Reconsidering Identity

Reconsidering IdentityAuthor(s): Martin SökefeldSource: Anthropos, Bd. 96, H. 2. (2001), pp. 527-544Published by: Anthropos InstituteStable URL: http://www.jstor.org/stable/40465555 .

Accessed: 19/11/2013 14:24

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Anthropos Institute is collaborating with JSTOR to digitize, preserve and extend access to Anthropos.

http://www.jstor.org

This content downloaded from 202.41.10.30 on Tue, 19 Nov 2013 14:24:17 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=anthroposinst

http://www.jstor.org/stable/40465555?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Ê Anthropos iTj 96.2001: 527-544

Reconsidering Identity

Martin Sökefeld

Abstract. - The paper discusses and defends the analytical usefulness of the concept of identity which has been pervasively criticized by authors like Richard Handler or Rogers Brubaker and Frederick Cooper. Starting with reviewing the problematic of concepts in social anthropology and continuing with discussing the rise of identity discourse, it is argued that concepts in social and cultural sciences are always suspended between their employment in scientific and in nonscientific discourse. This dual hermeneutics of concepts is, however, not a shortcoming which has to be overcome but a productive element that contrib- utes to their refinement. It is argued that in the case of identity dual hermeneutics leads to a reconceptualization of identity as qualified by the conditions of difference, multiplicity, and intersectionality. In the final parts of the paper, implications of this reconceptualization of identity for a concept of self are explored. [Theory, identity, self]

Martin Sökefeld, M. A. (Köln 1990), Dr.phil. (Tübingen 1996). He did field research on interethnic relations in Gilgit, Northern Pakistan between 1991 and 1993. From 1997 on he held teaching assignments at the Institute of Social and Cultural Anthropology, University of Hamburg. Since 1999 he is assistant professor at the same institute, where he is working on migrants from Turkey with a research project on Alevi identity in the German diaspora. - Publications: see References Cited.

1 Introduction

The intention of this paper is to explore the concept of identity generally, not limited to a specific ethnographic context, although I will draw on such contexts. The concept of identity has recently come under heavy attack. Two directions of critique seem to be of particular weight. The first of these denies that identity is a useful concept in cross-cultural context (Handler 1994; Rouse 1995), and the second argues that identity has been

charged with so many different meanings that it has ceased to be meaningful at all and should be replaced by a range of other, more specific concepts (Brubaker and Cooper 2000). Both arguments contribute importantly to the questioning of "identity" and they certainly have to be taken very serious. However, I am still of the opinion that "identity" does important work within the social and cultural sciences. My text will, therefore, be a defence of "identity" - if certainly a defence with a great deal of reservation derived from the understanding that "identity" today is indeed an overused concept that should be employed with much more caution and stricture. I argue that a concept of identity has to be seen in close connection with a concept of self and that both have political implications which have to be taken into account.

Before entering the debate, however, I will dwell a little on the general difficulty of concepts in social anthropology because in my opinion the quality of "identity" as a concept has an important impact on the contents and understanding of identity which has to be considered. I could also say that I am writing mainly about concepts in social anthropology, taking "identity" as example. What 1 am doing, then, is essentially the business of reflexivity.

2 Concepts in Social Anthropology

Social anthropology, like all social and cultural sciences, is to a great extent a trade in concepts. The proposition for which I am arguing here is that



528 Martin Sökefeld

concepts are nobody's property. Concepts are our tools to describe and analyse the world, or better, those sections of the world we are scrutinizing at a given time under particular circumstances. To put it in these terms, however, is a quite loose way of talking. We do not simply take concepts in order to describe or analyse a given section of the world, we employ concepts already to select, to cut out, to trim these sections. Consider the concepts "culture," "society," "religion," or even a concept that is seemingly much less problematic like "village." These concepts presuppose that the world can be divided into certain parts and aspects and they - apparently - show us how to do so. That is, concepts help us in making the objects we are scrutinizing, they are also constitutive of what we describe and analyse. If we were very strict logicians, we had to abandon the whole business because we are always guilty of a kind of petitio principii: The terms we are using to describe something already contain at least an outline of the structure of that description. However, logic is a discipline of pure thought the empirical contents of which tend toward zero whereas we deal with the empirical world which contains many logical shortcomings.

The difficulty of concepts is multiplied for social anthropologists because we are conscious of the fact that not all human beings employ the same set of concepts. Concepts are a matter of culture, and - this is the basic axiom of social anthropology - culture differs. Whereas for us it seems to be a simple task to say whether a given section of reality is to be termed religion or not (although it is considerably difficult to define "religion"), this may not be so for all people and all cultures. Others may use some label to group a phenomenon we would call "religion" together with other phenomena. For example, many Mus- lims make it an important point to emphasise that Islam - a matter we unhesitatingly call "religion" - is not only religion but includes elements we would categorize as economy, politics and, most importantly, law.1 Concepts originate from specific contexts in which they have more or less precise meanings. But anthropologists very often take concepts out of their frame of origin and apply them, for comparative purposes, to other realms. Concepts like "mana," "tabu," or "totem" are famous classical examples of this delocaliz- ing technique. Frequently, however, there is too little reflection about what is implicated in this

transplantation of concepts. To understand the effects of transplantation we have to recall the very important insight of structuralism that concepts do not simply refer to "things out there" but first of all to other concepts. The meaning of a concept is not simply its relation to some nonlinguistic entity (as supposed in referential theories of meaning) but its relation to other concepts. Concepts and meanings are inscribed into intricate and multi- stranded networks of other concepts and meanings. To take a concept out of its context of origin and to apply it elsewhere is to insert it into other networks of concepts and meanings, that is, to alter its meaning.

An important consequence of this fact is that concepts become increasingly difficult to define if they are employed in many different contexts. More and more semantic relations, examples, and differences have to be taken into account in order to arrive at a sufficient definition. Most frequently this means that concepts become increasingly un- specific. In the case of religion, this has resulted in the difficulty to find a definition which is able to accommodate the case of Buddhism. Religion is conventionally defined as referring to something which has to do with belief in superhuman or supernatural beings but Buddhism in its "pure" Theravada version does not recognize such beings. Still, most of us would call Buddhism a religion. A little trick to find a way out of this difficulty is to take this unspecificity into the concept, i.e., to define it not with reference to common elements and specific differences (as in the classical Aristotelian way of definition), but with reference to prototypes only (see Saler 1993 for a thorough discussion of the problem). This trick is increasingly applied in social anthropology. However, it implies nothing less than to give up the idea of precise definitions in the original sense of delimiting the applicability of a concept.

Beside this growing unspecificity, the application of concepts to other contexts has another important consequence: It results in their reification. When social anthropologists look for mana not only in the conceptual universe of Polynesia but elsewhere, they develop an understanding of mana "in itself." Before, however, there was no mana in itself but only mana for Polynesians. Now mana is not simply a concept of Polynesians but also one of anthropologists. Thus, the transplantation of concepts to other contexts resulted in many heavy volumes of books and papers attempting to define originally specific and more or less localized concepts in general ways foi comparative purposes. These difficulties apply noi

1 For a very brief discussion of this subject see Delaney 1991: 18.

Anthropos 96.2001



Reconsidering Identity 529

only to "foreign" concepts like taboo and totem but equally to "our" concepts, i.e., concepts originally taken from the context of "Western tradition" and transplanted elsewhere. "Religion" and, of course, "identity" are in this respect not different from totem and taboo.

In the 60s and 70s social and cultural anthropologists practiced a distinction which apparently saved them from the difficulties of this delocal- ization of concepts. They distinguished between ernie and etic models, concepts or understandings.2 Ernie concepts are those from within a culture, capturing its idiosyncratic meanings in particular cultural circumstances, whereas etic concepts are applied from outside, assigning meaning from a comparative and cross-cultural perspective, which, because it is not bound to a specific local context, is more generally applicable. Ernie concepts are cultural concepts, limited in their validity, whereas etic concepts are supposed to be culture-free or, simply, objective. Ernie concepts are local concepts, etic concepts are nonlocalized or metalocal concepts. The same distinction is frequently captured in the prefix "ethno-." Thus, ethnoscience is a science in inverted commas, a science representing a cultural preconception of the world but not its actual, "objective" conditions. Real science, in contrast, is never "ethno." The distinction between ernie and etic or "ethno-" and non-ethno apparently justified anthropologist's application of their concepts to other cultures (because their concepts are not localized), while at the same time it provided a renewed impetus for endeavouring to grasp, in Malinowskis (1922: 25) famous phrase, "the native's point of view."

