Divine Passions: The Social Construction of Emotion in India

By Owen M. Lynch

Naked holy men denying sexuality and feeling; elderly people basking in the warmth and security provided by devoted and attentive family members; fastidious priests concerned solely with rules of purity and the minutiae of ritual practice; puritanical moralists concealing women and sexuality behind purdah's veils—these are familiar Western stereotypes of India. The essays in Divine Passions, however, paint other, more colorful and emotionally alive pictures of India: ecstatic religious devotees rolling in temple dust; gray-haired elders worrying about neglect and mistreatment by family members; priests pursuing a lusty, carefree ideal of the good life; and jokers reviling one another with bawdy, sexual insults at marriages.
 
Drawing on rich ethnographic data from emotion-charged scenarios, these essays question Western academic theories of emotion, particularly those that reduce emotions to physiological sensations or to an individual's private feelings. Presenting an alternative view of emotions as culturally constructed and morally evaluative concepts grounded in the bodily self, the contributors to Divine Passions help dispel some of the West's persistent misconceptions of Indian emotional experience. Moreover, the edition as a whole argues for a new and different understanding of India based on field research and an understanding of the devotional (bhakti) tradition.
 This title is part of UC Press's Voices Revived program, which commemorates University of California Press’s mission to seek out and cultivate the brightest minds and give them voice, reach, and impact. Drawing on a backlist dating to 1893, Voices Revived makes high-quality, peer-reviewed scholarship accessible once again using print-on-demand technology. This title was originally published in 1990.

• Language: English
• Publisher: University of California Press
• Release date: Jul 28, 2023
• ISBN: 9780520309753

Book preview

Divine Passions

Divine Passions

The Social Construction of Emotion in India

EDITED BY

Owen M. Lynch

UNIVERSITY OF CALIFORNIA PRESS

Berkeley Los Angeles Oxford

University of California Press

Berkeley and Los Angeles, California

University of California Press, Ltd.

Oxford, England

© 1990 by

The Regents of the University of California

Library of Congress Cataloging-in-Publication Data

Divine passions: the social construction of emotion in India / edited by Owen M. Lynch: contributors, Peter Bennett…[et al.].

p. cm.

Papers presented at a conference held 12/1-14/85 at the University of Houston.

Includes index.

ISBN 0-520-06647-2 (alk. paper)

1. Ethnology—India—Congresses. 2. Emotions—Congresses.

3. India—Social life and customs—Congresses. 4. Love—Religious aspects—

Hinduism—Congresses. I. Lynch, Owen M., 1931-.

II. Bennett, Peter.

GN635.I4D58 1990

152.4—dC2O 89-4975

CIP

Printed in the United States of America 123456789

The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI Z39.48-1984. ®

For

M. N. Srinivas, anthropologist and guru,

and

David B. Kriser, philanthropist andfriend

CONTENTS 1

CONTENTS 1

PREFACE

Introduction: Emotions in Theoretical Contexts

ONE The Social Construction of Emotion in India

PART TWO Love and Anxiety in Intimate Familial Contexts

TWO The Ideology of Love in a Tamil Family

THREE To Be a Burden on Others Dependency Anxiety Among the Elderly in India Sylvia Vatuk

PART THREE Joy and Humor in Public Caste Contexts

FOUR The Mastram Emotion and Person Among Mathura’s Chaubes Owen M. Lynch

FIVE Untouchable Chuhras Through Their Humor Equalizing Marital Kin Through Teasing, Pretence, and Farce Pauline Kolenda

PART FOUR Erotic and Maternal Love in Religious Contexts

SIX Krishna’s Consuming Passions Food as Metaphor and Metonym for Emotion at Mount Govardhan Paul M. Toomey

SEVEN In Nanda Baba’s House The Devotional Experience in Pushti Marg Temples Peter Bennett

EIGHT Refining the Body Transformative Emotion in Ritual Dance Frederique Apffel Marglin

PART FIVE Conflicting Emotions in Cross-Cultural Contexts

NINE On the Moral Sensitivities of Sikhs in North America Verne A. Dusenbery

TEN Hare Krishna, Radhe Shyam The Cross-Cultural Dynamics of Mystical Emotions in Brindaban Charles R. Brooks

CONTRIBUTORS

GLOSSARY

INDEX

PREFACE

The papers in this volume were originally written for a conference on The Anthropology of Feeling, Experience, and Emotion in India held at the University of Houston on 1-14 December 1985. The conference was part of the Festival of India held in the United States during 1985-86. Nineteen highly provocative papers were presented; the nine in this volume were selected because they most directly addressed the conference’s theme. Generous support for the conference came from the American Institute of Indian Studies, the Ford Foundation, and the Government of India. The University of Houston was a gracious host. Pauline Kolenda deserves special thanks for creating a social and intellectual milieu crucial to the success of the conference in Houston and for working hard to see this volume in print. M. N. Srinivas was our honored senior participant and guru providing much pertinent, sage, and witty comment.

