In the early years of modern social anthropology, <->anthropologists announced their most important findings and theoretical advances to Section H of the British Association for the Advancement of Science. As 1997 president of this Section, I chose to address contemporary developments in one of our oldest concepts, 'culture', as a way of continuing that tradition.1

Why be so bold as to engage with a word which Williams (1976: 87) declared was one of the two or three most complicated in the English language and which in British, North American and European anthropology has had complex, contested and very different histories? By mid-century, Kroeber and Kluckhohn had found 164 definitions in their famous review of what anthropologists meant by culture (1952: 149). By the 1970s, when cultural anthropology was well established as one of the four fields of anthropology in the USA, in British anthropology 'culture' had nearly disappeared from view. In the last ten years, with the help of cultural studies, 'culture' has resumed centre stage in British anthropology. The aim of this paper is not to tally up how many definitions of 'culture' anthropologists have generated by the end of the century. Rather, the paper pursues Kroeber and Kluckhohn's observation that 'the occurrence of these [definitions] in time is interesting - as indeed the distribution of all cultural phenomena in either space or time always reveals significance' (ib.). The aim is to treat the prominence (or 'distribution' in Kroeber and Kluckhohn's terms) of 'culture' in the 1990s as itself a cultural phenomenon. What is the significance of culture's recent reappearance as a central concept in British anthropology? The issue is not confined to internal disciplinary debate. In the last decade, politicians and decision-makers have introduced 'culture' into the discourse of many different 'fields' (Bourdieu 1991) of contemporary society. Decision-makers and media commentators often claim legitimacy for their discourses by referring to 'culture, in an anthropological sense' - a phrase which closes off further exploration by claiming that there is one (their) meaning of culture which is at once too self-evident to warrant explanation and too deep to be delved into by non-anthropologists. How are decision-makers (whether they be anthropologists or claiming legitimacy from anthropology) politicizing 'culture' and deploying the concept in a range of fields of power? How can anthropologists use their new theoretical approaches to 'culture' to explore and reveal the effects of the current uses of this concept in contemporary politics?

I will start by discussing what I am calling 'old' and 'new' anthropological approaches to 'culture'. I will then use these approaches to examine how, and with what effects, decision-makers have introduced and deployed 'culture' in three different 'fields' the last fifteen years. First I will examine British right wing politicians' use of 'culture' to talk about nationalism in such a way that they can distance themselves from the taints of biological racism, yet reintroduce exclusive practices in an insidious cultural guise. Second, I will review how writers and consultants in organizational management use ideas of 'culture', which they attribute to anthropology, to propose new forms of organization. They claim 'de-layering' and 'flattening hierarchies' and the formation of 'flexible teams' of continually self-reskilling 'portfolio' workers will permit grass-roots creativity and workers' self management and empowerment. I will explore the unacknowledged costs of such 'empowerment' and how under the rubric of 'empowering corporate culture', there lurks an older idea of organizational culture as a tool of top down management control. The third field is overseas development where 'culture' is just entering the discourse (Wright 1997). Largely this is as a result of a UNESCO report Our Creative Diversity. This report was meant to do for 'culture' what the Bruntland Report did for the environment and development, but the report has so far gone largely unnoticed. Anthropologists played a major role in formulating the ideas of culture which this report proposes should be the basis for world ethics and development policy. Anthropologists of development have long sought such influence. Some would see the aims of anthropology as understanding the local, national and international processes by which impoverished people are marginalized and disempowered, in order to influence those processes, or promote the perspectives of those who are silenced, or enable them to speak and act more effectively for themselves. When anthropologists had an opportunity to act as policy makers and steer the culture bandwagon themselves, did they deploy a concept of 'culture' which would make any of these aims more achievable? In all three fields, politicians, officials and academic advisers are using 'culture' as a political tool. Whether the concepts are being deployed by anthropologists directly involved in influencing and writing policy (as in the third instance) or whether ideas are being attributed to anthropology for legitimation, in all instances, anthropology is implicated in the politicization of 'culture'. How can we use our understandings of political processes to reveal the ways decision-makers are using 'culture' in a growing number of 'fields', and analyse its effects on those who are marginalized and impoverished?

Old meanings of culture

In the early 20th century, ideas of 'culture' advanced by anthropologists took on a radical tone. Tylor's (1871) notion of culture as a whole way of life of a group or society marked a point of departure for modern social anthropologists:

'Culture' is that complex whole which includes knowledge, belief, art, morals, law, custom, and any other capabilities and habits acquired by man [sic] as a member of society (Tylor 1871:1:1).

If this was a point of departure, it was not a basis for consensus: anthropologists set off along divergent paths. Tylor's own approach was to combine Herder's romantic idea, that nations, groups within nations and peoples at different periods have distinctive cultures, with the enlightenment idea that each of these cultures was at a different stage in the evolution of civilization or in a progression towards European rationality. Boas rejected Tylor's social evolutionism. He stressed the particularity of each culture as a result of the group's responses to environmental conditions and their specific historical development. By treating 'culture' as the product of historical and social forces, not biology, he criticized racial determinism (Stocking 1974: 221). In Britain, Malinowski and his students advanced a different critique of the rationalistic Victorian conception of 'man' by arguing that far from being 'savage' and illogical, each of the 'peoples' in Africa, South Asia and the Pacific had a distinct, rational and legitimate way of life which should be valued: 'emphasizing the authenticity and coherence of distinct cultures was a way of resisting the civilising mission fundamental to the European colonial project' (Merry 1997). Anthropologists differed profoundly in their theories and in the aspects of western thought that they questioned, but they shared an idea of the world as made up of 'peoples', each with a coherent way of life, or 'culture'.

By the 1970s, far from being radical, this idea of 'a people' having 'a culture', was seen to have been a crucial element of colonialism. To critics, this idea of 'culture' created fixed entities in which the West could intervene. By measuring, categorizing, describing, representing and thereby supposedly 'knowing' others, the objects of that knowledge were made the subjects of new forms of power and control (Asad 1973, Said 1978). This once progressive idea was also taken up in regressive ways by extreme nationalists who used it not simply to champion claims for independence and sovereignty but also to pursue the politics of xenophobia, exclusion and ethnic cleansing.