Implicitly - but very importantly - another presupposition is made here, namely, that it is possible to distinguish neatly between natives and anthropologists. This distinction may have been unproblematic in Malinowski's time, but it is certainly not so today. Not only do many "natives" have escaped the spatial and geographical confine- ment that made them appear "native" in the first place (Appadurai 1988), for example, by migrating to Europe or the US. Further, a growing number of "natives" have themselves taken to social anthropology or similar disciplines, they have switched on their computers and compose busily all kinds of accounts the writing of which had appeared to be a privilege of Western researchers a few decades ago. This is especially obvious in a field of study like India where "native" scientists have reached

a degree of originality and quality that in many instances surpasses the achievements of "Western" researchers and that provide new perspectives and approaches not only for the limited field of Indian studies but for social and cultural sciences in general. Natives, then, have since long begun to define concepts themselves. The authority - or, to say it more bluntly, the power - to define and delimit concepts has become much more widely dispersed.

However, the distinction between "native's" and "anthropologist's" concepts between cultural and cross-cultural concepts has not only become blurred because natives have become anthropologists. Greater reflexivity in the anthropological ap- proach and changes of perspective that are generally subsumed under the label "postmodern" have challenged the anthropologists' power and monop- oly to define in a much more fundamental way. This challenge is not produced at the macro level, between "the West and the rest," it is a challenge at the micro level that undermines the very epistemological barrier supposed to exist between researcher and researched. In the conventional formulation, the researcher is the subject and the researched are the objects of research. Research objectifies, it deprives the humans that "are researched" (note the passive voice) of their subjecthood. In the process of research the researcher is the agent whereas those researched are only "reactive," they respond to the questions of the researcher, they are his or her informants. However, the epistemological barrier also implicates the allocation of agency and non-agency to researchers and their objects in exactly the opposite manner with regard to the practices of everyday life. Here, the researched are the agents, embedded in their worlds of practical concerns, whereas the researcher is the inert spectator looking onto this life world from the outside. Precisely because he is not really involved in the constraints of practice, he is able to develop a higher (objective, cross-cultural, metalocal etc.) vision of things. According to Pierre Bourdieu, the researcher belongs to the sphere of theory whereas the researched inhabit the sphere of practice.3 In much anthropological representation "the intellectual ooerations of natives are somehow tied to their

2 For a detailed discussion of the terms see Headland et al. 1990.

3 Bourdieu applies the distinction between the spheres of theoretical and practical logic with the intention to liberate practice from the shortcomings of theory in order to prevent the misunderstanding that theory could represent practice (1987: 28). However, his distinction (probably unintentionally) cements the epistemological barrier which presupposes the researched as detached from the sphere of theory, unable for the theoretical endeavour.

Anthropos 96.2001




niches, to their situations" (Appadurai 1988: 38); the intellectual operations of anthropologists are, however, not.

This is of course a gross simplification, and I would contend that a basic axiom of a more reflexive social anthropology must be that the epistemological barrier between researcher and researched is a distorted image of the structure of relations of production of knowledge in (not only) social anthropological research. Researcher and researched are interconnected in a web of relations of interaction (practice), and both researched and researcher are continually constructing knowledge and theories, i.e., generalized abstractions far exceeding the range of practical experience (cf. Moore 1996). Researcher and researched do not occupy principally different positions within the web of social and cultural relations. The difference between their positions is a matter of degree rather than of kind (cf. Fuchs 1999: 14; Sökefeld 1997a: 46-48). One is tempted to say that, if not a basic epistemological and ontological difference distinguishes researcher and researched, they at least belong to different cultures - the researcher being positioned in a culture of science. However, also the idea of cultural difference has become increasingly unclear and difficult to define. It seems that the image of different cultures, visualized like countries on political maps by delimited territories contrastingly coloured, is giving way to an idea of difference as standing out in diverse centers with large, continuous transitional border zones around and between them (Sökefeld 1999a).4 The culture of science would then consist only in a number of central values like objectivity and impartiality surrounded by zones of values and practices which are similar to those of nonscience culture.

The similarity in the positions of researchers and researched again challenges the power of cultural sciences to define general concepts. Those whom we study participate in the processes of constructing knowledge and interpreting the world,

and they too develop understandings and definitions of concepts that are meant to be generally valid and applicable. Finally, cultural/social scientific discourse and "everyday" discourse are not separate from one another in a waterproof way. Terms enter scientific discourse from everyday discourse and vice versa. Mana and taboo are examples for the first kind of transgression of discursive boundaries, "ethnic identity" is an instance of the second type. Such discursive transgressions may occur several times consecutively. Each of them most probably implicates semantic alter- ations: perhaps a multiplication of employment of a concept, a broadening of its meaning, or its in- sertion into new political and social contexts. This condition of concepts, called "dual hermeneutics" by Anthony Giddens (1976), is even more significant in social anthropology than in other social sciences, because discursive transgressions that have to be taken into account by social anthropology do not only occur between different discourses within a society (e.g., scientific discourse and everyday discourse) but also between discourses in different societies, related to diverse cultural circumstances.

I have collected so far a number of serious challenges to the special, authoritative position of social/cultural scientists and their concepts compared with their "objects" of research, including their concepts. These considerations I will take as background to discuss the critique of the concept of identity, voiced by Richard Handler and others, which denies that identity is a useful cross-cultural concept.

3 A Critique of Identity as Cross-Cultural Concept

Handler (1994) addresses the problem of reification. Social anthropological and generally academic discourse has increasingly become aware that many of its concepts are employed in reifying ways. Reflexive scientific discourse has taken pains to expose such reifications like "cultures," "traditions," etc. as social constructions (e.g., Hobsbawm and Ranger 1993) and frequently other concepts have been employed in order to escape the danger of reification. Identity certainly has become a favourite among such replac- ing concepts. "Cultures," writes Richard Handler (1994:27), "get constructed, deconstructed, and reconstructed as people pursue their identities," and he points to the obvious fact that identity is by no means immune against the dangers of reification. He, therefore, demands that the concept

4 An important factor in the conventional conceptualization of culture as delimited by clear difference is the coevolution of the two concepts nation and culture. Where a nation is a strictly territorially delimited political realm of power, the culture which defines that entity (as in the notion of national culture) is imagined in the same way. I would rather suggest to imagine cultures oriented at the model of prenational (and prenationalist) political entities, where centers of power, frequently invested in the person of the king, stood out. With growing distance from that center its power became smaller and the influence of the power of other centers increased, without a clear boundary in between. Thus cultures can be imagined as overlapping realms with central and peripheral regions.

Anthropos 96.2001




of identity must not be taken for granted but has to be explained in the same ways as those other concepts which are today employed with considerable reservation only. Identity, just like culture or tradition, is not simply a fact but a construct. Or, to put it in other words: Identity does not explain anything; identity has to be explained. Handler argues that, as a quite recent term of Western intellectual discourse, identity should not be applied unreflectedly to other cultures or times. He criticizes that today social anthropologists find "identity" in all cultures and times, thereby according "identity" the status of a human cultural universal. According to Handler, identity, a concept developed within the Western intellectual tradition, presupposes certain other ideas like that of a bounded and constant individual, an idea which psychological anthropology has shown not to be a universal. Further, identity in the West is usually referred to in the singular, i.e., a person has only one identity. To talk about the "identity" of humans of other cultures implicitly imposes, according to Handler, such preconceptions of the person onto them.5 If we subscribe, for instance, to Marriott and Inden's proposition that humans in India are not "m-dividuals" but "dividuals" because the Indian cultural model of the human being does not presuppose its boundedness, substantial unity, and constancy (Marriott 1976; Marriott and Inden 1977), we should not speak about the identity of such dividuals because that would be a contradiction in terms. Handler demands a very careful and restricted use of identity implying a sort of semantic purity. That is, he takes an original semantic structure of the concept and argues that it should be applied only where this structure makes sense.

Handler certainly raises important questions. The ethnographic foundation of his argument, however, seems quite hollow to me. It is based on an uncritical reading of ethnography, untainted by all questioning of ethnographic authority and representation. He refers to Marriott or Geertz without giving the slightest hint to the heated debate about the concepts of the person which these authors ascribe Hindus and Balinese respec- tively. That is, the ethnographic evidence on which Handler bases his argument has by no means remained unchallenged. It is surprising that on one side Handler argues against fictions of bounded cultures, nations, and all kinds of collectivities, but on the other he does not hesitate to refer to precisely such fictions ("the Ojibwa," "Hindus,"

"Balinese") in order to anchor his reasoning. In my view, such collective designations certainly imply "notions of boundedness and homogeneity" (Han- dler 1994: 30) which according to his argument should be avoided.6 In spite of these argumentative shortcomings the critique remains significant and should be subjected to a more detailed enquiry.