The use of words in Indian languages always presents problems of transliteration. Hindi words appearing in English language dictionaries and standard English spellings for proper nouns have been used as much as possible. Otherwise with one exception the system of diacritical marks presented by R. S. McGregor in his Outline of Hindi Grammar (Oxford University Press, 1977) has been used. The exception is that for Tamil alveolar stops and nasals a subscript dash (e.g., n, l, r) has been used. At the request of some authors, final silent a has been noted rather than dropped although in a few places local dialectical and spoken variants have been retained when appropriate. For easier reading, the terminal s indicating the English plural has been added to some Hindi words, although it does not so appear in Hindi. On first mention words in Indian languages have been written with diacritics and thereafter without. A glossary of the most important terms with diacritics is provided.

Two anonymous reviewers provided excellent suggestions for strengthening the book. Elvin Hatch offered an invaluable critique for improving its argument, and Barbara Metcalf first recommended it to the University of California Press. Thanks to them all and to our editors: Lynne Withey, whose interest, encouragement, and sponsorship made the book possible; Amy Klatzkin, who shepherded it through a complicated production process; and Lisa Nowak Jerry, who rescued the manuscript from mispellings, inconsistencies, and grammatical errors.

PART ONE

Introduction:

Emotions in Theoretical Contexts

ONE

The Social Construction of Emotion in India

Owen M. Lynch

Yogis lying on a bed of nails in search of detachment from all feeling, white- bearded gurus preaching meditation on the transcendental, close-knit families in which the aged and infirm live out their days happy and secure in the loving devotion of their children, and ritualists worshiping more by rote than by heartfelt devotion—these are some images through which the West perceives India and the emotional lives of its people. These, too, are the images that the essays in this book seek to replace with pictures of worship based upon deeply felt and deeply motivating ecstatic love, of elderly people anxious and afraid of impending physical deterioration and loss of independence, and of priests pursuing a carefree, lusty, and happy-go-lucky way of life. These new and different images are drawn from ordinary, everyday lives of next-door-neighbor Indians. They are painted by anthropologists who took the time to live with them, listen to them, and learn from them over many months of sharing and dialogue. Each essay in this book also portrays Indian emotional lives different in structure, meaning, and coloring from those of the West, yet all are so framed that they reveal, through dialogue, a common humanity.

Because in India the conception of emotions and of the capacity to lead emotional lives differs from that in the West, these essays raise problems for the West’s understanding of emotion, particularly when it is universalized into a theory and projected onto the Other. Cross-cultural encounters and problems of beliefs, theories, and presuppositions about the real, the natural, and the human are the questions upon which anthropologists thrive and through which they contribute to a critical knowledge of our Western selves. Recently some anthropologists have begun to pose those questions to the understanding of emotions.¹ All the essays in this book, then, have been written by anthropologists with an eye on not only India but also the development of a critical theory and understanding of emotion in the West.

If the passions are precisely those structures which connect and bind us to other people (Solomon 1976:19), then why until recently have anthropologists, who claim to study the structures that connect and bind us into social and cultural systems, either considered them irrelevant or failed to question their assumed nature and operation? Reasons for the neglect or failure are many, and they lie buried in the intellectual history of Western culture and its influence on anthropology’s founders and later theoreticians. To understand how the essays in this book till with the blade of a different plow the virgin soil of emotion in India, a brief answer to this question is necessary.

We must also be clear about what we mean by and understand to be emotion, if we are to understand an Other, such as India. In addition to many academic theories of emotion, there is a Western commonsense understanding of it (see Lutz 1986b). The social constructionist view underlying the essays presented in this volume runs contrary both to some Western commonsense notions and to some academic theories about emotion. Thus, * it is advisable to give a somewhat extended, although by no means comprehensive and adequate, survey of some of these ideas and theories.

