8

Creolization in Anthropological Theory and in Mauritius
Thomas Hylland Eriksen

A great amount of intellectual energy has been invested in cultural mixing during the last decades. Reacting against an idea of boundedness, internal homogeneity, and stability that has been associated with mainstream twentieth-century anthropology, hundreds—possibly thousands—of anthropologists have tried to redefine, reform, revolutionize, or even relinquish that abhorred “C” word—”culture.” The range of engagement is suggested in the apparent congruence between postmodernist American anthropologists (for example, Clifford & Marcus 1986) and their now classic critique of the Geertzian notion of cultural integration, and the older European critique of the structural-functionalist idea of social integration, which was led by people such as Barth (1966), whose rationalism and naturalism is everything but postmodernist. In both cases, presuppositions of integrated wholes, cultures or social structures, have been debunked. From being a discipline concentrating its efforts on understanding nonliterate societies, often implicitly positing the uncontaminated aborigine as its hero, anthropology increasingly studies cultural impurity and hybridity, and the dominant normative discourse in the field has shifted from defending the cultural rights of small peoples to combating essentialism and reifying identity politics. While this development has been important and necessary for a variety of reasons, the perspectives

154 CREOLIZATION developed risk being one-sided and inadequate.1 A focus on mixing and flows that does not take continuity and boundedness into account ends up undermining its own social theory: Every social scientist knows that cultural meaning is being reproduced and transmitted between generations and that natives do classify and create boundaries, often amidst powerful tendencies of cultural mixing. The views of culture as continuum and variation as endemic are uncomfortably uncontroversial today (see Fox & King, 2002, for a state-of-the-art overview). These views are problematic because the processes associated with socialization and institutional continuity in societies presuppose a wide range of shared understandings, which are often implicit and do not enter the arena of identity politics, and because groups remain bounded in a variety of different ways. It is a trivial fact that variation within any group is considerable, and cultural flows across boundaries ensure that mixing, in the contemporary world, is everywhere. However, the impression sometimes given that “everything” seems to be in continuous flux, that an infinity of opportunities seems to be open and that no groups, cultural identities, or ethnic categories are fixed, is due to a conflation of discrete phenomena. First: Strong identities and fixed boundaries do not preclude cultural mixing. Ethnic variation may well exist without significant cultural variation. Processes of cultural mixing say nothing about group identities and degrees of boundedness. Second: Fluid identities, conversely, do not preclude cultural stability or continuity. Cultural variation can exist without ethnic variation or other kinds of strong group boundaries. Culture is caused by varying degrees of shared meaning, whereas group identities result from clear, if disputed, social boundaries. Third: The political usages of cultural symbols do not mean that the people in question do not necessarily have anything in common. (Being paranoid is no guarantee that nobody is after you.) Historiography, it has been shown time and again, is necessarily a selective and biased discipline simply because far too many events have taken place in the past for any historian to give all of them a fair treatment. Yet its slanted narratives may become self-fulfilling prophecies in that they give people a shared frame of reference. Besides, the people described by nationalist historians or ideologists of group boundedness may not have that in

Different parts of this chapter have been presented at the Transnational Communities Programme, Oxford University in 1999, at the Creolization workshop at University College London in 2002, and at the Department of Social Anthropology, University of Cape Town in 2003. Thanks are due to the convenors and participants at these events, and in particular Charles Stewart, Zimitri Erasmus, Steven Vertovec, Laurent Medea, and Rose Boswell, for helping me to clarify the argument.

Creolization in Anthropological Theory and in Mauritius

155

common which their ideologists ascribe to them, but they may have other cultural elements in common, such as shared jokes and ideas about kin relatedness. What has interested writers on cultural mixing are the situations in which these frames of references do not function; where they are contested, nonexistent, or are being continuously rebuilt. But it may just happen to be the case, in other words, that ethnic boundaries coincide with certain cultural ones. (Moreover, the fact that something is socially constructed does not, of course, imply that it is unreal.) This means that the ambiguous grey zones, which can be located in the space between categories and boundaries under pressure, are privileged sites for studying the interplay between culture and identity. This is not because all boundaries eventually disappear, but because they are made visible through their negotiation and renegotiation, transcendence, transformation, and reframing. A world without social boundaries (a neoliberalist’s dream?) is sociologically unthinkable.

THE TERM “CREOLIZATION”
One of the more popular concepts used to deal with the increased complexity of the empirical fields now studied by anthropologists, is creolization (Drummond 1980; Hannerz 1992, 1996). Creolization is often used merely as a synonym for mixing or hybridity, but from the discussion below, it will become apparent that both the historical origins of the term and its contemporary usage in societies containing self-designating creoles suggest that a more restricted use of the term might be both necessary and analytically helpful. “Creolization,” as the term is used by some anthropologists, is an analogy taken from linguistics. This discipline in turn took the term from a particular aspect of colonialism, namely the uprooting and displacement of large numbers of people in colonial plantation economies. Both in the Caribbean basin and in the Indian Ocean, certain (or all) groups who contributed to this economy during slavery were described as creoles. Originally, a criollo meant a Spaniard born in the New World (as opposed to peninsulares); today, a similar usage is current in La Réunion, where everybody born in the island, regardless of skin color, is seen as créole, as opposed to the zoréoles who were born in metropolitan France. In Trinidad, the term “creole” is sometimes used to designate all Trinidadians except those of Asian origin. In Suriname, a creole is a person of African origin, whereas in neighboring French Guyana a creole is someone who has adopted a European way of life. In spite of the differences, there are resemblances between the conceptualizations of the creole. Creoles are uprooted, they belong to a New World, and are contrasted with that which is old, deep, and rooted.

156 CREOLIZATION A question often posed by people unfamiliar with these variations is “What is really a Creole?” They may have encountered the term in connection with food or architecture from Louisiana, languages in the Caribbean, or people in the Indian Ocean. The standard response is that whereas vernacular uses of the term “creole” vary, there exist accurate definitions of creole languages in linguistics and of cultural creolization in anthropology. There are nevertheless similarities, although there is no one-toone relationship, between the ethnic groups described locally (emically) as creoles and the phenomena classified as creole or creolised in the academic literature. A motto for the present exploration could therefore be be Gregory Bateson’s admonition that if one uses creative analogies, one ought to go back to the field from which the analogy was taken to investigate its internal logic. As he puts it: “the moment I begin to work out the analogy, I am brought up against the rigid formulations which have been devised in the field from which I borrow the analogy” (Bateson 2000b:75). In other words, it is worthwhile to take a close look at Mauritian creoledom to see if it could shed light on the theoretical applications of the term “creolization.”