4 Identity as a Globalized Concept: Academic Discourse

Although Handler argues that identity is not a universal concept and should not be employed as such in non- Western context, it has de facto doubtlessly become a globalized concept, ubiquitous not only in the different corners of the world but also in diverse kinds of discourse. The first thing which has to be explained about identity, I think, is this quite sudden ubiquity. A few decades ago the use of the concept was limited to the writings of philosophers, psychologists, and a few sociologists whereas since, beginning slowly in the 1950s and almost exploding since the late 1970s, the concept has pervaded all kinds of academic discourse. Further, it has very successfully transgressed academic boundaries and is mushrooming today in all levels of political and everyday discourse. Philip Gleason (1983), pointing especially to the pioneer- ing work of Erik H. Erikson, has lucidly explored the rise of identity in American academic discourse until the 1980s. Gleasons's paper does not mention its use in social and cultural anthropological discourse. Compared to social psychology, identity in social anthropology is rather a latecomer. This is surprising because works of anthropologists, especially Margaret Mead and Ruth Benedict, have been a highly significant source of inspiration for Erikson (Gleason 1983: 924 f.).

Because Erik H. Erikson' s work was a very important source for the popularization of identity in academic discourse I will take his understanding as a point of departure. Erikson writes: "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" (Erik- son 1980: 109). Identity as sameness reflects the

5 For a related critique see Rouse 1995.

6 Handler's critique of collective designation refers especially to all kinds of nationalist constructions (cf. Handler 1985). But because he himself quite convincingly points to paral- lels and similarities between anthropological and nationalist constructions of bounded collectivities, we should be very parsimonious with using such constructions in anthropological writing.

Anthropos 96.2001




meaning of the Latin word idem (the same) from which identity is etymologically derived. This "original" meaning of identity is also derived from the tradition of Western philosophy and logic (take for instance the clause of identity as developed from Aristotelian logic). The entrance of "identity" into social science discourse is marked, among other texts, by Erving Goffman's "Stigma" (1963), where "identity" replaced the term "self which Goffman had used in his earlier writings. Goffman does not explicitly define "identity" here but employs it in the sense of the collection of characteristics we use (often implicitly and unconsciously) to categorize persons (both others and ourselves). Here too, identity expresses a relation of sameness, sameness with a collection of attributes. This corresponds to what Handler takes as the central significance of identity: "Most scholarly usage of the term reflects our commonsense notion of identity, which I would gloss, colloquially, in this way: 'the identity of a person or group is what it really is, uniquely, in and of itself, in its inner being and without reference to externals.'" After quoting the entry under identity in the Oxford English Dictionary which conveys the same meaning, he continues: "Instances of 'identity' used in approximately these senses recur routinely both in social-scientific literature and in the discourse of cultural politics" (Handler 1994: 28).

Within social anthropology identity appeared first as "ethnic identity" or "ethnicity." Here, identity was not regarded as a problematic element of the concept; definitional endeavours rather fo- cused on the meaning of "ethnic." Already Fredrik Barth' s seminal text about ethnic groups and their boundaries (Barth 1969) employs the term "identity" without explicitly discussing its applicability and significance. Handler's critique therefore applies also to Barth who defines "ethnic ascription" as an ascription which "classifies a person in terms of his basic, most general identity, presumptively determined by his origin and background" (Barth 1969: 13, italics added). "Identity" is here taken to clarify what "ethnic" means, i.e., it is employed as a term that explains and that does not have to be explained. From this short quotation we can infer something about Barth' s notion of "identity." First, if ethnic ascription is the "basic identity," identity is not conceptualized in the singular. There have to be other identities in order that the classification into more or less basic identities makes sense, although we do not learn what other kinds of identity there are. Second, it is apparently possible to rank identities according to the degree of their "basicness" or generality. There is certain a tension

in Barth' s implicit notion of identity, not denying a plurality of identities but still affirming a basic singularity of the concept because a particular identity (here: ethnic identity) may be more basic than others.

Seen in the context of the development of ideas within social anthropology, we do great injustice to anthropologists' notion of identity if we simply insinuate that Barth and others uncritically and unreflectedly took to the term. We have to ask first what the achievement of talking about ethnicity and ethnic identity was, what other concepts were replaced by these terms and what epistemological shifts were implicated in this replacement.

The introduction of "ethnicity" into social anthropological discourse was already the outcome of a critique of reifying concepts. "Ethnic group" as a concept replaced "tribe" and this replacement of concepts signified a shift from an etic to an ernie perspective. "Tribes" were defined from outside, by the anthropologist who took certain "objective" characteristics like language, territory, social structure, ritual, or other cultural "traits" to distinguish one tribe from another. It was clear what a tribe was and, therefore, there was no necessity to talk about identity.7 In this context "identity" implicitly always signifies a problem, a need that is problematic because it may not be fulfilled once and for all. Identity is something that one does not simply have but that one has to achieve. Some social anthropologists lost their belief in "tribes" because they realized that it was indeed a problem to define and delimit such groups (cf. Moerman 1965; Leach 1954). Therefore, anthropologists started to pay attention to the question how members of a supposed tribe (or other kind of group) defined themselves. To ask for identity is essentially to ask how actors see themselves and others. It is not to take social groupings for granted but to look how they come into being, how they change, and how they are maintained (this, precisely, was the topic of Barth' s paper and edited volume). The shift from the "objective" delimitation of groups to their delimitation by asking for identities then already implicates the recognition of the construc- tive social creativity of human beings, some years before social and cultural constructivism became a major concern within anthropology.

The introduction of the term identity accordingly transfers the power to define what people are to these people, their contexts, and their respective "others." Identity is also a claim. Handler (1994)

7 For a last attempt to theorize such an objective delimitation of cultural groups see Naroll 1964.

Anthropos 96.2001




recognizes this but he says that the claim for identity is a kind of false consciousness because groups that claim an identity "are" not what they claim to be (a kind of thing, a matter, something fixed and bounded). They are always processes which are never fixed (1994: 30). This is of course true but the truth that identities - like groups - are never finally fixed nor even always consensually claimed does not simply belie claims for identity so that we could dismiss them. Handler's objection reminds of the expectation - expressed especially in the heyday of modernization theory - that the advancement of rationalization will simply wipe out all irrational "ethnic," primordial sentiments and attachments. History has proven this modernist expectation wrong (Comaroff 1996), and rather than to reassert that primordial, essentialist identities are mistaken we should try to understand why identities are claimed in this manner.

5 Identity as a Globalized Concept: Political Discourse

Identity has become globalized not only within academic discourse. It is a ubiquitous concept in many discourses and effectively escaped the control of the academy. People everywhere claim to "have" an identity. Among them are also people like "dividuals" in India, that is, human beings that according to Handler are not entitled to do so. Identity is, however, not the first globalized concept. We can learn something about identity if we first turn to another extremely successfully globalized concept of historically Western origin: the nation. Probably there are very few people today that would claim not to belong to a nation and not to have a nationality. Of course, nation and nationality remain poorly defined concepts, many attempts for definition notwithstanding.8 I do not want to repeat, compare, and analyse such definitions but recall the point made earlier, that science and the academy do not possess a mo- nopoly of the power to define concepts and the authority to control their use. People everywhere in the world may call political aggregates, even purely imagined ones, "nations" without caring whether they comply with academic definitional requirements or not. Originating in revolutionary and romanticist ideas of the late 18th and early

19th century, the discourse of the nation has grown constantly and spread all over the globe, especially since the middle of our century. Why did people outside of Europe take to the concept of the nation? Their motivation was of course not the theoretical- ly correct classification of political entities but a political purpose. Since its origin, the discourse of the nation is intrinsically connected with issues of power and rights (Calhoun 1997: 69). The notion of national self-determination, the idea that each nation has the right to govern itself, is inseparable from the idea of the nation.9 Politically involved individuals, especially from new elites, and movements took to the discourse of the nation in order to demand rights and power. This applies both to European nationalists in the 19th century and to non-European anticolonial nationalists in the 20th century. From a scientific vantage point we can try to deny that anticolonial struggles in cases like India were indeed "nationalist" movements if we are of the opinion that they do not match a European model of the nation.10 But that would be a lost game because we cannot control the employment of concepts and, as social scientists, have to take their actual discursive use into account. That means, we have to acknowledge the dual hermeneutics of the nation concept and accept that anticolonial movements have extended the meaning of the nation, including for instance also multilingual nations like India. More significant than conceptual purity and pollution is the fact that the nation concept has been adopted by nationalist movements from Western political discourse because it enabled to voice claims for independence and self-government in terms the colonial powers could not easily dismiss. The takeover of the

8 The most promising strategy to arrive at a reasonable and useful definition of "nation" is probably again a prototypic definition as exemplified by Calhoun (1997: 4 f.).

