Undoubtedly there is, but seeing it requires going beyond the alternatives of conceiving what we care about as either a matter of caprice and arbitrary choice or a matter of a necessity to which we cannot help submitting. By way of explicating the kind of necessity he has in mind, Frankfurt makes illuminating comparison between the necessities of love and care and the necessities of reason itself: “The volitional necessity that constrains us in what we love may be as rigorously unyielding to personal inclination or choice as the more austere necessities of reason. What we love is not up to us.” (49) And in this passage and the analogy it develops I think we can see that the relevant sense of something not being “up to us” is not that of something lying outside of our “immediate voluntary control” (44). For admitting a role for the person as agent in the process of caring or achieving wholeheartedness, even a role for the person as reasoner, does not 5

require the power of arbitrary choice over what we care about, any more than the person’s active role in theoretical reasoning requires the power of arbitrary choice over one’s beliefs. Surprisingly, despite the fact that “what we love is not up to us”, Frankfurt claims a role for the person as such in the shaping of what he cares about that is remarkably strong in its own way, and which raises further questions about the notions of activity and agency in his thought, and their relation to the themes of wholeheartedness, final ends, and ambivalence. By way of orienting ourselves with regard to the idea of caring about something, he offers the following supposition: Suppose we cared about nothing. In that case we would do nothing to maintain any thematic unity or coherence in our desires or in the determination of our will. [...] Various tendencies and configurations of our will would come and go; and sometimes they might last for a while. In the design of their succession and persistence, however, we ourselves would play no defining role. [...] It is through caring that we provide ourselves with volitional continuity, and in that way constitute and participate in our own agency. (16-17) The agent described here is a relative of Frankfurt’s wanton, the creature possessed of and moved by first-order desires, but without second-order attitudes towards his desires. But the infirmity described in this passage is explicated later in the book in terms of the importance of having final ends, goals that are pursued for their own sake and not merely as a means to something else. They are what make the difference between the sort of life described above and a life that is potentially meaningful and satisfying. One question that arises immediately here is that of the meaning and importance of what Frankfurt calls “thematic unity or coherence in our desires or in the determination of our will”. There is an obvious value in having enough coherence in one’s desires so that they do not interfere with each other, so they can be collectively pursued without self-defeat. But it is not clear what imperative there may be to any greater unity or coherence than this, or even what such greater unity would be. A person typically values and desires all sorts of things, large and small, short-term and long-term, personal preferences and social ideals. He values a range of different people as friends or family, but values them in quite different ways, he cares about winning in tennis, but not in the same way he cares about global warming or his SUV, enjoys both Schubert and John Lee Hooker, but not in the same situations, and he takes pride in doing a good job at work but also dreams about a different kind of life altogether. Such desires jostle against each other unsystematically and attract our energies and attentions in ways that are as much situationally determined as they are managed by the person himself. There can be a profusion of differences here, and it is not antecedently clear either when greater thematic unity or coherence is desirable for the person, or even just what greater thematic unity would mean. Undoubtedly there are people who are concerned that their various tastes, pursuits, friends, career choices, literary preferences and political commitments all hang together and display a kind of guiding theme, but such an ideal seems just one more preference among others, one that another person may reasonably be indifferent to. And if someone lacks that kind of concern, it’s hard to see what kind of convincing reason could be given to persuade him to bring greater thematic unity to his desires. The very absence of any over-arching unity may be something he values. And with regard to the 6