Theories of emotion in the West as they have been developed into paradigms for research are of two types: physicalist and cognitive. Until recently the physicalist theory has dominated academic circles.²

Physicalist Theory

Despite the cognitive overtones of his theory, Descartes was the most influential originator of physicalist theories; he ultimately reduced emotion to a subjective awareness in the soul of activities in the body, of passively experienced feeling. Descartes left unclear the relationship between the soul and the radically different and separate body. Hume elaborated on Descartes and considered emotions to be the registrations in the soul of particular feelings caused by primary sensations associated with some idea or perception. Emotion remained, then, a passive awareness, but in Hume it became a sensation mediated by perception into a particular feeling.

This Cartesian view of emotion took its most influential modern twist in the work of William James. For James, bodily changes follow directly the perception of the exciting fact and… our feeling of the same changes as they occur is the emotion (James 1890:449-450). He reverses the everyday notion that people cry because they feel hurt. Rather, they cry, and this physical change is the emotion they experience; emotion is a feeling of physiological changes. James tried to make psychology a science by turning from a method of introspective accounts of feelings in the soul to objective measurements of physiological changes in the body. And so, physiological psychology was born and with it a major modern paradigm of what emotions are.

Behaviorism added little to this paradigm except to shift observation and measurement to patterned responses or operant behavior created by physiological conditions elicited through specific stimuli. Yet emotion itself remained a physiological event. Freud’s early theory, too, is Cartesian because he considers emotions to be cognitively felt responses to physiological instincts or drives blocked by some early traumatic but unconscious event in the individual’s life. The source of anxiety and fear is in blocked drives, and the emotion itself is merely a safety valve to let off their energy or steam. Once again, emotions are passive experiences of ultimately physiological states.

I shall deal with some of the many objections to physicalist theories later in this essay, but it is important to note here that all of them take emotions such as fear, anger, and anxiety as paradigmatic and deal less well with the more subtle emotions such as hope, ennui, indignation, envy, and the like. Almost all physicalist theories consequently separate primary or basic from secondary or derived emotions.³ Moreover, physicalist theories, based as they are in physiology or drives, assume that at least basic emotions are universal.

From an anthropological point of view physicalist theories are interesting because they so well match the basic elements of Western common sense about emotion. For that reason they raise suspicions of Western bias and ethnocentrism. First, in Western common sense, emotions are passive: they are things that happen to us, we are overwhelmed by them, they explode in us, they paralyze us, we are hurt by them, and they threaten to get out of control. Emotional action follows a hydraulic metaphor of forces welling up inside of us or of psychic energy about to explode (Solomon 1984; see also Lakoff and Kovecses 1987). Second, emotions are irrational rather than rational, natural rather than cultural, and located in the lower faculties of the body where they are completely separate from the higher faculty of the mind, their master controller. As such, emotions can perform an important excusatory function, among others, in Western society. Just as someone may be excused for a minor peccadillo because she was upset, so, too, she may be treated more leniently for a major offense if it was a crime of passion or insanity. In the same way crimes committed under the influence of alcohol or drugs are treated as less culpable, particularly in American society (see Gusfield 1981). Third, in the commonsense theory of emotions the extension of the verb to feel from sensations to emotions essentializes them as things, as physiological states; just as one feels the heat of fire, so too one feels the heat of rage. Such a view fits in well with today’s drug culture; emotions can be bought in a pill on a back alley or at the local drugstore. Paradoxically, many medical professionals and addicts agree with the view that emotions are the chemical effects of the pill in the body, just as for some social scientists of the physicalist persuasion primary emotions are the action of neurochemicals on the autonomic processes of the body (Kemper 1987). Fourth, com monsense theory also assumes that at root people around the world are the same in their emotional dispositions; if people share nothing else, they at least share this aspect of humanity. One needs only empathy to understand the Other’s emotions. Finally, according to common sense, emotions are subjectively felt; they are the individual’s most intimate and private experiences. Although physiological, they are known by introspection into the hidden chambers of the self.

The consequences of both scientific and commonsense theories for the development of anthropological interest in emotion were and are profound. Emile Durkheim considered social facts as things; he did not say they were things. Yet emotions were considered things of the individual and, therefore, worthy of study by psychology or other disciplines. Because emotions belonged in the realm of the infrasocial, any explanation of a social fact based on them was wrong (Durkheim 1938:104). In his theory of ritual, emotional displays were social, not individual; therefore, they gave no evidence of individual emotional states. Only ritual’s symbolic and functional meaning could be understood.⁴

Functionalism, especially British functionalism, was set on a radically nonpsychological path by Radcliffe-Brown’s interpretation of Durkheim; emotions, therefore, were eliminated from consideration. Social sentiments were nothing more than structural principles determining the individual. When emotions did appear, as in some studies of witchcraft, some version of Freudian theory was used.