9 Referring to the perspective of early liberal nationalism, Calhoun writes: "People should not be loyal to such leaders [kings, emperors], but to their nations. It was as members of such nations that they could achieve 'self-determination,' both in the sense of democratic self-rule (or at least republican constitution-making) and in the sense of autonomy from the domination of other nations" (1997: 87).

10 Hobsbawm does so when he writes: ". . . apart from a few relatively permanent political entities such as China, Korea, Vietnam, and perhaps Iran and Egypt which, had they been in Europe, would have been recognized as 'historic nations,' the territorial units for which so-called national movements sought to win independence, were overwhelmingly the actual creations of imperial conquest, often no older than a few decades, or else they represented religio-cultural zones rather than anything that might have been called 'nations' in Europe. Those who strove for liberation were 'nationalists' only because they adopted a Western ideology excellently suited to the overthrow of foreign governments, and even so, they usually consisted of an exiguous minority of indigenous évolués" (1995: 137).

Anthropos 96.2001




that a colonized people is not a nation destined for self-determination because it lacks definitional requirements of a nation.

Although not necessarily apparent, Western conceptions of identity and the individual are indeed reflected in these politics of identity at a deeper level. In the predominating Western conception as referred to by Handler, the individual is an agent because he or she is identical, that is, consistent, bounded, and continuous. Rights in the Western understanding can be assumed only by such "identical" beings. Multiple personality syndrome and schizophrenia cause great problems for this conceptualization of rights because the parallel existence of several personalities in the same body raises questions of accountability. That is, people have to comply with that conceptualization of identity in order to be recognized as political and juridical subjects. The same holds true at the level of collectivities. An important stream of Western historiography accorded agency less to individuals than to collective subjects like classes or nations. Also in contemporary politics, both national and international, collectivities are recognized and their demands for certain rights and powers are accepted mostly if they are characterized by a certain degree of fixation, continuity, and institutionalization. Collective political subjects have to comply with important characteristics of the individual identical subject (as laid down in conventional Western conceptualizations of the self) because this is a necessary presupposition for their recognition within an (international) political system dominated - in a quite literal sense - by Western conceptions. For example, collectivities that seek support of the so-called "international community" in their struggle against oppressive political structures of the state, by which they are included, have to voice their demands in terms of identity and nation or at least protonational identity in order to make their claims recognizable. Fixed, bounded nations or other identical communities have to be constructed in order to articulate, and put forward as legitimate, demands for "self-determination."

It follows that the notion of identity in the sense characterized by Richard Handler is not only imposed onto other people by social anthropologists or other scientists, but that it is also assumed by these people themselves under the powerful conditions of global discourse and practice. What are we to make out of this self-identification and self-reification by turning to identity? Is it really simply a takeover of a conventional "Western" understanding? Before debating that question, I

"modular" (Anderson 1983) nation concept was an attempt - highly successful in the end - to beat the colonial powers with their own weapons. The nation concept is then not a descriptive category but an idea that demands a certain highly desired condition of rights and power - among others national self-government.

Therefore, writes Partha Chatterjee, nationalism "is inherently polemical, shot through with tension; its voice . . . betrays the pressures of having to state its case against formidable opposition. The polemic is not a mere stylistic device which a dis- passioned analyst can calmly separate out of a pure doctrine. It is part of the ideological content of nationalism which takes as its adversary a contrary discourse - the discourse of colonialism. Pitting itself against the reality of colonial rule - which appears before it as an existent, almost palpable, historical truth - nationalism seeks to assert the feasibility of entirely new political possibilities" (1998: 40). In a very different context, the intrinsic connection between nation, nationality, power, and rights is also painfully experienced by stateless persons or by immigrants who, because they are ascribed no or the wrong nationality, are denied significant rights of participation in their country of residence.

Much of this we can find again in the concept and discourse of identity. If people insist on a certain communal identity, they do so not because they entertain certain psychological or sociological categories but because issues of power and resistance are today intrinsically connected with identity. Identities are frequently voiced in subnational context where not full-fledged self-government is at stake but still the protection or allocation of more or less specific rights. To insist on an identity is also to insist on certain rights and to be denied an identity implicates a denial of rights. Thus, immigrants demand the right to "protect" or "preserve" their identity which may include demands for school instruction in, among other things, their mother tongue and religion. People may of course also claim a separate national identity, implicating the necessity of a more general autonomy or even political independence. The recognition that identity is a matter of claims, rights, and power has resulted in the replacement of a simple concept of identity in many cases by the notion of politics of identity. To argue that the concept "identity" may not be applied to others because they lack certain definitional characteristics presupposed by that application is to miss the political contents and intention of the concept, its character as a project. It resembles the assertion of colonialist discourse

Anthropos 96.2001




will discuss an important shift in the academic conceptualization of identity, a transformation of the concept which is missed by Handler.

powerful constitutive force" (Rattansi 1994: 29). It is significant, I think, that this emphasis on difference was voiced by theorists like Stuart Hall (1991) or Avtar Brah (1996) who, especially in the consequence of migration, have themselves experienced the effects of difference. They take pains to avoid the pitfalls of taking the insistence on difference simply as a point of departure to construct other, simply more "differentiated," essentialized identities. Instead they emphasise that identities are always "crossed-through" and intersected by differences. There are no pure identities. What was taken as constant, consistent, and singular identities before emerges now as temporary subject positions made of three conditions: difference, plurality, and intersectionality. This conceptualization puts Adorno' s dictum into theoretical practice that "the smallest rest of nonidentity sufficed that identity - a totalizing concept - is denied" (Adorno 1982: 33, my translation). It is very important that difference is supplemented by plurality and intersectionality because both conditions prevent a renewed essentialization. Plurality means that differences never exist in the singular. Everybody has to position her- or himself (or is positioned) with regard to a multiplicity of differences, and these differences do not simply run parallel and congruent but intersect each other. That is, this plurality may have a number of different effects: differences may (partially) erase each other but they may also reinforce one another - and they may do so only for a certain time and context. This is how I understand Martijn van Beek's idea of "dissimulation": Differences may be postponed and dissimulated in a certain context or for a particular purpose, resulting in a specific self-essentialization, but they are effective nonetheless and will impress their marks on the scene in certain other moments (Van Beek 2000).

6 From Identity to Difference

The both popular and academic notion of identity as sameness disregards an important shift in the present conceptualization of the concept. In much current discourse, identity is almost replaced by difference. The shift from identity to difference signifies that external references, which, according to Handler are rather excluded from the "classical" understanding of identity, are not only taken into account but occupy a central position in the structure of the concept. In anthropological conceptions of ethnic identity (cf. Barth 1969) this shift to difference was already prefigured because reference to others was included. Thus, the social anthropological idea of identity never simply took the "classical" notion of identity as ¿¿//-identity for granted. But anthropology cannot take the credits for the present emphasis on difference instead of identity. The most influential moment in the par- adigmatic shift was a series of critiques emerging from political and personal experiences of discrimination (i.e., difference) that were levelled against central, dominant theory from more "peripheral" perspectives. This questioning moment appeared first in the field of academic feminism where the existence of an identical category "women" was unmasked as a Western white middle-class construct. According to nonwhite critiques, the category woman is far from identical - it is subject to all kinds of differences like class, "race," and ethnicity (Barrett 1987: 30; Crosby 1992). There is no shared identity as "woman" but only a multitude of subject-positions marked by many distinctions (cf. Butler 1990): "The concept 'Woman' effaces the difference between women in specific socio- historical contexts, between women defined precisely as historical subjects rather than a psychic subject (or non-subject) . . . For . . . only as one imagines 'women' in the abstract, when woman becomes fiction or fantasy, can race not be seen as significant" (hooks 1993: 124). The insistence on difference challenges essentialized identities. This line of thought was extended from academic feminism to a general theoretization of "identities": "Alterity is important . . . because subjects and the social, and thus both individual and collective identities, are seen not as essentially given, but as constantly under construction and transformation, a process in which differentiation from Others is a

7 Politics of Difference and the Dual Hermeneutics of Identity

I have proposed that people struggling for power and the recognition of rights assume the discourse of identity (or the nation) in order to make their demands audible in a political system dominated by (already recognized) nations. This "strategic es- sentialism" (Spivak 1988) is not simply an unalter- ing acceptance of Western, dominant premises and "modular forms" (Anderson 1983), as Chatterjee has argued in the case of the nation concept (Chat- terjee 1998). The nation was not simply imagined in a "Western" style by anticolonial movements

Anthropos 96.2001




but it was reimagined.11 In my view, the emphasis on difference instead of identity is also - at least partly - the result of such a reimagination.

