Solomon, Robert C.
Among the Utkuhikhalingmiut ("Utku") Eskimos of the Canadian Northwest Territories, anger is a genuine rarity. In her book Never in Anger, anthropologist Jean L. Briggs suggests that even in circumstances that we would find intolerably frustrating or offensive, the Utku do not get angry. Where most of us would be resentful or furious, the Utku are merely resigned. Anger is not only unreasonable, it is all but unintelligible. According to Briggs, there is not even a proper word for it. (The Utku word used to refer to anger--e.g., in the behavior of foreigners--translates as "childishness.") Among the Ifaluk of Micronesia, murder is unknown, and the Tahitians were described two hundred years ago by one of Captain Cook's officers as "slow to anger and soon appeased." After two centuries of European fantasies and too many tourists, they are still slow to anger and soon appeased. Unlike the Utku, however, they have a rich and elaborate vocabulary for anger. They talk about it all the time, and they fear it as a demon. Anthropologist Robert Levy describes in some detail the curious contrast between the "hyper cognized" obsession with anger and its rarity. Sitting in rush hour traffic on Storrow Drive in Boston, by way of contrast, one finds that anger is neither rare nor slow nor easily appeased. Indeed, one would not be readily refuted in the hypothesis that ire is, in the heterogeneous and transient subculture that defines Boston traffic today, the dominant passion of life. And yet, anger is not so much talked about as clearly and frequently expressed, not only in verbal abuse but in physical violence or its threat, in lawsuits and in innumerable petty acts of vindictiveness and vengeance. It is often said, by way of both profundity and platitude, that people are, "deep down"--that is, in terms of their "basic emotions"--all alike. Psychologists (e.g., Paul Ekman and Carroll Izard)  have made this claim on the basis of extensive cross-cultural comparisons of facial expressions of emotion and their recognition. Neurologists (e.g., Paul MacLean and Jaak Panskepp) have made similar claims on the basis of the common structure and functions of the brain, especially its more "primitive" subsystems. Philosophers have long made similar claims just on the basis of common sense and everyday observation of what has traditionally and not without theoretical bias been called "human nature." Customs, laws, governments, marriage and mating rituals, table manners, and religious beliefs may vary from culture to culture, but emotions, at least, are said to be the same from Mount Abu to lower Manhattan, with variations appropriate to traffic and circumstances. It is because emotions are so "primitive"--to use one unflattering expression among many--that they are so common, indeed universal, and the same from culture to culture. Of course, what causes a particular emotion (e.g., anger) might well differ from culture to culture, as will the ease with which some emotions are triggered and the modes of expression and the contexts in which these expressions are thought to be appropriate. There may be considerable embellishment of "basic" (that is, universally "hardwired") expressions and elaborate differences in preferred courses of action, plans, and strategies. And, of course, different cultures will often describe their emotions quite differently--use different metaphors, different names--and in the expression of emotions some cultures display admirable verbal ingenuity and creativity. The language of emotions, the way they are talked about, and the ways in which they fit (or do not fit) into a culture's ethics and customs may vary enormously, and there will be any number of encouraging and discouraging features of the culture and the environment that will provoke, support, excite, suppress, or possibly even extinguish certain emotions or types of emotion. But the thesis remains, as psychologist Gardiner Lindzey insisted when he summed up this old idea in 1954: "emotions, as biological events, are the same the world over." The core of emotion and the basic emotional repertoire does not vary with culture, is not influenced by culture, and is the same from society to society. In this essay, I want to argue for a very different position. Emotions do have a biological basis, of course, and one can plausibly argue, on the basis of a growing amount of empirical data, that there are certain features of the psychoneurology of emotion, at least, that are prior to and fairly immune from cultural determination. Emotions, however, are far more than their biological substrata, and much of what is most important about emotion--in terms of interpersonal and cultural significance, in particular--is far from "primitive, " largely learned, highly conceptual and often conceptualized, and to some degree, at least, culture-specific. In this, I want to agree partially with Princeton anthropologist Clifford Geertz, who has long argued that "not only ideas but emotions, too, are cultural artifacts." This thesis as it stands is no doubt too strong or, at least, too one-sided. And I cannot here (and cannot competently, in any case) try to apply this thesis to the multitude of societies and cultures that display interesting and sometimes bewildering differences in their emotional repertoires, expressions, and language. But I would like to explore the groundwork for such an application and, where possible, suggest some examples that seem to me worth further exploring with other scholars and experts in the field. Philosophy and Emotions, East and West Philosophers have often contrasted "reason and the passions," typically favoring the former at the expense of the latter and defending philosophy itself as the love of reason, the control if not the extirpation of emotion. So, too, casual comparativists contrast the "East" and the "West, " caricaturing the former as a vast continent of mysticism, meditation, and odd Zen one-liners and seeing the latter as founded on a good solid foundation of science, reason, and argumentation. But on one point, at least, the two would seem to agree, and that is in their shared contempt for the passions. The ideal that emerges for philosophers both "East" and "West" too often tends to be a kind of calm, contemplative detachment, if not full-fledged apatheia or nirvana. The truth, of course, is that the distinction between reason and the passions is as suspect and ultimately as misleading as the dichotomy between "East" and "West." To begin with the latter, the cultures and philosophies of India are as different from one another as they are from the cultures and philosophies of Northern Europe, and there are multiple influences, crisscrossings, and similarities between India and Europe via the "Middle East." Moreover, China has little in common with Iran, Japan displays at least superficial similarities with the other "advanced" industrial nations, and what we call the three great "Western" religions all originate in what was formerly called "the Orient." Conquest and colonialism have forged many mergers and compromises, and, indeed, the East-West opposition has been played out within the East and the West, in East Asia, for example, when the infiltration of Indo-Aryan ideas transformed the culture of China in the first millenium,  and in nine-teenth-century Europe, where the primitive "East" (Germany) and the cosmopolitan West (England and France) were often in conflict as competing cultures. And, in Indian philosophy, there are traditions in hard-headed logic, the pride of the "West," that go back well over a thousand years. With reference to the emotions, similarly, no simple statement about philosophers' attitudes will suffice, from either East or West. There are Zen Buddhists who ridicule "dead sitting" and the dispassionate life. In Europe, that passionate movement which in the early nineteenth century received the name "romanticism" has its counterparts throughout the Western philosophical tradition, in Heraclitus and, even before him, in the various cults of Dionysus? Bengali life is as keen on the passions as any Byronic romantic adventure in the West. David Hume, in the middle of the Enlightenment, uttered the shocking pronouncement that "reason is, and ought to be, the slave of the passions." G.W.F. Hegel famously insisted that "nothing great is ever accomplished without passion" (as did Kant before him), and Nietzsche, a philosopher who in many ways belies the dichtomomy between East and West, also attacks the distinction between reason and the passions: "as if every passion did not contain its own quantum of reason," he argues in his notes in The Will to Power. In place of the false antagonism between reason and the passions, other distinctions emerge that are more informed, more promising, and more tantalizing. Some concern internal cultural divisions such as the difference between northern and southern and pre-Han and post-Han China; between Taoism, Confucianism, and Buddhism; and between different areas and philosophies of India and different schools of Zen Buddhism in Japan. In every society, some emotions are prized and honored while others are feared or condemned or held in contempt. Some emotions are considered violent, dangerous, and disruptive while others are considered essential social virtues. Indeed, in place of the general dispute between reason and emotions and the kind of questions typically asked of the emotions, the strategy is always to ask, "Which emotions? When and where?"--for there is no reason to think that emotions are all the same or can be treated sui generis under a single category. Emotions have usually been marginal if not foreign to Western philosophy, even in those philosophers (e.g., Hume and Spinoza) who both recognized their importance and spent considerable effort trying to understand them. (Eastern philosophy, in both China and India, for example, has been much more accommodating to emotion. Hindu philosophy has a great deal to say about the emotions, and Confucian ethics greatly resembles Aristotelian ethics in its concern for the virtues and the role that proper emotion plays in them.) Part of the reason for the marginality of emotions in Western philosophy, I would suggest, is the false dichotomy itself, reason and emotion, which manufactures distinctions and conflicts that hide the intricate intimacy between our more affectionate and excited states of mind and those insights, arguments, and strategies that typically pass as reason. The two most important steps in overcoming this dichotomy--and in overcoming the equally misleading dichotomy between East and West--are to look more closely at the phenomena in question, first of all, and then to develop a theory that does not presume precisely the separation that is being challenged. The best place to begin, accordingly, might be to look at one of the supposed terms under analysis, the category of emotion. Emotion as a Natural Kind Before we can begin to talk about the cross-cultural comparison of emotions, we first have to get clear about what we mean by "emotion." What family or family resemblances are to be included? And what are we to make of the fact that many cultures and languages do not share anything like our category of "emotion"? The Chinese qing, for example, shares some features of our category of emotion but not others, and Chad Hansen, for instance, has argued that there is no exact equivalence between Chinese and