involvement of the person as agent in the shaping of his desires, the concern with thematic unity may seem misplaced. Shouldn’t the constellation of the person’s cares and desires follow from his focus on the objects of care and desire themselves, and not on the question of whether the resultant array exhibits any thematic unity? There is another strand to Frankfurt’s argument here, however, which relates the question of unity among ones desires to the value of final ends. Final ends are important to us not only for the sake of the value of attaining them, but also because having final ends in the first place is a condition for the importance and meaning of what we do (58). The importance of having goals at all, even preliminary ones, means that “instrumentally valuable activity, precisely because it is useful, necessarily also possesses intrinsic value.”(59) However, he argues, we cannot be said to participate in our own agency if every goal we pursue has only instrumental value for us, is something pursued purely for the sake of something else. Frankfurt quotes Aristotle to the effect that desire is “empty and vain” unless “there is some end of the things we do which we desire for its own sake.”. He goes on to say, “We cannot make sense of what we are doing if none of our goals has any importance except in virtue of enabling us to reach other goals. [...] Otherwise our activity, regardless of how purposeful it may be, will have no real point. We can never be genuinely satisfied by it, because it will always be unfinished. Since what it aims at is always preliminary or a preparation, it will leave us always short of completion. The actions we perform will truly seem empty and vain to us, and we will tend to lose interest in what we do.” (52-3) Understood in one way, this invites the response that many activities that must remain forever unfinished needn’t seem empty and vain to us at all. Some activities, like being a parent or a musician, don’t admit of being finished or completed in this sense. Other activities that do admit of completion can be perfectly satisfying even when unfinished, even when it is known that they will not reach completion. Understood in another way, Frankfurt’s point is not so much about activities which must remain unfinished (which is, after all, the condition of mortals), but rather activities which are pursued exclusively for the sake of something else and have no value or satisfaction in themselves. On this interpretation, it is not clear whether this is a genuine possibility on Frankfurt’s own terms, given his claim that “instrumentally valuable activity, precisely because it is useful, necessarily also possesses intrinsic value” (59). The very having of a goal, even when it is not reached, is still valuable to the agent for its own sake. It is difficult to imagine what kind of life being imagined would only contain desires which lacked the kind of finality he has in mind. A person may check into his job day after day only because it pays the rent, and he may only be concerned with paying the rent because it provides him with a place to live, and he only cares about having a place to live because ... what? What is not clear is that we can continue this way and describe everything he does as something with no value to him in itself but only pursued for the sake of something else, which is also only instrumentally valuable to him. But this is a problem with the philosophical characterization, not a description of a bad way of life. If there is an incoherence in the description of someone utterly lacking final ends in this sense, then there is no position from which we can call such a life “empty and vain”. It may be empty and vain, of course, but not for lack of things pursued for their own sake. 7

There are other reasons for thinking that, if final ends are those things pursued for their own sakes, and not merely as a means to something else, then it is doubtful that they can do the kind of work that Frankfurt wants them to do. For on his account final ends are crucial to the unity of the will over time as well as to overcoming ambivalence and attaining the wholeheartedness that removes doubt and binds us to ourselves. ... without final ends we would find nothing truly important either as an end or as a means. The importance to us of everything would depend upon the importance to us of something else. We would not really care about anything unequivocally and without conditions. [...] It would then become impossible for us to involve ourselves conscientiously and responsibly in managing the course of our intentions and decisions. We would have no settled interest in designing or sustaining any particular continuity in the configurations of our will. (53) Passages such as this one make it clear that there must be more than one sense of ‘final end’ in Frankfurt’s story, at least one of which is much stronger than the idea of a something valued not merely as a means to something else. For we pursue all kinds of things for their own sakes, from playing tennis to caring for those we love, and the fact that the value in question is intrinsic rather than instrumental does not distinguish the quality or importance of our goals. Nor does it mean that the value in question is something pursued “unequivocally and without conditions”. Tennis may be pursued for its own sake, not as a means to anything else, and yet be pursued half-heartedly. It may also be an intrinsic value that we are easily distracted from, or that loses its savor over time, and hence fails to organize and bind the will over the course of one’s life. Further, such values and activities can conflict with each other, and contribute to the very ambivalence that final ends are meant to be the cure for. As Frankfurt himself notes, “Love comes in degrees. We love some things more than we love others. Accordingly, the necessity that love imposes on the will is rarely absolute.” (46) This suggests that Frankfurt’s real point is not simply about final ends as goals pursued for their own sake, but final ends in the sense of over-arching life goals, or core values that define a person’s life over time. But in that case we would need a separate argument to show that final ends in this much stronger sense are requirements of a good life or the meaningfulness of our pursuits. This claim will not follow, for instance, from the earlier argument that there is something incoherent to the idea of a life consisting purely of instrumental goals with nothing valued or pursued for its own sake. And further, when we think of lives we admire or that seem to us to be well-lived, we do not, I think, typically have or need a sense of a life shaped by a final end in this sense. In a given case this may of course be due to our lack of discernment about the person or the life, but it also seems that the subsequent discovery of a the shaping purpose of a final end would not change one’s mind about the admirability or success of the life or the person. This is hardly an infallible test, but it does, I think, point to the need for more argument and further exploration of the topic Frankfurt has opened up for us. Final ends, understood as goals pursued for their own sake and not merely as a means to something else, cannot by themselves provide us with something that is cared for “unequivocally 8