We have already seen that the critique of the identical category "women" in feminism was voiced from more "marginal" positions. In a short digression I want to present another example of the discourse of difference. It is taken from a nationalist discourse in the Northern Areas of Pakistan. A brief outline of the historical and political context of the Northern Areas is necessary to understand the point. The Northern Areas (NAs) are the high mountain area of the Western Himalaya, administered and controlled since the end of 1947 by Pakistan. Before, the area was under the influence of the British, with a small portion of directly administered territory and the greater part of indirectly ruled princely territories. The precise political circumstances during colonial rule were quite complex, but here it suffices to know that the whole territory was given under the administration of the Maharaja of Kashmir when the British left the region in July 31st, 1947. Political leaders in Gilgit, the central place of the NAs, were not happy with this new situation, and shortly after the Maharaja of Kashmir had declared accession to India a revolt was staged in Gilgit. The area was "liberated" and the local provisional government invited the government of Pakistan to take over the administration. The main motivation for this move was that the political leaders in Gilgit preferred the merger with the newly founded Muslim state of Pakistan. India, in contrast, was perceived as a Hindu state. After two weeks of indecision the Pakistani government sent a representative to Gilgit who took office in Gilgit (Sökefeld 1997b). In practice, however, the NAs continued to be administered like a colonial territory. The NAs were never recognized as part of the territory of Pakistan. They are still regarded a disputed territory only administered by Pakistan, pending the final solution of the Kashmir dispute.

A great part of the local population is not content with this state of affairs because it entails a number of juridical and political disabilities. For example, until today the people of the NAs have no right to cast their vote in the elections of the National Assembly of Pakistan nor do they have a kind of local legislative assembly.

Periodically, protest and resistance against these conditions erupts in the NAs, especially in Gilgit. Since the end of the 1980s, local political parties have started to voice a local nationalist discourse of the NAs, claiming that the region constitutes a nation, different and to be distinguished both from Pakistan and Kashmir. From this claim, the need of (national) self-determination and self-government is derived (see Sökefeld 1997a). What is of interest here is the question how "national identity" is conceptualized in this discourse. Any conceptualization of identity has to deal with the condition that the area is characterized by a very high degree of internal differentiation in historical-political and linguistic respect, as in the following quotation from a nationalist pamphlet: "Although, due to their particular geographical conditions, several parts of Gilgit Baltistan [a previous name for the NAs] have been divided into small independent states, and although the local rulers have often quarrelled among themselves, when during different historical periods Moghul, Kashmiri, Dogra armies and the British tried to conquer this region, the local rulers put aside their quarrels and fought against the foreign enemies and sometimes forced them to retreat. In this manner, the local rulers with their nationalist zeal have preserved the historical, geographical, political, and cultural identity of the country" (Anonymous 1996: 3 f.).

The Urdu word which I translated here as "identity" is tashakhus. Tashakhus is derived from the Arabic root shakhsa, to rise, to appear. The meaning of tashakhus is not identity in the sense of selfsameness but rather "difference." It is related to tashkhis, distinguishing, ascertaining. Tashakhus is what stands out, relative to others, what is distin- guishable. This may be an etymological accident but "identity" is indeed conceived here in terms of "difference from others" with whom those who are characterized by this tashakhus are indeed not identical. Identity is not selfsameness but sameness in terms of a shared difference from others. Once I discussed the issue of the nation of NAs with one of the leaders of the nationalist movement, Nawaz Khan Naji. He told me that this nation was indeed fundamentally different in its history, language, and culture from the nations of Kashmir and Pakistan. He even said: "We [the members of the nation] all speak one language." I was quite irritated, knowing that in Gilgit many different mother tongues are spoken. But he explained: "Of course, we speak many languages, but they are all the same because they are fundamentally different from the languages spoken by the surrounding nations" (i.e., Kashmir and Pakistan). The lan-

1 1 We also have to be careful not to essentialize a "Western" notion of the nation, as there are important differences in "Western" conceptualizations laid down, for instance, in the German "cultural nation" and the French "republican nation" (Koselleck 1992).

Anthropos 96.2001




guages are the same - or identical - because they are characterized by a common difference distinguishing them from others that do not belong to this nation.

The whole idea of a nation of the NAs rose in the context of experienced political differences, i.e., political discrimination by the government of Pakistan. Whereas the freedom struggle in 1947 aimed at a merger with the Pakistani state - identity with the Pakistani nation was already presupposed by virtues of being Muslims - the political struggle of the 1990s introduced the concept of a separate nation in order to voice the demand for self-determination because political participation apparently could not be achieved within the political system of Pakistan. An experience of difference was turned into the postulation of identity - a different identity of course. However, this difference is not unproblematical because the category and identity it engenders - the nation of the NAs - is affected and, in fact, divided by other differences, like descent, local origin and, most important and effective in terms of violent conflicts, sectarian affiliation. Just as the identity "woman" is not unproblematical in the perspective of those women whose identity is also marked by "racial" difference, the identity of the nation of the NAs is not perceived as unproblematical. There is not only a struggle between people of the NAs and the government of Pakistan, framed in nationalist terms, but there is also a perceived com- petition between different kinds of difference. For instance, the idea of a nation has to be established against the powerful dividing forces of religious difference. Although the difference of nation is not able to erase the difference of religion, it at least has to defer, to postpone, or to "dissimulate" it. A national identity is therefore not at all taken for granted, it is a project which, although its historical potentiality is perceived as given, has to be achieved against rivaling differences.

For my work in Gilgit, this local conceptualization required a reconceptualization of identity in terms of difference (Sökefeld 1997a, 1998). I distinguished different "dimensions of difference" intersecting one another and producing a field of forces in which individuals constantly have to position - and to reposition - themselves. I have to admit here that a local concept of identity - like tashakhus - was explicitly employed in Gilgit only in the case of the nationalist discourse. In all other cases, writing about religious or local identity, I am, according to Handler's critique, apparently guilty of imposing that term on the local people. The particular concept of identity I

imposed on them is, however, far away from what Handler gives as "the" Western notion of identity. Rather than presupposing selfsameness and a singular fixed essence, my concept of identity, developed in the empirical context of Gilgit, is characterized by the emphasis on difference, intersectionality, and plurality. Rather than imposing a fixed, Western idea of identity on non- Western people, I took their implicit conceptualization to reformulate what identity could mean. This, then, like the feminist reconceptualization, is a case of the recognition of dual hermeneutics of a social scientific concept. Far from being colonized by a supposedly universal idea, a local discourse is taken as point of departure to question that idea. Therefore, I would plea not guilty in the sense of Handler's critique.

In an earlier paper Handler has raised another important issue and warned social anthropologists not to become complices of nationalists or other fabricators of "identities" by taking their constructions and politics of identity for granted and sub- scribing to similar conceptions of identity (Handler 1985). Of course, my analysis of the nationalist discourse has to be cautious not simply to affirm the nationalist account. Although we as social anthropologists are inextricably part of the game if we research on identities we should not simply succumb to somebody's discourse. The danger to do so is not negligible because, as Handler has pointed out, social anthropological concepts like older, more essentializing notions of "ethnic groups," and nationalist constructions of nations are founded upon quite similar presuppositions. However, if, contrary to what Handler represents as the classical Western notion of identity, we take as an axiom of analysis that identities are constructed not upon selfsameness but upon difference, that identity does not exist in the singular but only in plurality, and that different identities intersect one another, we have already entered into a process of immunization against simply taking over particular constructions of particular identities.

8 Is Identity a Useful Analytical Category?

More fundamental than Handler's is the critique of identity raised recently by Brubaker and Cooper (2000). They not only question the cross-cultural applicability of the notion but its general usefulness as an analytical category. Their demand to drop identity as a concept of analysis rests on the diagnosis that in consequence of inflationary

Anthropos 96.2001




is turned into its opposite. This turn is driven to extremes in the author's subsequent assertion that also soft understandings of identity betray a hard core: "Even in its constructivist guise, the language of 'identity' disposes us to think in terms of bounded groupness. It does so because even constructivist thinking on identity takes the existence of identity as axiomatic. Identity is always already

fc there,' as something that individuals and groups 'have,' even if the content of particular identities, and the boundaries that mark groups off from one another, are conceptualized as always in flux" (2000: 27 f.).

However, for most contemporary theory identity is certainly never already there. In my example of the construction of a national identity of the Northern Areas of Pakistan it is possible to tell the precise story of its coming into being. It is even possible to give the time when certain designations for this imagined nation were invented, in spite of the fact that this nation was from its very beginning imagined as having (almost) "always already been there" (Sökefeld 1997a). This may not be the case in all or even most instances of identity - not because specific identities are axiomatically presupposed but only because too much history has passed and covered the precise origin of this identity.

