Current Anthropology Volume 40, Number 4, August–October 1999  1999 by The Wenner-Gren Foundation for Anthropological Research. All rights reserved 0011-3204/99/4004-0001$2.50

Debating Self, Identity, and Culture in Anthropology1

by Martin Sökefeld

This paper explores relations between ‘‘identity’’ and ‘‘self’’—con- cepts that tend to be approached separately in anthropological discourse. In the conceptualization of the self, the ‘‘Western’’ self, characterized as autonomous and egocentric, is generally taken as a point of departure. Non-Western (concepts of) selves— the selves of the people anthropology traditionally studies—are defined by the negation of these qualities. Similar to anthropolog- ical conceptualizations of identity, this understanding of non- Western selves points exclusively to elements shared with others and not to individual features. Consequently, anthropological dis- course diverts attention from actual individuals and selves. A dif- ferent approach is exemplified by a case from northern Pakistan in a social setting characterized by a plurality of contradictory identities. It is argued that an analysis of how a particular indi- vidual acts in situations involving contradictory identities re- quires a concept of a self as it emerges from the actions of indi- viduals that is capable managing the respectively shared identities. Besides any culture-specific attributes, this self is en- dowed with reflexivity and agency. This concept of self is a nec- essary supplement to the concept of culture in anthropology and should be regarded as a human universal.

martin sökefeld teaches at the Institute of Social Anthropol- ogy of the University of Hamburg (his address: Kamerstücken 28, 22589 Hamburg, Germany [martin.soekefeld@joice.net]). Born in 1964, he received his M.A. from the University of Cologne in 1990 and his Ph.D. from the University of Tübingen in 1997. He has done fieldwork in Gilgit, Pakistan (1991–92, 1992–93), and has published Ein Labyrinth von Identitäten in Nordpakistan: Zwischen Landbesitz, Religion und Kaschmir-Konflikt (Köln: Köppe, 1997), ‘‘ ‘The People Who Really Belong to Gilgit’: Theo- retical and Ethnographic Perspectives on Identity and Conflict,’’ in Transformation of Social and Economic Relationships in Northern Pakistan, edited by Irmtraud Stellrecht and Hans-Georg Bohle (Köln: Köppe, 1998), and ‘‘On the Concept ‘Ethnic Group,’ ’’ in Karakorum-Hindukush-Himalaya: Dynamics of Change, edited by Irmtraud Stellrecht (Köln: Köppe, in press). The present paper was submitted 12 vii 98 and accepted 14 x 98; the final version reached the Editor’s office 23 x 98.

1. I would like to thank Katrin Gratz, Beatrix Hauser, and Georg Stöber as well as the anonymous referees for critically reading an earlier version of the paper. The paper was much improved by their suggestions. For many discussions of the topic I am grateful to Beate Reinhold. The text is an outcome of reflection about field research in Northern Pakistan which was generously funded by the Deutsche Forschungsgemeinschaft.

417

The concept ‘‘identity’’ has undergone a paradigmatic shift in recent decades. Originally, its meaning was ‘‘sameness,’’ and in psychology this sameness meant ‘‘selfsameness.’’ Identity was understood as a disposi- tion of basic personality features acquired mostly dur- ing childhood and, once integrated, more or less fixed. This identity made a human being a person and an act- ing individual. Inconsistency of personality—that is, nonsameness of the self, nonidentity—was regarded as disturbance or even psychic illness. In social anthropol- ogy, the concept ‘‘identity’’ was used mostly in the con- text of ‘‘ethnic identity.’’ Here it pointed not simply to selfsameness but to the sameness of the self with oth- ers, that is, to a consciousness of sharing certain charac- teristics (a language, a culture, etc.) within a group. This consciousness made up a group’s identity. These under- standings were complementary rather than contradic- tory and fitted well together, as the group to which a person belonged constituted an important part of the so- cial environment in which and through which personal identity was formed. Erik H. Erikson combined the two: ‘‘The term ‘identity’ expresses such a mutual relation in that it connotes both a persistent sameness within oneself (selfsameness) and a persistent sharing of some kind of essential characteristics with others’’ (Erikson 1980:109). Common to the usage of the term ‘‘identity’’ in the two disciplines was that identity was basically one.