In a great flood of criticism, the components of this idea of culture were unpacked. British functionalists, for example, were criticized for having treated a 'culture' as a small scale, bounded entity organized through economic, social and political institutions which interacted as a self contained 'whole' sustained in a static equilibrium. This had clearly been a fiction when most of the places studied, however remote, were being visited not just by anthropologists, but by merchants, missionaries and colonial administrators. Societies were neither unchanging nor bounded, but part of a world order dominated first by colonialism and later by nation states, international capitalism and international agencies. These had been left out of a picture of 'cultures' as ahistorical, self-contained entities (Gough 1968).

Anthropologists of various persuasions were also criticized for treating 'culture' as if it were a set of ideas or meanings which were shared by a whole population of homogeneous individuals - which empirically was not the case.2 Asad (1979) criticized British anthropologists for seeking the unique 'authentic culture' of another society in the form of an integrated system of consensual 'essential meanings' which self-reproduced regardless of economic and political change. If anthropologists constructed the social order out of 'essential meanings' which did not change in new historical and economic conditions, how would social transformation occur? Instead, he argued, 'essential meanings' were discourses which some people in the society had managed to make authoritative by continually pre-empting the space of radically opposed discourses. The problem Asad thought anthropologists should address is how an authoritative discourse is produced in particular historical circumstances. In a paper which I take as a point of departure for the development of what I am calling 'new' approaches to culture,3 Asad argued that anthropologists had mistakenly endorsed, as 'authentic culture', historically specific dominant ideologies or authoritative discourses which were neither timeless nor uniformly shared.

Although anthropologists have developed new ways of thinking about 'culture', these 'old ideas of culture' have percolated out from academic discourse and, as will be shown below, are still in widespread use in public parlance. The main features of this, still-current 'old idea of culture' are:

bounded, small scale entity
defined characteristics (checklist)
unchanging, in balanced equilibrium or self-reproducing
underlying system of shared meanings: 'authentic culture'
identical, homogeneous individuals.

New meanings of culture

The changing political and economic conditions to which Asad referred were the end of European colonialism and the continued expansion into new areas of relations of production and exchange based on capital. Most recently, they would include the international organization of production and consumption, the spread of global communication networks, and the international integration of financial systems. These changes have provoked labour movements within countries and from the south to the north of the globe, as exemplified by a woman I met in my South Tottenham park recently. She is an Asian who grew up with an English education in Trinidad and has worked in England for 15 years in nursing and administration. She is learning Hindi at night classes so that she can converse with relatives she visits in India. Her and her family's experience of colonial labour migration, post colonial economic diaspora and 'roots' tourism speaks of what Hall called 'dislocated histories and hybridised ethnicities' (1993: 356). As anthropologists have argued for many years (Cohen 1974, Macdonald 1993), and more recently Hall and other exponents (Morley and Chen 1996) of cultural studies in Britain have made clear, cultural identities are not inherent, bounded or static: they are dynamic, fluid and constructed situationally, in particular places and times. This is not just a Western urban phenomenon of the 1990s. In a tribe in Iran where I did fieldwork in the 1970s, the population was made up of layers of refugees. Multiple identities were constantly negotiated; links with people in tribes from which they had fled were maintained or reinvented: there was no bounded, consensual, authentic, ahistorical culture. Theoretical developments in cultural studies, and in post-structural and feminist anthropology, have led us to understand that 'cultures' are not, nor ever were, naturally bounded entities.

The fracturing of social anthropology's central conceit has sent us back to look again at colonialism. Ortner (1984) questioned the original image of colonial power and 'the juggernaut of capitalism' impacting on, and inserting themselves into, an indigenous 'local culture'. She and others have been critical of the way both colonialism and 'local culture' appear as unitary entities in this image (Asad 1993: 5). What better choice of a site to challenge this image than the kind of location in which the old concept of 'culture' was founded: a remote island mid way across the Pacific Ocean? Merry (1997) studied 18th and 19th century Hawaii, and found a dizzying array of people from Norway to China were present in what she calls not a 'local community' but a 'contact zone'. In an unbounded site, this medley of people drew on the practices of their various places of origin, in the light of their current interests, to work out how to organize labour, trade and social relations. Contests took place between people in asymmetrical relations of power, over their multiple and contradictory cultural logics. Each actor endeavoured to manoeuvre, in unpredictable political and economic situations, to define or seize control of symbols and practices. Symbols and ideas never acquired a closed or entirely coherent set of meanings: they were polyvalent, fluid and hybridized. Key terms shifted in meaning at different historical times. When a coalition of actors gained ascendancy at a particular historical moment, they institutionalized their meaning of key terms in law.

Merry's is a good example of the new idea of culture as a contested process of meaning-making. The contest is over the meaning of key terms and concepts. How are these concepts used and contested by differently positioned actors who draw on local, national and global links in unequal relations of power? How is the contest framed by implicit practices and rules - or do actors challenge, stretch or reinterpret them as part of the contest too? In a flow of events, who has the power to define? How do they prevent other ways of thinking about these concepts from being heard? How do they manage to make their meanings stick, and use institutions to make their meanings authoritative? With what material outcomes?

Sue Reinhold (1993) poses these questions in order to reveal in detail the process of ideological struggle in 1980s Britain. The contest was over the power to define the state's attitude to homosexuality in Britain and make authoritative that definition through legislation. In the context of an atmosphere of homophobia and physical 'queer-bashing' attacks in London, a group in Haringey campaigned for 'positive images' to counter the negative stereotypes of gay people available to children in their schools. Their local opponents exercised their links to Conservative Members of Parliament, who appropriated and inverted the meaning of the key terms of the debate. The word 'promoting' was first used by the campaign to 'promote positive images' of homosexuality; MPs accused them of 'promoting homosexuality'. In successive parliamentary debates 'promote' was made to mean seduction of 'normal' children, which was equated with an attack on 'the family', the basis of order in the state, and thus with 'subversion'. The group of MPs succeeded in inserting a new clause into current legislation on local government outlawing actions or use of resources which would 'promote' the acceptability of homosexuality as a 'pretended family relationship'. This new meaning of 'promoting' and its associated cluster of terms, made authoritative through state legislation, had material effects: negative stereotypes were endorsed, and local authorities became timid about spending on services or issues for gay people which might possibly be interpreted as coming under the legislation in a test case. Reinhold (1993: 471-2) points to similarities between the contest over positive images and other campaigns against minorities during the Thatcher government. Right wing Conservatives used the authority of parliament to project negative meanings of key terms and symbols concerning ethnic minorities, miners and other categories which they marginalized, excluded from their dominant notion of 'Britishness' and demonized as a danger to order and subversive to the state.

