Home page Help page Preferences page
Search for specific termsBrowse by ThemesBrowse by Geographical RegionBrowse by MOST DocumentsBrowse by How ToBrowse alphabetical list of titles

Discussion Paper no. 13 - Some Thematic and Strategic Priorities for Developing Research on Multi-Ethnic and Multi-Cultural Societies
Open this page in a new windowDon't highlight search terms


Discussion Paper Series - No. 13

Some Thematic and Strategic Priorities for Developing Research on Multi-Ethnic and Multi-Cultural Societies

Juan Díez Medrano
Associate Professor
Dept. of Sociology
University of California, San Diego

Table of contents

The Analytical Framework
Graph 1 : Analytical Framework
Ethnic Conflict and Ethnic Democracy
The Accomplishment of MOST Tasks

Mass international migration, state centralization, the devotion of a growing army of intellectuals to the codification of the cultural diversity of our planet, the search for identity in an increasingly urbanized and anonymous world, and the political mobilization of ethnicity have pushed the problems associated with multiculturalism to the top of the political and research agendas. Whether societies are now more or less multicultural as before is an open issue. Although mass international migration has brought together people from widely diverse origins, state-guided cultural and political homogenization policies and improvements in means of mass communication have tended to reduce the world's cultural diversity. What distinguishes our age, however, is that we are more conscious of this diversity and that ethnicity has become politicized to such an extent that the major military conflicts of this century have been expressed as ethnic or national conflicts. This justified inclusion of the topic « Multi-ethnic and Multicultural Societies » as one of the themes which articulate the activities of the MOST Programme.

The present paper proposes a strategy to guide the development and funding of research projects around this topic. The strategy is conditioned by the spirit that guides the MOST Programme, which is to contribute both to the development of policies that facilitate ethnic co-operation within multicultural societies and to the strengthening of the multicultural character of these societies. This is a specially difficult problem to address because the cost of policies against ethnic conflict can sometimes be the undermining of the multicultural character of these societies and, reciprocally, multicultural policies can in some instances promote ethnic conflict. The strategy proposed here is also channelled to prioritize those lines of research for which the MOST Programme is better suited than other funding agencies. In particular, research projects that have a heavy comparative component.

The premise guiding this set of recommendations is that issues of policy cannot be divorced from issues of theory. That is, we cannot say whether a particular policy will be effective or not and under what circumstances, unless we reach these conclusions through carefully designed studies that control for other explanatory factors of ethnic inequality and ethnic conflict. In order to choose these controlling factors we need a theory that tells us what counts and what does not count. Unless one is aware about the need not to divorce our search for useful contributions to policy-making from general theoretical issues pertaining to ethnic conflict and ethnic equality, the research results generated by MOST research projects will only contain a good deal of information and unsubstantiated insights on problems of policy.

The Analytical Framework

We can situate the problem by presenting in graph form the major parameters of the task at hand (See Graph 1). As said above, the major goals to which research sponsored by the MOST Programme should be devoted is the development of policies (Policy 1 and Policy 2) which eliminate ethnic conflict (EC=Ethnic Conflict) and guarantee the protection of the multicultural character of the societies concerned, that is policies that guarantee equality of cultural, civil, political, and social citizenship rights for all cultural groups in a given society (ECR=Equality of Citizenship Rights).

The range of problems associated with the concepts Ethnic Conflict and Equality of Citizenship Rights is very broad and therefore invites a wide variety of research projects, theories, and policy proposals. For instance, it is useful when addressing the concept Ethnic Conflict to distinguish between violent (e.g. the Bosnian conflict) and institutionalized conflict (e.g. Québec Nationalism) and between ethnic conflict involving territorially concentrated indigenous groups (e.g. Tamils and Shinalese in Sri Lanka) and ethnic conflict involving ethnic groups which are not territorially concentrated (e.g. Africa-Americans and the White majority in the USA). The latter generally involves a host majority ethnic group and an immigrant or various immigrant minority groups, although immigrant groups can also confront each other (for instance Turks and Kurds in Germany). There are of course states which have to face both types of conflict. In Spain, for instance, conflict between different nationalities (e.g. Basques and Catalans) and the central state coexists with growing problems between the Spanish population and groups of immigrants from Latin America or Africa.