In the European history of ideas, the notion of the sameness of the self had had foundational qualities since Descartes’s Meditationes. The cognizing self (ego, I), certain of its existence through its own acts of cognition, became the warranty against an ambiguous and deceptive world of things. This warranty was valid only on the condition that ego remained the same, that is, identical. The self thus became subject in the dual sense of being subjected to the conditions of the world and, simultaneously, being the agent of knowing and doing in that world. The belief in this subject be- came the a priori for the possibility of knowing the world.

The metanarrative of the identical subject was finally destroyed by poststructuralist deconstruction. Michel Foucault analyzed the subject not as the source and foundation of knowledge but as itself a product or effect of networks of power and discourse (1979:35; 1980:98). In contemporary texts on identity, the concept seems not to exist in the singular. Whereas it was once defined by sameness and unity, both qualities have given way to difference and plurality. Psychology has turned its at- tention to multiple identities (Gergen 1994, Melucci 1997, Rosenberg 1997)—which, according to the word’s conventional meaning, is a contradiction in terms. The contemporary self is depicted as fragmented (Jameson 1984), essentially fluid and many-sided, as in Lifton’s ‘‘protean self’’ (1993), or populated by multiplicities, as in Gergen’s ‘‘saturated self’’ (1991). In the social and cultural sciences, what was once called ‘‘identity’’ in the sense of social, shared sameness is today often dis-

418 c u r r e n t a n t h r o p o l o g y Volume 40, Number 4, August–October 1999

cussed with reference to difference.2 Difference points to the contrastive aspect of identities and thereby em- phasizes the implicit condition of plurality. There can be identity only if there is more than one identity, and in this sense difference constitutes and precedes iden- tity. I will argue that the emphasis on difference calls into question anthropologists’ conventional assump- tion of shared identity and demands attention to a per- sonal or individual identity that is here called the self. Discourses about the self and about identity in anthro- pology are almost entirely distinct; I am calling for the establishment of a closer connection between them. This connection is in my view inevitable because in- creased attention to the social and cultural contexts of plural identities leads to a reworking of both concepts.

The significance of the self is greatly underestimated in anthropology. Two approaches to the subject can be distinguished. The first implicitly maintains that an- thropology’s subjects have an identity (shared with oth- ers, derived from a culture) instead of a self. The second analyzes the selves of these subjects by contrasting them with a paradigmatic conceptualization of the ‘‘Western self’’ (that is, the still mostly Western anthro- pologist’s own self), and because it denies the para- digmatic characteristics of the Western self to anthro- pology’s non-Western subjects it actually denies them a self. Anthropology’s treatment of the self of its sub- jects is an effective strategy of othering—positing a ba- sic difference between anthropologists and those they study.

To counter both approaches I will proceed as follows: After introducing my thesis of anthropology’s denial of others’ selves, I will present a case of plural and con- flicting identities in a town in the mountains of north- ern Pakistan. From this case I will draw the conclu- sion that it is impossible to conceive of the actions of individuals embracing a plurality of identities without referring to a self. After exploring meanings of and rela- tions between self and identity as they can be inter- preted from my material, I will discuss the relations be- tween anthropological understandings of culture and self.

Anthropology’s Denial of Others’ Selves

Anthropologists write easily about the identity of those they study in the sense of something shared with oth- ers, but they have much more difficulty in attributing a self to the people they are writing about. The attribu- tion of a nonself to these others is expressed in debate about intercultural variability of (concepts of) the self, in which the Western self is opposed to the self of the non-West (see Spiro 1993). In this debate the Western self is represented as an instance of the individual’s pro- viding it with boundedness, relative autonomy and in- dependence, reflexivity, and the ability to pursue its

2. On a related change in discourse from identity to difference in feminist studies, see Crosby (1992).

own goals. Clifford Geertz has characterized the West- ern concept of the self as ‘‘peculiar’’ in that it is ‘‘a bounded, unique, more or less integrated motivational and cognitive universe, a dynamic center of awareness, emotion, judgement and action organized into a distinc- tive whole and set contrastively against other such wholes and against its social and natural background’’ (Geertz 1984:126). This self fits neatly with the subject of Descartes’s Cogito.