English. Traditional Hindu and Buddhist teachings provide well-established categories that somewhat correspond to our category of emotions, but none of them (klesas, rasas) provides a comfortable fit with our categories. Indeed, what are we to make of the fact that there is and has long been disagreement about exactly which phenomena should count as emotions, not to mention the continuing disagreement over the nature of emotion as such? Is desire an emotion? Surprise? Depression? Boredom? Love has often been excluded from the list of true emotions, and faith has often been elevated to a version of reason rather than emotion. Emotions (like passions before them) have been defined in terms of their irrationality, in terms of physiology, in terms of perception, in terms of their underlying cognitions or desires, in terms of their expressions, in terms of their characteristic behaviors, in terms of their distinctive "feelings." At best, it would seem that we can say that "emotion" delineates a more or less agreed-upon set of phenomena in English, at the present time, with a considerable grey area around the margins and little agreement about what it is that qualifies a phenomenon as an emotion. It would seem that any question about the nature and categories of emotion must come to terms with the fact that emotions are in some sense to be identified as "subjective" phenomena, "in the mind" of the subject, though, to be sure, this language of "mind" and "subject" is itself culture-specific and, even within a given culture, often viewed with suspicion. What this means, in one common formulation, is that while we can observe the emotional behavior of others and listen to their verbal expressions and reports of their emotions, we cannot observe the emotion itself. That we can do, in some equally odd and hard-to-analyze sense, only for ourselves, in "one's own case." With regard to our friends, family, and neighbors, this inaccessibility of others' emotions, contrasted with the immediate accessibility we have to our own emotions, presents something of a philosophical puzzle. In fact, of course, it sometimes seems that we recognize other people's emotions better than they do, and they ours, and this in itself casts serious doubt on facile claims about subjectivity and accessibility. With people from a very different culture, however, the inaccessibility of emotion presents us with a genuine dilemma. How do we interpret their behavior, expressions, and reports without simply assuming (in the absence of evidence) that their feelings are the same as ours? How do we extrapolate from what we see and hear to what they feel, to the emotion itself? How would we recognize or understand differences? Anthropologists, long faced with this problem, have adopted a number of not always satisfactory solutions. Some have employed a "method" of empathy, getting to know the language and living in the culture. Others, of a more positivist bent of mind, restrict themselves to reporting what they can observe, abstaining, as much as possible, from interpretation and extrapolation. In practice, probably all anthropologists--and all of the rest of us, too--employ a mix of both methods, translating, observing, interpreting, and extrapolating from our own experience and forming one tentative hypothesis after another, drawing analogies to our own experience and trying to imagine experiences very different from our own. Translation represents a deep problem in such matters, however, although perhaps not so inscrutable or "indeterminate" as W.V.O. Quine once argued. Quine felt that even with reference to publicly observable phenomena, the fidelity of translation is open to question. Where the phenomenon in question is so personal and private--however public the expression and its behavior--the problem is even more complex. What one people mean by a word that we (or they) learn to translate (into English) as "anger" or "love" or "ecstasy" may nevertheless be a quite different phenomenon. Is what the Polynesians call "riri" really the same emotion as what we call "anger"? Is angst the same as anxiety or angoisse? (Why do French and American existentialists leave it in the German?) What one has to translate, in effect, is the whole culture, to see how that emotion--and the name for that emotion--fits into the systematic wouldview, language, and way of life of the culture. I am reminded of what the late J. L. Austin said of philosophical analysis in general, that it was essentially a form of "linguistic phenomenology" (which, he objected, "is quite a mouthful"). I have at hand no more digestible phrase, except perhaps the newly popular and somewhat New Age term "holism," but it is this global, systematic view of emotions, cultures, and language that I want to endorse and develop here. The subjectivity of emotion should not be confused, however, with an easily associated category, and that is the idea of emotions as distinctive feelings, where the word "feelings" shifts between physical sensations of a fairly readily identifiable sort and a very general term that means little more than subjectivity, that is "experienced." What is not essential to emotion, but is often taken as such, is the related notion of "excitement" (or "arousal," in the more scientific jargon of psychology). Some emotions, at least most of the time, may involve some such arousal, of fear and anger, notably, but it would seriously distort our investigation if this were to be assumed as a necessary feature of any emotion. There are enduring emotions that may involve little excitement, for instance those that Hume called "the calm passions." Buddhist compassion escapes from the usual list of the klesas (afflictions) for this reason. Indeed, one might dwell on the Buddhist notion of (what we call) "emotion" as "afflictions" for just this reason. How one understands the nature of emotion, what emotions are highlighted as paradigm cases, will have much to do with whether we tend or even try to think of all emotions as a single kind of phenomenon. If, on the other hand, one looks at the enormous range of emotions, both>West, the temptation to be so reductionist or simpleminded will be humbled. In the context of a cross-cultural comparison, emotion is not an ultimately defensible category, or what Aristotle called a "natural kind." The word "emotion" shifts its meaning from age to age, culture to culture-even where the language remains ostensibly the same. The word "emotion" has been in common use (as opposed to the older term "passion") fer only a few hundred years, and what counts as an emotion also changes. To be sure, some emotions--anger and fear, for example-seem to be more or less regulars. But others, especially various desires and sentiments, and many of the virtues and vices such as generosity and kindness or lust and greed, wander in and out of the category, depending on larger social and ethical issues. Sufficiently important passions, for example love, faith, and hope in traditional Christian theology, have been deemed too important to be counted as mere emotions. (Indeed, the argument that even ordinary secular love is not an emotion but something much more is a thesis that remains popular among philosophers and social theorists.) Indeed, the great rationalist Kant faced something of a problem along these lines himself, for having dismissed all sentiments and other "inclinations" from serious consideration in the estimation of moral worth, he then found himself somewhat at a loss how to understand "respect" and even "the sense of duty." Insofar as the passions have not been demeaned, they have been given an oppositional role in ethics. Thus Hume made the "moral sentiments" the core of his ethical theory, while both Aristotle and Confucius counted "appropriate" and "modest" emotion as virtue. Buddhism, according to the best-known version, is said to reject the emotions--along with desire--in all forms. In fact, Buddhism rejects only certain kinds of emotions, particularly those that we would be prone to call "passions." Again, it is not the nature of emotion that matters so much as the nature and place of particular (kinds of) emotions in a particular worldview. Within a worldview, there may well seem to be "natural" categories of emotion, although even there the taxonomy will more than likely be constantly evolving and there will be tensions and conflicts within. It is not as if emotions are simply biological or psychological entities, existing independently of the language that names, describes, and surrounds them. And insofar as emotions have been so reduced in status, the reason is not to be found in nature but rather in their accumulated place in society, particularly in ethics. The reason for the shift of meaning and status of various particular emotions and nonemotions has much to do with the all-important relationship between emotions and ethics. Emotions are not just disruptions of our otherwise calm and reasonable experience; they are at the very heart of that experience, determining our focus, influencing our interests, defining the dimensions of our world. "The depressed man lives in a depressed world, " wrote Wittgenstein (who evidently knew that world rather well). So, too, all emotions define a world, or worlds, whether frightful, joyful, or infuriating. Emotions are not just "phenomena, " to be analyzed and understood. They lie at the very heart of ethics, determining our values, focusing our vision, influencing our every judgment, giving meaning to our lives. But unlike Wittgenstein's rather solitary world of depression, most emotional worlds are constituted collectively, by the culture, its conception of itself and the world and by the nature of its cultivated interactions and the language in which it describes these interactions. The (Alleged) Universality of Emotion Are emotions (or, at least, the most "basic" emotions) universal phenomena, invariant from culture to culture? Do the various names for emotions in different cultures simply refer to the same phenomena or are they prone to those extraordinarily difficult translation and mutual understanding problems? Why does the universalist thesis seem so appealing (apart from the uninspiring answer that people simply assume what they do not know)? The metaphor that people are "deep down" alike is itself tantalizing, as are a great many metaphors concerning the emotions, but we understand here clearly enough what is being asserted. Stripping away the various layers of language and culture, different terms, different modes of expression, different idioms, different concepts and contexts, different religious beliefs and practices, different rules about property and propriety, different conceptions of right and wrong, and different social "temperaments" that are cultivated through breeding and upbringing, we seem to be left with a being that is in some sense perfectly "natural," who has only the wiring and the wisdom that God or nature provided, and whatever emotions remain have no longer been determined by the particularities of this or that culture or society. What seems so plausible is the idea that there are such emotions, although, to be sure, they will be (by their very nature) rather crude and unsophisticated. This very Rousseauian picture of "natural" human existence is, of course, itself the subject of one of the great philosophical debates of all times. It begins before Socrates and continues with the Stoics. It commands center stage