Three stages in these contested processes of meaning making can be identified in the above examples. The first is overt attempts by identified agents to redefine key symbols which give a particular view of the world, of how people should be and behave and what should be seen as the 'reality' of their society and history: in short, an ideology. A second stage is when such a view of the world becomes institutionalized and works through non-agentive power. Foucault has documented how knowledge about mental health, sexuality and criminality in the 18th and 19th centuries became the basis of new practices on which institutions were built. These institutional practices shaped perceptions, categories, values and behaviour.

A third stage is when a key term which carries a new way of thinking about one aspect of life enters other domains (outside the activities of the state) and becomes a diffused and prevalent way of thinking in everyday life. For example, Emily Martin (1994) found that 'flexible' first became a key term when people reacted to the AIDS/HIV virus by rethinking the immune system and the defence responses of the body. Surprisingly, 'flexible' and images of the immune system quickly entered the domain of employment to describe the attributes of post-Fordist, self-managed, self-improving and team-forming workers and companies. Within a short time, extreme versions of these flexible attributes, which had been symptoms of a mental illness, were reinterpreted positively as employment skills (Martin 1997). 'Flexible' moved quickly across three different areas of U.S. life - immunology, employment and mental health - and become a prevalent image of a new kind of self.

At its most secure, an ideology appears hegemonic. That is, it becomes so naturalized, taken for granted and 'true' that alternatives are beyond the limits of the thinkable. As Comaroff and Comaroff (1992) suggest, in its hegemonic dimension, culture appears coherent, systematic and consensual. It tries to look like an object, a thing beyond human agency, not ideological at all: in short, like the old idea of authentic culture. As mentioned above, anthropologists themselves had previously mistaken hegemonic ideologies for authentic culture and in the process, endorsed those in the community with the ascendant power to define the characteristics of their 'culture' and project it as timeless and objective.

No ideology, however hegemonic and entrenched in institutions and in everyday life, is beyond contest; 'culture' is a dynamic concept, always negotiable and in process of endorsement, contestation and transformation. Differently positioned actors, with unpredictable inventiveness, draw on, re-work and stretch in new directions the accumulated meanings of 'culture' - including old and new academic ones. In a process of claiming power and authority, all are trying to assert different definitions which will have different material outcomes. In sum the characteristics of new ideas of culture are:

'culture is an active process of meaning making and contestation over definition, including of itself' (Street 1993: 2) people, differently positioned in social relations and processes of domination, use economic and institutional resources available to them to try and make their defini- tion of a situation 'stick', to prevent others' definitions from being heard, and to garner the material outcome sites are not bounded - people draw on local, national, global links the way clusters of concepts form is historically spe- cific, and ideas never form a closed or coherent whole in its hegemonic form, culture appears coherent, syste- matic, consensual, like an object, beyond human agency, not ideological - like the old idea of culture.

Cultural racism

In British politics, this new view of 'culture' has itself been appropriated and redefined by the New Right. Led by Margaret Thatcher, the New Right represented an alliance between liberal economic and conservative political theories (King 1987). In economic affairs the state should promote private enterprise and encourage - even invent - markets. In political affairs the authority of the 'age-old' institutions of the central state should be upheld, supported by 'traditional' values in education and family life. In a study of the Salisbury Review, a principal journal of the New Right, Seidel (1985: 107) argues that the New Right appropriated one of the founding inspirations of cultural studies, Gramsci's ideas of hegemony. That is, (as set out above), ideology becomes hegemonic not only through the institutions of the state but by being diffused through all areas of everyday life. To unsettle and replace the dominant ideology since the Second World War, the New Right realized that they had not just to be active in politics, but to make interventions in 'culture'. They consciously engaged in the manipulation of words, especially the process of renaming and redefining key concepts. In particular the New Right focused on appropriating and reformulating the meanings of one semantic cluster - 'difference', 'nation', 'race', 'culture''.4

New Right authors seem to agree with the idea that the world can no longer be seen as a mosaic of discrete cultures, and that migration and diaspora have generated populations with multifaceted differences. They appropriated the anti-racist language about the need to respect cultural difference. This did not mean that they rejoiced in cross-cutting differences and fluid identities, or celebrated the creativity inspired by such hybridity, as Hall enjoined (1993). Instead, they inverted this meaning of 'difference'. They opposed the dilution of separateness which Hall relished, and turned difference into an essentialist concept to reassert boundaries: the distinctiveness of Englishness must be defended.

As Gilroy (1987: 60) pointed out, the New Right defined 'Englishness', as the hegemonic core of Britishness, through culture. They agreed with the anthropological idea that nations and cultures are historically constituted, not biologically or ontologically given. However, they used this idea not to erode but to reinforce exclusiveness. National identity was defined as a feeling of loyalty to persons of one's own kind (Seidel 1987: 50 quoting Casey). One's own kind, or 'we' was defined as those for whom a list of 'English' activities had pleasant associations or aroused enthusiasm. A quote from T.S. Eliot is used frequently:

[Culture] includes all the characteristic activities and interests of a people: Derby Day, Henley Regatta, Cowes, the twelfth of August, a cup final, the dog races, the pin table, the dart board, Wensleydale cheese, boiled cabbage cut into sections, beetroot in vinegar, nineteenth-century Gothic churches and the music of Elgar (Eliot 1948:31 quoted in Williams 1958: 230 and Casey 1982).

