This website will be closed down during 2014.

Thomas Hylland Eriksen's new site is now open at

You are now being redirected to the new Engaging with the world site.
























Us and Them in Modern Societies

Ethnicity and Nationalism in Trinidad, Mauritius and Beyond


Thomas Hylland Eriksen

Scandinavian University Press 1992

This book is a hybrid between a collection of articles and a theoretical monograph informed by original ethnography. Some of the nine chapters had been published in earlier versions in various professional journals; some were adapted chapters from my doctoral thesis; and the introduction was written entirely for this volume.

Drawing extensively on material collected in Mauritius (1986) and Trinidad (1989), two multi-ethnic societies which both display the whole range of ethnic relations at the micro level -- from intermarriage (frequent) to ethnic violence (rare) -- the book analyses ethnic phenomena at several levels: in its relationship to a supra-ethnic nation-building ideology, as a property of social relations and as a property of culture, in relation to peace and political stability, and in relation to power discrepancies. In several chapters it is shown that a certain degree of shared culture is required for a poly-ethnic society to be viable; notably, a shared set of political rules, a shared labour market and shared institutions of education. Thus the link between ethnicity and culture becomes tenuous, as one can have a great ethnic diversity and minor cultural variations (and vice versa, great cultural variations in ethnically homogeneous societies). In this sense, all societies are "plural" and not just the poly-ethnic ones.
The first chapter is available here in html format, while Chapters 1, 4 and 7 are available as this PDF: usandthem


Foreword by Bruce Kapferer 

1. Introduction: On the study of ethnicity and ethnicities
2. Ethnicity as a comparative concept: A justification
3. The cultural contexts of ethnic differences
4. Ethnicity versus nationalism
5. Containing conflict and transcending ethnicity?
6. Linguistic diversity and the quest for national identity
7. New World Indians: A comparison between Trinidad and Mauritius
8. Two nationalisms
9. Against pluralism: Ethnicity and modernity




1. Introduction: On the Study of Ethnicity and Ethnicities

Although every chapter in this book deals, to a greater or lesser extent, with aspects of society and culture in Trinidad and/or Mauritius, the book is chiefly intended as a contribution to the interdisciplinary theoretical discussion on ethnicity, nationalism and modernity. Chapters 2, 3 and 4 depart from, and elaborate on, current anthropological perspectives on ethnicity and nationalism; Chapters 5, 6 and 7 are more ethnographic in nature, although they, too, are meant to illuminate the theoretical discussions about the phenomena; while Chapters 8 and 9 are attempts to move beyond some of the current theorizing on "poly-ethnic societies". In this introductory chapter, I shall give an overview of the main issues to be tackled and the analytical framework employed in so doing. I will also briefly situate the present work in contemporary Anglophone academic discourse. First, however, I shall suggest why Trinidad and Mauritius deserve sustained attention by practitioners of the comparative social disciplines.

Why Trinidad and Mauritius?

Trinidad & Tobago and Mauritius are tropical island-states, located in the southern Caribbean and the south-western Indian Ocean, respectively. Neither has a pre-modern history; as societies, they were created by plantation colonialism and were thus contributors to the development of a capitalist world-system. (Trinidad, unlike Mauritius, did have an indigenous population, which has been brutally exterminated without leaving any visible trace.) They belong to a category of societies which has not been intensively studied by social anthropologists; they are neither "primitive" societies nor "our own" society. They represent varieties of modernity sometimes carelessly labelled "creole cultures". This term, parasitical on the more accurate linguistic term "creole language" (see e.g. Hancock, 1979), suggests the presence of an incongruous admixture of cultural traditions. This idea, if ultimately misleading, at least puts us on the right track. Both island-states, independent since the 1960s, contain populations of diverse origins, and are for this reason often classified as "plural" societies.