One can also distinguish different agents of ethnic mobilization and different types of demands, which can be raised violently, institutionally, or both. Conflict between territorially concentrated indigenous ethnic groups and the central state can express itself through demands for autonomy or secession by the ethnic minority or through centralization demands by the ethnic group that holds state power. Conflict between dominant indigenous ethnic groups and subordinate immigrant ethnic groups can be initiated by the dominant group or by the immigrant ethnic minority.

The arrows in the graph indicate causal relations. The graph simplifies the theoretical problems involved, by assuming unidirectional relationships. Dynamically speaking, however, many of these relationships are reciprocal. For instance, in a recent MOST publication (Multiculturalism: A Policy Response to Diversity) it is said that one of the explanations for the absence of ethnic conflict in Australia is the great equality of citizenship rights between the different ethnic groups which live in that country. At the same time, however, ethnic conflict can contribute to reduce the citizenship rights of particular ethnic groups and even the multicultural character of a particular society.

The graph distinguishes between intended (whole arrows) and unintended consequences (dashed arrows) of policies. Finally, the Category Other represents all variables, including policies which do not address ethnic conflict or multiculturalism, which can contribute to the goal of equality of citizenship rights for all ethnic groups and to the attenuation of ethnic conflict.

Using this graph as our starting point, the following sections outline some problems associated with the study of ethnic conflict and « ethnic democracy », suggestions about ways to contribute to the formulation of policies concerning these two major themes, and a list of priorities that could guide the funding of projects submitted to evaluation.

Ethnic Conflict and Ethnic Democracy

In order to understand the intensification of ethnic conflict in the twentieth century and, particularly, after W.W.II, we have to take into account a series of long-term historical processes studies by authors like Anderson (1983), Gellner (1983), Hobsbawm (1990), Mann (1993), Geertz (1963), and others. In particular, the multiplication of contacts between world cultures and state centralization and homogenization policies. These two processes obeyed the imperatives of capitalist development and inter-state competition, which required the creation of national markets, the search for new markets beyond the national borders, and the improvement of the economic and military efficiency of states.

The multiplication of contacts between world cultures originates in the massive colonization process of the late nineteenth century, in the mass migration movements from poorer societies to richer societies, which date back to the beginning of the century - the great waves of transatlantic migration - and gained impetus after the decolonization process that followed W.W.II, and the development of mass communications. These three major developments increased the salience of ethnicity and mobilized a legion of culture-diggers, obsessed with the discovery and classification of cultures. Not surprisingly, the emergence of anthropology as a discipline is directly related to the administration of colonial possessions. Mass international migration led also to the emergence of ethnically diverse labour markets, where socio-economic grievances were often expressed in ethnic terms.

State centralization and homogenization policies did at the state-level what mass migration was doing at the international level, that is to intensify contacts between previously isolated cultures within state boundaries through the creation of national markets. Simultaneously, state centralization and homogenization policies, and state growth in general, generated grievances among peripheral ethnic groups whose cultures and political institutions were threatened. They also signalled that the survival of a culture hinged on control of state political institutions. In this sense, the democratization of political life paved the way not only to class politics, but also to ethnic politics. In newly created states after decolonization, because of their often artificial character, the lack of a tradition of civil politics, the absence of a national market and of a well-developed class system translated in the primacy of ethnic politics, as Geertz has well demonstrated (1963).

Gradually ethnic politics became a mobilization force on a par with class politics. What is the origin of the strength of ethnicity as a mobilizing force is a question subject to debate. Why is it greater than that of height, weight, right- or left-handedness? (In a recent provocative and playful article, The Economist comments, for instance, on the prevalence of height discrimination in our society; there is no evidence, however, of a social movement centred around height.) According to socio-biologists, the emotional appeal of ethnicity (well documented by Walker Connor, 1994) has to do with the similarity between the concept of ethnic group and that of extended kin (Van de Berghe, 1981). Signs such as similarity of skin colour, language, religion, matter only to the extent that they are cues that indicate belongingness to this extended kin.

