Citizenship, Ethnicity, and Nation-States

Sinišay Malešević & John A Hall. The Sage Handbook of Sociology. Editor: Craig Calhoun, Chris Rojek, Bryan Turner. Sage Publication. 2005.


For most of human history, class, gender and social status were the central pillars of exclusion, polarization and conflict. Today, however, it is the question of legitimate membership in a particular state that determines an individual’s social standing. As every African or East European knows very well when approaching the European Union, passports determine social position. The speedy, control-free, blue line for European citizens (and unofficially for Americans, Canadians, Australians and members of other stable and ‘respectable’ polities) stands in stark contrast to the slow green line facing those who arrive from the rest of the world. At such moments a wealthy businessman from Morocco or Ukraine realizes how much worse is his social standing compared to that of a dole-dependent single teenage mother from Ireland or a New Age traveler from Belgium. The possession of a particular passport symbolizes the power of the modern state and its legal and material embodiment in citizenship. However, this profoundly contemporary legalistic underpinning of citizenship, important though it is, does not reveal the internal complexity of states. It frames nation-states as inherently stable and culture-free legislative entities. In order to understand the novelty of contemporary citizenship we will have to engage both historically and geographically with the question of internal cultural diversity.

A central presupposition of this chapter is that the relations between citizenship and ethnicity can only be understood once we realize—as is beginning to happen in contemporary sociology—that the nation-state is not some sort of static entity. We begin of course with abstract discussion of the three terms in the title. But attention then turns to three particular social realms in order to introduce historical and comparative evidence that will allow light to be cast on the theoretical issues raised. Something must be said about European history since its historical record did most to create the conceptual equipment at work in social science. Attention then turns to the United States, the most powerful nation-state in the history of the world, in large part to suggest that this social formation is not as far removed from European experience as its self-image might suggest. This discussion of the core of liberal capitalist society serves as a necessary backdrop to an all-too-brief consideration of the condition of the vast majority of humankind. There can be no more urgent need than that of determining whether the South is doomed to follow in the footsteps of the North. If there are reasons to fear, let it be said at once that there are rational reasons to hope. A final preliminary point must be made. A particular approach is taken here, namely, one that privileges political explanations on the ground that cultural forms are more consequence than cause of general social development; justifications for this view are offered throughout the chapter.


Historically, citizenship grew out of popular demand for civil, political and social rights. The classic account of T.H. Marshall (1963) saw this development in evolutionary terms—from acquiring the rights to free speech, worship, property ownership and justice (civil rights) in the eighteenth century via the securing of the right to vote and stand for office (political rights) in the nineteenth century to finally obtaining protection for disadvantaged groups via development of the welfare state (social rights) in the twentieth century.

Michael Mann critically extended this analysis by emphasizing historical particularities and contingency in the development of citizenship in Europe and America. While Marshall’s analysis had some empirical backing in the UK it could not properly explain development of citizenship elsewhere. The extension of citizenship rights, in Mann’s view, was historically determined by the interests of political, economic and military rulers who were in control of the particular state apparatus. Hence the political elites in the United States and the UK were constrained by the early development of economic liberalism and expansion of the civil rights (in the American case also due to the popular participation in revolution) which led to the development of the constitutional model of citizenship with the institutionalization of repression only for those who went outside the rules of the game’ (Mann, 1988: 192). In absolutist states such as Germany, Austria, Japan and Russia due to the dominance of agricultural production and the limited size of the working classes, the rulers (church, nobility and monarchs) were in a position to deny universal citizenship rights to all the other strata in society and were eventually forced only to concede limited civil (but not political) rights to the bourgeoisie. According to Mann, other European states moved from contested to merged models of citizenship through deep social and political conflicts between monarchists and clerics on the one hand and secular liberals and socialist revolutionaries on the other (as in Spain, Italy and France), or this struggle went through negotiated social change with the eventual victory of an alliance between small farmers, working classes and bourgeoisie (as in Scandinavia).

Bryan Turner (1994) has expanded this analysis even further, arguing that both Marshall and Mann have neglected the impact of social movements, different religious traditions and the possibility of creating a citizenship from below. In his view various forms of citizenship have developed in a dialectical and parallel interplay between the elite pursuit of control of the state and decisive actions of civil society groups. He builds his theory of citizenship around the dichotomies of private vs. public and active vs. passive, arguing that specific historical circumstances have determined the form and content of particular citizenship frames. Thus, American and French citizenship developed through revolutionary experience by popular pressure from below leading to an active understanding of citizenship; in contrast, the passive citizenship of England and Germany has its roots in the relatively peaceful way in which it was given from above (whether through the negotiation of competing elites as in England, or by a paternalist authoritarian state employing an instrument of modernization as in Germany). The historical routes taken by these states as well as the contents of particular religious traditions had a decisive impact on general attitudes to public and private spheres. Hence state-suspicious, privately oriented Protestantism had a direct impact on American citizenship, being at once active, individualist and apprehensive towards the state. In contrast, French Catholicism and secular Enlightenment-shaped collectivism privileged the public over the private sphere, and led to a collectivist and statist but very active model of citizenship. English Protestantism with no revolutionary tradition but with very developed civil society led to passive and private citizenship, whereas limited political rights coupled with Protestant ethics and authoritarian paternalism led to even more passive and private citizenship in Germany.