MAURITIUS AND ITS CREOLES
Mauritius, located near the Tropic of Capricorn eight hundred kilometers east of Madagascar, is a crowded, bustling, complex, democratic, hierarchical, and, of recent, prosperous island. Its total population is slightly over a million, giving it a population density of about 550 inhabitants per square kilometer. All the inhabitants are descendants of immigrants who have arrived during the last three centuries, from France, China, Africa, Madagascar, and India. About half the population are Hindus, but they are subdivided into North Indians (“Hindi speaking”), Tamils, Telugus, and Marathis, who function politically and culturally as separate ethnic groups. The largest ethnic group are North Indian Hindus, constituting about 40 percent of the population. Seventeen percent are Muslims of Indian descent, around 28 percent are creoles or nonwhite Catholics of African, Malagasy, or mixed descent, about 7 percent are Tamils, and 3 percent are Chinese and less than 2 percent Franco-Mauritians. It may be said that all the cultures of all the ethnic groups in Mauritius have been culturally creolized—uprooted and adapted to local circumstances—to a greater or lesser extent. For example, the Bhojpuri spoken by many of the Indo-Mauritians has been so strongly influenced by other languages that it is scarcely intelligible to modern Bhojpurispeakers in Bihar, and the Franco-Mauritians—like other Mauritians— eat spicy curries and lots of rice. Nearly every Mauritian speaks the local French-based creole language (Kreol or Morisyen) fluently, and it is the

Creolization in Anthropological Theory and in Mauritius

157

mother tongue of a substantial majority. Regarding lifestyle, consumption, and way of life in general, it is easy to demonstrate the effects of mutual influence among the ethnic groups that make up the Mauritian population, as well as cultural influence from the outside world—not merely from the West but also from India and East Asia. In spite of obvious cultural creolization evident throughout Mauritian society, it is chiefly the Mauritians of African and/or Malagasy descent who are classified locally as Creoles. Already in the 1850s, the Rev. Patrick Beaton entitled his book on Mauritius Creoles and Coolies (Beaton 1977), contrasting the two major groups of African and Indian descent, respectively. The ancestors of Mauritian creoles were slaves from different parts of Africa and Madagascar, brought there between 1715 (the beginning of French colonization) and 1810 (when the slave trade was banned). As in other plantation colonies based on slavery, slave owners in Île-deFrance (as Mauritius was called during French rule) mixed individuals from different ethnic groups, dissolving family structures and political organization. As a result, in a given compound, there were few shared collective cultural resources; no shared language, no shared kinship structure, cosmology, or traditional system of social organization that might have been transplanted and eventually reproduced. Thus the degree of cultural continuity in the slave groups was by default limited. As in similar setups in the Caribbean, a creole language developed quickly, using French vocabulary, a modified pronounciation and a simplified grammar. In Bernardin de St. Pierre’s travel book from 1773, Voyage à l’Île de France, fragments of the so-called patois spoken by the slaves are cited in a few places, and they appear to be similar to the creole spoken in Mauritius today. Some religious beliefs and practices have survived, in modified forms, although the slaves were converted to Christianity and their descendants are Catholics. However, the most significant “African survival” is in music and dance, where the séga, which has obvious African forebears, has attained a status as a national music of Mauritius. In other words, like groups known as creoles in other parts of the world, Mauritian Creoles have a history of uprootedness, and the connection with their places of origin was severed on arrival in the colony. This entailed the urgent necessity of crafting new cultural and social forms under conditions of extreme hardship.

THE NON-CREOLES
Let me now briefly contrast the situation of these Creoles with that of other Mauritian groups. The Sino-Mauritians, the most recent arrivals

158 CREOLIZATION to the island (most arrived during the first half of the twentieth century, a few after the Chinese Revolution), have changed their religion (to Catholicism) and to a great extent their language (to Kreol) but have retained both their kinship organization, many aspects of their material culture and important rituals, as well as active links with relatives in east Asia. Regarding the Franco-Mauritians, most of whom are descendants of Frenchmen who arrived in Mauritius in the eighteenth century, their kinship links with Europe have in most cases waned, yet they have always been the economically and culturally dominant group in the island. Even when the British conquered Mauritius during the Napoleonic wars, the Franco-Mauritians were promised the right to retain their customs, language, and religion. Regarding the ethnic groups originating from the Indian subcontinent, the conditions of their arrival could be said to resemble those of slavery (cf. Hugh Tinker’s influential book on the indentureship system, A New Form of Slavery, 1974). They were brought from Madras, Calcutta, and Bombay from the 1840s onward in order to replace the liberated slaves as plantation labor. When they arrived in the colony it turned out that they were interned in camps with restricted freedom of movement, and although they were nominally free, their situation may well justify the term “a new form of slavery.” However, relevant differences existed between slaves and indentured laborers. These people came from various parts of India: Bihar, eastern Uttar Pradesh, Andhra Pradesh, Tamil Nadu, and Maharashtra. They transplanted their beliefs and ritual practices, languages, kinship structures, food habits, a simplified caste structure, and rudiments of their political organization (including panchayats, village councils) to their new home, and even—in the case of the north Indian Hindus—invented a myth to the effect that the small, secluded lake Grand Bassin in southern Mauritius contained the holy waters of the Ganges. (Today, the largest Hindu festival outside India is said to be the annual Maha Shivaratree pilgrimage to Grand Bassin.) Soon after the arrival of the first batches of indentured laborers, temples were built, and as the indentureship period ended, they developed rural villages, replicating in no small degree the Indian countryside. In spite of this continuity, several aspects of Indo-Mauritian culture and social life result from cultural creolization—from aesthetic details such as clothing to more fundamental issues such as the omnipresence of the monetary economy and wage work; the jajmani system, important in rural India, vanished, and the caste system thus lost one of its most important functions. (There are only four operative castes among Hindus in Mauritius, and although politics, marriage, and interpersonal networks have a caste aspect, it is less important than in India.)

Creolization in Anthropological Theory and in Mauritius

159

Within the Indo-Mauritian communities, notably North Indian Hindus, North Indian Muslims, and Tamils, there are ongoing controversies regarding “cultural purity,” the relationship to Kreol and Western languages (English and French) and Western culture in general, questions of “roots,” and so on. More relevant in the present context are the different implications, at the level of political and cultural identity, of the histories of Creoles on the one hand, and the non-Creole Mauritians on the other. Non-Creole Mauritians have the opportunity to draw on enormous non-Mauritian cultural traditions in order to make political or existential statements about themselves. Although they, like the Creoles, can be seen as diasporic populations—uprooted, exiled, and homeless—their genealogical and cultural links with their ancestral country enable them to construe their past as an unbroken and continuous narrative that harks back to the mists of prehistory, and, even more importantly in an age of identity politics, their identity can be metonymically linked with a prestigious civilization—Chinese, Indian, Islamic, or European. These links are being reactivated in several ways, increasingly so since the early 1990s, which was not only a period of revitalized identity politics but also one of considerable prosperity in Mauritius. For example, the country has seen the rise of a moderately successful political Hindu movement along the lines of the Bharatiya Janata Party (BJP) in India itself; the Muslims, generally opting for an Islamic identity rather than an Indian (or Pakistani) one, have for years groomed their relationship with the Arab world and Mauritian Tamils, whose organisations have recently been very active in identity politics at the national level, employing architects and craftsmen from Tamil Nadu to build a spectacular Tamil temple near Grand’ Baie in northern Mauritius at the turn of the century. While Franco-Mauritians have always traveled regularly to France, IndoMauritians also increasingly travel to ancestral countries or spiritual homelands (in the case of Muslims), sometimes in search of their ancestral village.