The population segments which make up these societies are aware of their objective uprootedness; at the same time, they scarcely yearn for their ancestral lands (India and Africa, in most cases). Nation-building in Trinidad and Mauritius, in other words, is a complex project and frequently a thorny issue in domestic politics (see Chapters 4, 7 and 8; see also Eriksen, 1991e).Given the small territories of the islands, secession could never be an option for discontented groups. Further, nobody would be able to win a civil war. The uprooted populations of Trinidad and Mauritius have but two opportunities: emigration (which has been, and still is, common) or compromise. The latter option has largely been chosen in political life. During their brief period of independence, both societies have admirably avoided inter-ethnic violence, and both are functioning multi-party democracies. Lastly, both Trinidad and Mauritius are presently changing in ways which may (or may not) render ethnicity irrelevant in most practical contexts in a not too remote future.
In sum, then, Trinidad and Mauritius are tropical, densely populated, emphatically modern, poly-ethnic and democratic societies which change quickly, economically and culturally. What more could an analyst ask for? All of these issues will be discussed in the chapters to follow. For now I turn to an explication of the analytical framework to be employed.

Cornering the Elusive Fact of Ethnicity

Definitional quarrels concerning the concept of ethnicity and problems arising in this connection have led some scholars to discard the concept of ethnicity altogether (see, for example, Chapman et al., 1989), replacing it with a more comprehensive concept of classification. To make my position clear, I should state that in my view this is rather an overstatement of the issue. Instead of abandoning the ship, we might try to keep it afloat a while yet, to see whether or not the concept of ethnicity has been exhausted as a conceptual bridgehead towards a comparative understanding of social phenomena which are otherwise different.

Some of the contemporary confusion and resignation over the use and misuse of the concept of ethnicity arises, clearly, out of its being used for very different analytical (or ideological!) purposes, its being applied to human phenomena ranging from presumed biological dispositions (e.g. van den Berghe, 1981; 1986) or socio-psychological features of identity (e.g. Epstein, 1978; Liebkind, 1989) over situational analysis (e.g. Eidheim, 1971) and local political strategies (e.g. Cohen, 1969; 1974b) or minority strategies (e.g. Fishman, 1989) to comprehensive collective ideologies (e.g. Nash, 1988) on the one hand, and aspects of societal formations on the other (e.g. M.G. Smith, 1965.)1

In addition, ethnicity has entered the political vocabulary of our times, and the inaccurate usage current in the mass media may have a dangerously contagious effect on analytical conceptualizations. The academic discourse on ethnicity is multidisciplinary and frequently interdisciplinary, and the concept of ethnicity has lost some of its accuracy because of the lack of discipline sometimes implied by interdisciplinary work. I should therefore make it clear that I am persuaded that we need a shared, comparative concept of ethnicity which is so fashioned that it may shamelessly be applied to contexts which are otherwise enormously different. Ethnicity, then, should be taken to mean the systematic and enduring social reproduction of basic classificatory differences between categories of people who perceive each other as being culturally discrete. It has aspects of politics as well as aspects of meaning or identity.2

This concept of ethnicity will be discussed in several of the chapters to follow (see particularly Chapters 2 and 3). Below, I shall therefore limit myself to discussing a few of its implications not dealt with elsewhere.

"Kinds" of Ethnicity?

The still quite recent development in ethnic studies which can be referred to as the Barthian revolution, consists of a number of related insights developed in the volume edited by Fredrik Barth following a conference in Bergen in 1967 (Barth, 1969b). Barth and his Scandinavian colleagues stressed that ethnicity should not be regarded as a property of a group, but rather as an aspect of social relationship and process. In other words, it was seen as futile and misleading to distinguish ethnic groups through listing different "cultural traits" supposedly dividing "cultural groups", as had been common until the mid-1960s (and which is, incidentally, still common among non-specialists). Instead, Barth suggested in his celebrated introductory chapter, one should look for what was socially effective; that is the ethnic boundaries whereby socially relevant cultural boundaries were being reproduced. In Chapter 3, I discuss the Barthian perspective extensively, and I shall therefore leave it for now.

However, the insistence on formal aspects of social relationship as fundamental to ethnicity deserves a few comments in this introduction, not least as it is (I hope) to be read by some non-anthropologists. The issue deals with the relationship between form and substance in ethnicity. The programmatic insistence by Barth, Eidheim (1969, 1971) and others (which has, however, not always been followed up in practice) that all social phenomena involving ethnic boundary maintenance are in some relevant respect similar, no matter what their other characteristics, has led to great uneasiness, and has probably been partly responsible for the abandoning of the comparative concept of ethnicity on the part of a number of younger scholars, who prefer to slice up the social world according to different principles. For sheer common sense forces us to concede that ethnic groups in the Amazon forest are faced with problems different from those of ethnic groups in South London, and that the latter again are in important, analytically relevant respects different from secessionist movements in Canada or Sri Lanka. Can they meaningfully be regarded as the "same kind of group", and do they require the same analytical framework?