Feminists have also contributed to understanding of citizenship (Butler, 1993; Walby, 1994). Their emphasis is on describing the modern forms of citizenship in terms of the institutionalization of gender-biased norms. For one thing, the timetable by which women were accorded rights differed from that of men, thereby putting Marshall’s model in question. For another, feminists argue that social and political rights gained through the development and expansion of the welfare state in the West were historically linked to a male-centered life cycle and its corresponding norms that privilege continuous, uninterrupted full-time employment and with profound disregard for the feminine life cycle (with pregnancy, maternity, menopause and menstrual periods). The criticism has been particularly leveled against the strong classical liberal distinction between the public and private spheres where public was traditionally identified with active, productive and socially recognized work (male), while private was relegated to passive, unappreciated and unpaid domestic work (female).

Although all of these approaches have contributed significantly to understanding of citizenship they all share one pronounced weakness. The leading approaches on citizenship have focused primarily on class, gender, religious background and social status in attempting to explain individual differences between societies and have largely neglected the central question of the relationship between citizenship and cultural difference. There are many questions that need to be addressed here. Are universalist premises of citizenship incompatible with cultural particularities of ethnic and national group claims? Is multiculturalism a viable alternative to the melting pot ideology? What is the relationship between modernity and cultural homogeneity? It is questions such as these that make it essential to discuss the nature of ethnicity and nationalism—and then to provide a sketch for an historical and comparative sociology of the advanced and the developing worlds.


Although the term ‘ethnicity’ has its roots in the Greek ethnos/ethnikos, which was commonly used to describe pagans, that is non-Hellenic and later non-Jewish (Gentile), second class peoples, its academic and popular use is fairly modern. The term was coined by David Riesman in 1953 and it gained wider use only in the 1960s and 1970s (Glazer and Moynihan, 1975). However, from its inception ‘ethnicity’ has remained a ‘hot potato’ of sociology. Four distinct issues can usefully be highlighted.

First, although the term was coined to make sense of a specific form of cultural difference it generally acquired a rather different set of meanings. While Anglo-American tradition adopted ethnicity mostly as a substitute for minority groups within a larger society of the nation-state, the European tradition regularly opted to use ethnicity as a synonym for nationhood defined historically by descent or territory. At the same time both traditions shared a joint aim to replace until then a very popular, but with the Nazi experiment heavily compromised, concept of ‘race.’ Nevertheless, popular discourses in both Europe and America have ‘racialized’ the concept of ethnicity, that is ‘race’ was largely preserved (in its quasi-biological sense) and has only now been used interchangeably with ‘ethnicity.’

Secondly, the collapse of the colonial world in the 1950s and 1960s brought even more confusion on questions of race, culture and ethnicity. The homelands of former European colonizers have quickly become populated with the new postcolonial immigrants who were visibly different. Following now American popular and legislative discourse, these groups have also become defined as ‘ethnic’ thus simultaneously preserving old definitions of historical ethnicity by descent or territory (for example, Welsh, Flemish, Walloons and Basques) with the new definitions of ethnicity as an immigrant minority (for example, Pakistani, West Indian, Sri Lankan).

Thirdly, the fall of communism and the breakup of the Soviet-style federations along ‘ethnic’ lines and the emergence of ‘ethnic cleansing’ policies in the Balkans and Caucasus have further complicated these definitional issues. With wars on former Yugoslav soil, the term ‘ethnic,’ through the extensive and influential mass media coverage of ‘ethnic wars,’ has degenerated into a synonym for tribal, primitive, barbaric and backward.

Fourthly, the ever-increasing influx of asylum seekers, refugees and economic migrants to Western Europe, America and Australia who do not necessarily express visible or significant physical, cultural or religious difference to their hosts, and their legal limbo status (for example, waiting for the decision on asylum) have relegated the term ‘ethnic’ to a quasi-legislative domain where ‘ethnic,’ just as in the days of ancient Greece and Judea, refers again to non-citizens who inhabit ‘our land,’ that is, to second class peoples.

To clarify all these misuses and misunderstandings one has to explain who exactly is an ‘ethnic’ and what ethnicity stands for in contemporary sociology. First, ethnicity as used in contemporary sociology is a broad enough concept to accommodate distinct forms of social action defined in collective-cultural terms. Unlike ‘race,’ which is an epitome of a folk concept, often constructed in an ad hoc manner by social actors who are themselves trying to make sense of their everyday reality, the concept of ethnicity allows for sociological generalization without affecting particular instances of it. Although there is a clear genetic and physical variation between human beings such as skin colour, hair type, lip size and so on, as biologists emphasize, there are no unambiguous criteria for classifying people along the lines of these characteristics. Any such classification would artificially create groups where in-group variation would be greater than its presumed out-group variation. In other words ‘race’ is a social construct where phenotypic attributes are popularly used to denote in-groups from out-groups. Since there is no sound biological or sociological foundation for its use in an analytical sense one should treat it as no more than a special case of ethnicity. Hence, when the term ‘race’ is used in a popular discourse it cannot refer to a ‘sub-species of Homo sapiens’ (van den Berghe, 1978: 406) but is applied only as a social attribute.