LE MALAISE CRÉOLE
Such practices, which emphasise the organic connection between diaspora and metropole, simultaneously serve to strengthen group cohesion within Mauritius and to counteract a feeling of uprootedness or créolité. Regarding the Creoles, they have few opportunities to match the efforts of the other groups. To begin with, few Creoles recognize and cherish their African origins. Some intellectuals have tried to redefine the Creoles as “Afro-Mauritians” (Benoît 1984 see also Boswell 2003) along the same

160 CREOLIZATION lines as “Sino-Mauritians,” “Indo-Mauritians,” and so on , but with limited success. A handful of politically conscious parents have given their children Ashanti or Yoruba names, but French first names predominate massively. Direct contact between Creoles and Africans is limited, and although the booming Mauritian textile industry is now investing in Madagascar, the investors belong to non-Creole ethnic groups. Even if the effort to provide the Creoles with an African identity had been successful, it would have been difficult to give it a substantial content. Since the slaves came from different parts of West and East Africa, no Creole is able to point out where his or her ancestors came from. The Creoles lack a tangible precolonial past and are unable to draw on close links with a major civilization in their identity politics. Throughout the 1990s, one of the most pressing public issues in Mauritius was the cluster of social problems called le malaise créole, the “Creole ailment” (see Boswell 2003 for a systematic treatment; see also Eriksen 2004). Social change has been rapid in Mauritius since the mid-1980s, leading to a significant improvement of standards of living and educational achievements. In this process, it has become clear that the Creoles have been lagging behind. It has also been argued that the reasons for this can be found in Creole culture, which places a great emphasis on individualism, freedom, and consumption, and in Creole social organization, which lacks the strong kinship obligations characteristic of the other groups. These accounts are one-sided in that they fail to consider, among other things, the connections among Hindu political hegemony, kinship obligations, nepotism, and Hindu dominance in the state sector. Yet it must be conceded that their description of Creole values and way of life are not entirely inaccurate. As I have shown earlier (Eriksen 1988), moreover, that the powerful individualism among the Creoles has nothing to do with “African roots” but can be traced back to the social conditions under slavery, when family and kinship systems were destroyed, individual freedom emerged as the paramount value, and social relations were individualized and became contractual in nature. In the contemporary context of a democratic, competitive capitalist society, the Creoles are at a clear disadvantage because of their loose social organization and their relative lack of symbolic capital in Mauritian identity politics.

TWO WAYS OF BEING CREOLE
Some of the individuals who might seem the very embodiments of cultural creolization (genetically mixed, culturally familiar with several traditions, and frequently rejoicing in the cultural “mosaic” of Mauritius) are not defined as Creoles locally. However, there are also tendencies to the

Creolization in Anthropological Theory and in Mauritius

161

effect that local classifications more closely approach the analytical concept of creolization as it has been developed, for example, by Hannerz. A good example is offered by a ten-year old schoolboy of my acquaintance, who lives in a middle-class neighborhood in a Mauritian town. His father is a Tamil and his mother is a north Indian Hindu; he is, in other words, the offspring of a mixed marriage, although it would be considered “less mixed” than, say, a Hindu-Creole marriage. Since his father is a Tamil, the boy is expected to take Tamil lessons at school. This is not a language that has ever been spoken in his home. The way of life in his family could be described as very creolized; with regard to food, clothes, interior decorating, ritual, and music, both Indian and European influences are clearly present and are routinely and unquestioningly mixed. Not surprisingly, the boy did not want to take Tamil lessons and argued that he considered himself a Creole since his first language was Kreol. The parents, telling me about the son’s predicament, had no objections to his line of reasoning; to them, ethnic labels were unimportant and ancestral languages irrelevant; what mattered was the quality of the boy’s education. The boy, like many other Mauritians of non-African origin—and not just children of mixed marriages—saw himself as a Creole by virtue of speaking Kreol and not belonging to one of the distinct Asian or European communities in the island. This kind of process, which has been spoken of as “creolization” a few times in Mauritius, corresponds well to the anthropological concept of creolization, but not with standard ethnic classification in Mauritius, where Creole still means “a person of African/Malagasy or mixed descent.” A relevant aspect of Creole identity, as opposed to other ethnic identities in Mauritius, is its fluidity and openness. It is sometimes said that “many Creoles look like Indians nowadays,” and it is true that many Mauritians with Christian names and a “Creole” family structure, Creole networks and a creole way of life do look vaguely Indian. This is presumably caused both by conversions and by intermarriage. In general, Creoles are more tolerant of intermarriage than are other Mauritian groups, and it is to some extent possible to become a Creole within one’s own lifetime—while one cannot conceivably become a Hindu, a SinoMauritian, or a Franco-Mauritian. The fuzzy category of Mauritian Creoles thus includes both the traditional Creoles, that is, dark-skinned workingclass people most of whose ancestors were slaves, and a residual category of modern or postmodern Creoles, who are Creoles because,for various reasons, they do not fit in elsewhere. On this background, and given the increasing numbers of mixed marriages, some Mauritians actually envision a future when tu dimunn pu vini kreol—when everybody becomes a Creole. The notion of cultural mixing or impurity is important here, as is the notion of individualism.

162 CREOLIZATION The language Kreol (or Morisyen) must also be considered. Kreol, which evolved during slavery in the eighteenth century, was created by the ethnic category now called Creoles, but it is the lingua franca of all Mauritian communities as well as the mother tongue of most Mauritians. Attempts at making Kreol an official language have nevertheless failed; in 1982, the radical Movement Militant Mauricien (MMM) party, then in power for the first time, tried to implement it in the media and in schools but were met with massive resistance—not only from Indo- and Franco-Mauritians but also, perhaps surprisingly, from Creoles. In general, the groups working for a wider recognition of Kreol are small and considered left-wing. French and English predominate in the media, in the educational system, and in public administration, French being the main language of culture and English the main language of administration. The resistance to Kreol can be traced to three causes: First, it is still widely regarded as “the poor cousin of French,” as an impoverished, shallow, and context-dependent idiom. Secondly, its wider use at the expense of French and English might strengthen Mauritius’s isolation, since it is spoken only in Mauritius, Rodrigues, and the Seychelles. Thirdly, Kreol is still associated with the Creole ethnic group and/or creolization as it is understood locally, and Kreol has connotations of impurity and uprootedness. Interestingly, few of the activists who have struggled for recognition of Kreol are Creoles in ethnic terms; they must nevertheless be seen as Creoles in analytical terms—like the European left, some are in favor of strong versions of multiculturalism, whereas others reject cultural tradition altogether as a source of personal identity. Mauritian society is changing, and so are local perceptions of Creoles. While the standard definition of a Creole remains essentialist and racial, society currently accepts that particular forms of mixing may create new Creoles. Conversion to Christianity, commitment to mixing through marriage, and mixed parentage may, under certain circumstances, make a person Creole. During the last decades, there has been a tendency in the population censuses that increasing numbers state that both their commonly spoken language and their ancestral language is Kreol. This is significant in so far as a person whose ancestral language is Kreol (and not an Oriental language or French) identifies him- or herself as someone rooted in Mauritian society and not in an old world civilization; in other words, as someone belonging to a new society founded on the premise of uprootedness.