Allow me now to describe some characteristics of some different "kinds" of ethnic groups usually dealt with in the literature, in order to highlight their differences, to see if they have anything in common, and whether whatever they may have in common should either merit an extension of the Barthian perspective or contradict it. My own definition, as proposed above, is a variation on the Barthian theme; and it is also closely related to the heuristic concept of political­p;symbolic ethnicity proposed by Abner Cohen (1974a). The comparison between the four "types of ethnic groups" below is meant to indicate how and why substantial, empirical contexts and formal analytical contexts must be kept apart. It is also intended to show how comparison between substantial contexts (empirical, political or otherwise concrete societal phenomena) and the abstract classification of substantial contexts must be mediated by analytical contexts to be intelligible; that is, by our own inventions.

(1) Urban minorities. The Muslim immigrant populations of Western Europe may serve as a representative example of this category. Most of them have arrived since the Second World War in search of a livelihood. Although many second-generation immigrants of this category have lost their mother-tongue and have acquired citizenship, they remain self-consciously distinctive, and there can be no question of their status as ethnic minorities. Research, particularly in Britain and Scandinavia, has focused on problems of adaptation and, conversely, on discriminatory practices on the part of the host countries. More recently, questions of cultural identity and belonging have entered the research agenda. Some problems revealed in research on these minorities, and often mentioned by their spokesmen, are (i) discrimination in the labour market, (ii) cultural discrimination in the public sphere (re the Rushdie affair), (iii) marginality in relation to the formal political system, (iv) the loss of cultural identity; for example, the second generation's lack of a true mother country or mother-tongue. These minorities, which are nevertheless usually ideologically oriented toward an ancestral land, rarely or never demand political autonomy, and, of course, they never demand political independence. Their aim is to be as well integrated as possible into the labour market of the host country without losing their distinctiveness; many expect to return to their ancestral country eventually (and many do so, some even within a few years). Their strategies in relation to the political and educational systems of the host countries tend to reflect a concern to be accepted as valuable contributors to the economy on the one hand, and as a legitimate cultural minority on the other hand.

(2) Indigenous populations. "Indigenous populations" is a blanket term for aboriginal inhabitants who are politically non-dominant and who are not, or only partially, integrated into the dominant nation-state. This means that their language, customs, political practices and/or livelihood must be different from that championed by the state. Indigenous populations are also defined by their being acknowledged as such by international organizations such as IWGIA (International Work Group for Indigenous Affairs) in Copenhagen, Minority Rights Group in London, and their own non-localized council, WCIP (World Council of Indigenous Peoples). The Basques of the Bay of Biscay and the Welsh of Great Britain are usually not considered indigenous populations in these forums, although they are certainly as indigenous, technically speaking, as the Saami of northern Scandinavia or the Jivaro of the western Amazon. This is because their integration into the institutions of modernity is too complete; they take part in most of the practices instituted in, and sanctioned by, the nation-state. For one thing, the languages of "real" indigenous peoples should be chiefly oral, and their technology should be largely indigenous and non-industrial. As a rule, indigenous peoples are only partly integrated into, or claim the right of autonomy from, basic institutional dimensions of the modern nation-state such as capitalism, mass surveillance, militarization and/or industrialism (see Giddens, 1990:59). The concept "indigenous people" is not an accurate analytical one, but one drawing on broad family resemblances and contemporary political issues.
Scholars studying indigenous peoples implicitly assume that they need special protection and particular rights if they are to retain important aspects of their cultural heritage and develop some form of political autonomy. Features shared by indigenous peoples worldwide include: (i) territorial claims not respected by governments, (ii) threats of "cultural genocide", that is, enforced assimilation or physical extermination, (iii) a way of life requiring special measures in economic, political and or educational matters. Indigenous peoples do not, as a rule, intend to set up their own nation-states. On the contrary, they tend to stress that their cultural distinctiveness requires that they should be allowed (by the nation-state) to retain their original political system in some or all respects. In their political struggle, they often depict their loss of their ancient homeland as theft on the part of the immigrants. They may in this respect demand some form of retribution from the nation-state. Common to the groups assembled in the WCIP is also a non-modern traditional technology and non-state traditional social organization. In the study of indigenous peoples and in their political struggle, their cultural uniqueness is often contrasted with central aspects of modernity, although there are variations. (See also Chapter 4.)