Secondly, since it was commonly acknowledged that the classics of sociological thought had little or nothing to say about ethnicity, sociologists had to turn to anthropology and in particular to the seminal work of Frederik Barth (1969) in order to explain the power of cultural difference, both historically and geographically. Before Barth, cultural difference was traditionally explained from the inside out—social groups possess different cultural characteristics which make them unique and distinct (common language, lifestyle, descent, religion, physical markers, history, eating habits, etc.). Culture was perceived as something relatively or firmly stable, persistent and definite. Cultural difference was understood in terms of the group’s property (for example, Frenchmen have possession of a culture distinct from that of Englishmen). Barth’s work provided nothing short of a Copernican revolution in the study of ethnicity. Barth put traditional understanding of cultural difference on its head, that is he defined and explained ethnicity from the outside in: it is not the ‘possession’ of cultural characteristics that makes social groups distinct but it is the social interaction with other groups that makes that difference possible, visible and socially meaningful. In Barth’s own (1969: 15) words: ‘the critical focus of investigation from this point of view becomes the ethnic boundary that defines the group, not the cultural stuff that it encloses.’ The difference is created, developed and maintained only in interaction with others (for example, the Frenchness is created and becomes culturally and politically meaningful only through the encounter with Englishness, Germanness, Danishness etc.) Hence, the focus in the study of ethnic difference has shifted from the study of its contents (for example, the structure of the language, the form of the particular costumes, the nature of eating habits etc.) to the study of cultural boundaries and social interaction. Ethnic boundaries are explained first and foremost as a product of social action.

Thirdly, Barth’s research set a foundation for understanding of ethnicity in universalist rather than in particularist terms. Since culture and social groups emerge only in interaction with others, then ethnicity cannot be confined to minority groups only. As Jenkins (1997: 11) rightly argues, we cannot study minority ethnic groups without studying at the same time the majority ethnicity. The dominant structural-functionalist and modernist paradigm of post-Second World War sociology has traditionally viewed ethnicity as a parochial drawback from the past that will largely disappear with intensive industrialization, urbanization, universal national education systems and modernization (Parsons, 1975). Ethnic difference was understood in rather narrow particularist terms. But if ethnicity is understood more generally in terms of social interaction, culture and boundary maintenance, then there is no culturally and politically aware social group able to create a credible narrative of common descent, without ethnicity. In other words, as long as there is social action and cultural markers to draw upon (for example, religion, language, descent etc.), there will be ethnicity. And this is indeed where sociology comes into play.

Although Barth has provided a groundwork for the elementary understanding of ethnicity his approach fell short of accounting for political and structural repercussions in the organization and institutionalization of cultural difference. Why, when and how do individuals and groups maintain the ethnic boundaries? In trying to explain these questions post-Barthian sociology has drifted in different directions. Rational choice theory focused on individual motives and choices (Banton, 1983: Hechter, 1992). Viewing individuals as utility maximizes who struggle over limited resources, rational choice sociologists believe that ethnicity is no more than an advantage that can be used for individual gain. Speaking the same language, sharing the religious tradition, myths of common descent or any other form of cultural similarity help actors unite, making the price of collective action less ‘expensive.’ Michael Hechter (1992) argues that ethnic groups maintain their inter-group solidarity in two principal ways: by providing benefits to their members and/or by restricting and sanctioning their individual choices to prevent ‘free riding.’ Hence collective action on an ethnic group basis is most likely when individuals can benefit from it or when they fear sanctions from alternative behavior. Although successful in emphasizing the dynamic and manipulative quality of ethnicity, this approach has been criticized, among other things, for neglecting the structural conditions under which individual choices are made (Malešević, 2002b).

Working within the similar economistic tradition, neo-Marxist approaches emphasize what rational choice theory neglects—the structural determination of ethnic group behavior: the state’s role in reproducing and institutionalizing ethnically divisive conditions, the function of racist ideology in preventing working class unity or the relationship between economic inequality and ethnic identity (Miles, 1984). While traditionally Marxists have analyzed ethnicity as an ideological mask that only hides class antagonisms focusing almost exclusively on the capitalist modes of production, contemporary neo-Marxism is much more sensitive to autonomy of the cultural sphere. Recognizing limits of class analysis, contemporary Marxism (Solomos and Back, 1995) attempts to widen its analysis of ethnicity by directing its attention to the new social movements and identities other than class (Anthias, 1992). However, these are still, just as in rational choice theory, couched in antagonistic, economist terms where ethnicity remains a second order reality, a tool of exchange and coercion.