in Hobbes, in Locke, in Mandeville, in Hume, and in Rousseau. It continues in Sartre, Levi-Strauss, and the postmodernists. Are our "natural" emotions, as Hobbes contended, essentially selfish, prideful, defensive, and inconsiderate, or are they as benign and sympathetic as Rousseau and the sentimental capitalist Adam Smith suggested? However we answer such questions, it is hard to get rid of the presumption that there is some natural repertoire of emotions that constitutes "human nature," some set of "basic" emotions that, like the raw materials of an artist or a chef's ingredients, can then be combined and cultivated to provide the rich variety of cultures and temperaments in the world. The contemporary work of Ekman, Izard, and others, as well as the work of various neuro- and sociobiologists, suggests that there are such emotions, and these can be reduced to a small list, such as anger, fear, and dependency,  or a considerably larger list including, for example, guilt, joy, shame, disgust, and surprise. Although we may need to analyze virtually all emotions in terms of their embeddedness in a particular culture, we do not need to engage in any particularly difficult linguistic or anthropological maneuvers to understand and "empathize" with the basic emotions, those that we all share, according to this "universalist" view. It is easy enough to find persuasive examples: children running through the streets in terror away from an armed invader, a friend or family member betrayed and furious, or two friends or a family united after a long and painful separation. But it is by no means obvious, first of all, how much of our emotional lives can be so explained, nor is it obvious that these emotional scenarios are indeed the same, so summarily described. What seems to be terror or betrayal is circumscribed and in part defined by a surrounding network of relationships, expectations, and contextual assumptions. At best, we might subsume the emotion or emotional behavior in question within one of our broad, familiar categories, but then, it is important to note, we will very likely be classifying the emotion according to its initiating cause or condition rather than on the basis of features of the emotion itself. Or, following Ekman et al., we may find ourselves classifying similarities in facial expression, which at least leaves room for a certain skepticism: why should we think that the experience that accompanies (causes? ) such expressions is the same in every case? To be sure, such an argument becomes absurd as soon as it takes on the too-familiar "all or nothing" ("exactly the same, or totally different") logic of some of the current "multiculturalism" debates. But the strong bias toward "exactly the same" explanations prompts an equally strong temptation to deny that sameness, and if the differences turn out to be matters of nuance and perspective, that will hardly be surprising or disappointing. What matters are precisely the differences, if only because the similarities are already so familiar. The arguments for the universalist thesis, however, do not always turn on the examination of particular examples but rather tend to be a priori, that is, simply presumed from the outset rather than put to the test, subjected to exploration, experiment, and analysis, and opened to question, scrutiny, skepticism, and self-criticism. (It should be noted that the contrary thesis, that emotions necessarily vary from culture to culture, has also been taken as an a priori presumption, particularly among anthropologists and those philosophers who sometimes call themselves "relativists." The one dogmatism is not much improvement on the other.) I have often made a plea for empiricism in these matters, that is to say, simply an open mind. Whether or not there is anything that deserves to be called human emotion (and not just, in the trivial sense, that some humans have it) is something to be found out, something to be investigated, whether or not a final theory of "human nature" will ever be available. The question will always be open: do we in fact understand? It happens to most of us, from time to time, that we momentarily realize how little we understand even those closest to us. That is a heuristic (not to mention therapeutic) experience, and we should certainly keep it in mind whenever we are tempted to assert, with unquestioning confidence, what people in other cultures do or must feel. Biology is not destiny. Culture counts, too. As Rousseau pointed out in his own attempt to discern the nature of "natural man, " insignificant biological differences may nevertheless constitute very significant social and cultural differences. And if there are significant and seemingly inborn differences in abilities and temperaments as well, even between members of the same family, why should we not suppose--indeed, how can we possibly deny--that there may well be systematic and significant emotional differences between families, groups, tribes, and whole societies? Neither nature nor nurture need be in question here, for emotional temperament may well be (and no doubt is) the result of both genetics and upbringing. A culture may amplify (or it may suppress) one or another aspect of its emotional temperament. It may, for example, encourage displays of anger or grief or affection, or it may discourage not only the displays but the emotions themselves. In neurological terms, there seems to be little doubt that although the "reptilian" and other "lower" parts of the brain may be essential to emotional excitation, the cerebral hemispheres are also involved in any emotion or passion worthy of the name. Concepts and learning, not just simple "arousal," are part and parcel of every emotion. To make matters even more complicated, the brain itself develops with age and experience, and it seems to be premature if not splitting neurons (as well as abusing the now-prevalent computer metaphor) to debate whether these organic changes involve the "hardware" or "software" of the brain. But this is, in effect, to say that it is by no means clear what is meant by "the brain." The brain is not just a network of neurons but an organic, dynamic, functional part of a larger system misleadingly called "the body," which cannot, without severe paradox and confusion, be readily distinguished from the person, the living, feeling human being. Thus, in traditional Cartesian terms, it is by no means clear what is intended as "the body." It is not a dead but a living body, the body of a person. It is not just the subject of physiology but of psychology, sociology, and philosophy, too. Does it make any sense to talk of "pure physiology" where the emotions are concerned? I would argue that it does not, and as we interpret and embellish (but do not ignore) the findings of the psychoneurologists, what must be added to the workings of the central nervous system are not only such basic terms of psychology as "stimulus" and "reinforcement" but the exigencies of the context, the terms of the culture, and even the concepts of philosophy and religion. Buddhists refer to "the subtle body," and there is a wisdom here that is easily neglected or denied in the mechanical thinking of the Cartesian tradition. The subtle body is the ideal body, constructed through meditation and philosophy. It is developed not as so many muscles and aerobic systems but as a "psychological" entity, if that term, too, does not already presuppose precisely the dualism that is here being denied. Recent attention to alternative medical practices has made it quite evident that there are "maps" of the body--in terms of chakras, chi, and energy flows--that are quite at odds with the black and white drawings of Gray's Anatomy. Thus there are alternative understandings of the body and biology, and with these alternatives our understanding of the emotions may vary as well. But the important point is that biology does not entail the universality of emotions. If anything, biology would seem to establish the possibility of inborn temperamental differences between races, societies, inbreeding groups, families, and individuals. It could be true that all human beings except Clint Eastwood are hardwired to panic under certain terrifying circumstances, but very little of cross-cultural interest follows from this. We want to know: what circumstances? How much panic, and how does this differ from fear? To what extent can this inborn tendency be overcome through education, training, practice, or experience? Is panic acceptable in that society and those circumstances, or is the panic accompanied by shame or humiliation? Is the panic understood as a "natural" reaction or as a weakness? Is it experienced as an involuntary seizure or as a voluntary, albeit spontaneous, act? Quite apart from the more scientifically exciting questions of neuropsychology, there are the undeniable similarities of what has been called "the human condition." It is a simple fact about all of us that we are born into families; we are initially helpless and extremely vulnerable; we are prone to pain, to injury, to illness; and we need food, water, sleep, air to breath, and protection from the harsher elements of nautre. Moving quickly into the realm of philosophy: we are all going to die, and we know that we are going to die. We all have a need, like Harlow's monkeys, to be held, to be touched; and, as it happens, the continuity of the family and the species depends on reproduction, which is prompted by its own, often mysterious (and sometimes dangerous), biological urges to engage in that potentially delightful and much-embellished activity we summarize as "sex." Although courtship rituals, sexual mores, marriages, and the nature of the family vary dramatically from culture to culture, there are certain basic and possibly immutable structures that are essential to all cultures, even in the face of in vitro and surrogate pregnancy, "alternative" family structures, and various sexual revolutions and counterrevolutions. But what these structures might be, and how "immutable" they are, is a matter of considerable political as well as biological debate, and it has been persuasively argued that, in the midst of such controversies, there can be no neutral or "objective" research or arguments. More generally, all human beings seem to be more or less endowed with the ability to think, to imagine, to abstract; and on the basis of these biologically remarkable facilities, cultural plasticity has become the essential mark of the human, furthered and articulated by the various religions and philosophies. But there again, the interesting and by no means easy question is the extent to which these all have some common core, albeit nothing more than the nature of thinking, imagination, and abstraction as such, and there is always the temptation to elevate one form of thinking, imagining, and abstracting as the highest, truest, or only acceptable form. But the idea that there is some core of conditions, however minimal, that is both universally and distinctively human seems beyond reproach. That acknowledged, however, we once again face potential and frequent abuses and faulty inferences from this basic knowledge. Because some of these features of the human condition are universal, it is too quickly concluded that many or all are universal. From the fact that one-on-one male-female copulation is (for the time being) more or less essential to human reproduction and the fact that human infants