(3) Proto-nation-states ("ethnonationalist" movements). These groups, the most famous of ethnic groups in the news media in the early 1990s, include Kurds, Sikhs, Palestinians and southern Tamils, and their number is growing. They may be said to include diaspora or irredentist nationalists such as Kenyan Somalis, Northern Irish Catholics, Hungarians in Romania, Armenians in Nagorno-Karabakh and German-speaking Alto-Adigese; as a rule, however, they have no external nation-state to relate to. They are secessionists, claiming that their cultural uniqueness implies that they should have their own nation-state and not be "ruled by others". These groups, short of having a nation-state, may be said to have more substantial characteristics in common with nations in nation-states than with either urban minorities or indigenous peoples. They are always territorially based; they are differentiated according to class and educational achievement; they are neither more nor less modern than others. In accordance with a common usage of the term, these groups are "nations without a state".

(4) "Plural societies". The term "plural society" is usually used about colonially created states with self-consciously culturally heterogeneous populations (M. G. Smith, 1965; see Chapter 9). Typical plural societies, originally analysed by J. S. Furnivall (1948) and later by M. G. Smith, would include Burma, Indonesia and Jamaica. The groups that make up the plural society, although they are forced to participate in uniform political and economic systems, are regarded as (and regard themselves as) highly distinctive in other matters. According to Furnivall (1948) and Smith (1965), one group tends to dominate politics in the plural society. In the context of the typology of ethnic groups which I am presently trying out, the population segments of plural societies are distinctive in the following ways: (i) they have no external nation-state to relate to realistically; (ii) they are not strong nationalists, but rather tend to identify with their ethnic group; (iii) secessionism is normally not perceived as an alternative; (iv) each population segment is internally divided according to class and possibly other criteria of rank. According to Smith, these societies are deeply divided and potentially violent, but this view has been challenged repeatedly (see Ryan, 1990; see also Chapter 9). The relationship of the groups that make up plural societies to the modern institutions of the nation-state and the market, is not deemed an important variable in this approach. African nation-states and the United States alike are considered plural societies (M. G. Smith, 1986), although the groups that make up the former are much more heterogeneous in this respect than most of the groups that make up the latter. The general idea is that plural societies are faced with a constant threat of fragmentation due to group competition and group-based quest for power. Trinidad and Mauritius, which furnish the raw material for most of the analyses in this book, are both considered typical "plural societies".

A very wide formal definition of ethnicity, such as the one which I have proposed, would include all of these "kinds" of groups, no matter how different they are in other respects. Surely, there are aspects of politics (gain and loss in interaction) as well as meaning (social identity and belonging) in the ethnic relations reproduced by urban minorities, indigenous peoples, proto-nations and component groups of "plural societies" alike. Despite the great variations between the problems and substantial characteristics represented by the respective kinds of groups, the word ethnicity may, in other words, meaningfully be used as a common denominator for them. The distinctions that I have suggested merely refer to differences between particular historically contingent contexts of ethnicity. Besides, these distinctions are themselves highly problematic; notably, the idea of the plural society is in my view a dubious one (see Chapter 9).
An interesting empirical issue seems to be the fact that all of the "kinds of group" enumerated must relate politically to the nation-state, and stand in a problematic relationship to the nationalist ideology embodied by the state. Their mutual differences, from this point of view, seem to lie in their varying prospects for getting a nation-state of their own, and in their varying degrees of participation in the institutions of modernity (notably wage work, institutional politics, modern education and mass media use). The urban minorities often either have a nation-state of their own to relate to (albeit geographically dislocated, as it were), and/or identify themselves (to varying degrees) with the host country. The proto-nations aspire to have their nation-state. The indigenous populations tend to have the rejection of the nation-state at the top of their political agenda, while the constituent segments of the plural society may be expected to try to appropriate the state and nationalist ideology on behalf of their own group. On the other hand, the practices associated with the state are in some cases compatible with the demands of the ethnic groups, in other cases not. The crucial variable here seems to be modernisation, which indicates degrees of participation in, and control from, the institutions related to the state and market. On this score, however, there are important differences within the categories which I have suggested. Among indigenous peoples, for example, there is a great difference between the literate and politically articulate Saami of northern Scandinavia (Eidheim, 1971; 1985) and the largely illiterate and politically powerless Dyirbal of northern Queensland (Schmidt, 1987).