The problem with such a list is not just, as Williams observes (1958: 229-30), that Eliot is purporting to adopt from anthropology the notion of culture as 'a whole way of life' yet is only concerned with 'sport, food, and a little art' - characteristic of 'English leisure'. More to the point, these customs and practices are presented as expressions of homogeneous nationality (Gilroy 1987: 69) whereas, as Seidel points out, this list is decidedly white and Christian and frequently gender and class specific. Tebbit, the former Conservative party chairman, turned pleasure at the sound of leather on willow into a test of national allegiance when he asked, which side would Afro-Caribbeans applaud when the West Indies was touring Britain? The Tebbit Test threatened to make sentiments of attachment into instruments of policy. Hall discerned a danger that the answer to a question of identity would be used as a basis for conferring or withholding rights of citizenship. He was scathing at the idea of allegiance to the vagaries of English batting form being the price for drawing family allowance:

It should not be necessary to look, walk, feel, think, speak exactly like a paid-up member of the buttoned-up, stiff-upper-lipped, fully corsetted 'free-born Englishman' culturally to be accorded either the informal courtesy and respect of civilized social intercourse or the rights of entitlement and citizenship (Hall 1993: 358).

To the New Right, England stands or falls on the hegemony of a particular culture. Margaret Thatcher famously expressed a sense of threat of being 'swamped' by alien cultures that would dilute this exclusive version of Englishness. However, never could members of ethnic minorities be so attached to sentiments and values of Englishness that the New Right would accord them the right to participate in their definition and development. When some British Asians acted in terms of one of the professed core values of Englishness - tolerance and respect for different points of view - by proposing changes to the blasphemy laws during the Rushdie Affair, they soon found that their rights did not extend to shaping those core values. John Patten, the Minister of State at the Home Office, published an open letter through the press to British Muslims 'On Being British'. In a tone Asad finds reminiscent of colonial administrators addressing alien populations under their protection, Patten set out the essential components of Englishness at the core of British identity which he said they should learn. Apart from faith and family which he considers they already share, these are fluent English, understanding of the democratic processes, laws and system of government in Britain and the history that lies behind them - knowledge which few white Britishers could confidently claim to possess (Asad 1993: 242). Others add to such core values of Englishness a canon of literature and respect for authority.

This reformulation of nation in terms of culture rather than race was part of the New Right's attempt in the 1980s and 1990s to redefine racism out of existence. Like Enoch Powell before them, the New Right professed a revulsion for racism. They ridiculed the idea that the mosaic of human groups formed a fixed hierarchy based on grades of biological inferiority. By redefining race as a feeling of loyalty to people 'of one's own kind', they claimed race to be a moral and noble idea. To defend one's 'culture' from attack from people not 'of one's own kind' was legitimate self defence. In a neat inversion or denial of power relations (a form of blaming the victim), writers in the Salisbury Review accused people who sought equality for ethnic minorities of provoking racism by attacking whites. State institutions and 'traditional' values, for example in education, were at the core of the 'culture' which was to be defended. Those multiculturalists and anti-racists who sought to change the workings of state institutions or laws in the interests of treating all citizens more equally, did not recognize the distinction that Tebbit reiterated at the 1997 Conservative party conference, between nationality defined by culture and by political rights: between 'the English' and 'foreigners holding British passports' (Independent 8 October 1997). Multiculturalism, Tebbit claimed, was divisive (ib.). To writers in the Salisbury Review, anti-racists were also subversive, attacking 'our' institutions and values and threatening the order of 'our' nation. As Seidel points out, the use of 'we' and 'our' as a definer of nation drives a clear white wedge between black and anti-racist people, and the rest of the community (1985: 115). Writers in the Salisbury Review adamantly denied racism, yet their framing of nationalism in terms of 'our culture' cued a choice of policy recommendations for ethnic minorities - complete assimilation, retrospective guest worker status, or removal by repatriation - which were in implication and effect racist.

In summary, the New Right appropriated the new ideas of 'culture' from cultural studies, anti-racism and to a lesser extent social anthropology, and engaged in a process of contesting and shifting the meanings of 'culture', 'nation', 'race' and 'difference'. They mobilized 'culture' to reinforce exclusion, using it as a euphemism for renewed racism, with profound implications for public policy and people's lives (Kahn 1995: 6).

= Corporate culture

In the early 1980s, 'culture' became a buzz word in management studies. Deal and Kennedy (1982) discovered 'corporate culture' and Peters and Waterman (1986) claimed that excellent companies were those that had a 'strong' culture. Soon a corporate culture, often equated with a mission statement, had become the sine qua non of any serious organization. This literature attributed the culture concept to anthropology: Geertz (1973), Turner (1974), Bateson (1972) and Douglas (1987) were the most frequently quoted. Both researchers in organizational studies and practising managers looked to anthropological ideas of 'culture' for a metaphor for new forms of organizing in the 'post modern' political economy. There has always been a close relationship between academic research on organizations and the thinking of practising managers, such that organization researchers have played a central role in 'making' organizations (Calas and Smircich 1992: 223). This interchange between academics and practitioners has increased in the 1990s as managers have called on researchers and consultants to provide 'training' to change organizations. It is not unusual for anthropologists researching organizations to find managers asking them for references to publications which would extend their repertoire of metaphors to manage by (Martin 1994) and staff referring to anthropological ideas acquired through training courses.

Companies are using both old and new ideas of 'culture' as tools of management. Some managements emphasize that the company is a clearly demarcated entity, with a boundary against its environment, containing specified groups of people, organized hierarchically, each with a checklist of the behaviours which constitute company culture. For example, McDonalds marks out its space and identity with the golden arches logo and standardized decor and food containers. The core beliefs of the company culture - Quality, Service, Convenience and Value - are drummed into managers at Hamburger University to bond the far-flung franchisees together (Deal and Kennedy 1982:193). Counter staff have to follow a checklist of standardized behaviours in performing each task - right down to when to make eye contact and at what points to smile at a customer during a transaction. In this example, the old idea of 'culture' as a bounded entity with a fixed identity and checklist of characteristics is deployed in a centralized system of command and control.