are extremely helpless and must be raised by adults, it is too quickly concluded that what is oddly known as the "nuclear family" is as such one of the essential features of human nature. From the fact that all human beings are capable of abstraction and imagination it is too quickly concluded that human existence has a built-in or readily cultivated sense of "transcendence," and that "spirituality"--almost always defined in some restricted and sectarian way--is essential to human nature. Even on a more minimal level, the need to eat, for example, allows of only limited universalization. Not only do dietary preferences and prohibitions vary enormously from culture to culture (termites are a delicacy in East Africa; pork is prohibited in Palestine), but nutritional requirements seem to vary enormously, too. The meaning of food and of certain foods also varies, prompting emotional reactions from delight to disgust. True, all human beings need to eat, and our basic biological apparatus is such that the food has to go in one end and waste has to emerge from the other, but around that minimal universality the emotional range of particular practices is remarkable indeed. My point here has simply been to cast some doubt on the idea that the universality of emotions is a thesis so obvious that it can simply be taken for granted, without for a moment suggesting that the opposite thesis, that different cultures necessarily have different emotions, is any gore plausible. Social Construction, Concepts, and Contextual Behaviorism One alternative to a priori universalism, which has come into prominence in recent years, goes by the name of "the social construction of emotions." It has been of particular interest, naturally enough, to anthropologists, but it has proved to have appeal to any number of theorists in philosophy and the social sciences as well. It is the theory, as the name implies, that emotions are not "natural," that there is nothing that is "essentially human," that emotions are "constructed" within and by a society. Unfortunately, however promising as an investigatory theory, social-construction theory lends itself to the same uncritical dogmatism as its universalist counterpart. It is often assumed by radical postmodern and residually Marxist critics, for example, that social-construction theory simply preempts any form of psychological or biological determininsm where emotions are concerned. It is presupposed that there must be a cultural explanation for emotional behavior, even where the commonality of that behavior and its seeming independence from local cues and conditions would seem to be evident. As an investigatory theory, however, social-construction theory promises to open our eyes to precisely those aspects of emotion that are determined and even "taught" by a culture, and to those aspects of a culture that are defined by its emotions as well. The mere idea that emotions are "constructed" within and by a society, of course, leaves out a great deal that is of considerable importance. It does not follow, for example, that societies therefore construct emotions differently. The human condition or our neurology might be such that we rather routinely develop much the same emotional responses, no matter how embellished or how complicated. Nor does it follow that there are not some emotions, at least, which might not be the same the world over. There may be some emotions that are so "basic," whether by virtue of our condition or our neurology, that they appear, virtually unembellished, in every society. Emotions may be "constructed" but they are constructed out of something, from raw material that is, first of all, to be found in human experience, in the human body and in the human condition. Whatever else they may be, emotions are experiences, experiences of circumstances, some of which we share. Some of these shared experiences are bodily experiences, and it is with some good reason that so many theorists of emotion have taken emotions to be bodily experiences or, what is not the same, experiences of the body. William James, in particular, defined an emotion as a perception of a bodily disturbance, and he asks us, rhetorically, whether we can imagine anger without shaking and tension, including an "impulse to vigorous action." So, too, we might ask to what extent we can imagine anger in a creature without a body (though, to be sure, we can imagine an infuriated ghost with a merely ghostly body). We need not accept James' strong identification of emotions and (perceptions of) physiological disturbances to appreciate the plausibility of his account. Human bodies are indeed quite a good deal alike, whatever the culture, and the characteristic bodily experiences associated with strong emotions-consider fear, anger, jealousy, shame--will presumably be very much alike as well. But the extent of this similarity can easily be overestimated, and the cultural influences on and interpretation of such experiences can easily be underestimated. Different cultures--and different individuals, for that matter--have very different thresholds and experiences of even so "pure" a sensation as pain, and the nature and meaning of the pain will vary considerably with its context, its significance, and the evaluations that surround it. A pain suffered without explanation is a different experience from the pain suffered voluntarily, say, in an initiation rite in which one is called upon to demonstrate one's physical courage. So, too, the physiological experiences of an adrenaline rush or any other hormonal changes typical of strong emotions will vary considerably according to the context, the significance of the emotion, and the many evaluative concerns that define that emotion's place and propriety in the society. Getting angry or being afraid in inappropriate circumstances is a different experience--a different physical experience--from getting angry or being afraid when one is expected or supposed to do so. What the social construction theory sees quite clearly (and what too many physiological theories neglect or ignore) is the extent to which such questions of significance and evaluation play an essential role in our emotions and our emotional life. Concepts, however crude, are in evidence in even the most primitive emotional reactions. Social construction theory and its allies--for example, Lazarus' "appraisal theory"--naturally put proper emphasis on the role of concepts, in particular normative and evaluative concepts. But here we want to be cautious in drawing our conclusions, especially in drawing the conclusion that concepts, and therefore emotions, are cultural creations, "artifices," and not natural or bound to biology. The concepts in question may be the product of social learning and sophisticated religious or philosophical teaching, but they may also be a basic component of the human condition-consider the concept of food, the concept of exhaustion, the concept of a physical threat or a threat to one's territory, the concept of kin, the concept of death. If physiology-based theories tend to go wrong in putting too much emphasis on biology and too little on social psychology, social construction theory may too easily be tempted to over-emphasize socially learned concepts and ignore or neglect the possibility of universal concepts based on the contingencies of the human condition as such, perhaps even "instinctual" reactions, which are by no means simply to be dismissed from explanations of human behavior. I have argued elsewhere that an emotion consists, at least in part, of ways of consciously being in the world, which I call "judgments." It is easy to see, I hope, both how emotions thus understood might be clear candidates for social determination and, perhaps slightly less obvious, how they might be largely determined by circumstances and, if one accepts either of the very different ideas of a priori judgments or biologically determined judgments, how they might nevertheless be universal. Judgments require concepts, and the question, then, is what concepts are required for emotions, or, rather, what concepts are required for which particular emotions? Some concepts (e.g., the concept of morality embedded in moral indignation) are clearly dependent on language and culture. Such philosophically sophisticated emotions as moral indignation, angst and anguish, passionate faith, and outrage against injustice seem to require quite complex philosophical concepts (of morality, nothingness, transcendence, and justice, respectively). Other emotions, notably those that one can readily imagine expressed in the eager, disappointed, or anxious face of one's dog, obviously require much less sophisticated concepts. lhich of these more common concepts are socially constructed, socially learned, and socially based? Certainly not all of them. It is an old philosophical prejudice, happily now on the wane, that insists that only human beings, that is, only beings with a language, have concepts. This confusion is, I believe, part and parcel of the confusion of consciousness with self-consciousness and of selfhood with self-reference and self-reflection. But in any case, it seems obvious to me that dogs, for example, make distinctions, evaluate situations, follow (whether or not they formulate) strategies, and recognize people and their intentions as well as objects and their meanings. They also have some sense of themselves, which is to say, they employ concepts, even self-referential concepts. But, of course, not all concepts precede language. The concepts of mathematics (as opposed to the recognizing of certain patterns, which dogs can demonstrably do) require language, a language acquired by human beings only over a sizable stretch of time and with considerable effort and ingenuity. Certain (but by no means all) self-references and concepts of self require a language, and much of what we call self-consciousness requires a very sophisticated language with various syntactic and metalinguistic properties as well as a complex (and controversial) metaphysics. Not every human language has these properties, or, in any case, they vary considerably. Most philosophical concepts, from death to divinity, require language--and what a language! As Nietzsche once wrote, "I am afraid we are not rid of God because we still have faith in grammar." Our language dictates our concepts and thus many of our emotions. Moral indignation requires the concepts of morality. Romantic love, I would argue, requires not only an extremely complex language of the self but the narrative of romance as well. My dogs may adore me, but they do not, cannot, alas, love me. No loss to them, only to me, for only I, not they, appreciate the language and the concept of love. What concepts are required for which emotions? And what concepts require culture as their prior condition? There are both conceptual and content issues here. Many emotions--indeed, I would argue, most human emotions--presuppose participation in society, established relationships, and open possibilities for relationships that can then be ful-filled, promised, violated, or betrayed. Shame is an obvious example, but there are hundreds if not thousands more. One cannot feel shame or be ashamed outside the context of society (although one can, of course, feel or be ashamed when no one else is around; so, too, for embarrassment). One cannot feel any of those emotions of love, loss or nurturing (cf. the Ifaluk fago) unless there