In order to explain ethnic conflict, we do need to understand, however, the mobilizing force of ethnicity. We can take it as a given, for the same reason that in order to explain class conflict, we do not need to develop a theory of why material interests are so important for individuals. What is important, however, is to recognize with Breuilly (1993) that ethnic political mobilization is, like class mobilization, a form of politics, which under the ethnic discourse, or side-by-side to it, often encompasses class or plain political interests that have little to do with ethnicity. Since class and ethnicity have been the two major political mobilization forces of our century, the greatest challenge for those interested in ethnicity is to explain why people sometimes mobilize on the basis of class and sometimes on the basis of ethnicity.

Unfortunately, we do not have a clear answer to this question yet, partly because of the multi-faceted character of the problem, partly because of the paucity of systematic empirical research. We have a pretty good idea of the causes for the historical process that led to the emergence of ethnic conflict as a major problem in contemporaneous societies. It follows pretty much the logic sketched above. There is also a pretty good understanding of the ethnic identity phenomenon. Although vaguely conscious ethnic groups may have existed for centuries, as defended by authors like Smith (1986) or Armstrong (1982), the proliferation of relatively large ethnic groups probably dates back to the invention of the printing press and to its use for the publication of books in the vernacular (Anderson, 1983). Print capitalism in vernacular languages facilitated the coalescence of large ethnic imagined communities out of the infinitely greater cultural variety which existed world-wide. The nineteenth- and twentieth- century state was the second major agent in the development of self-conscious ethnic groups, through its practices of cultural homogenization (see Weber, 1976), its statistical definition and classification of ethnic groups, and its manipulation of cultural distinctions for political purposes (Laitin, 1985). And then there are other factors cited in the literature, such as geographic-bureaucratic pilgrimages by educated subjects of the colonial powers in Asia and Africa (Anderson, 1983), processes of ethnic occupational segregation (Barth, 1969; Hechter, 1978), or the discovery and invention of traditions belonging to different ethnic groups by intellectuals (Hobsbawm and Ranger, 1983).

We are much less sure, however, of the causes that lead to ethnic conflict in some multicultural societies and to peaceful co-operation in others. We do not know why people mobilize sometimes on the basis of class and sometimes on the basis of ethnicity. part of the problem resides in the lack of distinction between different manifestations of ethnic conflict, be it nationalist demands, anti-immigrant mobilization, violent conflict, peaceful expression of ethnic grievances, demands for secession, demands for autonomy, etc. These are widely different phenomena. Nevertheless, they tend to be treated in the literature as interchangeable phenomena that ought to be explained by one single theory.

At bottom, the theories that have been proposed to date tend to be structural and can be roughly divided into two camps. One camp that claims that ethnic conflict, whatever its expressions, arises out of a cultural division of labour (Modernization Theory and the Theory of Internal Colonialism - of course, from different analytical perspectives). A cultural division of labour exists when one dominant ethnic group monopolizes the good positions and a subordinate ethnic group is relegated to the bad positions. To the second camp belong those who think that, on the contrary, ethnic conflict emerges when a cultural division of labour breaks down, that is when different ethnic groups compete for the same positions (Ethnic Competition Theory and Split Labour Market Theory - also from different analytical perspectives).

The debate between these two contradictory theoretical positions matters in light of the effect that state efforts conducive to equalize the socio-economic conditions of different ethnic groups can have on ethnic relations. If competitive theories are correct, there would be a potential contradiction between policies aimed at equalizing the citizenship rights of different ethnic groups and the goal of minimizing ethnic conflict. Whereas the Australian example of multicultural policies can be used to show that equality, or competition, leads to ethnic co-operation, the increasing mobilization by whites in the United States against Affirmative Action programmes aimed at improving the life conditions of ethnic and other minorities in the United States tends to support the competitive argument.