Anthropological characterizations of ‘‘the other’’ are often inversions of European self-images (Fardon 1990: 6). This certainly applies to understandings of others’ selves. In the conceptualization of non-Western selves, the Western self3 was taken as the starting point and the non-Western self was accordingly characterized as its opposite: unbounded, not integrated, dependent, unable to set itself reflexively apart from others, unable to dis- tinguish between the individual and a role or status that individual occupies, unable to pursue its own goals in- dependently of the goals of a group or community. Ef- fectively, this characterization involved the negation of all the definitional qualities of the self, that is, of those that point to the differentiation of the self from others. We can conclude, then, that by being denied a Western self, anthropology’s others were denied a self at all. Many or even most anthropologists would hesitate to subscribe to this rather polemical proposition, but it simply carries the logic of this anthropological approach to its logical conclusion. A well-known instance of this perspective is Geertz’s (1973) description of the self in Bali. According to his analysis, the Balinese person is extremely concerned not to present anything individual (distinguishing him or her from others) in social life but to enact exclusively a culturally prescribed role or mask. Similarly, in the anthropology of South Asia there is an ongoing debate about whether the people of the subcontinent are endowed with anything compara- ble to a self (in the paradigmatic Western mode), that is, whether they can be spoken of as individuals.4

3. The concept of the Western self requires a discussion in itself which cannot be undertaken here for want of space. The under- standing of the Western self as autonomous and integrated is mostly derived from (selected [see Murray 1993]) Western written philosophical traditions and not from analyses of experiences of people in the West. Some recent studies suggest that there are dif- ferent kinds of Western selves and that the differences between Western and non-Western selves may be much less than is usually assumed (e.g., Holland and Kipnis 1994, Ouroussoff 1993, Stephen- son 1989). It seems, then, that the concept of the Western self that pervades anthropological analyses of non-Western selves is the re- sult of a particular kind of Occidentalism (Carrier 1992, 1996a). 4. According to Dumont (1965, 1970, 1980), from the South Asian cultural perspective individuals are always elements of larger so- cial units in which all value is vested. Marriott takes the con- trasting stance that there are no individuals but only ‘‘dividuals’’ in India, because human beings are only temporary and constantly transformed compositions of elementary coded substance (1976, 1989; Marriott and Inden 1977). Both positions have been heavily criticized from different perspectives (see Mines 1988, 1994; Morris 1978). Mines (1994), exploring both public and private aspects of individuality in Tamil Nadu, shows convincingly that individual- ity and personality are indispensable to social life in South India, although they are conceived differently than in Western societies. The same conclusion is reached by McHugh (1989) in her analysis

s ö k e f e l d Self, Identity, and Culture 419

In spite of the psychological definition of identity as selfsameness, this denial of a self does not amount to a denial that, for instance, Indians have an identity in the anthropological sense. Dumont and Marriott, who ar- gue in their own ways against the individuality of peo- ple in India, would hardly deny that Indians have identi- ties. In this understanding, then, Indians have an identity instead of a self. Contrary to the meaning of ‘‘identity’’ in psychological usage, identity here refers not to an idiosyncrasy of the individual—a bundle of features that distinguish her or him from others—but to identity with others, with one’s cultural/social group. This idea pervades Lévy-Bruhl’s conceptualiza- tion of the ‘‘primitive soul.’’ 5 We have to keep in mind that the people who are ascribed such an identity are always others and the objects of anthropological dis- course. Anthropologists generally do not apply this un- derstanding of identity to themselves and members of their own society. People who have only such an iden- tity are not autonomous, independent, and pursuing their own goals; they are dependent on their cultural/ social group and behave according to the prescriptions and interests of that group.6 Cultural and social deter- minism lurks behind this conceptualization of non- Western selves.

In anthropological discourse, the question of identity is almost completely detached from the problem of the self. In the vast body of literature about ethnic identity the self is rarely mentioned, and in writings about the self a relation between the self and identities is some- times noted but remains unexplored (e.g., Morris 1994: 1). However, if we look at analyses of non-Western con- cepts of the self, it cannot go unnoticed that these con- cepts of the self (as represented by anthropologists) are modeled precisely on the anthropological understand- ing of identity: they are sociocentric (Mageo 1995, Read 1967, Shweder and Bourne 1984), just as identities are social and shared.