CREOLE ESSENTIALISM
Mauritian notions of creoledom are traditionally associated with language and origins, which only partly overlap. Although most Indo-Mauritians

Creolization in Anthropological Theory and in Mauritius

163

speak Kreol, their language of reference still tends to be an Indian language, which means that although they live in a society based on uprootedness, migration, and mixing, they retain a rooted self-identity based on notions of purity, continuity, and boundaries. Creoles as an ethnic group have no fixed criteria for membership. Creoledom means impurity, openness, and individualism. The Kreol language is seen as an oral idiom lacking history and literature, and as rather superficial and limited compared to the great civilizational languages of English, French, Hindu, Urdu, and Mandarin. Its utility lies in its ability to unite otherwise very different groups in a shared field of communication. Now, the anthropological use of the concept cultural creolization closely approximates Mauritian usage. Creolization is seen as a process whereby new shared cultural forms, and new possibilities for communication, emerge owing to contact. It highlights the open-ended, flexible, and unbounded nature of cultural processes, as opposed to the notion of cultures as bounded, stable systems of communication. In Mauritian public discourse, notions of change, flux, personal choice, and hybridity are routinely contrasted with tradition, stability, commitment to fixed values and purity. These debates closely resemble the aforementioned debates in the academic community regarding stability and change, boundedness, and openness. In the Mauritian context, the phenomena classified as creole—whether the ethnic group Creoles, the language Kreol, or people who have been “creolized”—nearly always represent points of view that are consistent with the creolization perspective of culture. As I have shown, the problems faced by Mauritian Creoles, in a society dominated by essentialist, “rooted” identities, are weak internal organization and a chronic problem of leadership, lack of myths of origin that can match the others, as well as external stereotyping as being morally and culturally opportunistic. There are tendencies for Creole organizations to try to match the other ethnic groups by fashioning a Creole identity that is no less essentialist, no less rooted and bounded than the others. Simultaneously, a movement in the opposite direction amounts to the creolization of non-Creoles—that is, a growing commitment to the “mongrel” culture of Mauritius itself, borrowing and new juxtapositions—at the expense of “ ancestral cultures” (for example, some Indo-Mauritians nowadays play the séga) and openness to change. Both tendencies coexist and delineate a major field of political discourse. The relevant parameters are depicted in Figure 1: The debates concern, on the one hand, cultural similarity versus variation—in this regard, Mauritian politicians, unlike their European counterparts, favor difference rather than homogenisation—and, on the other hand, the relationship

164 CREOLIZATION between notions of purity and notions of mixing. As Figure 1 suggests, this field of discourse extends far beyond Mauritius and could shed light on Western political ideologies as well (plus, perhaps, academic debates on the nature of culture). Figure 1 Positions in discourses about culture and identity. In Mauritius, controversies chiefly concern the relationship between multiculturalism and creolization. Ethnic nationalism is scarcely an option, and cosmopolitanism is politically uncontroversial. Seen as an instance of Douglas’s (1970) grid-group model, the options can be described in the following way, clockwise beginning in the upper-left corner: low grid, high group; low grid, low group; high grid, low group; high grid, high group. There are Creoles who are against creolization (who favor multiculturalism, that is, strong group identities and clear boundaries), just as there are non-Creoles who favor creolization (mutual influence among cultural groups, individualism).
Similarity

Ethnic nationalism

Westernization or cosmopolitanism

Purity

Mixing

Multiculturalism

Creolisation

Difference

Creolization in Anthropological Theory and in Mauritius

165

CREOLIZATION AND DIFFUSION
Creolization, certainly in the Mauritian context, does not refer to any kind of mixing but to those adaptations, dislocations, and cultural dynamics, resulting from contact, that do not result in a fixed belonging to a bounded, historical, cultural tradition. If we want cultural creolization to be theoretically useful, a first step must consist in distinguishing between cultural mixing and collective identities. The second step should consist in distinguishing between different forms of mixing. Terms such as “hybridity,” “creolization,” “métissage,” “complexity,” and “syncretism” are used by academics to describe processes of mixing, but they are rarely defined in relation to one another. To give an accurate meaning to a term such as “creolization,” it is necessary first to study its meaning in linguistics, that is the origin of the analogy of cultural creolization. A preoccupation with cultural mixing is, incidentally, far from new. The anthropological notion of “cultures in the plural,” which has been so severely criticized in recent years, is historically and conceptually related to the biological notion of species. Darwin was himself an eager observer of everything alive (and fossilized), and although he saw the study of culture as an aside from his main work in evolutionary theory, he was fascinated by “pure savages.” In his Voyage of the Beagle (1997/ 1837), he describes the “Fuegian” (Ona) culture of Tierra del Fuego in Hobbesian terms, seeing the life of the natives as a brutally competitive, and in many ways debased, form of existence. Later visiting Samoa, he was pleased to learn of the civilizing work of missionaries but was otherwise uninterested in the local culture, because it had been contaminated. Just as he was uninterested in the giant tortoises of Galàpagos, believing them to have been imported from Alhambra in the Indian Ocean, he did not view the mongrel culture of Mauritius with the same keen interest as he had observed “savages” in South America. He privileged purity and continuity over mixing and disjuncture (this is discussed in greater detail in Eriksen 1997b). A key notion in Darwinism is the importance of intraspecies variation for natural selection, yet the boundaries between species are assumed to be absolute. Or are they? In The Origin of Species (Darwin, 1985/1859), Darwin writes about animal hybrids that they tend to be sterile if they are crosses between “two animals clearly distinct being themselves perfectly fertile” (p. 86). Interestingly, he ends the chapter about variation under domestication by stating that “the accumulative action of Selection, whether applied methodically and more quickly, or unconsciously and more slowly, but more efficiently, is by far the predominant Power” (p. 100, my italics).