As stated before, research has rarely tried until now to differentiate between types of ethnic conflict. Therefore, there is very little theorization of those phenomena that are of maximum interest for the MOST Programme, that is the explanation of violent expressions of ethnic conflict and the explanation of maximalist ethnic demands tending to disrupt the social order in which they are formulated. Here one needs to move beyond broad structural explanations and introduce elements that belong to the political sphere. It is not that structural factors do not count; it is to be expected that the same factors that lead to ethnic conflict will also contribute to the emergence of a context prone to violence. Many other factors can, however, determine whether people mobilize on the basis of class membership or on the basis of ethnic group membership and whether conflict manifests itself violently or radically.

For instance, two highly similar societies in terms of economic development, cultural singularity, presence of immigrants from other regions, exposure to cultural repression in the past, such as the Basque Country and Catalonia, have show two very different types of nationalism. Whereas Catalan nationalism has generally been moderate, Basque nationalism has been much more radical and violent. Recent research shows that the radicalization of Basque nationalism in the 1970s, compared with the moderation of Catalan nationalism, had something to do with differences between the Basque and Catalan structures of political competition within the democratic opposition to Franco and with the different political dynamics that there two different structures of political competition facilitated during the 1970s (Díez Medrano, 1995). These political dynamics reflected the way Basque and Catalan nationalists formulated their demands during the 1970s and how the government, before and after Franco's death, responded to the qualitatively different challenges posed by the Basque and Catalan nationalist movements.

What is significant is that whereas in 1969 Catalans expressed more radical political demands than did Basques, by 1980, the situation had been dramatically reversed. This indicates that political processes matter, even though they are constrained, like they were in the Basque and Catalan cases, by long-term socio-economic and political structural factors. MOST should therefore concentrate not only on long-term factors leading to ethnic conflict or co-operation but also on those short-term factors related to the micro-management of ethnic conflict by governments. Some studies have shown, for instance, that consociational politics (Lijphart, 1968) or political arrangements that promote bargaining and representation of conflicting interests can reduce the potential for violence. Both a diversified political competition structure, in which some non-violent and non-radical groups carry a large part of the ethnic demands, or a bargaining attitude from the part of the political authorities can contribute to separate the moderates from the radicals.

The second goal pursued by the MOST Programme should be the development of policy recommendations that would contribute to the development of multicultural democratic societies in which the different ethnic groups enjoy equal citizenship rights. The Australian case provides a particular optimistic example of how such multicultural policies are seen not only as a matter of justice but also as a matter of economic rationality and efficiency. In the area of ethnic equality, the role played by policies is potentially large, and even in societies like the US, with a relatively underdeveloped welfare state, policies such as Affirmative Action have contributed to the improvement of the economic situation of ethnic minorities. That some have benefited more than others reveals only that the survival of a particular culture or the economic conditions enjoyed by a particular ethnic group do not depend on policy alone.

Some aspects in the study of equality of citizenship rights, and in the study of the impact of policy on such equality, are easier to investigate than others. Equality of cultural rights is perhaps « easier » to study. Whether people know their language, are able to practice their religion, or dress like they like, is relatively easy to determine. Research focused on the impact of different cultural policies is also rendered feasible because other variables likely to considerably affect the learning and transmission of culture are relatively few and less important.

To study the equality of economic rights across ethnic groups is much more difficult, however, especially when one compares indigenous with immigrant groups. Ethnic inequalities in a particular social structure reflect among other factors the fact that the indigenous groups were already there, whereas immigrants may have been directed to particular occupations. Therefore, the point of entry for immigrants matters and studies of ethnic inequality must remove this factor, as well as other factors such as differences in age, gender, level of education, acculturation, etc.

The study of the effect of particular policies on ethnic stratification carries even greater complexity. This added complexity results from the difficult methodological problems involved in the study of social stratification and social mobility in general, such as the need to take into account the roles of expanding versus stagnant labour markets or the role of changes in the job structure. Unless these factors are removed, one cannot fairly assess the role of multicultural policies on the attainment of ethnic equality of citizenship rights.

The Accomplishment of MOST Tasks

The MOST Programme should therefore devote its efforts to the analysis of the effects of alternative long-term and short-term policies on the attainment of the goal of equality of citizenship rights and on the prevention and solution of ethnic violence through democratic means. Particular attention should be devoted to understand the relationship between policies aiming at equalizing citizenship rights across ethnic groups and violent ethnic conflict.