Symbolic interactionist perspectives are overtly critical of such a view. Blumer and Duster (1980), Lal (1995) and other interactionists argue that social action is often more symbolic than economic and that ethnicity can most adequately be studied and explained by focusing on the individual and collective subjective perceptions of reality. In this perspective ethnicity is analyzed as a social process through which individuals and groups acquire, maintain, transform or change their ‘definitions of situation.’ In Lal’s words (1995: 432) the perceptions of ethnic ties are ‘influenced by the situation in which we find ourselves, the presence of real or imaginary significant others, and “altercasting” as well as positive or negative value we assume a particular identity will confer in a particular context.’ Ethnic groups operate through the ‘collective definition of situation’ on the basis of which they participate in the ongoing processes of interpretation and reinterpretation of their experiences (Blumer and Duster, 1980: 222). As often stressed by interactionists, objective unequal distribution of economic rewards or political power between the ethnic groups does not necessarily result in group conflict. It is rather the nature of their mutual symbolic interpretations and collective perceptions that determines inter- and intra-group relations.

The view that human beings are predominantly symbolic, cultural creatures who create their own worlds of meanings has been put under scrutiny by sociobiologists. Sociobiology starts from a simple and apparent fact that humans are made of flesh and blood, that they need to eat, drink, sleep and copulate, which are features shared with the rest of the animal kingdom. Culture is regarded as important but is seen as being subordinate to nature since it has developed from nature and is dependent on changes in nature. According to sociobiologists, just as animals, humans are genetically programmed to reproduce their genes. When direct reproduction is not possible one will reproduce indirectly—through kin selection. P. van den Berghe (1981) has persistently argued that ethnicity is no more than an extension of kin selection. Ethnic groups are defined by common descent and are seen as being ascriptive, hereditary and generally endogamous. Since ethnic nepotism has biological origins, it is argued that ‘those societies that institutionalized norms of nepotism and ethnocentrism had a strong selective advantage over those that did not’ (van den Berghe, 1978: 405). Sociobiology is the only sociological tradition that explicitly takes a primordialist stance in the explanation of ethnic relations. Its view that ethnic groups are biologically determined for in-group favoritism has been subject to the critique of most other research traditions, but power elite theory has provided the most sustained criticism of primordialist positions.

Power elite approaches argue that what is crucial for understanding of ethnic relations is focus on human beings as political rather than biological animals. Brass (1994), Cohen (1981) and others speak of ethnicity in instrumentalist terms. Nevertheless, this instrumentality is not of an economic (as in rational choice theory) but rather of a political nature and it focuses more on the role of individuals and groups in positions of power than on the randomly picked utilitarian agents. Power elite theories are developed around the two spheres of human activity—power (politics) and symbolism (culture). Their argument is that cultural markers are for most of the time arbitrary and what matters in ethnic relations is how, when and by whom can these symbols be manipulated to mobilize social groups. Symbols are considered to be powerful mechanisms of elite control because of their ambiguity and emotional intensity. In this perspective conflicts based on ethnicity are explained as something that ‘arise out of specific types of interactions between the leaderships of centralizing states and elites from non-dominant ethnic groups especially, but not exclusively, in the peripheries of those states’ (Brass, 1994: 111). Although clearly able to accommodate some propositions of symbolic interactionism (symbolism), rational choice theory (instrumentality of cultural markers) and neo-Marxism (unequal position of social groups), this position has been criticized for treating ‘masses’ in a passive, conformist and submissive way and for neglecting the study of motives and values behind the ethnic mobilization.