and without conditions” or provide a thematic unity or coherence to our desires. This may not be such bad news, if it is true, as I suspect, that a life of love, care, commitment, and meaning, need not involve such unity or such unconditionality. Frankfurt is of course aware that the priority he gives to the avoidance of ambivalence will not strike everyone as obvious, and near the end of the book he makes his most direct response to the question of the special value of wholeheartedness, and makes provocative comparison with rationality in one’s beliefs. Self-love consists, then, in the purity of a wholehearted will. But so what? What reason is there for us to be particularly interested in wholeheartedness, or eager for it? [...] One thing in favor of an undivided will is that divided wills are inherently selfdefeating. Division of the will is a counterpart in the realm of conduct to selfcontradiction in the realm of thought. A self-contradictory belief requires us, simultaneously, both to accept and to deny the same judgment. Thus it guarantees cognitive failure. Analogously, conflict within the will precludes behavioral effectiveness, by moving us to act in contrary directions at the same time. Deficiency in wholeheartedness is a kind of irrationality, then, which infects our practical lives and renders them incoherent. (96) A divided will can mean several different things. It may mean that a person has various conflicting desires that he knows he cannot jointly satisfy; or it may mean that the person is ambivalent in his relation to a single object of desire, finding it both attractive and aversive; or it may mean that although his will is not in conflict with some contrary value or desire, nonetheless in acting on it he is chronically uncertain and plagued by doubts. One difficulty in the analogy with contradiction in belief is that, while it is true that two contradictory beliefs cannot both be true, two conflicting desires may well both be for something worth desiring. And because the will aims at changing the world, in a way that is not true of belief, there need be no irrationality in continuing to pursue two conflicting aims in the attempt to alter the conditions that made them conflict in the first place. The person who discovers that two of his aims cannot be satisfied together does not yet have reason to abandon either of them. They may still be accurate reflections of what is genuinely desirable and good in the world, and as such, he would be wrong to act as if the recognition of their incompatibility somehow discredited their worthiness to be cared about or pursued. In favor of ambivalence, then, we could say that it is not always clear in life when we are ‘trying to have it both ways’ in the sense that we criticize and which involves genuine incompatibility between goals or values. When things look that way at first, it may only be the willingness to live in ambivalence for some time that motivates and enables one to see how the goals are really compatible after all, or how the values in question are at a deeper level in harmony with each other. The person with no tolerance for ambivalence will sometimes abandon one of the goals or values too quickly, or fail to see how compromising on some third goal will enable him to coherently pursue both of the first two. And we may argue in a similar spirit with respect to ambivalence as a conflicted relationship of the will to one thing, rather than a matter of distinct incompatible goals. When a person is ambivalent with respect to his job, or his country, or his parents, there is always room for the argument that if incoherence is to be located somewhere it is to be located in the object of one’s ambivalence, and not in the will 9

itself. In this way ambivalence may be seen as a positive good, or part of a healthy relationship of the will to the complexity and ambiguity of the world it confronts. A final worry about seeking a cure for ambivalence is that, as with the frustration of desire, there is both the theoretical and the practical temptation for ambivalence to be treated through reduction in what one cares about, withdrawing support from the value or goal that lost out in the struggle, and hence wholeheartedness being purchased at the cost of reduction of self. In this familiar type of response to ambivalence, conflict is avoided not through resolution of the elements of the will, but through elimination of the complexities of the will that created the conditions for ambivalence in the first place. For this reason a practical policy of avoiding occasions for ambivalence is no more to be recommended than the analogous epistemic policy of avoiding error or contradiction at all costs. We can equally avoid the risks of cognitive failure by steady reduction in our cognitive interests and commitments, but surely that condition is a cognitive failure of its own kind. The cognitive quest for deeper or more comprehensive explanations of the phenomena of the world requires a willingness to make one’s belief system more rather than less vulnerable, to extend hypotheses to the point of risking contradiction or disconfirmation. Looked at this way, ambivalence may not be a disease of the will anymore than doubt or uncertainty are diseases of the intellect. To be sure, doubt and ambivalence have their chronic forms, but so do such virtues as steadfastness and confidence. I have only touched on a few of the themes of this book, which is a deep and challenging account of the more troubling and fulfilling aspects of human life. It may be read with profit on its own, but it will serve even better as an introduction to the essays published elsewhere that make up the later philosophy of Harry Frankfurt. For Frankfurt, the fact that we are by nature reflective creatures means that we are ineluctably faced with the question of taking ourselves seriously, a question and a task which the other animals are spared. In this book, and in all his writing, Frankfurt provides us with a stirring enactment of the possibilities and difficulties of such seriousness.1

In preparing this review, I have benefitted from discussions with Borgna Brunner, Nick Halpern, and Martin Stone. 10

1