are others upon whom to bestow that love, others to lose, others to nurture. But there are also questions about how one might learn this or that concept, possibly apart from the tutelage of a culture. Could one master an emotion by mastering the language and concepts of a society from an anthropology book and a language-learning tape? Does a person learn to be angry? Does one learn to feel affection? Infants deprived of hugging will not easily learn to hug later on. Birds deprived of other birds will not learn to sing, and chimpanzees deprived of other chimpanzees will not learn how to mate. How much of our emotional life proceeds by imitation, by emulation, through the internalization of concepts which can only be taught by the culture of which they are a part? "Assertiveness training" programs teach people how to get angry (or do they just "give them permission" to do so?). (Romantic) love is learned from books, movies, and television, but what could be learned without them? The questions here are as subtle as they are difficult, and the idea of socially constructed emotion, like the idea of natural emotions, is not a fit subject for dogmatic pronouncements. It is hard to believe that there would be no emotions that are shared by every society, despite the cultural embellishments and interpretations, perhaps, and expressed in different ways and subject to different belief systems. We have already suggested several candidates for such emotions. Fear is an obvious candidate, and some sort of affection or other "attachment" and kinship emotions, although even here we have to appreciate just how deeply cultural symbols and interpretations cut into whatever sort of affection we take to be shared. What is meant by "family" varies considerably from culture to culture, and "romantic" love is a quite specifically culturally defined and cultivated emotion. It has recently been argued that even motherly love is a cultural artifact, created and not merely "natural, " and some sociobiologists and many cynics have long made the same point about fatherhood. Compassion, or some sort of distanced concern for others, seems to be the presupposition of every society and it is built into virtually every philosophy. (Thus the near universality of the so-called "Golden Rule.") Ecstasy would seem to be an emotion whose nature is precisely to cut through or leap beyond the ordinary social-emotional constructions, and its universal nature--if such can be defended--might therefore consist in its transcendence of each and every social scheme. Even so, however, the nature of that transcendence would itself have to be expressed in terms of the standards of the culture, so that ecstasy in Buddhism would be very different from ecstasy in Christianity, and would be very different from the ecstasy involved in making a great scientific discovery or winning the Olympic four-hundred-meter run. The ecstasy of orgasm, which is obviously not universal (or for that matter singular) even in its crudest description, shifts both its significance and its essential nature depending on the cultural mores, myths, and prohibitions surrounding the experience. Insofar as an emotion is constructed within and by a society, which is it that is constructed? An emotion, I suggested (and it is generally accepted, except by particularly rabid behaviorists), is first and foremost a mode of experience. This does not mean that it has to be acknowledged, named, or reported. This does not mean that the subject knows better than anyone else what he or she is feeling. This does not mean that the subject of the emotion even knows "what is going on." What it does mean is that the subject must be conscious, that is, must be experiencing something, and that something must be defined in part in terms of the emotion. But how does the social-construction view capture the experience of emotion? It is not that it cannot explain the nature of emotional experience, and I will argue (as I have argued) that the social-construction view provides many of the tools we need to understand how we come to have the emotions that we do have. But the social-construction view as such too easily tends to be agnostic on the question of consciousness and, accordingly, ignores the experience (the "phenomenology") of emotions. What worries me in particular is the tendency of some social-constructivist views to lean sympathetically toward behaviorism, as if what is learned is merely a pattern of social behavior, not a mode of personal experience. To be sure, the two are intricately intertwined, as William James so clearly recognized. Unfortunately, James confused and conflated his insights concerning the intimate interrelation between emotions and expressive behavior with a very different physiology-based view of emotions. The question--which was not a question for James--was whether the emotion is also, indeed essentially, an experience. For James (and for me) the answer is an obvious "yes." One cannot be content with a theory that explains only the learning of emotional behavior (and the conditions and contexts in which it is appropriate to display that behavior) but does not account for the experiential content of emotion. Which brings us back to the question of cross-cultural comparisons of emotion. What the social-construction theory does very well for us is to remind us to let go of a priori prejudices about the universality of emotions and look carefully at the different contexts and causes, the different conditions, the different expressions of emotion, the different emotions that are "appropriate" in different societies. But once one subtracts all of these differences, what is left? One is tempted, at least, to say: the emotion--the emotional experience--itself. But the social-construction theory provides us with an excellent way of approaching the problem of mutual understanding without falling into the pitfalls and a priori presuppositions of "empathy." It is not enough to empathize with people from a very different culture. One has to know the rules, the conditions, the mores, the local myths, and popular expectations. One has to understand the society and not merely the emotion. Understanding--or "empathizing"--does not mean sharing (actually having) the experience of others. What it does mean is comprehending enough of the context and its meanings to appreciate the kinds of concepts and evaluations that are definitive of (and "the reasons for") the emotion. To understand your Jamaican neighbor's panic about losing a tooth and not being able to find it, perhaps you need to learn something about certain common Jamaican superstitions; and whether or not you come to believe them, at least you will be able to understand why he or she should be so upset about something that (to you) seems without significance, the mere inconvenience and expense of having to go back to the dentist. If you do not understand why your Japanese visitor is so visibly shaken by your asking him whether or not he is hungry, perhaps you should learn something about Japanese etiquette and the utter inappropriateness of such a question. To understand the suicidal behavior of the Chinese student who failed your course, perhaps you should learn something about the pressure he or she feels from the family, that sense of disgrace which, perhaps unfortunately, is virtually unknown to most American students. If you don't understand why your African American coworker gets so indignant when you continue to use the word "Negro, " perhaps you should learn something about the psychological legacy of slavery and the importance of group autonomy, minimally, the right of a people to name and identify themselves. According to the contextual behaviorist, it is neither the pattern of behavior nor the facts of the matter that are in themselves significant. It is the pattern of behavior and the facts of the matter in context and as understood by the participants. This is, perhaps, considerably weaker than the full-blown social-constructionist view but also much richer than the standard behaviorist view. It is, however, the key to a workable theory of the cross-cultural comparison of emotions, precisely because it puts the cultural context and culturally defined experiences in first place, displacing the usual role given to "feelings" and physiology or to mere patterns of behavior, pure and uninterpreted. It is in this context, I believe, that we might best appreciate the force of a familiar argument to the effect that the Chinese sense of ritual (li) wreaks havoc and ultimately undermines the familiar Western distinction between the subjective and the objective. In ritual--but not only in ritual-it is the behavior that counts, though not just the behavior. Confucian doctrine--like the Aristotelian doctrine of the virtues in ancient Greece--is rich in admonitions concerning the proper state of mind and the appropriate emotions in such behavior. In such behavior, the distinction between what is "in" the mind and what in intended, meant, or "expressed" in the behavior is clearly beside the point. What the Confucian insists upon is what the contextual behaviorist hypothesizes: that the total meaning of the performance is to be found in the performance itself, in its proper context, performed in the appropriate manner. There is nothing "left over, " no ghostly performance behind the performance, like a dance coach behind the curtain going through the motions with her pupils. But neither could a ritual, or for that matter any expressive behavior, be carried out by a robot, "mindlessly." (This common translation of "effortlessness" in ritual and habitual action leads to a great deal of misunderstanding.) According to the contextual-behaviorist point of view, cross-cultural variation in emotion is only to be expected, but on a "deep" level, that of interpretation and significance, not on the "superficial" level of behavior as such. Thus we can accept and admire Paul Ekman's painstaking research on the apparent universality of facial expressions of emotions without accepting the universalist and reductionist theory such research is sometimes employed to advance. An emotion is not merely its physiological, perhaps hardwired, behavioral expression, and even if its bodily behavioral expression were demonstrably universal, that does not imply that the emotion as such is universal, for the emotion is its significance, not just its expression? And once we move away from the minimal expressions of emotion in winces, frowns, smiles, and raised eyebrows--just as we moved from the minimalist-sensation aspects of emotion of Rylean "thrills, twinges, pangs, throbs, wrenches, itches, pricklings," and so on--there is by no means any assurance that emotions will be the same the world over and every reason to think that they will not be. In different climates and in the context of different customs and expectations, in political hierarchies, dictatorships, and egalitarian societies, in the light of different religions and philosophies, "the same" acts have very different meanings and involve very different emotions. A French kiss is an offense in England and therefore may be associated with shame as well as romantic passion. Munching a muffin (or anything else) with your left hand in Yemen is considered disgusting and therefore may be an object of ridicule and humiliation. As that great jail-cell anthropologist the Marquis de Sade pointed out, not without self-justification, in the years