Interfaces of Modernity

Apart from conforming to my proposed definition of ethnicity, there seems, thus, to be nothing uniting the different "kinds" of ethnic groups, except their all having to relate actively to the nation-state as ethnic groups. This empirical fact would support Giddens' (1985; 1990) and others' claim that the contemporary world is profoundly a modern one (Giddens rejects the term "post-modern"), where the nation-state is the "pre-eminent power container". A shared interface, which could be a useful analytical bridgehead, is therefore the nation-state (see Chapter 4).

However, the lumping together of, for example, "plural societies" and "indigenous peoples" as categories of ethnic groups, seems analytically unfortunate, since their mutual differences may prove more significant than their similarities. Moreover, there are, of course, also other ways of distinguishing between "kinds of ethnicity" or "kinds of ethnic contexts". Some are tried out in later chapters in this book; some have been tried out by others (see, for example, Yinger, 1986). Seen as such, ethnicity as a comparative concept is devoid of substantial content, let there be no doubt about that. The dimensions along which we choose to distinguish between kinds of ethnic phenomena, therefore, are contingent upon the questions we ask as analysts. The typology tried out above is constructed along the dimension of differential incorporation into the nation-state. If one were chiefly interested in the importance of ethnicity in comparative social classification, it would be natural to develop a typology of contexts where the ethnic element ranged from the very important to the almost insignificant. If, again, one were chiefly interested in accounting for the presence of ethnicity in a particular society, one would need to distinguish between societal levels and try to assess the importance of ethnicity at each level, as well as depicting the interlevel connections. Such a set of distinctions could, for example, look like this:

(1) State organization
(2) Political organization
(3) Property and the division of labour
(4) Patterns of settlement
(5) Casual intercourse
(6) Marital ideologies and practices

In some societies, thus, ethnicity may have an important bearing on virtually all aspects of social organization. In others, only rules of endogamy (which are followed to a varying degree) serve to reproduce ethnic boundaries socially. The semantic density of ethnicity varies enormously. At one extreme, ethnic difference could be intrinsically connected with cultural idioms related to almost every conceivable social situation (one could think of the heavily ethnically flavoured contexts of Israel, the Eastern Cape or the US South); at the other extreme, ethnicity is relevant only once a year in connection with the celebration of a national festival. The distinctions are clearly important if one wishes to locate ethnicity accurately in social time­p;space. And one might go on, inventing a host of further kinds of distinctions between ethnic contexts, tailored for dealing with particular sets of assumptions or analytical questions. Such distinctions, no matter how "concrete" and "empirically founded" we may claim them to be, are ultimately our own inventions, and are as such contingent on the questions we wish to examine. Let me now, therefore, turn to the substantial issues with which this volume is concerned.

Ethnicity and Nationalism in the Contemporary World

Following the change in the dominant analytical perspective on ethnicity usually attributed to Barth, the interest in ethnicity and ethnic phenomena has grown enormously in social anthropology and related disciplines. This has also come about as a reaction to changes taking place in the world outside of academia. As aspects of modernity become dominant and begin to penetrate the very heartlands of anthropology, the discipline needs to respond to these changes. This has partly been undertaken through a change in the dominant empirical focus from "tribe" to "ethnic group", and additionally, most contemporary anthropologists do in some way or other account for the influence of the nation-state and the commodity market on the contexts which they study. In this sense, the world has shrunk. Moreover, conflicts and political alignments in the contemporary world tend to be expressed through ethnic idioms. Culture has in other words become ideologized; it has become a kind of symbolic system prone to conscious manipulation through politics. An increasing number of the world's inhabitants become self-consciously aware that they have a culture; in a sense, they thereby invent their culture. The kind of tradition that one desperately tries to revive and revitalize has, of course, a different content, and a different political function, from that of one's great-grandparents, who never objectivated their culture as something detachable from themselves. Cultural innocence has been irretrievably lost (cf. Eriksen, 1991f; 1991g).