The MOST goal of providing policy-makers with the tools to implement policies that will achieve the double role of promoting ethnic equality of citizenship rights and ethnic co-operation, can be achieved only if the projects that are sponsored contain clear explicit information about questions to be answered, theory, data, and methods. Whether one thinks of research projects that aim at testing hypotheses, historico-comparative studies aimed at generating new hypotheses, or of case studies aimed at describing and/or explaining a particular instance of multicultural society or ethnic conflict, one needs to make explicit what theoretical frame guides the observation and measurement of reality. Otherwise, there is a risk of ending up with collections of reports about different cases which do not contribute to the accumulation of knowledge or provide scientifically based policy guidelines.

At the level of methods, MOST is particularly well suited for the development of comparative projects. Given this, it is of great importance to encourage the standardization of data used to compare cases. it is quite unlikely that strictly comparable data from secondary sources exist for the different cases that are to be analyzed. Therefore, successful comparisons will be those that use whatever standardized information is available and/or collect standardized information from scratch. MOST should therefore encourage applicants to provide guidelines about the type of information that will be used, in which form it exists at the moment, and what needs to be collected.


MOST should promote research this is not normally funded by traditional funding agencies or in places that have normally little access to those funding agencies. The geographical criterion is especially important. Most academic institutions and funding agencies are located in the wealthier countries. This leads to a bias in the selection of countries that are considered in investigations, and thus to a bias in the theoretical implications derived from such studies and to the underdevelopment of the research capacities of poor countries where ethnic conflict is a fundamental problem.

At the infrastructural level, MOST could contribute to compensate for this by promoting projects that aim at creating an adequate academic structure for the study of ethnic conflict. That means, promoting for instance the development of research institutes and academic programmes in poorer countries where the basic theoretical and methodological tools for the study of ethnic conflict and citizenship rights would be provided. The connection of these research centres with other centres around the world through Internet would guarantee a two-way flow of information, that would both benefit the entire scientific community and help providing policy guidelines to policy-makers in these poorer multicultural societies.

Alternatively, MOST could make a more modest contribution to our understanding of poorer multicultural societies, by sponsoring projects whose primary goal is the collection of theoretically guided information on those societies: On the social structure by ethnic group, on the relations between the groups (as measured through historical analysis, interviews to policy makers, attitudes surveys), and on the policies developed to address the issue of multiculturalism in both its socio-structural and ethnic relations aspects. This information could be used by the governments concerned and international organizations to monitor the potential for violent ethnic conflict and the degree of ethnic inequalities in these societies, and if well distributed could provide incentives to researchers to include these countries as part of their research projects.

A second infrastructural resource that MOST could promote is the collection of standardized data on ethnic stratification and conflict across the world. Unfortunately, such information is available only for a small subsample of wealthy countries. This partly explains why, for instance, Western nationalist movements have received more attention in the scientific literature than have far more serious instances of group conflict in many states of the developing world. It would be very useful to fund projects that pursue the following related goals:

a. Determining what objective empirical data on group conflict and on group characteristics within multicultural societies is available in different countries;

b. Based on the literature on ethnic stratification and conflict, elaborate a list of variables for which it would be fruitful to obtain information in as large a number of countries as possible.

c. Facilitate the obtention of such information in as large a number of countries as possible.

In particular, a major objective should be to promote the collection of standardized census information on ethnic group membership and on the population's distribution by ethnic group and occupation. The MOST Programme is particularly suited to provide a lead in this direction because of its integration in the UN structure and the potential to achieve this goal by collaborating with the Population Division of the United Nations and individual national governments. The pay-off that this effort would have cannot be emphasized enough. Standardized stratification data by ethnic group would allow to identify the level of ethnic inequality prevailing around the world. Without this information we cannot begin to study the effects of multicultural policies on levels of ethnic inequality across societies. We cannot identify either the effect of ethnic inequality on ethnic conflict, an issue which, as shown above, has occupied most research on ethnic conflict for the past thirty years. Regardless of the hypotheses tested, no study aimed at providing policy guidelines to promote ethnic co-operation can do without controlling for the effects of ethnic inequality.