In other industries, managers are using new ideas of 'culture' as an image for new forms of organizing. This is especially in industries where products are designed, manufactured, distributed and marketed all in different countries. To stay competitive, products are continually redeveloped, and the sites of production, the employees, and relations between them are forever changing. Harvey describes companies 'whose material presence might be no more than a box of contracts, the enumeration of those people who belong, temporarily and for the duration of a particular service, to the network which generates wealth and power for another equally disparate and dispersed group of investors' (1996: 6). Where is 'the organization'? No longer does an architectural monument symbolize the company or contain the workforce. Work is organized through teams or alliances, operating across boundaries and rapidly reforming in new circumstances. Such companies look for staff who are prepared continually to 're-skill' themselves, engage in 'personal reinvention' to cope with risks and new situations, and acquire a 'portfolio' of experiences and contacts to help them 'hop' from job to job (euphemisms for workers on short-term contracts with no job security or career structure, who have periodically to retrain at their own expense and are handling high stress levels). In order to harness workers' knowledge, managers want staff to feel empowered to participate in mixed teams of managers and workers and to put forward new ideas for products or ways of organizing.

In this context, the idea of differently positioned actors being active participants in a process of meaning-making - a version of the new idea of 'culture' - is attractive to managers. The image is associated with rhetoric about empowerment. Workers and managers are 'trained' to make decisions in teams taking everyone's perspective into account. Their attention is also 'trained' on this highly visible, apparently transparent decision making, as if power were dispersed and the organization decentred. Martin's work in the United States (1994) and my work (Wright 1991) and students' dissertations in the UK indicate that workers are often ambivalent, experiencing empowerment in some respects, yet perceiving the gap between corporate rhetoric and the frequent reorganizations, 'shakeouts', 'de-layering' and re-locations, imposed from the top down. Just as the rhetoric of 'organization as culture' highlights participation and empowerment, yet workers see another material reality in the shadows, so Harvey noticed that at the Expo'92 corporations highlighted certain aspects of their 'culture' for the consumer, yet other aspects were obscured. Corporations used new technologies to display transparently and reflexively how 'culture' was constructed through multiple perspectives, connectedness and networking. What they excluded from the representation of this world, where according to Fujitsu 'the only frontiers are in your mind' (Harvey 1996:111), was the organization of relations of production. Similarly, the use of 'culture' in organizational management has a partial effect: it encourages reflexive analysis of the supposedly empowering relations between workers, but does not analyse how these relations are situated within an international organization of capital and power.

This relationship between the highlighted foreground of localized participation and empowerment and the not-completely-obscured political and economic background, is echoed in the management literature. Even among those writers who most avidly espouse 'organization as culture' (e.g. Schein 1991, Smircich 1985) there is a sliding of definitions, from the new idea of 'culture' as a continuous process of meaning making into the old idea of 'culture' as a 'thing' which managers could define from above and act upon in a system of command and control. I will examine how Geertz, the anthropologist most quoted in organizational studies, is used in this literature in order to indicate how this elision occurs and what are its effects.

One phrase from Geertz is used above all in organization studies and by training consultants:

man [sic] is an animal suspended in webs of significance he himself has spun. I take culture to be those webs (1973: 5).

Geertz used the above phrase in an article about a sheep raid in Morocco. His aim was to interpret the different constructions that the actors - a Jewish merchant, Berber tribesmen and a French colonialist - placed on a sequence of events. Each sought to make their interpretation of events definitive as they 'tripped over' one another's purposes: pursuing trade, defending honour and establishing dominance. The three actors were in unequal relations of power and had different personal abilities to impose their meanings on events. Geertz makes clear that he was studying the interaction between three ways of making significance from one sequence of events. He specifically was not trying to isolate the elements of 'a culture', nor specify the relations between those elements, nor characterize the whole as a system organized around core symbols (1973: 17). He was not suggesting that all three actors were caught in the same way in one web.

Geertz used this sequence of events to illustrate how a merchant and dissident tribes challenged yet succumbed to French dominance at the early stages of colonialism. It is appropriate for organizational researchers to refer to this article when looking to anthropology for new ways of analysing 'organization as culture' in a period of equally momentous global economic and political change. However, in this literature an anthropological focus on contestation and power is absent. For example, Smircich (1983) starts off in language precursive of Street's (1993, quoted above) when she suggests that that 'culture is something an organization has', rather than 'something an organization is' (1983: 347). To advance this view, she describes Geertz's approach which, she accurately concludes, enables organizational analysts to problematize the concept of organization 'for the researcher seeks to examine the basic processes by which groups of people come to share interpretations and meanings for experience that allow the possibility of organized activity' (1983: 351). At this point, there is a sliding from new to old ideas of culture. She claims Geertz's focus has much in common with organizational leaders', as both are concerned with 'how to create and maintain a sense of organization, and how to achieve common interpretations of situations so that coordinated action is possible ... leadership can best be understood as the management of meaning and the shaping of interpretations' (ib.). Geertz has been appropriated as a tool of management and his idea of 'culture' which had some of the elements of contestation and process developed by the new ideas of 'culture', has been converted into the old idea of 'culture' as an entity to be acted on from above. Where ideas of 'culture' are being used to manage self-motivated, flexibly-networking and team-forming staff through ideas of empowerment, it is even more important that analysts should not, as organizational studies have tended to do in the past, take a manager's perspective on workers as the objects of study (Wright 1994). The focus should be on how managers are deploying both old and new ideas of 'culture' in order to gain workers' active participation in new ways of organizing production, profit and power.

Culture and development

In my third case, 'culture' is entering a new domain, overseas development, with the help of anthropologists. Two examples are used, which both refer to old ideas of 'culture'. In the first example, an international agency, UNESCO, in its vision of a new ethical world order, maps out a world made of 'cultures' as discrete entities, without engaging with the issue of contestation over the power to define. In contrast, in the second example Kayapo leaders have used ethnographic film to assert their own definition of their 'culture' and used the strategies others have used against them to challenge the processes that have marginalized them.