In recent years, committed arguments have been made for the self as a basic human condition. The out- standing example here is Anthony Cohen (1994), and it is significant that his book on self-consciousness is sub- titled ‘‘an alternative anthropology of identity’’ (my italics). Cohen argues that the disregard for others’ selves in anthropological accounts results in distorting fictions, denying self-consciousness to the people an- thropologists study. Most frequently this disregard is expressed in a premature subsumption of individuals under the groups to which they belong. He maintains that to attend to the self in anthropological inquiry is

of individuality among the Gurung of Nepal. However, there are Indian writers who argue that there are no bounded individuals in India (e.g., Vaidyanathan 1989). 5. According to Lévy-Bruhl, the members of ‘‘primitive’’ groups possess not an ‘‘individual individuality’’ but rather a ‘‘shared indi- viduality’’ and a ‘‘quasi-identity’’ to the extent that they are almost interchangeable (Lévy-Bruhl 1963[1927]:99). 6. This understanding of non-Western ‘‘selves’’ is carried to ex- tremes by much fiction writing, where the non-European is simply an ‘‘atom in a vast collectivity’’ (Said 1995[1978]:252).

not to indulge in individualism but to problematize what is otherwise taken for granted: the relationship between the individual and the social. How does the self relate to groups, to society, or, in the present con- text, to shared identities? These questions have to be asked, Cohen argues, not ‘‘to fetishize the self but, rather, to illuminate society’’ (p. 22). Cohen was fol- lowed by Rapport (1997), whose call for attention to self and individuality is even more radical. For him such an approach is an ontological necessity, because ‘‘it is in individuality that the roots of the social and cultural lie’’ (p. 2). Cohen’s and Rapport’s arguments still seem to be in the minority. I intend to support this approach by exploring the relationship between the self and plu- ral identities.

What if the people anthropologists study cannot be categorized by shared identities? What if their identity is not one but, as in a plural society, a collection of dif- ferences setting them off from varying groups of others? if each individual is characterized by a specific pattern of differences from and commonalities with certain oth- ers? if what is shared with some conflicts with what is shared with others? The implicit parallelism of self and identity in anthropology as something essentially shared does not fit into the conceptualization of plural societies. My conclusion from the following case will be that there can be no identities without selves. This case from northern Pakistan is not about a cultural con- cept of the self. It is about whether we can discern something besides the changing identities of the ‘‘em- pirical agent’’ (Dumont’s term)—that is, whether we have to conceive of an acting self in addition to its iden- tities.

My field research in the multi-‘‘ethnic’’ town of Gil- git in the Northern Areas of Pakistan, was on ‘‘eth- nicity.’’ Educated anthropologically in the tradition of Durkheimian and social anthropological approaches, I was determined to focus on groups rather than on acting individuals or selves. The episode related here forced me to acknowledge the importance of acting individu- als in a study of ‘‘ethnic’’ identity. It is an example of the struggle to act and to present oneself as a consistent self in a situation of plural and contradicting identities related to intense social conflict.

Conflicting Identities and a Self in Gilgit

Gilgit, a town of about 50,000 inhabitants, is the admin- istrative and commercial center of the Northern Areas of Pakistan. The population of the area is highly differ- entiated by a number of dimensions of identity. Migra- tion to the urban area has made the composition of the population very complex. Some important dimensions of difference (i.e., identity) are religious affiliation, lan- guage, regional belonging, qōm,7 clan, and kinship

7. The meaning of qōm is quite broad and ranges from kinship- based groups to the political nation (Sökefeld n.d.a). In the present context it can be translated as ‘‘quasi-kinship group.’’

420 c u r r e n t a n t h r o p o l o g y Volume 40, Number 4, August–October 1999

(Sökefeld 1997a). In the present context two of these, religion and kinship, will play the major part.

For two and a half decades the town’s population has been split by a militant conflict between Shiites and Sunnis, with periodically erupting tensions between members of the two groups. From 1975 onward, weap- ons were used in these encounters, which regularly re- sulted in losses of life. When I was staying in Gilgit be- tween 1991 and 1993, the population was effectively polarized by the experience of almost 20 years of vio- lence.8 Intermarriage between Shiites and Sunnis had ceased when the conflict began. Kinship relations across the religious boundary, which before had been very common, were being neglected if not altogether de- nied. A considerable relocation of settlement in Gilgit and the surrounding villages had resulted in a number of ‘‘pure’’ Shiite or Sunni neighborhoods where mem- bers of the two sects had earlier lived together. Some parts of the bazaar could now be identified as being ei- ther Shiite or Sunni. Commensality between Shiites and Sunnis had ceased almost totally. This applied espe- cially to meals including meat, because the sharı̄’a, the Islamic law, prescribed that only meat of animals slaughtered by Muslims was permitted. The refusal to eat meat provided by the other sect amounted implic- itly to the judgment that the others were not really Muslims but infidels (kuffār). Separation always reached its highest pitch during and just after a period of tension.