American cultural anthropology did little better than British functionalism. In its openness to psychological interpretation, American cultural anthropology took inspiration from Freudian theory. Therefore, the assumed nature of emotions, as physiological and universal, went unquestioned. Cultures were seen as variously working on, channeling, and shaping universal emotions, and basic personality was the result of cultures working on an assumed universal emotional base. The interpretation of emotions … [was] quite distinct from the emotion itself, thus leaving the emotion proper outside the realm of anthropology (Solomon 1984:239).

To the extent that Max Weber’s influence has been felt in anthropology, it has maintained the separation of emotion from rational thought. His identification of one form of social conduct as emotional, resulting from immediate satisfaction of an impulse, left the nature of emotion in a realm similar to that of the drives or instinct. Yet he felt uneasy about this type of emotional social action and asserted that it often overlaps with value-related conduct, that is, conduct explainable by some cultural value (Weber 1963). Weber’s uneasiness might precisely have led him to a social theory of emotion itself, but this was overshadowed by his task of showing the progressive rationalization of social action with its implication that emotion itself was impulsive, thus less meaningful and less worthy of study.

Finally, fieldwork itself was permeated by naive Western ideas concerning emotion as natural and universal; paradoxically fieldwork confirmed what it should have questioned. That was the problem with both Briggs’s (1970) study of anger as almost totally controlled among the Utku Eskimo and with Carstairs’s (1967) psychoanalytically oriented study of Indian Rajputs. They assumed that they could understand behavior as expressions of Western definitions of universal emotions, but that assumption was precisely the hypothesis to be verified (Solomon 1984:247). Anthropologists have been less reticent about imputing universal emotional abilities to others then [sic] they have been about projecting particular cognitive abilities to all humans (Lutz 1986a:297; see also M. Rosaldo 1984:137).

In a world of strange customs, odd practices, different logics, and alien moralities, it was comforting to assume that others were familiarly human when they laughed, cried, loved, and raged. Especially was this so when loneliness threatened the expatriate field-worker. Empathy assumed that human emotions were universally the same. Therefore, one could understand the Other’s emotions as reflections of one’s own and use without question American English categories for emotion in descriptions of them (Lutz 1988:42). Empathy made fieldwork tolerable as well as scientific, but, as rationalization for a method of cross-cultural understanding, it short- circuited the questioning and problematizing of emotional life itself as well as the possibilities of investigating its cultural construction.

An Alternative Approach: Cognitive Theory

Claude Levi-Strauss signaled a different approach to emotion in anthropology when he said:

Actually, impulses and emotions explain nothing: they are always results, either of the power of the body or of the impotence of the mind. In both cases they are consequences, never causes. The latter can be sought only in the organism, which is the exclusive concern of biology, or in the intellect, which is the sole way offered to psychology, and to anthropology as well. (Levi-Strauss 1962:71)

Although he accepts a Cartesian absolute distinction between mind and body, Levi-Strauss departs from the Cartesian heritage; he considers emotions, although in the body, to be products of the mind. He further explains: Men do not act, as members of a group, in accordance with what each feels as an individual; each man feels as a function of the way in which he is permitted or obliged to act. Customs are given as external norms before giving rise to internal sentiments (Levi-Strauss 1962:70). Levi-Strauss was, perhaps unwittingly, adverting to an alternative theory of emotion in Western culture, cognitive theory, with roots tracing back to Aristotle’s Rhetoric.

A cognitive theory is one that makes some aspect of mental activity—a belief, a thought, or a judgment—essential to emotion in general and to identifying separate emotions in particular. Aristotle says:

Take for instance, the emotion of anger: here we must discover (i) what the state of mind of angry people is, (2) who the people are with whom they usually get angry, and (3) on what grounds they get angry with them. It is not enough to know one or even two of these points; unless we know all three, we shall be unable to arouse anger in anyone. The same is true of the other emotions. (Aristotle 1941:1380)

Aristotle is saying that the cause of anger is primarily in a state of mind, in beliefs about others and the reasons that activate those beliefs. Such beliefs have grounds in evaluations concerning the angry person’s relations with others. Clearly, a judgment implicates the feelings that are part of the emotion (Lyons 1980:33-35).⁵ Artistotle’s thought is particularly modern because he places emotion not merely in the mind but also in a sociocultural context.