UNESCO's (1995) report Our Creative Diversity marks the culmination of the UN decade for culture and development. This was an opportunity for anthropologists to have an overt influence on the use of the concept 'culture' and several world famous anthropologists contributed to its definition.5 The report argues for two definitions of 'culture'. First, it takes up the argument made by development anthropologists that 'culture' is not just one domain of life (like economics, politics, religion) but is 'constructive, constitutive and creative' of all aspects of life including the economy and development. Second, it argues that the world is made up of discrete 'cultures' or peoples. The neglect of 'culture' in the first sense within 'cultures' in the second sense has caused development efforts to fail (1995: 7). Frustrated expectations coupled with globalization, and the collapse of the bipolar world order (1995: 9, 28), it is argued, have led to confrontations between narrow group identities over scarce resources (1995: 9) which have been manipulated into violence (1995: 16). Whereas failed development gives rise to this destructive aspect of cultural identities of 'peoples', successful development would result in a flourishing of culture, creativity and progress.

This argument rests on a particular view of cultural diversity. An introductory quote from Marshall Sahlins defines culture as 'the total and distinctive way of life of a people or a society' (1994 quoted in UNESCO 1995: 21). This old view of 'culture' is supported by a paper by L<130>vi-Strauss (initially written for UNESCO in 1952 and revised in 1973),6 from which the title of the report is drawn. Levi-Strauss put forward what Eriksen (1997) calls an archipelago vision of the world as made up of 'peoples' each with a radically different 'culture' like a string of separate islands (the view contested by Merry, above). In the report, sometimes a 'people' is equated with a country, although it is also said that the world consists of 10,000 distinct societies in 200 states (1995: 16). Unfortunately, according to the report, people are mixing as never before (1995: 9). Instead, their distinctiveness should be encouraged, as it is by looking across boundaries between distinct cultures that people gain ideas for alternative ways of living. The report's recipe for creativity, experimentation, innovation and the dynamic of progress is a diversity of distinct entities with clear boundaries (1995: 15). Human civilization depends on creative diversity.

L<130>vi-Strauss has provided UNESCO with a map of a flat world. The mosaic of cultures is reminiscent of 1930s social anthropology. It misses the dimension of 'culture' as a process of contestation over the power to define organizing concepts - including the meaning of 'culture' itself. In the report an unidentified voice does the defining and disguises or disclaims its own power as common sense. It is envisaged that in this plural world, nation-states, rather than trying to create nation-wide cultural homogeneity, should encourage diverse ethnic groups within their borders to contribute to a civic community with shared values. Similarly, cultural diversity in the world should be protected by a code of global ethics, on which the report thinks the world can reach consensus. In setting out the parameters of this global ethical code the undefined voice of the report begins to make value judgments. Only 'cultures' that have 'tolerant values' (whose idea of tolerance?) would be respected and protected by the global code. Of course, 'repulsive' (in whose view?) cultural practices should be condemned (1995: 54). A reported criticism of human rights for fostering an individualism which is alien to non-western values, receives the reply 'Human rights is not unduly individualistic [by whose criteria] - it is just an appropriate way to regard all humans as equal' (1995: 41). UNESCO's vision of a code of global ethics to order a plural world rests on a contradiction between respecting all cultural values, and making value judgments about acceptable and unacceptable diversity.

In contrast to UNESCO's top-down grand plan for a pluralism of bounded cultures, even these old ideas of 'culture' work very differently when their definition is in the hands of indigenous people. Wagner (1975) argued that in the very act of fieldwork anthropologists 'invent' a 'culture' (in the old sense) for a people. Anthropologists plunge into situations which are beyond their interpersonal and practical competence. To cope with this, they encourage themselves by thinking that they are dealing with a 'thing' and they can learn how it 'works'. Some people in the host society gain insight into the anthropologist's perspective - often whilst trying to control and domesticate her or him - and for the first time perceive their daily life as a thing that works in patterned ways. The anthropologist proceeds as if what is being studied is 'a culture'. In the process, what people had hitherto experienced as an embedded way of life becomes objectified and verbalized - in Wagner's terms, invented - as 'culture'.

Terence Turner provides an example from his fieldwork among the Kayapo in Brazil. Twenty-five years ago, he found 700 of the 800 members of one group had died of disease. Missionaries provided medicine in exchange for the Kayapo's adopting western clothes, building their village along a street, and suppressing their ceremonials. A state organization controlled their trade and communication with the outside, and embezzled their cash from the nut crop. The Kayapo felt dependent and in a situation over which they had no control.

Turner saw his role as an anthropologist as 'uncovering the authentic social and cultural system beneath the corrosive underlay' (1991: 291). He found his authentic culture in the surviving social and ceremonial rituals which, to him, reproduced Kayapo as social persons in a moral universe. The Kayapo did not see it like that: it was just the way they did things. They did not have a concept through which to objectify and label their everyday life as a 'culture'. He argued that they needed such a concept to deal with their situation: to give them an identity and distinguish themselves as a 'culture' on a par with other indigenous people and vis-<133>-vis the dominant national society in an inter-ethnic state system.

Turner says that the Kayapo were visited by many anthropologists 25 years ago who respectfully sought to learn and record Kayapo 'culture'. He says that anthropologists were innocent of the political implications of their participant observation. However, the Kayapo realized that what missionaries and state administrators used as justification for subordination and exploitation, another set of Westerners valued highly. 'Culture', which had seemed an impediment, now appeared as a resource to negotiate their co-existence with the dominant society.

After a Disappearing World documentary was made, the Kayapo sought further documentaries so as to reach the sympathetic elements in the west. When they arranged to meet the Brazilian government to oppose the Altamira dam, they choreographed themselves for the western media in order to gain support of the western audience and add pressure on the government. Gone were the shorts, T-shirts and haircuts that had appeased the missionaries; with men's bare chests, body ornament and long ritual dances, the Kayapo performed their 'culture' as a strategy in their increasingly confident opposition to the state.