The polarization of Shiites and Sunnis found its strongest expression in the mutual accusation that the others were kuffār (singular kāfir). Kuffār are nonbeliev- ers. They are impure. Muslims can have no real com- panionship with them. It is impossible to share a meal with them, let alone family ties. By no means all Shiites called Sunnis kuffār and vice versa; to accuse someone of being a kāfir was so strong and dishonoring a re- proach that most people hesitated to voice it publicly, and many people were certainly not of the opinion that the others were indeed kuffār. Still, the reproach was sometimes made in public, and graffiti expressing it could sometimes be seen on the walls of the town. These incidents always caused intense disturbance and resentment among those being called kuffār.

Although the difference between Shiites and Sunnis had become very sharp, it was not devoid of ambiguity arising from the fact that each person embraced identi- ties derived not only from religious affiliation but also

8. Kuper’s concept of polarization fits the scene in Gilgit well: ‘‘Po- larisation is conceived here as involving mutually hostile action. . . I reserve the term for an intensification of conflict by aggressive action and reaction. Polarisation then is a process of increasing ag- gregation of the members of the society into exclusive and mutu- ally hostile groups, entailing the elimination of the middle ground and of mediating relationships. Episodes of conflict accumulate. There are corresponding ideologies . . . , presenting simplified con- ceptions of the society as already polarised into two antagonistic groups with incompatible and irreconcilable interests, rendering inevitable the resort to violence’’ (Kuper 1977:128).

from others, for example, to clans, localities, or lan- guage communities. In spite of the religious polariza- tion which had pushed the religious identity to the fore, other identities continued to be relevant. The religious identity might conflict with other identities of the same person, and this meant conflict between opposed values and objectives. At times, the value of conforming and being loyal to one’s sect proved irreconcilable with the value of keeping one’s family together and caring for one’s relatives because a relative was regarded as kāfir. The self, then, embraced conflicting identities.

Still, people tried to reconcile these values. They pur- sued various and at times contradictory objectives, sometimes simultaneously. How did they develop a sense of a continuous and consistent self in this con- text? Or did they do without one? I want to explore these questions in an episode of fieldwork which I expe- rienced with Ali Hassan,9 the man who taught me much of what I learned about culture and society in Gilgit.

Ali Hassan

Ali Hassan was a Shiite, a member of the Yeśkun qōm and within it of the Catōrē clan. He was a respected el- der of his qōm and of the town as a whole. He possessed considerable agricultural land both in the part of Gilgit where he lived and in a more distant place on the edge of town. This peripheral area was cultivated by a tenant, since Ali Hassan had suffered a heart attack in 1992. The tenant belonged to the Gujur10 qōm and was a Sunni. Ali Hassan also owned three houses and nine shops which he rented out, and more shops were under construction.

Ali Hassan had been his father’s first son after eight marriages, and of course his father had feared losing him. Therefore he had been given as a ‘‘milk son’’ to a Sunni Gujur family from a neighboring valley, where he had stayed for three years. At the age of 15 he had mar- ried a girl of another Yeśkun clan. When I met him, he was in his sixties and still versatile and innovative. Be- ing very generous, he helped his relatives and others out with firewood, produce from his lands, and assistance in dealing with the authorities. His conception of ‘‘rela- tives’’ was very inclusive. If it suited his purposes, he could conceive of a relation with almost anyone. He spent a considerable part of his life attending to rela- tionships.

Being a very pious Shiite, Ali Hassan visited the main mosque every day to offer the midday prayer. He was completely convinced that his sect was right in the reli- gious conflict. He always reiterated that it was not Shi- ites who had begun the conflict and that they were very much against sectarian violence. In spite of his loyalty

9. Personal names are pseudonyms. 10. The Gujur mainly practice animal husbandry and menial labor, and their prestige is quite low.

s ö k e f e l d Self, Identity, and Culture 421

to the Shiite cause, he maintained a quite independent stance. Conscious of many contradictions between ri- wāj (local custom) and sharı̄’a, Ali Hassan preferred many norms of the former, for example, those regulat- ing landed property and inheritance. Only if forced by an explicit judgment would he have complied with the norms of sharı̄’a in these matters.