just before the Revolution, "how can I still retain any feeling of guilt for having committed... a crime in France which is nothing but a virtue in China? Or should I make myself miserable or be prodigiously troubled about practicing actions in France which would have me burned in Siam?" " Moral science, " he concludes, "is simply geography misconstrued." What is wrong with the social-construction theory of emotions and with contextual behaviorism as well is the fact that, for all of their virtues, they tend to leave out subjectivity, "experience." Sometimes, this is just an oversight, a shift in emphasis. Meanings, social-construction and contextual-behaviorist theorists will rightly argue, are public and not private affairs. More often, the problem is a certain familiar timidity, an unwillingness to talk about one's own experience (for who else's experience would one talk about), or a hesitancy to discuss something that too readily seems undiscussable, "ineffable." If we can discuss in detail the context, the behavior and its meanings, then why should we have to talk about anything else, namely the way it seems (or "feels") to the subject? At the extreme, of course, social-construction theory and contextual behaviorism are simply a reintroduction, under a more respectable name, of the old behaviorist prejudice that not only is there nothing to say about "inner" experiences but there are no such experiences, and any self-respecting scientist or philosopher would not be caught untenured talking about them. But this is a mistake, for whatever else the emotions may be, however caused, however learned, however cultivated and contextualized, they are, first of all, experiences. And it is to the nature of these experiences that I now want to turn. Emotions as Judgments: The Subjectivity of Emotions Defended What tends to get lost in social-construction theory is subjectivity, experience, the irreducibly personal aspect of emotion. A retreat to "feelings" does not answer this objection, for the appeal to feelings then sacrifices not only the sociocultural aspect of emotions but, in an important sense, their personal aspect as well. Insofar as the "feeling" involved in emotion is some set of sensations, notoriously the "Jamesian" affects of physiological disturbances emphasized in the James-Lange theory of the turn of the century, there is nothing particularly "personal" about them, despite their much-celebrated Cartesian properties of immediacy and privacy. What makes an emotion personal is the fact that it is a particular person's outlook on the world, embodying his or her peculiar perspective, background, hopes, fears, and concerns. Love is personal, not because one suffers from those sometimes insufferable physical symptoms but because the lover has a very special and personal perspective on the beloved and perhaps a history as well. Anger is personal not because of the sensations of rage but because of the fact that one has been personally offended or frustrated. What counts as "personal" (and, of course, what counts as an offense, etc.) will vary considerably from culture to culture, but to understand an emotion in any cultural context is to be able to appreciate how the person who has that emotion sees the world. To simply repeat what is by now well-charted ground, emotions are by their very nature intentional--they have intentional objects in the world and are not merely "inner" occurrences--and they consist (among other things) of evaluations. These evaluations are not just "Yea! " or "Nay! '--as the popular positivist A. J. Ayer once suggested--but they cover a wide variety of judgments, from very personal embarrassment, for example, to the objective ethical judgments involved, notably, in moral indignation and feelings about justice or, more often, injustice. An emotion, in other words, is not a feeling and not simply "inner"--if by that we mean something concrete and readily specified, as James intended, a mere "affect," a sensation caused by visceral commotion, an "epiphenomenon." But if we mean something more by "feeling" than what is loosely suggested in the colloquial equation "emotions are feelings," then we had better be able to specify what that is. It is a tiresomely old objection, which I and others have answered many times, that a purely "cognitive" theory seems to leave out the "distinctively affective dimension" of emotions. To be sure, there is still much to say on this topic, but the one response which serves no purpose other than to hide the issue is to proclaim the necessity of some sort of "emotionally relevant feeling," "affectivity," or "affective tone," for this does no more than to repeat the fact to be explained, that it is emotion that interests us and not simply belief. But to understand what a proper response should be, we should go back to the much demeaned notion of "subjectivity" and ask with some care exactly what is meant by the claim that emotions are subjective. Emotions, I claim, are a species of judgment. There are a number of immediate qualifications, which I have elaborated on elsewhere but will only mention here. First and foremost, judgment is not the same as belief, and the fact that critics typically exchange the one term for the other is responsible for a number of confusions. Second, judgment need not be overly intellectualistic. It need not be conscious, reflective, or articulate. We make kinesthetic judgments all the time, for example, usually without being aware of the fact that we are making them, much less articulating them (to ourselves or to others) as we go. Third, emotional judgments (like kinesthetic judgments) are always evoked from a perspective, defined in part (as in kinesthetic judgments) by one's physical embodiment but, more generally, by one's place in the world, one's cultural context, one's status and role(s) in that cultural context, one's personal situation. In this, they are distinctively opposed to mere beliefs, which in the attenuated if central role they now occupy in cognitive science and philosophy/B> of mind are distinctively not perspectival or, properly articulated, indexically located. (Thus such a routine judgment as "It looks like rain" is reconstructed in terms of "a belief in the probability of rain given evidence e at time t at place p.") Furthermore, judgments are not necessarily (and perhaps not even usually) propositional, as beliefs always are. Fearing Satan or being angry at John is not necessarily translatable into any set of propositions of the sort "fearing that Satan will...." or "angry that John...." What importance this may have for cognitive science is beyond the scope of this essay, but the significance for the analysis of emotion can be simply stated. The sorts of judgments that are essentially involved in emotions, as opposed to the beliefs and propositions that they may entail or suggest, are personal, perspectival, and, as such, not particularly vulnerable to the charge that they are merely cognitive and devoid of affect. What do we mean by "affect," assuming that we are not simply retreating to repeat that it is emotion and not belief that concerns us and without treating affect (as Freud sometimes did) as nothing more than the various bodily sensations and other physical feelings that sometimes (but by no means always) accompany emotion? What we mean is that the "feeling" in question is irreducibly personal--it is not anger that is in question here but my anger--that it can be understood only from my perspective (or one sufficiently like it), that it involves a matter of considerable concern to me and is not a matter of abstract indifference or mere curiosity. Perhaps I should quickly say that these first-person pronouns are by no means necessarily first-person singular, and it is not unintelligible--and in some societies it is essential--that the first-person role is fulfilled by a "we," an actual or potential group who either are together or could be in the emotional situation specified. "Subjectivity," in the most minimal sense, simply refers to the first-person standpoint, "pertaining to the subject and his or her particular perspective, feelings, beliefs, and desires, " as she or he otherwise defines it. It does not follow, however, that a subjective judgment is about the subject, that is, that the subject is the object of his or her own emotional judgment. To say that emotions are judgments is to say that they are modes of construal, ways of viewing and engaging in the world, including, sometimes, ways of construing the self. Like most judgments (leaving aside that rather problematic set of Kantian a priori judgments), they are culturally taught, cognitively framed, but implemented by the individual. They are not opposed to but intrinsic to experience. (Here, a lesson from Kant.) They are not just descriptive but constitutive of the world, our world, as fearsome, offensive, appealing, hopeful, painful, devastating, or devastated. They are not (usually) deliberative, and they are often spontaneous, habitual, unthinking, "natural." In Buddhism and Confucianism, there is considerable importance placed on the notion of effortless "nonaction." Emotions, on the view that I have briefly suggested here, are usually the same sort of effortless, spontaneous activity. This means, on the one hand, that they are not full-blooded, intentional actions, but, more important, they do not just happen to us either. They are an essential part of our repertoire of responses to the world, "subjective" in the sense that they are distinctively personal and perspectival in the senses discussed above, defined by their significance. Talking about Emotion We do not just have emotions. We also recognize that we have them. We think about them, talk about them, speculate about them, theorize about them. And this is not the activity of social scientists and philosophers alone. We all do it, and our modes of recognition, thought, talk, and theory are then played back into both the language of emotion and, as important, the emotions themselves. Discussion of emotion in general, however, is for the most part an academic pursuit, parasitic on the very general noncategory of "emotion" and largely confined to the Western tradition. Most discussions of emotion, in academia as well as in the "real world," tend to be focused on particular emotions. In our society, conversations often tend to focus on outrage, envy, resentment, and, in a very different mood, romantic love. We talk very little about grief and very little about gratitude, although these two emotions form the foundation of a great many extended conversations in a great many other cultures. Among the Kaluli of Papua New Guinea, for example, grief and gratitude form two of the central themes of the entire culture, while American males, to be very specific, seem to feel very uncomfortable with both of them. What is particularly interesting, however, is that the amount of talk about an emotion may have very little to do with its actual frequency of occurrence. In Tahiti, people seem to be obsessed with the topic of anger even though it is extremely rare to see anyone get angry. In America, I would suggest with some hesitation, talk about love tends to be much more pervasive than the actual emotion. Robert Levy has suggested that we recognize the differences and the connections between such emotion talk and the emotions themselves with the concept of hyper- and hypocognized emotions. Anger is hyperecognized