The Kayapo were exceptional in the Amazon area in not only obtaining funding for their own video cameras and training for their film crews, but also in surviving in sufficient numbers and having the economic and physical strength to resist their oppression. Turner says that by the 1990s the Kayapo had obtained videos, radios, pharmacies, vehicles, drivers and mechanics, an aeroplane to patrol their land, and even their own missionaries. Supported by machinery hitherto associated with dependency, these now-consummate ethnic politicians had learnt to objectify their everyday life as 'culture' (in the old sense) and use it as a resource in negotiations with government and international agencies.

Kayapo politicians seem to have been fully aware of the constructedness of 'culture'. They seem to have dealt with contests among themselves over the power to define. They exploited the way the old idea of 'culture' masks power differentials within groups and they borrowed western filmic tropes of realism and authenticity which deflect attention from questions like how is author-ity constructed, who controls the technology, who holds the camera, who is depicted as active and who as passive and marginal? (Moore, R. 1994). They presented themselves as a homogeneous and bounded group, 'the Kayapo', so successfully that even the anthropologist, who should have noticed the process by which they contested and constructed their communal 'authentic voice', does not mention it.

They defined 'culture' for themselves and used it to set the terms of their relations with the 'outside world'. In a history spanning forty years, missionaries, government officials, the Kayapo, anthropologists, international agencies and non government agencies had all competed for the power to define a key concept, 'culture'. Missionaries and government agencies initially had used the concept to define an entity that could be acted upon, producing disempowerment and dependency among the Kayapo. The Kayapo strategy to wrest control of this concept from missionaries and government officials and turn it against them was part of a struggle not just for identity but for physical, economic and political survival.

Turner shows that 'culture' can be used to very different effect, depending on who is doing the defining. The UNESCO Report, Our Creative Diversity, seems to be seeking the positive outcomes from the autonomous definition of culture evident among the Kayapo. However it neglects to see7 that the flows of creativity that it associates with vigorous 'cultures' is a product of continuous assertion of the power to define in a political process involving local, national and international actors. This political dimension of meaning making, well understood by Kayapo politicians, is a dynamic which is absent from the UNESCO report.

Conclusion

I have distinguished between two sets of ideas about culture in anthropology: an older set of ideas which equates 'a culture' with 'a people' which can be delineated with a boundary and a checklist of characteristics; and new meanings of 'culture', as not a 'thing' but a political process of contestation over the power to define key concepts, including that of 'culture' itself. Earlier this century, anthropologists used the old ideas of 'culture', the construction of an objective classification of people, as a strategy for appearing outside of politics. Now anthropologists who adopt new ideas of 'culture' are compelled to recognize that academic definitions of 'culture' are themselves positioned and political and therefore a resource for anthropologists and others to use in establishing or challenging processes of domination and marginalization.

'Culture' in both its old and new senses has been introduced into many new domains in the 1980s and 1990s, including cultural racism and multiculturalism, corporate culture and culture and development. Sometimes anthropologists have been directly involved, as in preparing the UNESCO report or filming the Kayapo. Sometimes politicians or managers have appealed to 'anthropological ideas of culture' for legitimacy. Either way, anthropologists are implicated in the politicization of 'culture'.

In the political strategies explored in this paper, actors have deployed 'culture' in a number of different ways and with different material effects. British New Right politicians have appropriated the new idea of 'culture', turned it into a euphemism for race, and mobilized it to reinforce exclusion and marginalization. In 'corporate culture', old and new ideas of 'culture' have been used as tools of management, often sliding from one to the other, in strategies to harness workers' active participation in a process of meaning-making which managers ultimately reserve the power to define and control. The Kayapo provide an example of indigenous politicians asserting their own definition of 'culture' and using it to set the terms of their relations with the outside world. They were consciously using old ideas of 'culture' with an appreciation of the politics of its construction. The voice of Kayapo politicians, presenting an apparently consensual 'authentic culture' of 'the Kayapo', has succeeded in being heard in national and international forums. The UNESCO report aspired for 'cultures' in the old sense to have the creativity and dynamism of the Kayapo. However, the report did not confront the central issue in the Kayapo case: that they were engaged in a struggle with the state and international agencies over the power to define. Instead, both the UNESCO report and the British New Right's cultural racism deploy a disembodied voice, 'we', to authorize a top down definition of 'culture' as if it were common sense or 'natural'. This strategy, like the old anthropological strategy of objectification, tries to mask or erase the politicization of culture.

It is disappointing that the opportunity provided by the UNESCO report, for anthropologists to make an impact on the political use of 'culture' in ways which would benefit the disadvantaged and marginalized, was not used more effectively. If we aim to influence local, national and international processes by which people are impoverished and disempowered, it behoves us to reflect on our own anthropological analyses of how politicians, policy advisers and decision-makers are deploying old and new ideas of 'culture'. We might learn from our analyses of the political strategies of others how to intervene more effectively ourselves in the politicization of 'culture'. In the context of recent laments about anthropology's loss of authority and diminishing relevance to the study of contemporary cultural processes (due in part to the advance of cultural studies, GDAT 1996), such reflection might also help restore a much needed critical edge to the discipline.

Bibliography

Asad, Talal (ed.) 1973. Anthropology and the Colonial Encounter. London: Ithaca P.

----1979. Anthropology and the analysis of ideology. Man 14: 607-27.

----1993. 'Multiculturalism and British identity in the wake of the Rushdie Affair' in Genealogies of Religion. Baltimore: John Hopkins U.P.

Bateson, G. 1972. Steps to an Ecology of Mind. New York: Chandler Publishing Co.

Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge: Polity P.

Calas, M. B. and Smircich, L. 1992. 'Using the 'F' word: feminist theories and the social consequences of organizational research' in Mills, A. J. and Tancred, P. (eds) Gendering Organizational Analysis. London: Sage.

Casey, J. 1982. One nation: the politics of race. Salisbury Review 1 (Autumn): 23-8.

Cohen, Abner (ed.) 1974. Urban Ethnicity. ASA Monograph 12, London: Tavistock.

Comaroff, John and Jean. 1992. Ethnography and the Historical Imagination. Boulder: Westview P.