The demise of functionalism and the attack on positivism as a philosophical justification opened the way for the development in the social sciences of this alternative cognitive tradition concerning emotion. It has had particular appeal to anthropology because of its openness to the centrality of culture as constitutive of emotional life rather than as an overlay, an interpretation, or a reflex of an assumed biological or physiological universal base. Cognitivism, as an approach to the study of emotion, has developed into many variations, some of which retain a universalist perspective. One variation of cognitivism, social constructionism, is particularly influential in anthropology. Social constructionism modified by insights from deconstructionism is the theoretical perspective within which most essays in this volume must be understood.

Social Constructionism

The literature on social constructionism in the study of emotion is by now rich, open to further development, and varied in nuance, assumption, and method.⁶ Nevertheless, I think most constructionists agree with certain basic propositions:⁷

1. Emotions are essentially appraisals, that is, they are judgments of situations based on cultural beliefs and values. If you tickle an Indian, he will laugh. But, if Indians of the Chuhra caste joke with their sister’s groom at a wedding, they, too, will laugh (Kolenda, this volume). Laughter is not the emotion of humor; it is a metonym whereby the physiological effects of the emotion stand for the emotion (Lakoff and Kovecses 1987:196). Idiots may laugh, but a certain intelligence and a native’s grasp of a culture are necessary to understand a joke and experience humor. The joke is not a stimulus to a universal emotion; rather, getting the joke’s point—that is, appraising it correctly— constitutes the emotion.

2. Emotional appraisals are constitutive for the individual and deeply involve, even move, the self in its relationships to social others, things, or events. We speak of our deepest emotions as meaningful. The feeling does not move us, rather, the emotional appraisal is so full of meaning that it constitutes a moving experience for us.

3. As cultural appraisals, emotions are learned or acquired in society rather than given naturally. They are, therefore, culturally relative, although theorists differ on the degree to which this is so.

4. As appraisals, different emotions are identified by their intentional object, that is, by the object as understood by a cultural interpreter, either self or other. Emotions, then, implicate in some way agent responsibility. For example, Jane’s secretary, John, comes into her office and spills coffee on her new dress. She can either feel anger because his behavior is careless and clumsy or feel pity because he is distracted by his child’s serious illness. Her reaction will depend upon what she knows ofjohn’s situation and how she evaluates it. Likewise, at a marriage sisters-in- law mercilessly tease Indian Chuhra grooms (Kolenda, this volume). The grooms, depending upon how they appraise the situation, respond with silent sullenness, tears of humiliation, or retaliatory good humor. There are probably no universal, objective situations that, without agent appraisal, automatically trigger in humans innate emotional responses such as humor or fear.⁸

5. As appraisals, emotions also involve moral judgments about prescribed or expected responses to social situations. When Jane feels pity for John because of his worry about his child, she implies that his behavior is excusable. When Vatuk’s (this volume) Indian elderly express anxiety about impending old-old age, they imply criticism of family members who may withhold the seva or service that Indian culture expects from them.

6. Finally, emotions, because of their moral content, have consequences for the way individuals relate and for how social systems are variously constructed and operate; they have functions. In short, emotions presuppose concepts of social relationships and institutions, and concepts belonging to systems of judgement, moral, aesthetic, and legal. In using emotion words we are able, therefore, to relate behaviour to the complex background in which it is enacted, and so to make human actions intelligible. (Bedford 1986:30)

Some ways in which the social constructionist approach differs from the physicalist approach throws further light upon its distinctiveness and, more important, upon its reorientation both of how we look at emotions and of how we can innovatively research them. First, emotions are not passions; they are not things that happen to us insofar as we are actively involved in making the appraisals essential to them. When Jane feels either pity or anger at John’s spilling the coffee on her, her feeling either of pity or of anger is an active appraisal of the situation. The physiological sensation that may come with the spilling of the coffee is not the emotion because that sensation could accompany either pity or anger. Pity or anger is Jane’s appraisal of John’s action; it is her self-conscious evaluative reaction to him. To claim that someone does something out of anger, love, grief, or the like is to maintain the myth of the passions, without satisfactorily accounting for human action (see Sarbin 1986:94).

Second, from a constructionist point of view emotions are rational, not irrational, uncontrollable eruptions from within the natural self. Emotions are essentially cognitively based appraisals of situations, and this allows that they can be subjected to rational persuasion and criticism (Armon- Jones 1986:44). If I am hurt at not being invited to Mary’s wedding, I can reasonably be expected to change my feelings (that is my appraisal) when someone points out that only close relatives were invited because Mary’s father had recently died and his long illness had exhausted the family’s funds and put the family into debt. When Brooks’s (this volume) American Hare Krishnas in India change their appraisal of the devotion (bhakti) experienced by non-Hare Krishna Indian holy men, they themselves begin to experience devotion to Krishna in a way different from that of their sect. In short, emotions are often socially negotiable experiences (Lutz 1988).