Although in his opinion the conflict between Shiites and Sunnis was the Sunnis’ fault, he did not condemn them indiscriminately. The guilty parties, those who slandered Shiites as kuffār, were the ‘‘new’’ Sunnis— Tablı̄ghı̄, Wahabı̄, and Deobandı̄.11 In his estimation, the real, righteous, traditional Sunnis, whom he termed Brelwı̄ or Hanafı̄,12 disapproved of the conflict as much as the Shiites. Further, he distinguished Sunnis ac- cording to kinship or qōm membership. There were his ‘‘own,’’ xāndānı̄13 Sunnis—those with whom he shared qōm membership or kinship or who were at least proper people of Gilgit—and there were other, ‘‘foreign’’ ones—nonrelatives, people of other qōm, and especially immigrants such as Pashtūn and Kashmı̄rı̄. The separa- tion established by differential religious affiliation could in his view be at least partially erased by some other relation. All the relations established by religion, descent, kinship, etc., were possible bases of commu- nity. For example, Ali Hassan told me that he rented his shops and houses only to Shiites and Ismailis because both basically belonged to one religion and no dispute would arise with them. However, as we have seen, one of his tenants was a Sunni—Gujur and a relative of his ‘‘milk’’ family. On another occasion Ali Hassan told me that Pashtūn were always pressuring him to rent them a shop, but he resisted because they were foreign and not xāndānı̄, and he emphasized that he was prepared to rent a shop to xāndānı̄ Sunnis.

While we were talking about the shops, one of Ali Hassan’s affines who was present, another Yeśkun, told me that many Sunnis belonged to their qōm and that a number of his sisters and female cousins were married to Sunnis. Then he added, ‘‘But for us Shiites, Yeśkun religion is more important than qōm. This is very differ-

11. Tablı̄ghı̄ are the followers of the Tablı̄ghı̄ Jamāt, a lay mission- ary movement founded in the 1920s in northern India that aims to bring Muslims back to the right path, that is, to the prescribed observances such as daily prayer (Ahmad 1991). Deobandı̄ are the followers of the famous North Indian Islamic seminary at Deo- band, which propagates an orthodox version of Sunni Islam (Met- calf 1989). Most Sunni ulema in the present-day Northern Area be- long to the Deoband school. Wahabı̄ are the followers of Saudi Arabian Wahabiyya. 12. Brēlwı̄ are the followers of the school of Bareilli, the rival of the seminary of Deoband, which propagates a kind of folk Islam holding Sufism and local saints in high esteem (Metcalf 1989). Hanafı̄ are followers of the law school ( fiqh) of Abu Hanifa. Here Ali Hassan’s categorization is incorrect; all the types of Sunnis mentioned are Hanafı̄. Shiites, in contrast, are Jafrı̄, followers of the fiqh of Imam Ja’far as-Sādiq. 13. Xāndānı̄ is derived from xāndān, ‘‘family,’’ but it means not only belonging to a kinship group but also sharing basic traditions and value orientations. For a detailed analysis, see Sökefeld (1997a, 1998).

ent among S· ı̄n. For them, a S· ı̄n is in the first place S· ı̄n, not Shiite or Sunni. This is very good; it should be like that also among Yeśkun!’’ 14

At the same time, Ali Hassan and other respected Yeśkun from Gilgit and other places were endeavoring to organize a meeting of their qōm in order to overcome the religious cleavage within their group and to gain strength for the Yeśkun in confronting other qōm, espe- cially the S· ı̄n. Sunni Yeśkun were part of this effort. I was often told by Yeśkun that it was only their own qōm that was divided and thus weakened by religious conflict. Some Yeśkun even attributed the whole con- flict to a conspiracy of S· ı̄n against Yeśkun. For Ali Has- san, the benefit of an assembly of Yeśkun was self-evi- dent. It was a value in itself to promote the solidarity of one’s qōm. He had told me several times that he was Shiite in the first place and that religion was most im- portant for him, but in the situation described here this was obviously not the case. The aim of the whole en- deavor was precisely to subordinate religion to qōm.

Do then these contradictory basic identities—qōm, kinship, and religion—create separate ‘‘compartments’’ of the person? Do they prevent the development of a sense of a consistent and continuous self? Not quite, as we shall learn by accompanying Ali Hassan to the wed- ding of one of his relatives and listening to his reflec- tions on that visit.