Deal, T. and Kennedy, A. 1982. Corporate Cultures. The Rites and Rituals of Corporate Life. Harmondsworth: Penguin.

Douglas, Mary 1987. How Institutions Think. London: Routledge and Kegan Paul.

Eliot, T. S. 1948. Notes Towards the Definition of Culture. London: Faber

Eriksen, Thomas Hylland. 1997. 'Our Creative Diversity' paper to conference on 'Culture and Rights' Sussex University, 15-16 July.

GDAT 1996 'Cultural Studies will be the death of anthropology' Manchester: Group for Debates in Anthropological Theory.

Geertz, Clifford. 1973. The Interpretation of Cultures. New York: Basic Books.

Gilroy, Paul. 1987. There Ain't No Black in the Union Jack. London: Hutchinson.

Gough, Kathleen. 1968. 'New proposals for anthropologists' Current Anthropology 9: 403-7.

Hall, Stuart. 1993. 'Culture, community, nation' Cultural Studies 7: 349-63.

Harvey, Penelope. 1996. Hybrids of Modernity. London: Routledge.

The Independent. 1997. Tebbit questions the loyalties of 'two-nation' immigrants. 8 October.

Kahn, Joel S. 1995. Culture, Multiculture, Postculture. London: Sage.

Keesing, Roger. 1974. Theories of culture. Annual Review of Anthropology 3: 73-97

King, Desmond. 1987. The New Right: Politics, Markets and Citizenship. London: Macmillan.

Kroeber, A. L. and Kluckhohn, Clyde. 1952. Culture. A Critical Review of Concepts and Definitions. Cambridge, MA.: Papers of the Peabody Museum XLVII:1.

Levi-Strauss, Claude. 1973 [in English 1997]. 'Race and history' in Structural Anthropology II London: Allen Lane.

Macdonald, Sharon (ed.) 1993. Inside European Identities. Oxford: Berg.

Martin, Emily. 1994. Flexible Bodies. Boston: Beacon P.

----1997. 'Managing Americans: policy and changes in the meanings of work and the self' in Shore, Cris and Wright, Susan (eds) Anthropology of Policy. London: Rouledge.

Merry, Sally Engle. 1997. Law, culture and cultural appropriation. Yale Journal of Law and the Humanities (forthcoming).

Moore, Rachel. 1994. 'Marketing alterity' in Lucien Taylor (ed.) Visualizing Theory. London: Routledge.

Morley, David and Chen, Kuan-Hsing. 1996. Stuart Hall. Critical Dialogues in Cultural Studies. London: Routledge.

Ortner, Sherry. 1984. Theory in anthropology since the sixties. Comparative Studies in Society and History 26 (1): 126-66.

Peters, T. and Waterman, R. 1986. In Search of Excellence. Lessons from America's Best-Run Companies. New York: Harper and Row.

Pondy, L. and Mitroff, I. 1979. 'Beyond open system models of organization' in Cummings, L. and Staw, B. (eds) Research in Organizational Behaviour. Greenwich CT.: JAI P.

Reinhold, Sue. 1993. Local conflict and ideological struggle: 'positive images' and Section 28. Unpublished D.Phil. thesis, U. of Sussex.

Sahlins, Marshall. 1994. 'A brief cultural history of "culture" ', paper prepared for the World Commission on Culture and Development, August.

Said, Edward. 1978. Orientalism. Harmondsworth: Penguin.

Schein, E. 1991 'What is culture?' in P. Frost, L. Moore, M. Louis, C. Lundberg and J. Martin (eds) Reframing Organizational Culture. London: Sage.

Seidel, Gill. 1985. 'Culture, nation and "race" in the British and French New Right' in Levitas, Ruth (ed.) The Ideology of the New Right . Oxford: Polity.

----1987. 'The white discursive order: the British New Right's discourse on cultural racism with particular reference to the Salisbury Review' in Zavala, Iris, van Dijk, Teun and Diaz-Diocaretz, Miriam (eds) Approaches to Discourse, Poetics and Psychiatry. Amsterdam: John Benjamins.

Shore, Cris and Wright, Susan (eds) 1997. Anthropology of Policy. Critical Perspectives on Governance and Power. London: Routledge.

Smircich, L. 1983. 'Concepts of culture and organizational analysis' Administrative Science Quarterly 28 (3): 339-58.

----. 1985. 'Is the concept of culture a paradigm for understanding organizations and ourselves?' in Frost, Peter, Moore, Larry, Louis, Maryl Rees and Lundberg, Craig (eds) Organizational Culture. London: Sage.

Stocking, George. 1974. The Shaping of American Anthropology, 1883-1911: A Franz Boas Reader. New York: Basic Books.

Street, Brian. 1993. 'Culture is a verb: anthropological aspects of language and cultural process' in Graddol, D., Thompson, L. and Byram, M. (eds) Language and Culture. Clevedon, Avon: British Association for Applied Linguistics in association with Multilingual Matters.

Turner, Terence. 1991. 'Representing, resisting, rethinking' in Stocking, George (ed.) Colonial Situations. Madison, Wisconsin: U. of Wisconsin P.

Turner, Victor. 1974. Dramas, Fields and Metaphors. Ithaca, NY: Cornell U. P.

Tylor, Edward B. 1871. Primitive Culture. New York: Harper.

UNESCO. 1995. Our Creative Diversity. Report of the World Commission on Culture and Diversity, Paris: UNESCO Publishing.

Wagner, Roy. 1975. The Invention of Culture. Chicago: U. of Chicago P.

Williams, Raymond. 1958. Culture and Society, 1780-1950. Harmondsworth: Penguin.

----1976. Keywords. London: Fontana.

Wright, Susan. 1991. 'Evaluation of the Unemployment Strategy'. Middlesbrough: Cleveland County Council.

----1994. 'Culture in anthropology and organizational studies' in S. Wright (ed.) Anthropology of Organizations. London: Routledge.

----1997. 'Culture in development', paper to Social Development Advisers' International Network, Overseas Development Administration, London, 28 January.

Return to editorials and articles