What for Brubaker and Cooper is the reason to reject identity is for me an essential element of its usefulness as an analytical category: the duality of essentialist and constructivist readings, of identity as a category of practice and a category of analysis. The concept identity enables us to point to certain constructions in social reality while simultaneously conveying the essentialist implications of actors and the (de-)constructivist stance of social scientists. This Janus-faced semantic structure of identity - if fully realized - is an insurance against the conflation of categories of practice and of analysis. It reminds us strongly that identities, although posed by actors as singular, continuous, and bounded can be shown to be subject to the condition of plurality, intersectionality, and difference.

use both in academic discourse and in "real life" the concept has lost specific contours. Identity is "too ambiguous, too torn between 'hard' and 'soft' meanings, essentialist connotations and constructivist qualifiers, to serve well the demands of social analysis" (Brubaker and Looper 2000: 2). This dualism of "hard'Vessentialist and "soft'Vcon- structivist understandings of identity is at the core of their argument. Initially, they equate the hard understanding with identity as a category of practice and the soft one with identity as a category of analysis. They demand not to blur the boundary between both kinds of categories and not to take over essentialist identity into analytical discourse simply because essentialist, reified readings of identity are employed in the politics of the people we study. Their conclusion that "analysts of this kind of politics should seek to account for this process of reification" and that "we should avoid unintentionally reproducing or reinforcing such reification by uncritically adopting categories of practice as categories of analysis" (Brubaker and Looper 2000: 5, original italics) is valid if somewhat close to kick at an open door because today few ethnographers of identity simply take over their subject's conceptualization. Brubaker and Cooper continue by distinguishing a number of different uses of identity as analytical category. They come to the conclusion that these diverse understandings "point in sharply differing directions;" that identity ". . . bears a multivalent, even contradictory theoretical burden" (2000: 8). Such a diversity of readings of a concept based on diverse and even contradictory theoretical approaches is in my understanding a quite normal state of affairs in the social and cultural sciences and not a reason to abandon such concepts. Returning to the distinction of hard and soft understandings of identity, their argument takes a surprising turn. Taking the hard understanding as the "real" meaning of the term because it is derived from everyday usage, they reject the soft understanding of identity because ". . . it is not clear why weak conceptions of 'identity' are conceptions of identity. The everyday sense of 'identity' strongly suggests at least some self-sameness over time, some persistence, something that remains identical, the same, while other things are changing. What is the point in using the term 'identity' if this core meaning is expressly repudiated?" (2000: 11, original italics). Here, identity as an analytical concept is rejected on the grounds that its soft reading is sharply distinguished from its understanding as a practical category. That is, the authors' earlier demand not to conflate practical and analytical concepts

9 The Self

This reading of identity, however, leaves us with another difficulty. If identity - as analytical category - does not refer to an essential selfsameness but to a multiple social web of intersecting strands of difference, what about the individual then? Are individuals only ever changing, ever newly tied

Anthropos 96.2001




knots in this web? This image does not simply accord the individual human being a passive role in that web only, it outright dissolves the idea of the individual. This conceptualization very much fits with sociocentric conceptions of the individual prevalent in much social anthropological analyses of others, i.e., non-Western human beings (Söke- feld 1999a). Handler refers to such supposedly other cultural concepts of self in order to support the argument that "identity" in the sense of a fixed, bounded, and homogeneous entity is not cross-culturally applicable (1994: 30-34). I am not convinced by his argument because he considers only one type of imposition of "Western" concepts on others: being critical of mistaken conceptual assimilation, he warns us not to impose "our own" concepts of self and identity on others. He is, however, not critical of othering, i.e., the imposition of a conceptual a priori difference on these others. But the critical and reflexive debate of the 1980s has shown that the difference of others, that is, non- Westerners, is often an artefact of our premises and approaches (Fabian 1983). As a famous example of "other" concepts of self Handler refers to Geertz's writings about the person in Bali (Handler 1994: 32 f.; Geertz 1973). Fredrik Barth has argued since that Geertz's construal of the person in Bali is more a result of a selective evaluation of cultural material than of an ethnography of actual persons on the island: "Ideally the performance part of the person, conforming to conventional roles, could indeed - using Geertz's expression - be called 'depersonalized;' but that is neither in the Balinese view a complete model of the person, since it lacks impulse and emotion, nor is it capable of accounting for the behaviors that emerge among Balinese. The North Balinese understand a great number of events and social facts to be the product of the hidden, individual features of people's hearts and the imperfections of their moral discipline. What is more, knowing that an imperfectly bridled heart, turbulent with desires, emotions, and evil, hides in every fellow human being, the Balinese orient themselves accordingly. Contrary to Geertz's understanding and description, they are exceptionally, though usually quite discreetly, alert to every slightest clue to each other's individuality. This makes good sense: only by observing closely can one hope to read the true import of acts and reactions, with the avoidance of offense and fear of sorcery reprisals as critical concerns that permeate nearly all social intercourse" (Barth 1993: 169). Not only the Balinese case nurtures doubts about the validity of the difference between Western and non- Western selves that in

many accounts appears as a dichotomy a priori and that not only overgeneralizes and overhomoge- nizes non- Western selves as sociocentric but, also, Western selves as egocentric (Spiro 1993).

According to Barth, there are even in Bali individuals with individual concerns. That is, the reading of identity as deconstructed by difference has to stop short of completely erasing the individual agent. After all, the supposedly Western conception of a fixed, essential identity had to be abandoned because we recognized that identities are constructed. Identities are constructed not by impersonal historical or social forces but, of course, by human individuals within particular social and historical conditions and constellations. These individuals also have to deal with the plurality of differences in which they are suspended or, in other words, with the plurality of identities they assume or are ascribed in particular moments. Therefore, I regard indispensable a concept of self endowed with agency and reflexivity as pertaining to every human being. At first sight it may seem that with the idea of such a self I am reintroducing through the backdoor the kind of Western identity which just before I have dismissed. But this is not the case. This concept of self does not presuppose a certain cultural conception. It does not presuppose ideas like individual autonomy, clear boundedness or individualism. It just assumes that every human being is able to act. This ability of agency does not mean anything more than that every human being is able to take initiative (Arendt 1958), that is, to make beginnings with its deeds and communications. Agency of course takes place within a field of social, cultural, historical, and economic forces, but agency is underdetermined by these forces. Agency also includes reflexivity, that is, the ability to reflect consciously about the field in which action takes place, to evaluate outcomes of action and to attempt to prefigure the outcome of further action - all this, however, does not implicate that action actually achieves the previewed outcomes. There may of course be particular cultural conceptions of the self but I regard agency and reflexivity as human universais, to employ this rather discredited term. According- ly, even in cultural environments which promote strictly sociocentric, anti-individualistic conceptions of the self, actual selves are not deprived of agency and reflexivity.

One may think that if reflexive agency is regarded as a human universal, social and cultural anthropology has nothing to do with it because we are interested in human beings as cultural beings. But our ideas about human selves are

Anthropos 96.2001



Anthropos 96.2001


very important premises for our understanding of humans as cultural beings, for our understanding of culture indeed. The idea of reflexively acting selves is a powerful check against deterministic pretensions of culture. It forces us to reflect in- tensely about how relationships between culture and selves are to be conceived, how selves acquire, enact, create, and transform culture. This requires us to pay strict attention to selves in our ethnographic work and, further, to represent selves as selves in our ethnographies.

10 The Politics of Concepts of Self

In my view, there are also very important political reasons to insist on such a conception of reflexively acting selves in social anthropology. During the greater part of its historical development, the discipline of social anthropology has neglected selves in favour of collectivities like groups, societies, or cultures. Besides, psychological anthropology has quite successfully taken pains to point out that cultural conceptions of the self may be very different from what is supposed to be the Western conception of self and individual. A dichotomy of Western selves as egocentric, individualistic and non- Western selves as sociocentric has emerged and sociocentrism almost became a premise for anthropological conceptions of others' selves. This argument of fundamentally different selves was the base of Handler's rejection of the applicability of the concept of identity at non- Western people. But this dichotomy is politically not innocent. Concepts of selves are also far-reaching if mostly not explicit premises of political conceptions and rights, and the non- Western sociocentric self can be appropriated for the purpose of repressive political aims and practices. In what follows I tentatively want to explore the relation between conceptions of selves and politics by discussing the issue of universal human rights.