in Tahiti and gratitude is hypocognized in America. But these differences between talk and emotion are not irrelevant to the emotion, as the chatter of a press-box kibitzer is irrelevant to the action below. Anger is so deionized in Tahitian society that people are too terrified to get angry (and when someone does, it tends to fall into the extremely pathological category of "running amok"). Gratitude presupposes so many judgments about debt and dependency that it is easy to see why supposedly self-reliant American males would feel queasy about even talking about it. Part of the emotional portrait of any culture, accordingly, is an account of how much they talk, how often they talk, and how they talk about the different emotions. And, again, although we might write as if there is or could be a single overarching conception of emotion (or particular emotions), the varieties of emotion chat make this, too, rather dubious. I began by denying, rather perfunctorily, that "emotion" names a single coherent category and refers to what Aristotle recognized as a "natural kind." More important, the changing categories of emotion reflect changing views about emotion, and these in turn lead to sometimes dramatic changes in the experience of the emotions themselves. As Amelie Rorty has argued, rather convincingly, concepts of the mind have changed considerably over the ages, and, in particular, conceptions of emotion have changed. "Instead of being reactions to invasions from something external to the self, passions became the very activities of the mind, its own motions.... During this period [roughly, from Descartes to Rousseau], emotions also cease to be merely turbulent commotions: among them appear sentiments, ways of feeling pleasures and pains as evaluations, and so as proper guides to action." But if we find such dramatic shifts in the meaning of emotion within our own recent intellectual history, how much more dramatic might the differences be between another culture's categories and our own. And as the conception of emotions in general change, Rorty reminds us, the prime examples of the passions change, too. Thus we should expect--and we are not disappointed--to find that the primary emotions, those of the greatest concern, vary considerably from culture to culture. Indian classifications and distinctions between emotions, passions, and afflictions display a very different structure from our own common taxonomies. And even when certain emotions remain superficially the same, they may have very different status and play very different roles in social interactions. Anger is considered merely fear" in Utku culture, and fear is inevitably mixed with humiliation in a warrior society. Family affections, which would seem essential to virtually every society, are very different depending on the conception of family, the conception of family roles, and so forth. And these conceptions, needless to say, undergo change, both gradual and violent, as in the several social revolutions of the past few decades. In contemporary Sinology, a similar debate is raging about the historical changes in the meaning of "qing." While "qing" has never meant "emotion" or referred to any such category, its meaning nevertheless requires some subtle association with certain emotions. The problem, among others, is that we take the category of emotions to be something settled, and then wonder whether or to what extent the Chinese category corresponds to our own. But, of course, it does not, not only because we do not have any such settled category ourselves but because there is no reason whatever to think that the Chinese concept--which seems to have more to do with what we might call "authenticity" than with a particular realm of "the mind"--bears any interesting relationship to the kind of classificatory category that "emotion" is supposed to have in English. So, too, the complex comparative taxonomies of emotions and emotion language in Sanskrit and Bengali distinguish between sattva (lightness), rajas (movement), and tamas (heaviness) as the key attributes of various mental modes. ("Positive" and "negative" are obviously inadequate to represent sattva and tamas, and our excessive sense of emotions as "inner" tends to block our understanding of rajas, which, again, destroys that too-easy dichotomy.) Looking over the specifications and examples of sattva (lightness: cheerfulness, joy, equanimity, bliss, dispassion, nobility) , rajas (movement: egoism, mendacity, lust, anger, desire, vanity, pride, hatred, agony, grief, greed, intolerance), and tamas (heaviness: delusion, paranoia, indolence, oversight, non-discrimination, languor), we realize that the emotional experience of ancient India is being carved up in a way that is not at all familiar to us. Many of the examples above are not even candidates for emotion, according to our more familiar taxonomies, and many of our favorite emotions don't seem to make the list. The Indian psychologist Caraka employs something of a Jamesian model, using pleasure and pain and the harmony and disharmony of bodily humors (flatulence, bile, and phlegm) to characterize and categorize emotions. But though Western physiopsychologists, too, have accounted for the emotions in terms of various "humors" (bile, gall, choler, and spleen, for instance), it should be evident to us that something very different is confronting us here and it is not simply a different choice of fluids any more than it is a simple problem of translation. The notion of "flatulence" as not only an occasional cause but as the basis of emotion will strike the Western reader as odd, to say the least, and the inclusion as emotions of languor, confusion, deceit, hypocrisy, and doubt and the very idea of lightness and heaviness as major categories of emotion obviously points to a conception of the emotions and their role in human life that is very different from our own. Perhaps the most striking thing to note, in ourselves as well as in other cultures, is the prevalence of metaphor in our theories. This is not necessarily a weakness in a theory (the theory of "particles" in physics is, first of all, a series of metaphors), but it is worth noting the extent to which even our supposedly most "scientific" theories are founded on metaphors. More to the point, metaphors are not just "figures of speech." They provide the medium for our experience as well. Accordingly, the most prevalent imagery of the emotions, what I have called "the hydraulic metaphor," is worth considerable speculation. It is the treatment of emotions as fluid quantities, capable of filling, overflowing, and being channeled in the psyche. Freud, of course, accepted this metaphor quite literally, calling the affects variously "Q" and "psychic energy" flowing through the tubing (a.k.a. "neurons") of the "psychic apparatus." The classic notions of cathexis, catharsis, and sublimation are part and parcel of this metaphor, and the new, hip language of "flow" is just another version. It is particularly fascinating to find similar metaphors in other cultures, but with significant differences. Freud, a creature of romanticism, took the passions to be explosive and dangerous, and in one of his most famous diagrams he portrays the psychic apparatus as a rather unstable boiler system. In Buddhism, the fluid metaphor is rather that of the ocean (citta) (a metaphor also adopted by Freud), suggesting calm and boundlessness. In the usual hydraulic metaphor, emotions keep us in a constant state of tension. In the oceanic metaphor, however, emotions represent a transient disruption, a "fluctuation" or "agitation" in an otherwise peaceful sea. The differences here are significant. Thus, in the Western tradition, a great poet like Goethe could write, "With most of us the requisite intensity of passion is not forthcoming without an element of resentment, and common sense and careful observation will confirm the opinion that few people who amount to anything are without a good capacity for hostile feeling upon which they draw freely when they need it." Compare that to the "joy of quietness" in the Dhammapada, and then compare both to the "fire and water" metaphors in Bengali discussions of emotion. Are these differences in theory? Differing poetic images? Or are these not rather differences between cultures, ways of life, and no mere metaphors, no mere theories. Many of the most important metaphors having to do with emotions are those that concern the body. Compare the traditional European discussion of emotion in terms of the bodily humors (and the updated versions in terms of neurology and computer analogies) with the Tantric notion that the emotions are located in the chakras, in the "subtle" non-corporeal body, or with the Yogic suggestion that even the grandest emotions are but intellectual disorders, fluctuations, afflictions (klesas), psychic "sedimentations." Consider, again, the Chinese conception of chi "power" or "energy," and consider what different meanings the body and the emotions take on with such a conception. The hydraulic model is, of course a bodily metaphor, but so are a number of more aesthetic, expressive conceptions of the body, including a number of challenging feminist conceptions, which are just now beginning to work their way into the literature. It is taken for granted, in virtually every culture, that the body has something to do with emotions, that emotions are, in some way, embodied, but these different conceptions of body are by no means the same, and the emotional experiences that involve them cannot simply be supposed to be the same either. Talk about emotion is not just commentary on emotion. It is also, in part, constitutive of emotion. Philosophical and psychological theories of emotion, including those that try to take into account the variations in emotion and emotion language around the world, contribute not only to our understanding of emotion but our emotional experience as well. Given the potential richness of this experience, and given the poverty of any theory that prefers to ignore or deny such differences, why should we settle for any view of emotions that offers us any less? NOTES Parts of this essay were presented at the East-West conference at Mount Abu, Rajasthan, January 1990, and in Brisbane, Queensland, in 1992. An earlier, longer version of this article has been published in Joel Marks and Roger T. Ames, eds., Emotions in Asian Thought: A Dialog in Comparative Philosophy (Albany: SUNY Press, 1994). 1. Jean L. Briggs, Never in Anger (Cambridge: Harvard University Press, 1975). 2. Robert Levy, The Tahitians (Chicago: University of Chicago Press, 1971). 3. See, e.g., Willard Gaylin, The Rage Within (New York: Viking-Penguin, 1984). 4. Paul Ekman, "Expression and the Nature of Emotion," in K. R. Scherer and Paul Ekman, eds., Approaches to Emotion (Hillsdale, New Jersey; Erlbaum, 1984); Carroll Izard, Human Emotions (New York: Plenum Press, 1977). 5. Paul MacLean, "Sensory and Perceptive Factors in Emotional Functions of the Triune Brain," in Amelie Rorty, Explaining Emotions (Los Angeles: University of California Press, 1980); Jaak Panksepp, "Toward a General Psychobiological Theory of Emotion," Behavioral and Brain Sciences 5 (1982): 407-467. 6. Clifford Geertz, The Interpretation of Cultures (New York: Basic Books, 1985). 7. In Europe, the "East" once referred to Germany and the other more backward cultures of Europe. Once Germany had proved its cultural, political, and military prowess, "East" moved eastward to Poland and Russia, and now is used by the Russians to refer to Siberia. 