166 CREOLIZATION The problem of distinguishing between varieties and “true species” is a main issue in Origin, for obvious reasons: the book’s theoretical agenda consists in accounting for “biological decreolization,” that is, the emergence of new, fixed species through interaction between varieties of old species under shifting environmental circumstances. Where to place the boundary between species was a main problem, and unlike Linnaeus who believed—at least as a young man—that species were created by God, Darwin knew that there were some deeply disturbing continuities between individuals deemed to belong to different species. One of Darwin’s lasting legacies for the working biologist consisted in developing a new set of criteria for classifying plants and animals and mapping out their mutual interrelatedness, replacing morphology with evolutionary lineage. One of the contributions of classic, Boasian, cultural anthropology a few decades later consisted, similarly, in replacing a racial classification of humans with a cultural one. Today that model seems to be exhausted because of overwhelming internal variation and boundary breakdown, but it has not been superseded by an alternative model simple enough to account for the increasingly complex facts that are now collected by ethnographers. Already in the generation after Darwin (and Tylor), alternative models of cultural dynamics were developed. Classic diffusionism, informed by the nineteenth-century German humanist tradition including comparative linguistics, problematized the idea of autonomous, relatively closed cultures—and it is reminiscent of, and encounters some of the same problems as, current research on globalization. But the American notion of the melting pot goes even farther back. It seems to have been used first in Crèvecoeur’s Letters from an American Farmer (1904 [1782]), where the author asked “What is the American, this new man?” and answered “here individuals of all nations are melted into a new race of men, whose labours and posterity will one day cause great changes in the world.”2 Emerson spoke about “the smelting pot” in the mid-nineteenth century, and with Israel Zangwill’s popular drama “The Melting Pot” from 1908, the term became a label of self-description for many new Americans. The members of the Chicago school of urban sociology, who set out to do research soon after the staging of Zangwill’s play, generally assumed that “acculturation” would eventually replace ethnic entrenchment, except in the case of the blacks (see Hannerz 1980). Another famous analysis of “acculturation” from the same period is Bateson’s article about culture contact and schismogenesis, published in 1935 (Bateson 2000a), where Bateson argues that contrary to what many expect, group differences may just as well be accentuated as reduced in situations of contact. Neither the Chicago school nor Bateson discussed

Creolization in Anthropological Theory and in Mauritius

167

mixing. Ralph Linton, in his introductory textbook from 1936 (The Study of Man), talks about hybridity first in a biological sense (noting that human hybrids are viable and fertile creatures) and, later, with reference to culture. Showing that virtually everything that is usually seen as “100 percent American” is imported, he nevertheless presumes that this does not affect group identity, cohesion and boundedness.3 The issues that have been raised recently under labels such as “creolization” and “hybridization” have been more actively debated in earlier periods than commonly assumed, often in similar ways. Although he does not develop the point, Linton clearly distinguishes between culture and identification. To him, there is no contradiction between cultural diffusion and a strong sense of group identity, although he was concerned that his compatriots were becoming nationalist bigots because they failed to recognize the extent of foreign influence on their everyday lives. In other words, cultural diffusion and the recontextualization and reintegration of “borrowed items” into preexisting cultural repertoires have long been recognized, even if the dominant research agendas of the mid-twentieth century tended to bracket diffusion. It is not sufficient to point out that mixing does take place; it is necessary to distinguish between different forms of mixing. Notwithstanding cultural diffusion, which in the era of mass communication often takes place without encounters between groups of different origins, there are a number of different outcomes from long-term encounters between distinct groups. Sometimes, one group is absorbed into the other; sometimes it is absorbed culturally but not socially (the ethnic boundaries remain intact); sometimes the groups merge to create a third entity; sometimes a hierarchical complementary relationship or a symmetrical competitive relationship occurs; sometimes, again, one group eventually exterminates the other (see Harris 1995 for a typology of forms of mixing in the New World).

THE LINGUISTIC SOURCE
The term “cultural creolization” is associated with Hannerz’s (1992; 1996) important contributions to an anthropology of global flows from the mid-1980s onward. However, the analogy between linguistics and anthropology implied by the term was, as acknowledged by Hannerz, explored earlier by Lee Drummond (1980) in a study from Guyana. In his article about the “cultural continuum” in polyethnic Guyana, Drummond (1980) notes that Guyanese routinely apply ethnic ascription in social classification and cultural stereotyping. He then shows that their usage

168 CREOLIZATION is situational and often self-contradictory (Drummond 1980:368). Rather than accepting the emic view (which is shown to be inconsistent) of a society made up by distinctive ethnic-cum-cultural groups, Drummond proposes a view inspired by creole linguistics, arguing its relevance for culture theory in general:
If variation and change are fundamental aspects of cultural systems, then we must consider the possibility that ethnographic studies of small, postcolonial, ethnically fragmented societies such as Guyana illustrate creole processes found in societies everywhere. (Drummond 1980:370)

The main problem arising from use of the creole metaphor, Drummond then notes, is that of descriptive fragmentation, leading the ethnographer to “exhaust himself describing and comparing every little pocket of informants” (p. 371) —a problem that would soon become familiar in postmodern anthropology, and often resolved through recourse to terms such as “multiple voices,” “polyphony,” “discourses,” and the like. Drummond’s own solution, drawing on Derek Bickerton’s creole linguistics, consists in seeing culture as a single entity that cannot be pluralized; as “overlapping sets of transformations or continua” (p. 372). A similar view is voiced by Hannerz when he states:
A cultural theory adequate to the task of understanding complex cultures must be able to deal with the fact that the division of labor is in large part a division of knowledge, making very problematic the notion that culture is by definition shared. (Hannerz 1986:363)

This view represents an advance over the earlier view of the world as an archipelago of cultures (Eriksen 1993, 1994). However, it remains to be explained why boundaries continue to be reproduced through reflexive identity politics, movements aiming to purge local culture of insidious influences from outside, and through everyday practices that are often not verbalized. Cultural flows may be everywhere, but so are cultural continuities. Likewise, hybrid and anomalous identities presuppose the existence of bounded, unambiguous identities. A similar point is made by Hannerz (1990) when he points out that cosmopolitans depend on locals in order to be cosmopolitans. Transgressing boundaries is impossible unless boundaries are being diligently reproduced. Now recall Bateson’s advice to go back to the context the analogy was taken from, looking for unresolved problems in the original context, assuming that similar problems might arise in analogous usage. In order to assess the significance of the term “cultural creolization,” it is therefore necessary not only to look at the emic uses of creolization, as I have

Creolization in Anthropological Theory and in Mauritius

169

done in the first part of the chapter, but also to examine the original linguistic usage of the term. Bickerton once distinguished pidgins from creoles by stating, bluntly, that “pidgins constitute a grossly handicapped case of second language learning while creoles show a partial recourse to the faculté de langage” (Bickerton 1976:183–84). In general, linguists have defined creoles as former pidgins that have become mother tongue to a group of people, thereby increasing in structural complexity and semantic richness. Creoles, moreover, are standardized to a certain degree, given that they are the main language of a community of people (and often the lingua franca of a larger population). Another process that may take place in linguistically complex situations with creoles and standard languages existing side by side is decreolization, whereby the creole increasingly approximates the standard form. The Guyanese situation described by Drummond may, by virtue of being a cultural continuum, be seen as a postcreole one, where boundaries have already become fuzzy. The question that remains to be explored is whether these terms— creolization, standardization, decreolization, postcreole continuum—may be helpful in refining the anthropological usage of “creolization” and related terms.