Reason’s distinction from, and elevation over, emotion is part of both Western intellectual and commonsense traditions. The wisdom of reason against the treachery and temptation of the passions has been the central theme of Western philosophy (Solomon 1976:10; see also Lutz 1986b). It is important to note that the Western hierarchical distinction of reason over emotion implies the further hierarchical distinctions of human over animal and culture over nature. Yet the separation of reason from emotion is not so easily made in other cultures as Lynch, Marglin, Toomey, and Bennett (this volume) note for India, Lutz (1988) notes for the Ifaluk of Micronesia, and Parkin (1985) notes for the Giriama of Africa. Indeed, deconstructionists would argue that reason requires emotion as a supplement, allowing the discourse about reason in the first place. Moreover, finding emotions in animals, as did Darwin, and in the behavior of decorticated cats, as do some psychologists, is reasoning by anthropomorphic metaphor rather than by any actual identity; emotion words, as signs, are assumed to have positivistic referential meaning and are decorticated from the free play of differences in which they are embedded.

Third, because emotions are social constructions, they are as variable as any other cultural phenomena. Attempts to understand them through empathy are no more than projections of one’s own ideological assumptions about emotional reactions onto the Other. Understanding emotions as social constructs, however, offers a much more refined, subtle, sociologically informative, and individually significant picture of emotions in the lives of others. From this point of view, all emotions are of equal value; there is no need for separation into primary and secondary emotions. Moreover, given differences in the social construction of emotion, the way in which those differences shape selves and the way in which they connect and involve selves in social systems are diverse, as are experiences of emotional life itself. Thus, whereas the affect ‘shame’ may everywhere concern investments of the individual in a particular image of the self, the way that this emotion works depends on socially dictated ways of reckoning the claims of selves and the demands of situations (M. Rosaldo 1984:149).

Finally, social constructionism raises the problem of feeling in a way different from physicalist theories and common sense. Indeed, feeling has been a major stumbling block in the way of Westerners’ understanding the role of emotion in their own lives and the Other’s life. Westerners’ question—

"What are they really feeling?"—is based on the assumption that ultimate psychological reality is internal; what may be ignored in the process of focusing on that question are indigenous epistemological notions about what can be known, what is worth knowing, and where a problem really lies. (Lutz 1985:73)

In the English language the verb to feel is so intimately linked to understanding of emotion that one can scarcely imagine the Other without it. Yet in Hindi and the Dravidian languages there is no such specific equivalent verb, and Indians get along quite well without it.⁹

Emotions are not and cannot be accurately identified by specific feelings. Schacter and Singer (1962) in their review of the evidence for physiological indicators of various emotional states opined that the evidence is inconclusive. The kinds of feelings or sensations one has, say, for anger, may be the same as those for fear or rage. There is nothing specifically in feeling itself that distinguishes fear from either anger or rage; the seeming difference comes from using different emotion words to appraise the situation. In the English language much difficulty comes from eliding the feeling of emotions with feeling of sensations as though they were the same. Yet most everyone will agree that feeling the hurt of an insult is not the same as feeling the hurt of a cigarette burn. Emotions as feelings can be said to be unreasonable, unjustified, or inappropriate in a way that sensations as feelings cannot. Therefore, the use of the word feeling is homonymous; emotions are not essentially sensations.

In their own experiments Schacter and Singer injected two groups of subjects with adrenaline, a substance known to create physiological sensations similar to those occurring in certain emotional states. One group was told of the physiological reactions they would have; the other was not. When the latter group began to feel the effects of the drug, they sought some explanation for their condition and found it in another person present in the room. This person was really an actor behaving in an either elated or angry way. The subjects, then, interpreted their physiological state according to their social context. Interestingly enough those in the group warned about what reactions to expect were unable to interpret their feelings emotionally; knowing that the cause was the drug, they could not use the label, emotional, for their feelings. In other words, the subjects had to have some belief or cognition about a socially defined event in order for them to interpret it as emotional (Schacter and Singer 1962; see also Lyons 1980:115-129; Solomon 1976: 150-170; Armon-Jones 1986; Averill 1980:327-329). Although the methods of Schacter and Singer’s study have been challenged incisively…, its conclusions are consistent with a large number of subsequent studies (Gordon 1981:573).¹⁰ Again, the important point is that feelings do not tell us what our emotion is; rather our appraisal of the situation, our emotion, tells us what we may feel. There is no single unique feeling, essence, or thing that goes with and identifies each and every emotion.