The approach that is most sensitive to the criticisms raised above is a Weberian approach to ethnicity (Collins, 1999; Jackson, 1982/3; Stone, 1995). In fact, contrary to the commonly held view, Weber has provided a fairly developed and articulated theory of ethnicity. Moreover, Weber provided a definition and analysis which allowed for a non-essentialist view of ethnicity long before Barths path-breaking study. If one reads Weber properly, it is possible to see that Weber did not conceive ethnicity in terms of a ‘group property’ but rather in terms of social action. Following his ideal-type methodology Weber perceived all social groups as quasi-groups, emphasizing their amorphous and dynamic potential. In the same way ethnicity is understood as a potential social attribute not as an actual group characteristic. Weber defined ethnicity in terms of two key factors—(a) a belief of social actors in common descent based on cultural differences and (b) a political action through which this belief becomes socially meaningful (1968: 385-98). What is crucial here is his view that ‘it is primarily the political community, no matter how artificially organized, that inspires the belief in common ethnicity’ (Weber, 1968: 389). Hence, this position anticipates Barth’s emphasis on boundaries and even goes a step further, accounting for a group mobilization and linking it to some propositions raised by power-elite theories. Furthermore, by introducing the concept of ‘monopolistic closure’ Weber’s theory of ethnicity has room for economic instrumentality broader than rational choice type of analysis. Weber argued that ethnicity can often be explained by looking at how individuals tend to close relationships by using ‘any cultural trait’ to ‘ensure economic opportunities’ for their group. This monopolistic social closure of groups ties well into the symbolic interactionist emphasis on symbolism since the Weberian approach stresses the link between ethnicity and status. Ethnicity often becomes a mechanism for the monopolization of status honor since the sense of an ethnic group’s honor is rooted in a belief of the group’s superiority. As Weber has shown, quite often low economic group standing is coupled with high ethnic group status and vice versa (for example, white manual workers vs. ‘blacks’ in the United States, Fijians vs. Indians in Fiji, Serbs vs. Albanians in Kosovo). These inter-group relations can also undergo swift transformation with the advent of charismatic personalities who are often able to draw on the power of emotional and value-rational social action to initiate dramatic social change. The link between charismatic authority and value rationality is key for understanding the power of popular appeal that ethnic nationalism can quickly generate (Malešević, 2002a). In this way Weberian tradition is able to explain individual and group motives behind the ethnic mobilization. The greatest advantage of Weberian tradition over its competitors comes from a simple but crucial idea that although a universal sociological theory of ethnicity is possible, there is a multiplicity of ‘ethnic situations.’ Ethnicity can overlap with status, class, legal or political rights or with caste. As Rex (1986: 14) points out, ethnic groups ‘may be arranged in a hierarchy of honor, they may have different legal rights and they may have different property rights.’ Weberian tradition is the most systematic and synthetic approach that anticipates the original Barthian argument which explains ethnicity through the social (inter)action.


The distinction which has had the longest intellectual career within the theory of nationalism is that between civic and ethnic nations. This was first introduced by Hans Kohn (1967) and it has recently been given new life by Rogers Brubaker (1992) in an impressive comparative study of French and German citizenship laws. The distinction contrasts French and American nationalism with, above all, those of Eastern Europe. In the former one can become a citizen easily, by accepting local laws and customs—with citizenship being given as of right to anyone born on the territory of the state. In the latter, citizenship rights are reserved to those of a similar ethnic background. Although the situation has just changed, a clear example of this latter situation was Germany’s acceptance of ethnic Germans from Eastern Europe and Russia who could not speak the language—and its near refusal to give citizenship to Turkish workers, even though they could speak German and had quite often lived the whole of their lives within that country. It is not surprising that civic has come to be associated with good, and ethnic with nasty—a view particularly clearly articulated by Eric Hobsbawm (1990).

There has been conceptual advance beyond this stark binary opposition. To begin with, we should not accept everything that is implied in the formula ethnic/bad, civic/good. There is nothing necessarily terrible about loyalty to one’s ethnic group, whereas civic nationalism is not necessarily nice: its injunction can be ‘join us or else.’ Of course, ethnic nationalism is indeed repulsive when it is underwritten by relativist philosophies that insist that one should literally think with one’s blood. Further, civic nationalism becomes more liberal when it moves towards the pole of civility, best defined in terms of the acceptance of diverse positions or cultures. Whether this move is so to speak sociologically real can be measured by asking two questions. First, is the identity to which one is asked to accede relatively thin, that is, does it have at its core political loyalty rather than a collective memory of an ethnic group? Second, are rates of intermarriage high? All this is obvious. Less so, perhaps, is a tension that lies at the heart of multiculturalism. In the interests of clarity, matters can be put bluntly. Multiculturalism properly understood is civil nationalism, the recognition of diversity. But that diversity is—needs to be, should be—limited by a consensus on shared values. Difference is acceptable only so long as group identities are voluntary, that is, insofar as identities can be changed according to individual desire. What is at issue is neatly encapsulated when we turn to the notion of caging. If multiculturalism means that groups have rights over individuals—if, for example, the leaders of a group have the power to decide to whom young girls should be married—then it becomes repulsive. Such multiculturalism might seem liberal in tolerating difference, but it is in fact the illiberalism of misguided liberalism, diminishing life chances by allowing social caging. This view is, of course, relativist, and it is related to ethnic nationalism in presuming that one must think with one’s group. Importantly, the link to ethnic nationalism may be very close indeed. If there are no universal standards, and ethnic groups are held to be in permanent competition, then it is possible, perhaps likely, that one group will seek to dominate another.