The Universal Declaration of Human Rights of 1948 conceives of individual subjects as the subjects of human rights; subjects which are supposed to be "equal" and endowed with agency, freedom and individual dignity.12 The purpose of the Declaration of Human Rights is precisely to protect these individual endowments. Obviously,

the kind of individual or self that is presupposed as the subject of human rights is more akin to the projected egocentric Western self than to a sociocentric non- Western self. Therefore, social anthropologists sometimes take the idea of non- Western sociocentric selves as point of departure to challenge the universality of human rights (Wilson 1997: 6). Unfortunately, such an argument is not restricted to theory and academic debate. The authorities of countries like China, for example, often point to the assumed sociocentricity of their subjects' selves and conclude that a declaration of human rights should not emphasize individual rights but the obligations of the individual toward his or her community. Here, anthropological conceptions of the self can easily be misused as legit- imation of repressive political action. In a similar vein, Western cultural relativists' challenges of the universality of human rights implicate the same kind of non- Western selves - selves subordinate to cultures that differ from Western culture especially in not buttressing individuality.13 They allege that human rights conceived of as universal are ethno- (Euro-) centric or culture imperialist and endanger the viability of other cultures. Rhoda Howard quite polemically writes: "Denial of the individuality of members of these [non- Western] societies merges them into an amorphous whole, in which cultures are at risk of being destroyed (for example by introducing new norms of human rights) but actual people are not. When the people who comprise these romanticized cultures are considered not real individuals with their own needs, wants, and desires but rather living anthropological exhibits, then their human rights can go unheeded. The primitive by definition is natural and cannot have socially constructed desires for human rights that Westerners have as refined, alienated social beings. The primitive is not capable of abstract thought, of stepping out of her environs to consider the nature of social life or the ethics of her group. Thus to introduce the ideal of human rights even into verbal discourse with a primitive is to be an

12 See for instance Article 1 of the Universal Declaration of Human Rights: "All human beings are born free and equal in dignity and rights. They are endowed with reason and conscience and should act towards one another in a spirit of brotherhood."

13 For such a position see Pollis and Schwab (1979: 8 ff.). Wilson criticizes this relativism: "Yet the most serious contingent 'reality' is that relativism itself is an integral part of the meta-narrative of those governments who actively oppose the applications of international human rights to their polities. The tolerance of relativism here has a directly conservative political implication - the maintenance of highly inegalitarian and repressive political systems. An undeniable truth is that many governments around the world continue to carry out abominable acts against 'their' populations, and relativism is the most useful available ideology which facilitates international acquiescence in state repression" (1997: 8 f.).




imperialist, to set off a process of social change that may well wreck the indigenous social order" (1993: 328). The reverse side of cultural relativism is an ideology of authenticity that conjures up cultural purism as highest value - regardless of the fact that culture never is (and was) pure but "tainted" by hybridity.

The argument criticised by Howard points to a reified understanding of culture quite similar to traditional anthropological notions that conceive of culture as bounded and shared, which oppose culture and self and which finally relegate the self to a position of lesser importance because it is only a product of culture. In the famous declaration on Human Rights of the American Anthropological Association, drafted by Melville Herskovits, this position is clearly expressed: "If we begin, as we must, with the individual, we find that from the moment of his birth not only his behavior but his very thought, his hopes, aspirations, the moral values which direct his action and justify and give meaning to his life in his own eyes and those of his fellows, are shaped by the body of custom of the group of which he becomes a member" (American Anthropological Association 1947: 539 f.). Consequently, the rights of a culture almost replace individual rights: "The individual realizes his personality through his culture, hence respect for individual differences entails a respect of cultural differences" (1947: 541). The major shortcoming of cultural relativist arguments in relation with the issue of human rights is that they are unable to specify who is capable of determining what the particular culture of a given individual or group is. Consider Indian culture and society. According to Dumont (1980) and Marriott (1976), there are no valued individuals from the perspective of Indian culture either because the ("empirical") individuals are always understood as parts of a group, or because they are no real units of society but they themselves (as "dividuals") are subdivided into coded substances. From the Indian cultural perspective it is then perfectly possible to justify certain "cultural" ways of interaction which implicate practices of gross inequality and that are repressive for specific kinds of individuals from the universalist point of view, especially for so-called Untouchables. But who is in the position to decide what is "Indian" and what is not, which ways of action are allowed because of their "Indianness" and which are not? Is the competence for this decision vested in certain cultural experts (Brahmans? anthropologists?) or in sources like texts? Is it a matter of statistics, i.e., of the normativity of facts? Or is every empirical Indian

individual entitled to decide that question for him- or herself? The last possibility is in fact impossible because what is at stake are conceptions of relationships between human beings, and this is not an individual issue. However, there are conceptions of culture and individuality in India held by Indian empirical individuals which are quite at odds with for instance Dumont' s understanding of "Indian." I am of course alluding here to positions expressed by the underprivileged of Indian society that hold their own concepts of the person (Vincentnathan 1993) and of values (Mencher 1974) and that denounce what is considered as generally "Indian" by Dumont as a position held by certain, dominant sections of Indian society. In their view it is just the "brahmanical view of caste" (Berreman 1979). This inner-Indian dispute mapped onto an anthropological debate points to the fact that understandings represented as "cultural" are subject to power relations and intense social struggle.14 If we understand culture as process and struggle and selves as reflexively acting beings involved in this struggle, the argument of cultural relativism (or "cultural absolutism," as Howard calls it) against universal human rights dissolves. No culture can then be turned as an absolute value against the struggling individual.15 Again, we consequentially have to accept that individuals/selves are indeed reflexively acting (struggling, making decisions, dissenting, putting themselves in opposition to groups or dominant views, constructing and deconstructing identities, etc.) and accordingly we have to give importance and attention to these acting selves and to affirm their rights.

11 Conclusion

In this paper I have gone a long way from the difficulties of anthropological concepts to the idea of

14 This difficulty becomes especially obvious if applied to gender affairs. Mostly, "cultural" rules curtailing the op- portunities of women are made by men. Men are no doubt frequently in a position to decide what is the "culture of women." But their position is of course a result of power relations and not of "cultural expertise."

15 Such a position also lifts the suspicion of cultural impe- rialism from the idea of universal human rights. Human rights that are already a product of cultural struggle in the West then appear as an instrument in the struggle, for example of Dalits in India, for the improvement of their social conditions. It is anyway worthwhile to reflect upon the fact that human rights are accused to be imperialist mostly by those dominant in non-Western societies, not by those who labour to improve their conditions of living against such domination.

Anthropos 96.2001




universal human rights. Taking Handler's critique of the reifying imposition of a Western concept of identity on others as a point of departure, I have briefly discussed the globalization of identity discourse in both the academy and the political realm and pointed to a new conceptualization of identity as difference - a conceptualization quite at a distance of what Handler gives as "the" Western notion of identity. This conceptual change is part of the dual hermeneutics of identity: It is not simply a result of academic debates but of intersec- tions of political and academic discourses. Identity as difference indeed questions the core characteristics of Handler's conventional meaning of the concept. Identity is not identical anymore. Rather than to liberate anthropology's "others" from the reifying imposition of the concept of identity, we should free that concept from the encumbrance of certain reifying presuppositions. Further, I have argued that, contrary to the critique voiced by Brubaker and Cooper, the value of identity as an analytical category is its ability to accommodate the duality of essentialist and constructivist readings. The acknowledgement of human's ability to construct and deconstruct identities requires us to conceive of human individuals as reflexively acting selves - a notion that contradicts a great deal of anthropologists' construais of others, i.e., non-Western, selves. With the debate about the cultural relativity or universality of human rights I have finally entered politics a second time.

The concept of identity is a striking example that shows how central concepts of social anthropology are embedded in political ramifications. In my opinion, the analytical usefulness of our concepts cannot be ensured by trying to maintain a conceptual purity that struggles to keep them out of compromising political contexts. To enable and maintain the analytical value of our conceptual instruments, we have to reflect upon their condition of being appropriated into non-disciplin- ary discourses and the ensuing transformations of meaning. We cannot escape the dual hermeneutics that is a basic condition of our science.

References Cited

Adorno, Theodor Wiesengrund 1982 Negative Dialektik. Frankfurt: Suhrkamp Verlag. [1966]

American Anthropological Association 1947 Statement on Human Rights Submitted to the Com-

mission on Human Rights, United Nations. American Anthropologist 49: 539-543.

Anderson, Benedict 1983 Imagined Communities. London: Verso.

Anonymous 1996 Manshur. Gilgit. [Photocopied typescript]

Appadurai, Arjun 1988 Putting Hierarchy in Its Place. Cultural Anthropology

3: 36-49.

Arendt, Hannah 1958 The Human Condition. Chicago: The University of

Chicago Press.

Barrett, Michèle 1987 The Concept of "Difference." Feminist Review 26:

29-41.

Barth, Fredrik 1969 Introduction. In: F.Barth (ed.), Ethnic Groups and

Boundaries; pp. 9-38. Bergen: Universitets Forlaget. 1993 Balinese Worlds. Chicago: The University of Chicago

Press.

Beek, Martijn van 2000 Dissimulations. Representing Ladakhi Identity. In:

H. Driessen and T. Otto (eds.), Perplexities of Identifica- tion. Anthropological Studies in Cultural Differentiation and the Use of Resources; pp. 164-188. Aarhus: Aarhus University Press.