Part of the commonsense view of emotions explains behavior by reference to emotional feelings. For example, James hit Anne because he felt angry with her, or again Mary refused to come to the party because she felt insulted. Yet the reference to feeling in such statements does little or nothing to explain behavior. Rather one expects James to have a reason for his anger, just as one assumes that someone must have insulted Mary. In other words, the actions are explained by some unstated appraisals of situations in which they each found themselves.¹¹ If emotion words merely named some inner experience that preceded or accompanied behaviour, to explain behaviour by using them would not give the insight that it does (Bedford 1986:29).

Medick and Sabean (1984) rightly criticize historical and anthropological studies of the family that treat emotions and interests as opposites; thus, traditional families marry for property, and modern couples marry for romantic love. Such studies assume a progressive sentimentalization of the family. Yet in romantic love and emotional relationships material interest is hidden in the concern to communicate at the level of the individual and the emotional (Medick and Sabean 1984:11). Sentimentalization, as explanatory, is a dead end barring further investigation into the cultural appraisals and evaluations made under the guise of marriage for romantic love.

Some ways in which the verb to feel is used are of particular interest. One may say either I am angry with you or I feel angry with you. The former statement is emotionally stronger because it implies some censure of the other person; the latter, even though it contains the verb to feel, is weaker because no such negative evaluation is implied. The so-called feeling of anger is not as important as the implied moral evaluation of the relationship (Armon-Jones 1986:51-54). Statements about emotions, then, communicate information not just about persons but also about the social context in which they are used.

Just as feelings cannot identify particular emotions, so too behavior cannot be used as an adequate criterion to identify an emotion. Two people may look at the same behavior and interpret it in different emotional terms. For example, John walks into the classroom, sees Mary sitting in the first row, and immediately blushes. Jim interprets the blush as John’s being in love with Mary. Nancy interprets it as embarrassment because he is about to jilt Mary for Nancy herself. John himself says that he feels ashamed, not embarrassed, because he knows he will be responsible for Mary’s imminent rejected feelings. Correct identification of the emotion requires culturally relevant information about its social context and the appraisal made in it.¹²

Using commonsense theory we talk as though emotions were private, unique inner experiences, known truly only through introspection. Yet Wittgenstein (1958:243-264) has demonstrated how this is impossible. One would have to give an inner experience its own unique and private emotion word in which case it would have no meaning because meaning is essentially part of a public and social language. One cannot know what the private emotion word means because to know the meaning of a word is to know how to socially use it. But there is no criterion of correct use of a private word. Inner feelings make sense only because people already have words for them known to us by use in public language. Use of emotion words pivots essentially on the social evaluative aspect rather than on their identification of some inner essence.¹³ Indeed, it is most likely that one comes to know what it is to feel angry after learning from others what it is to be angry (Bedford 1986:16-19).

At least two problems arise from the discussion of emotions presented thus far. The first is whether a social constructionist perspective is committed to a purely cognitive interpretation of emotion, and, thus, to the mind side of the Cartesian mind/body dichotomy. The second problem concerns commonsense theories of emotion that, as I have tried to show, may be in error. What sense can an anthropologist committed to the Other’s point of view make of such theories?

Concerning the first problem, social constructionist theoreticians disagree about whether bodily feeling of some sort is an essential part of the meaning of emotion. Bedford (1962, 1986) and Armon-Jones (1986) argue that bodily feeling is only contingently related to the use of emotion terms and therefore is not essential to them. Solomon, too, despite his phenomenologically oriented study of emotion, considers feeling or experience nonessential to emotion when he says that feeling is the ornamentation of emotion not its essence and "emotions are self involved and relatively intense evaluative judgements" (Solomon 1976:158, 187; 1984:249). Just how evaluative judgments, matters of cognition, can be relatively intense is not clear.

On the other hand Perkins (1966) argues that emotions essentially involve nonspecific bodily states of feeling (cf. Armon-Jones 1986). Lyons confines the paradigmatic case of emotion to occurrent, not dispositional, states that

include the person’s beliefs about his or her persent situation, which may or may not be caused by a perception of some object or event, but which are the basis for an evaluation of the situation in relation to himself or herself. This evaluation in turn causes the wants or desires which lead to behaviour, while the evaluations and wants together cause abnormal physiological changes and their subjective registering, feelings. (Lyons 1980:57)

By abnormal physiological states Lyons means nothing more than a stretching or dampening down of our more usual physiological processes and states (1980:60). In this theory the evaluation or appraisal causes the physiological state that is also essential to an occurrent emotion. Michelle Rosal- do (1984.* 138) in a more intuitive way argues that it will make sense to see emotions not as things opposed to thought but as cognitions implicating the immediate, carnal ‘me’—as thoughts embodied.