If these are ideal typical positions, a powerful stream of modern social theory in effect suggests that some have greater viability than others. A series of thinkers, interestingly all liberal, have insisted that homogeneity, whether ethnic or civic, is a must’ if a society is to function effectively. John Stuart Mill made this claim when speaking about the workings of democracy, insisting that the nationalities question had to be solved in order for democracy to be viable (Mill, 1975). The great contemporary theorist of democracy Robert Dahl has reiterated this idea (Dahl, 1977). The notion behind all this is straightforward. Human beings cannot take too much conflict, cannot put themselves on the line at all times and in everyway. For disagreement to be productive in the way admired by liberalism, it must be contained—that is, it must take place within a frame of common belonging. Very much the same insight underlies David Miller’s view that national homogeneity is a precondition for generous welfare regimes (Miller, 1995). This is correct: the generosity of Scandinavian countries rests on the willingness to give generously to people exactly like oneself. But the great theorist of the need for social homogeneity was of course Ernest Gellner. As it happens, the explanation he offered for this ever more insistently—that of the necessity of homogeneity so that industrial society can function properly—is rather question-begging. But even the most cursory consideration of his life suggests that he captured something about the character of nationalism. Born into Kafka’s Czech-German-Jewish world and forced into exile in 1939, he returned in 1945 to find the Jews murdered and the Germans being expelled. A second period of exile ended when he returned when communism fell—to witness on that occasion the secession of the rich majority from the Slovaks. Visceral experience underlay his image of political space moving from the world of Kokoschka to that of Mondrian—that is, from a world in which peoples were intermingled to one in which national homogeneity was established (Gellner, 1983: 139-40).

The claim of those variously stressing the need for homogeneity amounts to saying that we are very unlikely to have civil nationalism, that is, that multinational entities are an impossibility. This is to say that constitutional schemes—federal, confederal and consociational—from which civil nationalism hopes so much are very unlikely to work. That has certainly been a key part of the experience of Europeans, as we shall see, for this has been the dark continent of modernity, with homogeneity being achieved through repulsive means—through population transfers, ethnic cleansing and genocide much more than by voluntary assimilation (Mazower, 1998). The key analytic question within the theory of nationalism—a question with immediate consequences for ethnicity and citizenship—is whether civil nationalism is a realistic social possibility. As noted, this leads to a further question, that of whether the rest of the world follows the European example. If so, the future of world politics looks set to bring us catastrophe, given the complex ethnic intermingling of many states, particularly some of those in the developing world.

As it happens, there is a counter-argument to the pessimistic view associated with Gellner’s predominantly socio-economic causation. Advances in sociology suggest that the character of social movements results overwhelmingly from the nature of the state with which they interact. This political sociology may well apply, as noted, to working class behavior. Liberal states that allowed workers to struggle at the industrial level avoided creating politically conscious movements; in contrast, authoritarian and autocratic regimes so excluded workers as to give them no option but to take on the state. This general notion—that the barricades are so terrifying that reform is habitually more attractive than revolution—has very large applications. The case against Gellner is that the politics explain nationalism as much or more than socio-economic factors. More particularly, the secessionist nationalism privileged by his definition of nationalism results more from a reaction to the authoritarianism of empires than from the social inequality faced by a culturally distinct group. Liberalism before nationalism may allow for containment, that is, respect for historical liberties might allow multinational frames to exist.

It is important to stress here that it is liberalism which is at the core of the position that stands as an alternative to Gellner’s sociology—for we should not uncritically romanticize democracy. Tocqueville long ago pointed out that majorities could in theory be tyrannical. Whether he was correct or not about the United States, there can be no doubt that in numerous instances—for example, Protestant hegemony in Northern Ireland from 1922 to 1969—democracy has been exercised freely and fairly, and at the expense of minorities. More generally, democratic participation is not always a good in and of itself, despite the recent vogue for civil society and civic virtue. This suggests an equally important corollary. Bluntly, democracy matters less than liberalism. Liberal regimes may achieve very great stability by diffusing various conflicts through society rather than concentrating them at the political center. Pure democratic participation will destabilize unless it is channeled through social institutions which tend to contain, manage and regulate conflict. The Balkan Wars of the past decade have demonstrated that democratization does not necessarily bring peace and prosperity. However, the collapse of communism did not lead to violence in every instance, suggesting that attention be given to two variables (Snyder, 2000). First, political leaders who imagine that a new world can only bring their downfall may well be tempted to play the nationalist card in order to stay in power (for example, Miloševic vs. Klaus). Secondly, democracy may well lead to violence if it lacks the institutional framework that allows it to control its passions, that force it to reflect. Snyder stresses in this context that democratization clearly leads to violence when news comes from a single authority. And all this is to say that in our own time a multinational state, even with the benefits of the purported lessons of the past, utterly failed to successfully transform itself.

The paradox at work can be underscored. The presence of institutions of conflict regulation can shape and channel, even perhaps tame newly emergent popular pressures. In contrast, authoritarian regimes are likely to create social movements armed with total ideologies. The contrast is between societies in which liberalism came before democracy and those in which democracy came before liberalism. Our position as a whole is that of Tocqueville (1955), the central tenets of whose masterpiece are that liberalism and authoritarianism are self-perpetuating. We need not be quite so pessimistic, for some authoritarian regimes have become liberal democracies, but we should be aware how difficult is that transformation—and that the advent of democracy does not necessarily entail sweetness and light.