CULTURAL DECREOLIZATION?
Decreolization is the process whereby the boundary between the standard (written) form of a language and the creole is gradually blotted out, and an analogic range of intermediate forms replaces the digital eitheror; the boundary is replaced by a frontier area, and the creole forms begin to approximate the standard form. The result can be a postcreole continuum, as can be witnessed in parts of the English- and Frenchspeaking Caribbean, where the distance between standard and creole varieties is fluid, and where spoken language is not necessarily farther removed from the standard language than are certain dialects and sociolects in the metropolitan countries. Another possible outcome of decreolization is the eventual obliteration of the creole forms, sometimes applying only to a particular segment of the population, and a resulting strengthening of boundaries between that segment and the rest of the population. Cultural decreolization evidently occurs; the first kind of example that comes to mind is that of identity politics seeking to purify cultural forms “contaminated” by foreign influence—sometimes, as in the cases of France and Iceland, focusing on language, but more often pertaining

170 CREOLIZATION to physical markers of identity (clothes, food and so on), norms (often relating to religion), and knowledge (of, for example, history). Such efforts at cultural purification do not result in a postcreole continuum but rather to the strengthening of boundaries and removal of ambiguity. We seem to be faced with three typical phases in linguistic creolization: the development of creoles; a stable situation of diglossia with fixed creole and standard forms; and the eventual dissolution of fixed boundaries, that is, decreolization and the emergence of postcreole continua. These stages can rarely be identified in cultural processes, where the cultural continuum is a fact from the beginning, notwithstanding the possible existence of strong social boundaries. A related question is that of standardization. Creole languages historically did become standardized even if they remained oral. As noted earlier, historical sources documenting usage of a French-based Creole in Mauritius (for instance, Beaton, 1977 [1859]; St. Pierre, 1983 [1773]) indicate that there has been considerable continuity in the language for two hundred years up to the present. Although creole languages tend not to be standardized with respect to spelling conventions, state-sanctioned grammars, and so on, their ability to change is clearly restricted by their conventionalized everyday usage by the community of native speakers. Regarding cultural creolization, we may ask, similarly, under which circumstances the process of mixing ends and standardized, reified idioms are established. Some may react against this way of phrasing the problem, claiming with Deleuze and Guattari (1980), Urry (2000), and others that the main characteristics of the human condition consist in movement and change. This position must be resisted. Of course, there is movement, mixing, and change; but there is also stability and continuity. The systemic model developed by Morin (for example, 2001), where organization emerges in the interface between order and chaos through the selforganizing properties of recursive interactions, covers the facts better than does any proposition to the effect that either everything changes or nothing does. There are everywhere degrees of boundedness and stability, just as there are degrees of openness and change. Standardization of cultural forms can be presumed to happen through political agency, through the sudden isolation of a group or through its externalization in writing, museums, or similar institutions. The cultural form in question will then (again presumably) have been codified in a certain, authoritative way recognized as a standard. Typical examples from European cultural history are linguistic standardization via the printing press and nationalism (Anderson 1983), and standardization of musical expression through notation (Treitler 1981), which served to enhance complexity at the cost of “freezing” evolving traditions.

Creolization in Anthropological Theory and in Mauritius

171

The parallel between musical notation and linguistic standardization (or decreolization) are evident, just as there is much evidence of musical (and linguistic) “recreolization” taking place as global communication and migration to the West set the stage for intensified contact across boundaries and a consequent weakening of the national project of cultural homogenization.

TOWARD A DEFINITION
In many anthropological writings on creolization, including some of my own, the term is used more or less interchangeably with hybridity and mestizaje/métissage: it refers to displacement or mixing. Sometimes it results from the enforced or voluntary displacement of groups or individuals; sometimes it simply amounts to the local appropriation and incorporation of foreign influence. The incorporation of Country & Western music into the standard cultural repertoire of rural southern Norway can accordingly be described as a process of creolization, just as the complex cultural dynamics, involving interaction among various groups of Europeans, Asians, Africans, and Native-Americans, leading to the emergence of a distinctive Caribbean cultural intersystem during and after slavery. The concept of cultural creolization has been criticized from two perspectives: it is accused of being too wide and general—if every cultural process is creole in character, the term seems superfluous—and for implicitly positing the existence of pure forms existing prior to creolization. Neither Boas nor Kroeber, neither Malinowski nor Mauss was a stranger to the idea of demographic movement and cultural mixing. The assumption of the relatively isolated society was chiefly a methodological device, and Radcliffe-Brown himself worried about the increasing difficulty of isolating the entities for study (Radcliffe-Brown 1952:193). Network analysis, multisited fieldwork, and similar methodological reforms have responded to this challenge without changing the basic premises—the Manchester school, to mention an obvious example, remained loyal to the Evans-Pritchard who wrote The Nuer. The recent emphasis on flows, change, variation, and mixing may therefore be seen as less radical than often assumed. An important point made by Hannerz (1996:67), however, is that there are important differences in degree, and that the choice of terminology directs the attention toward processes of particular analytic interest. Nobody would deny that it enhances our understanding of, say, contemporary immigrant life in Berlin to look at the cultural dynamics of Turkish-German relationships, or that contemporary black British culture is an outcome of a dynamic relationship among Africans,