8. Graham Parkes, "Nietzsche and Zen Master Hakuin on the Roles of Emotions and Passion," in Marks and Ames, Emotions in Asian Thought. 9. E. R. Dodds, The Greeks and the Irrational (Boston: Beacon Press, 1957). 10. June McDaniel, "Emotion in Bengali Religious Thought: Substance and Metaphor," in Marks and Ames, Emotions in Asian Thought. 11. Hegel, Reason in History, Trans. R. S. Hartman (Bobbs-Merrill), p. 29; Nietzsche. The Will to Power, trans. W. Kaufmann (Random House). 12. Chad Hansen, "Qing in Pre-Buddhist Chinese Philosophy," in Marks and Ames, Emotions in Asian Thought. 13. On the denial of love as an emotion, see, e.g., Camus, in his Myth of Sisyphus, and O. H. Green, in "Emotions and Belief," American Philosophical Quarterly 6, where he argues that love isn't an emotion but a long-term, complex disposition. Jim Averill similarly argues that love is a "commitment," and, on the popular front, M. Scott Peck argues that "mature love" (as opposed to the "infantile regression" of "falling in love") should not be construed as any kind of mere "feeling." On faith elevated to reason, see almost any of the great Christian philosophers from Augustine to Aquinas, and Immanuel Kant, in Religion within the Bounds of Reason Alone. 14. William James wrote one of his more provocative essays under the title, "Does Consciousness Exist? " and his answer, essentially, was, "no." Heidegger attempted to develop his philosophical language of "Dasein" precisely in order to bypass the Cartesian terminology of "the mind" and "the subject." The whole of behaviorism, both in philosophy and in the social sciences, is similarly based upon a deep distrust (or outright rejection) of all talk about the mysterious "mind." Much of postmodernism, though vehemently opposed to such sciences, comes to the same conclusion, that there is no such unified subject as that which is often discussed (and taken for granted) in traditional philosophy. 15. E.g., Margaret Mead, in her classic The Coming of Age in Samoa (Chicago: University of Chicago Press, 1932); Jean Briggs, in her Never in Anger; and Catherine A. Lutz, in her Unnatural Emotions (Chicago: University of Chicago Press, 1988). 16. Marvin Harris, e.g., criticizes Mead's method in particular, in his Rise of Anthropological Theory (New York: Crowell, 1968). 17 - W.V.O. Quine, Word and Object (Cambridge: MIT, 1960). 18 - In his Grounding of the Metaphysics of Morals, Kant makes the dubious claim that "love as an inclination cannot be commanded; but beneficence from duty, when no inclination impels us,... is practical and not pathological love. Such love resides in the will and not in the propensities of feeling, in principles of action and not in tender sympathy" (Kant, Ethical Philosophy, trans. James Ellington [Hackett, 1983]; original German edition, p. 399). 19. Amelie Rorty has traced some of these shifts and relationships in her book, Mind in Action (Boston: Beacon Press, 1988), and, in particular, in her article "From Passions to Emotions and Sentiments," in Philosophy 57 (1982): 159-172. An excellent, very specific study of such changes and relationships is the account by historian Peter and psychologist Carol Stearns, Anger: The Struggle for Emotional Control in America's History (1986). 20. I have argued this thesis at length in my book, The Passions: Emotions and the Meaning of Life (Indianapolis: Hackett, 1993). 21. John Watson, Behaviorism (Chicago: University of Chicago Press, 1930). 22. Cf. Izard's (1971) list of ten (anger, contempt, disgust, distress, fear, guilt, interest, joy, shame, and surprise), Ekman's (1982) list of six (anger, disgust, fear, joy, sadness, and surprise), Panksepp's (1982) more modest (brain-based) list of four (expectancy, fear, rage, and panic). 23. Andrew Ortony, Gerald Klore, and Allan Collins, The Cognitive Structure of Emotion (Cambridge: Cambridge University Press, 1988). 24. I would like to thank Phil Shaver of U.C. Davis for some entertaining discussions on this particular point. 25. E.g., in "Getting Angry: The Jamesian Theory of Emotion in Anthropology," in R. Levine and R. Schweder, Culture Theory (Cambridge: Cambridge University Press, 1984), pp. 238-254. 26. Rousseau, Discourse on the Origins of Inequality, trans. D. Cress (Indianapolis: Hackett, 1983). 27. A putative counterexample to this claim in fact suggests a fascinating area of understudied complexity. Rage is sometimes cited by neurologists as an emotion that can be neurologically induced, without any particular context or conceptual content whatever. But on closer scrutiny, even such seemingly neurological phenomena must involve some minimal cognitive content and distinct modes of perception and response as well as the characteristic physical manifestations so readily visible, for instance, when cats are used as the experimental subject. And it is important to distinguish rage from anger, where the latter is by its very nature "intelligent" (no matter how inappropriate or wrongheaded). A similar argument can be made for panic (cf. fear) and, perhaps, certain sorts of affective need or longing (cf. love). 28. For an exhaustive (and exhausting) survey of this literature and the various phenomena it explores, see Michael Murphy, The Future of the Body (New York: Tarcher, 1990). 29. Rom Harre, ed., The Social Construction of Emotions (Oxford: Blackwell, 1986). 30. If I seem to be harsh on this particular dogmatism, it is because I was once guilty of it myself. As Goethe famously said, one is no more fiercely opposed to any view than to one which he himself has recently abandoned. 31. James, I would argue, does not actually accept this account either, but the plausibility of physiological "arousal" as an essential component of emotion has influenced many psychologists ever since, notably Schachter and Singer, in their classic "Cognitive, Social and Physiological Determinants of Emotional State" (Psychology Review 69 (5) (September 1962). 32. The most thorough description I know of the physical experiences involved in emotion is in Nico Frijda's Emotion (Cambridge: Cambridge University Press, 1989), esp. chap. 4. 33. See Richard Lazarus, "Toward a Cognitive Theory of Emotions," in Feelings and Emotions, ed. M. B. Arnold (New York: Academic Press, 1970). 34. See Vicki Hearne, Adam's Task (New York: Random House, 1987). 35. Catherine Lutz, Unnatural Emotions. 36. I am ignoring here the very difficult matter of defining "consciousness," except to point out, again, that consciousness should not be confused with self-consciousness, as it has been through most of modern philosophy and, consequently, through much of cognitive psychology, too. See, for a good recent discussion, Daniel Dennett, Consciousness Explained (Cambridge: MIT, 1992). 37. Again, see Frijda, Emotion. 38. I would argue that James has at least three different and sometimes radically opposed theories of emotion. In his 1884 essay, "What Is an Emotion?" he argues both for a physiological and for a behavioral conception of emotions, both of these conflated with the experiential "sensations" that accompany bodily disturbances and movements. But a visceral disturbance or a flushed face is very different from a gesture or a full-blown action, although certain muscular tensions may seem to fall in between. James also proposes a very different view of emotions in his religious writings, notably in The Varieties of Religious Experience, such that emotions clearly emerge as a species of spiritual phenomena and hardly physical or physiological at all. l9. Ekman, Paul, "Biological and Cultural Contributions to Body and Facial Movement in the Expression of Emotions," in A. Rorty, Explaining Emotions (Los Angeles: University of California Press, 1980). See also James Russell's critique of Ekman's arguments in Psychological Bulletin 113. 40. Marquis de Sade, Justine, trans. R. Seaver and A. Wainhouse (New York: Grove Press, 1965), p. 696. 41. "It is not the least virtue of a theory that it can be refuted," wrote Nietzsche, and the virtue of James' theory was precisely that it was sufficiently precise so that it could be refuted, as it soon was, by W. B. Cannon in the early part of this century and by Schachter and Singer more recently. 42. Robert C. Solomon, "On Emotions as Judgments," American Philosophical Quarterly, 1988, and "Phenomenology, Emotions, and the Self," in Emotions in Ideal Human Development, ed. Cirillo B. Kaplan and S. Wapner (Hillsdale, New Jersey: Erlbaum, 1989), p. 67. 43. An excellent argument to this effect is Annette Baier's reply to Donald Davidson's essay, "Hume's Cognitive Theory of Pride," Journal of Philosophy 76 (1978). 44. Ronald de Sousa has helpfully outlined four different meanings of subjectivity in his The Rationality of Emotion (Cambridge: MIT, 1987): there is the response to the phenomenological question, "what it's like to be [angry, in love, jealous]"; there is the projection of one's own attitudes upon the world; there is the fact that some properties require a relationship between a subject and the world ("relativity"); and there is the importance of perspective, the fact that all emotions imply a point of view. 45. I belong to an international society for research on the emotions (ISRE, for short), and one of the perennial concerns of the organization has been the preponderance of American and European psychologists and the comparative paucity of scholars from other fields and, even more so, the almost total absence of researchers from other parts of the world. But the problem is not difficult to diagnose. The very category in question has either no equivalent or not much significance in other parts of the world. Indian scholars, who tend to be well versed in both traditions, have a certain advantage, but still their numbers are small. Chinese and Japanese scholars tend to focus on the physiological aspects of emotion, thus passing awkward questions about culture-dependency and translatability. 46. Steven Feld talks extensively about the Kaluli and their practices in Sound and Sentiment (Philadelphia: University of Pennsylvania Press, 1982). Shula Sommers has done an extensive analysis of attitudes toward gratitude, in particular, in American (and other) societies, in her 1984 study, "Adults Evaluating Their Emotions, " in Malatesta and Izard, eds., Emotion in Adult Development (Beverly Hills: Sage, 1984) . On American attitudes toward grief, see Colin Parkes, Bereavement: Its Psychosocial Aspects (New York: International University Press, 1972) ; also, B. Schoenberg, ed., Bereavement (New York: Columbia University Press, 1975). 47. Levy, The Tahitians. 48. Ibid. 49. Amelie Rorty, "From Passions to Emotions and Sentiments," p. 159. 50. See, e.g., Ortony et al., Cognitive Structure of Emotions, and James Russell, "Culture and the Categorization of Emotion," in Psychological Bulletin 110: 426-450. 51. Chad Hansen, "Qing in Pre-Buddhist Chinese Philosophy." 52. Solomon, The Passions. 53. Goethe is quoted in Francis Richardson, The Psychology of Pedagogy and Anger(Baltimore: Warwick and York, 1918), p. 84, and in W. Gaylin, The Rage Within, p. 79. Moving from the Classical-Romantic to the current and cynical, we might quote a similar line from a popular American children's cartoon show, Ren and Stimpy: "I love getting angry. It makes me feel great!" 54. June McDaniel, "Emotion in Bengali Religious Thought."