Changes in the actual world have contributed to bridging gaps between academic disciplines in this respect. Traditionally the domain of historians and political scientists, the comparative study of nationalism has recently become close to the concerns of anthropologists and sociologists studying ethnicity ­p; in a sense, it has forced itself upon them. International relationists concomitantly realise the importance of what they call "internal" (or domestic) conflicts and the need for anthropological perspectives (see Ryan, 1990, for a recent statement). Ernest Gellner's concise theoretical monograph on nationalism (1983) has in this regard served as a stimulus comparable in impact to that of Barth with respect to ethnicity. Gellner's thesis was that nationalism has developed as a Gesellschaft ideology trying to mitigate the socially fragmenting effects of industrialization and large-scale social organization. He points out that there is an infinite number of possible nationalisms, and, by implication, that nationalisms are inventions; their claims of historical continuity are always dubious and must be analyzed as expressions of ideology. Similar points were made by Benedict Anderson (1983) and Eric Hobsbawm (1983), and contemporary discourse on nationalism accordingly tends to focus on the ideological aspects of nations as imagined communities (Anderson's phrase) tailored to suit the social organization of industrial society.
Studies of ethnicity as well as nationalism are thus at a relativizing stage, where the social construction of identities and the relativity of "historical truths" are focused upon. In this book, particularly Chapters 4, 6, 8 and 9 are intended as critical contributions to the interdisciplinary discussion of nationalism. In Chapters 4 and 6, I discuss the relationship between ethnicity and nationalism; in Chapter 8, different aspects of nationalism are distinguished between; and in Chapter 9, a model of post-national and post-ethnic social identity is outlined.

Power and Domination

Sometimes analysts distinguish between violent and non-violent ethnic conflicts. In my view, one might in many of these cases discard the predicate "ethnic" and simply talk of violent versus non-violent conflicts. To characterize a particular conflict as an "ethnic" one is relevant if and only if one talks comparatively about forms of political organization and process which encourage either the improvement or the deterioration of inter-ethnic relations. From a political perspective, this is clearly the most important field for interdisciplinary research on ethnicity and nationalism. Since much previous research has, in my view, been tainted by insufficient analytical tools (such as "pluralist theory" and other reifying conceptualizations of "cultural groups" and the like), both conceptual rethinking and fresh research are called for. The most important questions dealing with political systems in so-called poly-ethnic societies addressed in this book are these two:

(1) What are the conditions for peace in poly-ethnic societies? My choice of Trinidad and Mauritius as foci for comparative research on ethnicity and nationalism was strongly influenced by the fact that both were emphatically poly-ethnic, and yet had avoided violent ethnic conflict since moving to independence in the 1960s. Most of the following chapters contribute to explaining how this can be; see particularly Chapters 4, 7 and 8. In my view, anthropologists have not paid sufficient attention to the manifestly destructive aspects of social identities; I have in mind phenomena such as violent racism and chauvinist nationalism (see, however, Kapferer, 1988; see also Jenkins, 1986). These phenomena need careful analysis. My own contribution, for what it is worth, consists chiefly of critical analyses of programmatically non-violent, non-chauvinist ideologies of cultural unity.

(2) Is it fruitful to talk of poly-ethnic societies at all, or does such a terminology both misrepresent social reality and serve to justify crude ethnicism and or brutal chauvinist nationalism? If the social disciplines are to yield any new insights, they must be critical in the sense that they do not appropriate folk conceptualizations of society without investigating the social reality to which they refer. If nationalisms and ethnicities are seen as "natural" entities which are not dealt with critically by investigators, then they will not be able to understand how social realities can be social products and in what ways they are ideological. If they fail to regard folk concepts of national and ethnic identity critically, analysts can easily become the hostages of nationalists wishing to justify violent and discriminatory practices. The analytical deconstruction of ethnicity and nationalism can therefore be politically important. The most fundamental deconstruction of these concepts, which are nevertheless debated throughout the book, is to be found in Chapter 9. The relevance of this deconstruction for the contemporary anthropological discussion concerning the concept of culture is made explicit there and, to some extent, in Chapter 3.