It is important to emphasize that census are better suited than survey data in order to study this problem. Surveys, unless the number of cases is very large, cannot represent adequately the occupational structure of even medium-size ethnic groups. On the other hand, a workable cost-saving compromise would be to promote the collection of census data on occupation by ethnicity in a relatively large, randomly selected number of countries, in order to cover a broad range of geographical, economic, and cultural contexts.

The third infrastructural contribution that the MOST Programme could make to policy-makers is the sponsoring of research projects aimed at compiling and distributing detailed information o policies of prevention and solution of ethnic conflict and policies that try to achieve the goal of ethnic democracy in multicultural societies. This information could be collected by interviewing policy-makers in different countries, analyzing legislation, and analyzing processes of ethnic violence through newspaper analysis or other sources of primary information on the unfolding of crises.

Beyond supporting projects that contribute to the development of research infrastructure, MOST could contribute to policy formulation by sponsoring projects whose aim is to analyze the role of multicultural long-term and short-term policies in achieving ethnic equality of citizenship rights and ethnic co-operation. As stated above, particular emphasis should be placed on the micro-management of ethnic conflict episodes, by carefully comparing instances of ethnic conflict successfully and unsuccessfully addressed by the political instances.

At the level of research design, MOST should attempt to correct two major biases in the published literature, which are the privilege of cases where there is ethnic conflict over those where there is no ethnic conflict, and the privilege of Western societies over non-Western societies. It should also favour comparative research projects over case studies. By comparative projects one should understand, however, projects guided by a clearly-defined question, and projects where an attempt is made to collect standardized information on theoretically determined variables, whether static or dynamic. This does not mean, however, that research projects ought to be quantitative. The crucial issue is the design of such comparative research projects. The ad hoc juxtaposition of cases, based on conducting similar projects in different countries will be of little use. Comparative projects do not need to be overly ambitious. What matters is that they control for theoretically relevant variables, so that one can unambiguously identify the role of policy and how its effectiveness interacts with the social and economic context where the policies are implemented.


Anderson, Benedict 1983. Imagined Communities. London: Verso.
Armstrong, John. 1982. Nations Before Nationalism. Chapel-Hill: University of North Carolina Press.
Barth, Frederick. 1969. Ethnic Groups and Boundaries. Boston: Little Brown.
Breuilly, John. 1993. Nationalism and the State. Chicago: Chicago University Press.
Connor, Walker. 1994. Ethnonationalism: The Quest for Understanding. Princeton: Princeton University Press.
Díez Medrano, Juan 1995. Divided Nations: Class, Politics, and Nationalism in the Basque Country and Catalonia. Ithaca, NY; Cornell University Press.
Gellner, Ernest. 1983. Nations and Nationalism. Ithaca, NY: Cornell University Press.
Geertz, Clifford. 1963. The Interpretation of Cultures. New York: Pantheon Books.
Hechter, Michael. 1978. « Group Formation and the Cultural Division of Labor », pp.293-318 in American Journal of Sociology 84.
Hobsbawm, Eric J. and Terence Ranger 1983. The Invention of Tradition. New York: Cambridge University Press.
Hobsbawm, Eric J. 1990. Nations and Nationalism since 1780. New York: Cambridge University Press.
Laitin, David. 1985. « Hegemony and Religious Conflict: British Imperial Control and Political Cleavages in Yorubaland », pp.285-317 in Peter Evans, Dietrich Rueschemeyer, and Theda Skocpol, eds., Bringing the State Back In. New York: Cambridge University Press.
Lijphart, Arend. 1968. The Politics of Accommodation; Pluralism and Democracy in the Netherlands. Berkeley: University of California Press.
Mann, Michael. 1993. The Sources of Social Power, Vol.II. Cambridge: Cambridge University Press.
Smith, Anthony. 1986. The Ethnic Origins of Nations. New York: Blackwell.
Van den Berghe, Pierre. 1981. The Ethnic Phenomenon. New York: Elsevier.
Weber, Eugen J. 1976. Peasants into Frenchman. Stanford, CA: Stanford University Press.

To MOST Clearing House Home Page

To see the document in English
Pour voir le document en
Para ver el documento en