172 CREOLIZATION West Indians, and native Britons. In other words, creolization (or a similar term) is useful for describing a certain kind of cultural process, which has proven to be rather common, in all probability increasingly so in the present era of accelerated communication and demographic mobility. In addition to the aforementioned, creolization can also be taken to refer to dislocation, alienation, localizing foreign influence, anomalous or ambiguous ethnic groups such as douglas (people of mixed AfricanIndian origin) in Trinidad and Guyana, or traveling groups (tatere) in Norway and Sweden (people of mixed Gypsy-Scandinavian origin). Different parts of a cultural environment, and of people’s life-worlds, are being affected by influence from outside at different speeds and to differing degrees. Sometimes people are acutely aware of changes taking place in their immediate environment and take measures to stop them, to enhance them, or to channel them in their preferred direction. At other times, people may be unaware of these processes, even if foreign influences and cultural mixing may change their cultural environment profoundly. These are some of the intricacies of contemporary cultural processes that need disentangling if we are to be successful in studying them accurately. Merely stating that mixing is an inherent feature of contemporary culture is no more enlightening than saying that cultural diffusion is a fact. More precision is needed (see Khan 2001). Thus: • Cultural pluralism directs the attention of the researcher toward the relative boundedness of the constituent groups or categories that make up a society. It is a close relative of multiculturalism. • Hybridity directs attention toward individuals or cultural forms that are reflexively—self-consciously—mixed, that is, syntheses of cultural forms or fragments of diverse origins. It opposes multiculturalism seen as “nationalism writ small.” • Syncretism directs attention toward the amalgamation of formerly discrete world views, cultural meaning, and, in particular, religion. • Diasporic identity directs attention toward an essentially social category consisting of people whose primary subjective belonging is in another country. • Transnationalism directs attention, rather, to a social existence attaching individuals and groups not primarily to one particular place but to several or none. • Diffusion directs attention toward the flow of substances and meanings between societies, whether it is accompanied by actual social encounters or not. • Creolization, finally, directs our attention toward cultural phenomena that result from displacement and the ensuing social encounter

Creolization in Anthropological Theory and in Mauritius

173

and mutual influence between/among two or several groups, creating an ongoing dynamic interchange of symbols and practices, eventually leading to new forms with varying degrees of stability. The term “creole culture” suggests the presence of a standardized, relatively stable cultural idiom resulting from such a process. Cultural decreolization occurs when, in the case of group-based power differentials and inequalities, the subordinate group is socially or culturally assimilated into the dominant one (for example, cholos becoming mestizos in Latin America) or when a creolized idiom is “purified” and made similar to a metropolitan or “high culture” form. I propose a definition of cultural creolization, thus, that is faithful to its linguistic origins but that does not restrict itself to societies where “creole” is an emic term or where linguistic creolization has taken place (see, for example, Willis, 2002, on creolization in the Pacific; Archetti, 1999, on creolization in Argentina). The Mauritian example presented in some detail above nevertheless refers to a society where the term “creole” (kreol) is commonly used to designate both a language and a cultural group.

MAURITIAN MEANINGS OF CREOLIZATION
In Mauritian society, a main tension is that between creolization and multiculturalism (Eriksen 1997a). The Catholic Archbishop of the Mascareignes expressed a common view when he said, in the early 1990s, “Let the colours of the rainbow remain distinctive so that it can be beautiful.” Opposition to cultural mixing and mixed marriages within Mauritian society is strong. In the eyes of many Mauritian politicians as well as ordinary citizens, peace is maintained on the crowded, culturally heterogeneous island only because there is a precarious numerical equilibrium and functioning politics of compromise between the ethnic groups. Any “upsetting of this balance” would ostensibly threaten the peace. An alternative view is posited by Mauritians who reject ethnic classification as a set of primordial identities and who demand the right to be mixed in every sense. In the 1970s, this view was forcefully developed by the politician, scholar, and poet Dev Virahsawmy, who wrote poetry and plays in Kreol in spite of his Telugu origins, and who even devised his own religious rituals blending influences from the great world religions present in the island. The term “creole” itself has a fairly unambiguous meaning in Mauritius. It refers to those Mauritians who are Catholic and have African, Malagasy,

174 CREOLIZATION or mixed origins and/or who are seen by others and see themselves as Creoles. As shown in earlier ethnographic work on Mauritius (Boswell 2003; Eriksen 1988, 1998), Creole culture is perceived as stable and fixed, although it lacks—unlike the other cultural groups of the island— an illustrious past. At the same time, the Creole ethnic category is more open to new recruits than other ethnic groups in the island. Let us now consider the four main features of creolization identified above, in the context of Mauritian society: New forms resulting from contact between/among two or several groups; eventual standardization; decreolization; and the postcreole continuum. New cultural forms clearly have arisen, and continue to arise, out of intergroup contact. This is evident, among other places, in food, music, religious beliefs, and linguistic practice, if not necessarily at the level of identification. Standardization has partly taken place in Kreol, but there is no commonly agreed-on orthographic standard. At the level of culture, some effort has been invested into creating a standard Creole high culture on a par with the Asian and European cultures already recognized in Mauritius. There is no general agreement as to its content, however. Some argue that it ought to have strong African elements, while others argue that the history of slavery and mixing is what constitutes the unique Creole contribution to Mauritian culture (Boswell 2003). Some Mauritian Creoles even search for cultural authenticity in Rastafarianism—a Jamaican religious movement (Wathne 2003). Creole essentialism is far from unknown in Mauritius. Occasionally, Creoles claim that they are the only vrais Mauriciens, real Mauritians, since they are the only group who, as it were, emerged from the Mauritian soil. As is evident, this version of Creole identity is different from the attempt to anchor it in an African past, but both are attempts to fix and standardize a collective identity. Decreolization takes two forms in Mauritius: the “purification” of a cultural form perceived to have been creolized in the past, and attempts to fix and anchor cultural forms and social identities hitherto associated with mixing and impurity. Many urban middle-class Mauritians of French or mixed origin are worried about the Kreol (and more recently English) influence on Mauritian French, and try to purge their language of mauricianismes, that is, Creolisms. Similar concerns are voiced regarding the “contagious” effects of other aspects of Creole culture. Similarly, Mauritian Hindus emulate metropolitan forms by bringing pandits from India to “purify” their ritual practices; while Mauritian Tamils have taken steps to ensure that their new temples conform to the standard in Tamil Nadu. The most widely acknowledged form of decreolization in Mauritius

Creolization in Anthropological Theory and in Mauritius

175

is nevertheless the process sometimes disparagingly described as fer blan (“to make white”), whereby people with a Creole background “Gallicize” their way of life, begin to address their children in French and change their lifestyle to make it conform to the norms of the dominant FrancoMauritian group. The other, rare, form of decreolization consists in Creoles insisting on representing a rooted, uniquely Mauritian culture with a respectable peerage and no present indebtedness to foreign influences. They would argue that creolization, in the sense of borrowing, has ceased and that Creole culture is no longer a creole culture but a standardized form. The remaining question is whether the notion of the postcreole continuum makes sense in the Mauritian context. I should argue that it does, existing alongside and often predominating in relation to, processes of decreolization. Creole identity is associated with mixing, diffusion, and borrowing, and many Creoles have manifestly mixed origins. This implies that any Mauritian who does not obviously belong to one of the bounded, historically anchored communities will be seen as belonging to the residual category of Creoles. Notwithstanding the minority view that Creole identity has been or should be decreolized, there are in fact multiple points of entry into the category and no fixed essence in Creole identity. *** I hope the foregoing discussion has shown that the linguistic analogy is useful, but paradoxically, this is not least the case when flows are stemmed and fixity is implemented, either at the level of reified culture or at the level of group boundaries. The concept of cultural creolization not only tells us about mixing and boundary transgression but also calls attention to the formation of new standardized forms and new boundaries. As the Mauritian material shows, there are many points of entry into Creoledom. Some are born creole, some achieve creoledom and others—such as the children of certain mixed marriages—have creoledom thrust upon them. The key terms are “dislocation,” “fuzzy boundaries,” and “intergroup cross-fertilization.”