State and Nation in Europe

Although it is important to note that nationalism was associated with horror in European history, cognitive advance depends upon explaining why this was so. After all, in the middle of the nineteenth century Europe was at the pinnacle of its power, confident that it represented progress. The European balance of power depended on the interactions of Austro-Hungary, Wilhelmine Germany, Imperial Russia, Great Britain and France. The fate of the Ottomans was very much part of the mental world of these great powers; the position of the United States came slowly to assume great significance, especially for Great Britain. If all this suggested ebbs and flows of power and influence, no hint was present that this was the scene for a new, great Peloponnesian War—a conflict so visceral that it knocked Europe off the perch that it held briefly as the leader of the world. What were the essential contours of this conflict? Further, does understanding these variables allow us to suggest that the link between nationalism and nastiness was contingent rather than absolutely necessary?

The rivalry between these states was such that the most immediate structural element at work was that of the need to industrialize. An obvious consequence that troubled ruling elites was the emergence of working classes. In fact, a whole series of sectoral divisions amongst workers meant that no unitary class existed inside a particular state, let alone between them—at least when workers were left to themselves. Extreme repression of radicals combined with liberal treatment of the rest famously created in the United States a world in which workers began to consider themselves as middle class. Something of the same pattern had put paid to the Chartists in England, but the presence of some, albeit very limited, state interference—that of the Taff Vale court decision which for a short period prevented union organization—ensured that class loyalty was created, that is, socialism was avoided but a Labour Party was created. In contrast, regime exclusion did create socialist class unity. Anti-socialist laws in Wilhelmine Germany created a movement with political and industrial wings, formally wedded to revolutionary ideas but in fact made reformist by the speedy abolition of the laws in question. In Imperial Russia autocracy differed from authoritarianism in being at times even more suspicious of capitalism (McDaniel, 1988). The fundamental factor at work was regime policy. Militancy varied precisely in relation to state actions: reformists came to the fore as the result of the political opening of 1905, whilst revolutionaries triumphed inside the movement once concessions were abandoned. The end result of these policies was the creation of the only genuinely revolutionary working class in human history. In a nutshell, the historical record does indeed support the political sociology of class that was outlined at the start.

To consider industrialization only in terms of its impact on class would be a mistake. Every state sought an exactly similar set of industries in order to maintain its geopolitical independence, and this in turn led to economic tensions. The importance and character of imitative industrialization is captured in the marvelous demonstration by Gautam Sen (1984) that every industrializing state in the nineteenth century sought to have the same basic portfolio of heavy industries—so as to ensure its capacity for geopolitical independence, that is, its ability to produce its own weapons. Differently put, states interfered with markets. In this context, the elements of historical sociology that concern us here revolve around three factors that explain the nature of Europe’s twentieth-century disaster. Each factor can be seen as an extension of the beliefs of Max Weber, namely his visceral nationalism, his commitment as a Fleet Professor to an imperial policy, and his insistence that the empire’s conduct of German foreign affairs was disastrous. And it should be said clearly that these factors were at work in all the countries involved.

First, developmental states characteristically felt weak when they ruled over a mass of different ethnic and national groupings. For one thing, Britain seemed to gain strength from its homogeneity- although this perception faded once Home Rule politics made it clear that Britain was in its way as composite a state as were other empires. But the determination to copy the ethnic homogeneity of leading European powers had a further element to it, namely that of seeking to strengthen the legitimacy of the state by playing the national card against socialism. Accordingly, nationalism comes to the fore at the end of the nineteenth century as much from above as from below. Perhaps curiously, nationalism had not been enormously successful in the years before 1914. Geopolitical interference stood behind the cleansing of perhaps 5 million Muslims from the new Balkan nation-states (Mazower, 2000). This suggested of course that the stakes of any general conflict, should it occur, might well be very great indeed (Kaiser, 1990: pt 4). But as long as the balance of power remained in operation, nationalism had great difficulty in breaking the established mould of state borders. A clear contrast can be drawn between the logic of the situations facing different empires (Lieven, 2000). Austro-Hungary quite simply had no chance to become a modern nation-state: the dominant ethnicity was simply too small to serve as a Staatsvolk. What evolved in consequence was a situation, in Count Taaffe’s words, of ‘bearable dissatisfaction’ (Lieven, 2000: 191). If the Magyars were content, the Slavic nations within the Austrian half were not terribly treated—for all that they hoped that the monarchy would move towards greater constitutionalism. Demands were contained, however, by a clear awareness of geopolitical realities. As early as 1848 the Slavs had realized that to become small but unprotected nations was to risk annihilation should Germany or Russia be drawn into a power vacuum.

The second factor can usefully be introduced by saying, again, that nationalism is an essentially labile force, able to connect with and deeply influenced by the social forces of any particular historical moment. The reference to Max Weber as Fleet Professor brings to attention the crucial fact that nationalism was, in this period, linked to imperialism. There is a sense in which Weber himself should have known better. As Adam Smith had stressed long ago, colonies could be more of a millstone than an advantage. But it is very often the case that what matters socially about economics is less the facts in and of themselves than what people believe to be the facts. In this case, imperial dreams had a very considerable rationale. When Lord Roseberry admitted that the British empire did not pay at the time, he went on immediately to say that it might none the less be absolutely necessary in the longer run.