Centrifugal and Centripetal Forces in Globalization

The contemporary interest in ethnicity and nationalism, and the currently vivid exchange of views across academic boundaries, are largely caused by changes having taken place in the external world; the fact that nationalism and ethnicity, as foci of personal identity and of social organization alike, are empirically of great importance to many of the inhabitants of a world about to become thoroughly modern. The next important analytical step to be taken should in my view be a renewed, comparative focus on social identities. Since ethnic and national ideologies are of highly varying importance worldwide, it is highly pertinent that we try to account for the "negentropic" variations developed within, and in response to, the culturally and socially universalizing idioms of modernity. Why is it that ethnic ideologies are more important in some contexts than in others; what are the other identities available, and under which circumstances are they relevant? What exactly does it mean to be a citizen? I am not claiming that this is an unexplored field.3 However, we seem to lack a unified conceptual framework for the comparative study of social identities in this sense of the word. In this book, a main concern lies in the search for social determinants in the construction of social identities and differences in a world that increasingly appears as a seamless one. The concept of identity itself is not, however, dealt with critically.

Again, the study of ethnicity and nationalism is being caught up with by the world. The tendencies sometimes described as globalization (see Featherstone, 1990; Giddens, 1990), which create entirely new socio-cultural configurations in time­p;space (to use Giddens' terminology), are highly relevant in this regard. The fact that knowledge, culture and even social organization no longer need to be confined to a particular location, clearly has important effects on the constructions of social identities. Migration, the spread of global mass media, mass education and of the main international language (English), the increasing power of the nation-state in most of the world, and the increasing dominance of monetary economies, together indicate profound social changes in the contemporary world. It seems that the agenda of modernity is about to be realized on a global level, at least at the level of symbolic representations. Whether or not Eric Hobsbawm is correct when he suggests that nationalism has thereby had its day,4 it is doubtless true that important aspects of contemporary social identities are non-localized. The universal languages of pop music, soap operas and consumerism, or, for that matter, the global appropriation of Kafka's, Marquez's or Ngugi's novels, cannot be directly linked with particular ethnic or national identities; they smooth out differences and create the impression that the world is seamless. To this extent, they may seem to transcend territorially based identities. On the other hand, these processes in some areas create counterreactions in the form of ethnic, nationalist, linguistic or religious revivalism desperately trying to control indigenous cultural resources and maintain not only social boundaries but also the subjectively perceived cultural content of the group. However, such reactions may credibly be seen as confirmations of the hegemony of modernity, both since they tend to use the language of modernity for their own ends (they use the mass media and appeal to people's cultural self-consciousness), and since they relate ideologically to the ideology of modernity as simple negations of it.
Personally, I would like to believe that the contemporary upsurge in ethnic animosities and violent nationalist sentiment seen in parts of every continent is but a transitory phenomenon; a counterreaction directed against the irreversible social changes and cultural homogenization brought about by different forms of modernization. Although it would probably not be wise to hazard the guess that ethnic sentiments will eventually disappear, there are reasons, indicated above and in Chapter 9, for believing that their command over individuals may eventually diminish. On the other hand, conflicts between poor and rich countries may easily turn violent, and will in that case probably be justified by forms of nationalist ideology; that is, ideologies stressing the cultural differences between us and them.

In our endeavour fully to understand these and related contemporary processes of change and continuity, the combined efforts of scholars from various academic disciplines will be required. Not least for this reason, it is to be hoped that the lively interdisciplinary discourse on nationalism, ethnicity, the nation-state and globalization in the world of high modernity will continue as we uneasily approach a new millennium.

Notes to Chapter 1

1See the contributions to Rex and Mason (1986), particularly Jenkins' and Yinger's papers, for an overview of some current approaches.

2This definition has benefited from conversations with Harald Eidheim.

3See, for example, Epstein's (1978) and Roosens' (1989) fine comparative studies of ethnic identity. Social psychologists have for years investigated identity using quasi-quantitative methods; see, for example, Weinreich (1986) and the contributions to Liebkind (1989).

4 Hobsbawm (1990) cites Hegel to the effect that "the owl of Minerva flies at dusk", in this assuming that our contemporary interest in nationalism is a symptom of its imminent disappearance...

©Thomas Hylland Eriksen 1992

The book was published by Scandinavian University Press , and can presumably still be purchased here.