Notes
1. See Sahlins (1999), de Heusch (2000), and Bader (2001) for critiques which are, incidentally, less charitable than the perspective I am developing here. I nevertheless agree with these authors that it is a travesty to reduce culture to the politics of culture, since culture is largely implicit, embodied and nonverbalised. This does not imply that studying the politics of culture is uninteresting, only that it does not tell us much about that which “goes without saying because it comes without saying.”

176 CREOLIZATION
2. I owe this example to John Davidson. 3. Linton’s justly famous ethnographic vignette begins thus : [D]awn finds the unsuspecting patriot garbed in pajamas, a garment of East Indian origin; and lying in a bed built on a pattern which originated in either Persia or Asia Minor. He is muffled to the ears in un-American materials: cotton, first domesticated in India; linen, domesticated in the Near East; wool from an animal native to Asia Minor; or silk whose uses were first discovered by the Chinese. All these substances have been transformed into cloth by methods invented in Southwestern Asia. If the weather is cold enough he may even be sleeping under an eiderdown quilt invented in Scandinavia.

References
Anderson, B. 1983. Imagined communities. London: Verso. Archetti, E. 1999 Masculinities. Football, polo and the tango in Argentina. Oxford: Berg. Bader, V 2001. Culture and identity, Ethnicities, 1(2):251–273. . Barth, F. 1966. Models of social organization. London: Royal Anthropological Institute, Occasional Papers, 23. Bateson, G. 2000a [1972a]. Culture contact and schismogenesis. In Steps to an ecology of mind. Chicago: University of Chicago Press, pp. 61–72. ——— 2000b [1972b]. Experiments in thinking about observed ethnological material. In Steps to an ecology of mind. Chicago: University of Chicago Press, pp. 73–87. Beaton, P 1977 [1859]. Creoles and coolies, or five years in Mauritius. London: Kennikat. . Benoît, G. 1984. The Afro-Mauritians. Moka: MGI. Bickerton, D. 1976. Pidgin and creole studies, Annual Review of Anthropology 5:169–193. Boswell, R. 2003. Le malaise créole: A critical study of ethnic identity in Mauritius. Ph. D. thesis, Vrije Universiteit Amsterdam. Clifford, J., and G. Marcus, eds. 1986. Writing culture. The poetics and politics of ethnography. Berkeley, CA: University of California Press. Crèvecoeur, J. H. St. John. 1904 [1782]. Letters from an American Farmer. New York: Fox, Duffield. Darwin, C. 1997 [1837]. Voyage of the Beagle. London: Wordsworth. ——— 1985 [1859]. On the origin of species by means of natural selection. Harmondsworth: Penguin Classics. Deleuze, G., and F. Guattari. 1980. Mille plateaux. Paris: Odile Jacob. Desmond, A., and J. Moore. 1991. Darwin. Harmondsworth: Penguin. Douglas, M. 1970. Natural symbols. Harmondsworth: Penguin. Drummond, L. 1980. The cultural continuum: A theory of intersystems, Man 15(2): 352–374. Eriksen, T. H. 1988. Communicating cultural difference and identity: Ethnicity and nationalism in Mauritius. Oslo: Occasional Papers in Social Anthropology, 16. ——— 1993. Do cultural islands exist? Social Anthropology 2(1):133–147. ——— 1994. Kulturelle veikryss. Essays om kreolisering (Cultural crossroads: Essays on creolisation). Oslo: Universitetsforlaget. ——— 1997a. Multiculturalism, individualism and human rights: Romanticism, the Enlightenment and lessons from Mauritius. In Human rights, culture and context. R. Wilson, ed. London: Pluto, pp. 49–69. ——— 1997b. Charles Darwin. Oslo: Gyldendal. ——— 1998. Common denominators: Ethnicity, nation-building and compromise in Mauritius. Oxford: Berg.

Creolization in Anthropological Theory and in Mauritius

177

Eriksen, T. H. 2004. Predicaments of multiethnicity: Lessons from the Mauritian riots of 1999. In Ethnicity, nationalism and minority rights. S. May, T. Modood, and J. Squires, eds. Cambridge: Cambridge University Press. Fox, R., and B. King, eds. 2002. Anthropology beyond culture. Oxford: Berg. Friedman, J. 1994. Cultural identity and global process. London: Sage. Hannerz, U. 1980. Exploring the city: Inquiries toward an urban anthropology. New York: Columbia University Press. ——— 1986. Small is beautiful? The problem of complex cultures, Comparative Studies in Society and History 28:362–367. ——— 1990. Cosmopolitans and locals in world culture. In Global culture. Nationalism, globalization, and modernity/. M. Featherstone, ed. London: Sage, pp. 237–52. ——— 1992. Cultural complexity. New York: Columbia University Press. ——— 1996. Transnational Connections. London: Routledge. Harris, O. 1995. Knowing the past: The antinomies of loss in highland Bolivia. In Counterworks: Managing diverse knowledges. R. Fardon, ed. London: Routledge. Heusch, L. de. 2000. L’ ethnie. The vicissitudes of a concept, Social Anthropology 8 (2): 99–116. Khan, A. 2001. Journey to the center of the Earth: The Caribbean as master symbol, Cultural Anthropology 16(3):271–302. Linton, R. 1937. One hundred per-cent American, American Mercury 40:427–29. Morin, E. 2001. La Méthode V: L’Identité humaine. Paris: Seuil. Radcliffe-Brown, A. R. 1952. Structure and function in primitive society. London: Routledge and Kegan Paul. Rushdie, S. 1991. Imaginary homelands. London: Granta. Sahlins, M. 1999. Two or three things that I know about culture, Journal of the Royal Anthropological Institute 5(3):399–421. St. Pierre, B. de. 1983 [1773]). Voyage à L de France. Paris: Maspéro. ’Ile Tinker, H. 1974. A new form of slavery: The export of Indian labour overseas 1880–1920. Oxford: Oxford University Press. Treitler, L. 1981. Oral, written and literate process in the transmission of medieval music, Speculum 56(3):471–91. Urry, John. 2000. Sociology beyond societies. London: Routledge. Wathne, Kjetil. 2003. Mauritius, a kaleidoscope of peace? M.A. dissertation, University of Oslo. Werbner, P and T. Modood, eds. 1997. Debating cultural hybridity. London: Zed. ., Willis, D., ed. 2002. The age of creolization in the Pacific: In search of emerging cultures and shared values in the Japan-America borderlands. Hiroshima: Keisuisha.