What is at issue here, and why is it important? Almost all these authors agree that emotions in some way implicate and involve the self in some nonordinary way. Pure cognitivism, in its attempts to distinguish feeling from sensation and to show that feelings, as sensations, do not identify emotions, goes too far in making the self a bodyless, unfeeling, purely logical mind. Yet the self is not merely a mind; it is a totality of mind and body. Emotions affirm what they assert. They assert an appraisal, and they affirm this by grounding it in the reality of the bodily self. In this way they are simultaneously body-mind as well as individual-social, thus giving them their great importance to the social scientist. They connect and bind us to other people in a most social way; at the same time they seem the most individual and personal reactions. Social constructionists, as far back as Robertson- Smith, have found some of the power of ritual and religion in its grounding cultural ideas in nature and its clothing them, as Clifford Geertz (1966:24) says, with… an aura of factuality. Emotions, as moral appraisals, are grounded in the nature of our bodily selves, securing for them their bedrock commonsense character. In Hinduism, as the essays in this volume point out, an added factuality is given to emotions by grounding them not merely in the self but also in nature in the form of food, music, and scent.

From a deconstructionist point of view, cognition’s supplement is bodyemotion upon which it depends so that cognition itself can be present. Like physicalism, cognitivism—pure, self-sufficient, and referring to an identifiable concept—is more an attempt to establish a bounded, authoritative, and controlling discourse than an attempt to understand human emotions as caught in a historical play of difference.

By way of an anthropological aside, it is interesting to note that many, if not most, cultures locate emotions somewhere in the body. Among the Ifaluk of Micronesia emotion words are identified as being about our insides even though emotion words are identified and sorted as statements about situations of relationship rather than as internal states (Lutz 1986a 1268). Among the Pintupi aborigines of Australia emotions function primarily as moral displays of, or appeals to, one’s relatedness to others, but they also take place in the stomach where the spirit is located (Myers 1986:107). Among the Giriama of East Africa the heart, liver, kidney and eye are the seat of conjoined reason and emotions in general just as in Shakespearian England the heart ‘thinks’ as well as ‘feels’ and the liver is the seat of the passions (Parkin 1985:145). The widespread tendency to locate emotions in various parts of the body creates, it seems to me, metonyms, not metaphors, for the total mental-bodily self that is moved by moral involvement in the world, not in its private world. Such a movement of moral involvement is identified as an emotion. Mr. Spock was not human, because, although he could make absolutely rational appraisals, his self was unmoved by them; he was a computer masquerading in a human body.

Total self-involvement, I think, is part of the meaning of Wittgenstein’s remarks concerning emotion:

the gasp of joy, laughter, jubilation, the thoughts of happiness—is not the experience of all this: joy? Do I know that he is joyful because he tells me he feels his laughter, feels and hears his jubilation—or because he laughs and is jubilant? Do I say ‘I am happy* because I feel all that? … The words ‘I am happy* are a bit of the behaviour of joy. (Wittgenstein 1980:151)

The second problem raised by a social constructionist approach concerns how an anthropologist, committed to presenting the Other’s socially constructed reality, is to deal with the erroneousness of folk beliefs concerning emotion. Are commonsense notions about emotion wrong, or are they right insofar as those beliefs constitute the reality in which people live? One solution to this problem has been to interpret emotions functionally in the sense of what they do, how they are used, and what they are saying in social situations and events.¹⁴ Beliefs about what emotions are may be false from a scientific and a logical point of view, but that is not what emotions are about; rather, they are cultural appraisals of the social situations and events that they constitute. As I have already noted, we sometimes use emotion terms with an exculpatory function: when we say he did it because he loved her, his love operates to excuse his foolish action. Among the Pintupi aborigines of Australia compassion (ngaltu) presupposes an idea of relatedness and can be best understood as the possibility of being moved by another’s wishes or condition (Myers 1986:115). It functions, therefore, in situations where food and other things ought to be shared.

In this volume Bennett, Marglin, Toomey, Trawick, and Lynch show in various ways that many Indians believe that food literally is a form of emotion, particularly love. Bennett shows how this belief functions in such a way that exchanges of food unite members of the Pushti Marg sect in love and are also the most fundamental form of their relationship to the deity. Food offerings, offerings of love, also symbolically