An Uneasy Visit at a Wedding

One of Ali Hassan’s elder sisters had married into a re- spectable Sunni Yeśkun family at Napura, an old village on the western edge of Gilgit. Years ago this sister had died. Her son Malik Amman was of about the same age as Ali Hassan. Ali Hassan was quite concerned to be on good terms with these Sunni relatives, and I had accom- panied him on visits to them several times. In January 1993 Malik Amman’s granddaughter was going to marry a man from the Punjab. Ali Hassan did not ap- prove of this marriage, but he invited me to accompany him to the wedding. He told me, ‘‘I am very angry that she is to marry a Panjābı̄, but Malik Amman did not ask me. There are plenty of suitable boys in Gilgit!’’

The marriage was to take place on a Friday. When I arrived at Ali Hassan’s house, he told me that his wife’s cousin Sher Khan would come with us. He had formerly been the general secretary of the Shiite community as- sociation in Gilgit but, after receiving a personal threat in the religious conflict, had resigned from that office and left. He had returned to Gilgit only a few days be- fore. We went to his place and met him with his son. Because Sher Khan left his house only under the protec- tion of arms, both his son and Ali Hassan’s son, who accompanied us, carried Kalashnikovs.

14. S· ı̄n is the other important indigenous qōm of Gilgit. Rivalry between S· ı̄n and Yeśkun is at times quite strong (see Sökefeld 1994).

422 c u r r e n t a n t h r o p o l o g y Volume 40, Number 4, August–October 1999

It was still morning when we arrived at the home of Malik Amman. We were welcomed very cordially by a number of family members and led into the house. Malik Amman was sitting there with his younger brother Sadiq. After the customary exchange of compli- ments, he told us that the marriage ceremony would not take place until afternoon. Sher Khan remarked at once that he was unable to stay so long because he had to attend a meeting. Sadiq and Malik Amman urgently attempted to persuade him to stay at least until lunch, but Sher Khan showed no inclination to change his mind. While the exchange between them was going on, we were served sweet tea and dried fruit. After a very short time Sher Khan told us that he had to leave. Again Sadiq insistently begged him to stay without success. Before we got up, Sadiq said a prayer in which he in- voked the ahl-e bayt (the family of the Prophet, includ- ing his son-in-law Ali). This invocation is a Shiite prac- tice of which ‘‘orthodox’’ (i.e., Wahabı̄ or Tablı̄ghı̄) Sunnis are very critical. From their perspective, invok- ing humans in prayer places them close to God and vio- lates the fundamental dogma of monotheism. Immedi- ately after the prayer, we said good-bye and left the house. One of Malik Amman’s sons followed us to the jeep, continually pleading with us to stay. Finally, he exclaimed with disappointment in his voice, ‘‘Are we contractors of a religion or are we relatives?’’ But we got into the jeep, left, and returned to the center of town.

The next day I asked Ali Hassan what kind of Sunni Malik Amman was, and he responded, ‘‘He is Hanafı̄. But his brother Sadiq is Tablı̄ghı̄ and strict Wahabı̄!’’ Be- cause I had been discussing commensality with other informants, I then asked him, ‘‘Is it true that Shiites never eat meat provided by Sunnis?’’

Ali Hassan: Yes, they do not share meals in our houses, and therefore we do not eat in their houses.

I: Was this the reason that we left yesterday be- fore the meal was served?

Ali Hassan: No, we left because the marriage was to take place only in the afternoon and because Sher Khan had to attend a meeting.

I: If we had stayed for lunch, what would you have eaten?

Ali Hassan: I would have eaten only ghi. I would have eaten only soup, without meat.

I: Do you refuse to eat meat in the houses of all Sunnis or only in the houses of Wahabı̄?

Ali Hassan: Oh, I will never eat meat with Wa- habı̄. But I will eat with Hanafı̄.

I: But you told me that Malik Amman is Hanafı̄!

Ali Hassan: Yes, but this Tablı̄ghı̄ [Malik Am- man’s younger brother Sadiq] was present! And I am an elder of the Shia, and Shiites would have re- sented it very much if I had eaten there.

I: Yesterday, Sadiq invoked the ahl-e bayt in his prayer . . .

Ali Hassan: This he did only for us.

I: I thought that Hanafı̄ Sunnis also paid respect to the ahl-e bayt.

Ali Hassan: They only say that. Orally they do, but not with the heart! And Wahabı̄ do not do it at all.

I: Your son Rasul told me that he had stopped vis- iting their house.

Ali Hassan: He is angry with them because they gave their granddaughter to a Panjābı̄. I was angry, too, and I asked them, ‘‘Couldn’t you find some- body here?’’ So far away, that is bad for the family. It is bad.