Berreman, Gerald D. 1979 Caste and Other Inequalities. Meerut: Folklore Institute.

Bourdieu, Pierre 1987 Sozialer Sinn. Kritik der theoretischen Vernunft. Frank-

furt: Suhrkamp Verlag.

Brah, Avtar 1996 Cartographies of Diaspora. Contesting Identities. Lon-

don: Routledge.

Brubaker, Rogers, and Frederick Cooper 2000 Beyond "Identity." Theory and Society 29: 1-47.

Butler, Judith 1990 Gender Trouble. London: Routledge.

Calhoun, Craig 1997 Nationalism. Minneapolis: University of Minnesota

Press.

Chatterjee, Partha 1998 Nationalist Thought and the Colonial World. A Deriv-

ative Discourse? Minneapolis: University of Minnesota Press. [1986]

Comaroff, John L. 1996 Ethnicity, Nationalism, and the Politics of Difference in

an Age of Revolution. In: E. N. Wilmsen and P. McAl- lister (eds.), The Politics of Difference. Ethnic Premises in a World of Power; pp. 162-183. Chicago: The Uni- versity of Chicago Press.

An earlier version of this paper was presented as a guest lecture at the Institute of Ethnography and Social Anthropology, University of Aarhus, Nov. 16, 1999. I thank Martijn van Beek for the invitation to deliver this lecture. Subsequently the paper has profited much from the very lively debates in my seminary on the concept "identity" at the Institute of Social and Cultural Anthropology in Hamburg, summer 2000.

Anthropos 96.2001




Crosby, Christina 1992 Dealing with Differences. In: J. Butler and J. W. Scott

(eds.), Feminists Theorize the Political; pp. 130-143. New York: Routledge.

Delaney, Carol 1991 The Seed and the Soil. Gender and Cosmology in

Turkish Village Society. Berkeley: The University of California Press.

Dumont, Louis 1980 Homo Hierarchies. Chicago: The University of Chica-

go Press.

Erikson, Erik H. 1980 Identity and Life-Cycle. New York: Norton.

Fabian, Johannes 1983 Time and the Other. How Anthropology Makes its

Object. New York: Columbia University Press.

Fuchs, Martin 1 999 Kampf und Differenz. Repräsentation, Subjektivität und

soziale Bewegungen. Das Beispiel Indien. Frankfurt: Suhrkamp Verlag.

Geertz, Clifford 1973 Person, Time, and Conduct in Bali. In: C. Geertz, The

Interpretation of Cultures; pp. 360-411. New York: Ba- sic Books.

Giddens, Anthony 1976 New Rules of Sociological Method. London: Hutchin-

son.

Gleason, Philip 1983 Identifying Identity. A Semantic History. Journal of

American History 69: 910-931.

Goffman, Erving 1963 Stigma. Notes on the Management of Spoiled Identity.

Englewood Cliffs: Prentice-Hall.

Hall, Stuart 1991 Old and New Identities, Old and New Ethnicities. In:

A. D. King (ed.), Culture, Globalization, and the World System; pp. 41-68. London: Sage Publications.

Handler, Richard 1985 On Dialogue and Destructive Analysis. Problems

in Narrating Nationalism and Ethnicity. Journal of Anthropological Research 41: 171-182.

1994 Is "Identity" a Useful Cross-Cultural Concept? In: J. Gillis (ed.), Commemorations. The Politics of Nation- al Identity; pp. 27-40. Princeton: Princeton University Press.

Headland, Thomas N., Kenneth L. Pike, and Marvin Harris (eds.) 1990 Ernies and Etics. The Insider/ Outsider Debate. New-

bury Park: Sage Publications. (Frontiers of Anthropolo- gy, V) .

Hobsbawm, Eric J. 1995 Nations and Nationalism since 1780. Programme, Myth,

Reality. Cambridge: Cambridge University Press.

Hobsbawm, Eric, and Terence Ranger (eds.) 1993 The Invention of Tradition. Cambridge: Cambridge Uni-

versity Press.

hooks, bell 1993 Sisters of the Yam. Black Women and Self-Recovery.

Boston: South End Press.

Howard, Rhoda 1993 Cultural Absolutism and the Nostalgia for Community.

Human Rights Quarterly 15: 315-338.

Koselleck, Reinhart 1992 Volk, Nation, Nationalismus, Masse. In: O. Brunner und

W. Conze (Hrsg.), Geschichtliche Grundbegriffe. Vol. 7; pp. 141-431. Stuttgart: Klett-Cotta.

Leach, Edmund R. 1954 Political Systems of Highland Burma. A Study of

Kachin Social Structure. London: G. Bell and Sons.

Malinowski, Bronislav 1 922 Argonauts of the Western Pacific. An Account of Native

Enterprise and Adventure in the Archipelagoes of Mel- anesian New Guinea. London: George Routledge and Sons.

Marriott, McKim 1976 Hindu Transactions. Diversity without Dualism. In:

B. Kapferer (ed.), Transaction and Meaning. Directions in the Anthropology of Exchange and Symbolic Behav- ior; pp. 109-142. Philadelphia: Institute for the Study of Human Issues.

Marriott, McKim, and Ronald B. Inden 1977 Toward an Ethnosociology of South Asian Caste Sys-

tems. In: K. David (ed.), The New Wind. Changing Ide- ntities in South Asia; pp. 227-238. The Hague: Mouton.

Mencher, Joan P. 1974 The Caste System Upside Down, or the Not-So-Myste-

rious East. Current Anthropology 15: 469-493.

Moerman, Michael 1965 Ethnic Identification in a Complex Civilization. Who

Are the Lue? American Anthropologist 67: 1215-1230.

Moore, Henrietta L. 1996 The Changing Nature of Anthropological Knowledge.

An Introduction. In: H. L. Moore (ed.), The Future of Anthropological Knowledge; pp. 1-15. London: Rout- ledge.

Naroll, Raoul 1964 On Ethnic Unit Classification. Current Anthropology 5:

283-312.

Pollis, Adamantia, and Peter Schwab 1979 Human Rights. A Western Construct with Limited Ap-

plicability. In: A. Pollis and P. Schwab (eds.), Human Rights. Cultural and Ideological Perspectives; pp. 1-18. New York: Praeger.

Rattansi, Ali 1994 Western Racisms, Ethnicities, and Identities in a "Post-

modern" Frame. In: A. Rattansi and S. Westwood (eds.), Racism, Modernity, and Identity on the Western Front; pp. 15-86. Cambridge: Polity Press.

Rouse, Roger 1995 Questions of Identity. Personhood and Collectivity in

Transnational Migration to the United States. Critique of Anthropology 15: 351-380.

Saler, Benson 1993 Conceptualizing Religion. Immanent Anthropologists,

Transcendent Natives, and Unbounded Categories. Lei- den: E. J. B. Brill.

Anthropos 96.2001




Sökefeld, Martin 1997a Ein Labyrinth von Identitäten in Nordpakistan. Zwi-

schen Landbesitz, Religion und Kaschmir-Konflikt. Köln: Rüdiger Koppe Verlag.

1997b Jang âzâdl Perspectives on a Major Theme in Northern Areas' History. In: I. Stellrecht (ed.), The Past in the Present. Horizons of Remembering in the Pakistan Himalaya; pp. 61-82. Köln: Rüdiger Koppe Verlag.

1998 "The People Who Really Belong to Gilgit" - Theo- retical and Ethnographic Perspectives on Identity and Conflict. In: I. Stellrecht and H.-G. Bohle (eds.), Trans- formations of Social and Economic Relationships in Northern Pakistan; pp. 93-224. Köln: Rüdiger Koppe Verlag.

1999a The Concept of Culture between Politics and Social An- thropology. From Difference to Continuity. Ethnologie Heute 3. [Available online: http://www.uni-muenster.de/ EthnologieHeute/eh3/culture.htm]

1999b Debating Self, Identity, and Culture in Anthropology. Current Anthropology 40: 417-447.

Spiro, Melford E. 1993 Is the Western Conception of the Self "Peculiar" within

the Context of World Cultures? Ethos 21: 107-153.

Spivak, Gayatri Chakravorty 1988 Subaltern Studies. Deconstructing Historiography. In:

R.Guha and G.C. Spivak (eds.), Selected Subaltern Studies; pp. 3-32. Oxford: Oxford University Press.

Vincentnathan, Lynn 1993 Untouchable Concepts of Person and Society. Contri-

butions to Indian Sociology 27: 53-82.

Wilson, Richard A. 1997 Human Rights, Culture, and Context. An Introduction.

In: R. A. Wilson (ed.), Human Rights, Culture, and Context; pp. 1-27. London: Pluto Press.

Anthropos 96.2001



Reconsidering Identity

Documents

Transcript of Reconsidering Identity