It is the third factor, the nature of foreign policy-making inside imperial courts, to which attention must be given for an explanation for the breakdown of order that then allowed nationalism and imperialism to cause disaster. A preliminary, scene-setting point is simply that the late nineteenth-century European great powers were engines of grandeur, whose leaders habitually wore military uniform. The difficulty that such rulers faced, however, was that making foreign policy was becoming ever more difficult. Jack Snyder has usefully suggested that foreign policy-making tends to be rational when states are unitary (Snyder, 1991; cf. Mann, 1993: ch. 21). Examples of such rational states include the rule of traditional monarchs, the collective domination of a revolutionary party so much in control of a late, late developing society as to have no fear of popular pressure, and the checks and balances on foreign adventures provided by liberal systems. In contrast, late developing societies—which combine authoritarianism with genuine pressures from a newly mobilized population—tend to lack the state capacity necessary to calculate by means of realist principles.

The First World War was not a Clausewitzian affair, in that statesmen lost control of policy-making. Industry applied to war in part explains this, but still more important was the fact that a war of peoples needed justifications other than the merely dynastic or territorial. The chaos that resulted exhausted the European fabric. It was this factor which made the peace treaty disastrous. The lack of genuine geopolitical agreement encouraged the politics of economic autarchy. The failure to solve the security dilemma cemented the link between nationalism and imperialism. This was the world of Hitler and Stalin, of the horrors of ethnic cleansing, population transfer, mass murder, and of total war between the two great revolutions of modernity.

The First World War had ended badly despite the making of formal treaties. In contrast, the Second World War ended well without formal agreements. What mattered most of all was consideration given to power politics, that is, the creation of a secure frame within which economic and social forces could then prosper. Spheres of influence were established between two great superpowers which very rapidly came to understand each other extremely well, not least because the presence of nuclear weapons forced them to be rational. Nationalism was ignored, stability created. There were two elements at work in the reconstitution of Europe (Maier, 1981; Ruggie, 1982). Europeans themselves made a very major contribution. Fascism was thoroughly discredited, beaten in its own chosen arena of military valor. More particularly, French bureaucrats, aware of the devastation caused by three wars with Germany within a single lifetime, effectively changed France’s geopolitical calculation. If Germany could not be beaten militarily, it could perhaps be contained through love. The origin of what is now the European Union came from a decision by the two leading powers to give up their geopolitical autonomy, by establishing genuine interdependence in coal and steel—that is, in giving up the capacity to make their own weapons. This move was made possible by the second factor, the presence of American forces. Europeans of course did a great deal to pull Americans in—with Lord Ismay famously arguing that foreign policy should seek to keep the Americans in, the Russians out and the Germans divided.

As Milward (1992) explains, European states had sought, between 1870 and 1945, to be complete power containers, unitary and in possession of markets and secure sources of supply. The fact that this led to complete disaster produced humility—which is not to say for a moment that state power somehow lost its salience. Rather, states discovered that doing less proved to give them more, that interdependence within a larger security frame allowed for prosperity and the spread of citizenship rights. Differently put, breaking the link between nationalism and imperialism enhanced rather than undermined state capacity. However, liberalism in Europe, from the Atlantic to Ukraine, and including most of southeastern Europe, is made easier because very great national homogeneity has been established, in largest part thanks to the actions of Hitler and Stalin.

The fundamental change in geopolitical realities after the collapse of the Soviet bloc certainly played a part in key developments within the European Union, most notably that of binding Germany within Europe by avoiding any German economic hegemony through the Bundesbank. Still, continuities are more important than new developments. For one thing, this liberal democratic league has the capacity, not least given that one cannot be a member without respecting minority rights, to consolidate liberal democracies in Central Europe just as it did in Southern Europe a generation ago. For another, statist calculations remain at play: the Franco-German condominium survives, whilst French determination to balance Germany has led it virtually to rejoin the NATO command structure. Perhaps most important of all, there is no sign of fundamental change to the rules of the geopolitical game. The mere sign of worry, let alone any threat of withdrawal, on the part of the United States has seen Europeans own up to the fact that they wish the American presence to continue, despite its varied imperfections.

Fascism had been defeated in the hottest of wars. In contrast, the Cold War ended with a whimper. The period since 1989 has made crystal clear that the Soviet developmental model was deeply flawed. For one thing, whatever the benefits of initial heavy industrialization and social modernization, there is now no doubt that the absence of market mechanisms doomed Soviet style economies to waste and inefficiency. Socialism as a power system had sought to establish its own channels of control, thereby in effect continuing Tsarism’s distrust of independent civil society. When power was absolute, command-administrative methods had great force. Once softer political rule came to the fore, it became obvious that force was linked to rigidity. If the lack of flexibility caused problems, the inability to decompress—that is, the inability of socialism to emulate some authoritarian capitalism regimes in liberalizing from above—resulted from another facet of an atomized society, bereft of social institutions. Liberalization processes depend upon the striking of bargains, often in some round-table negotiations. Gorbachev’s difficulty was that there were no leaders of independent organizations, able to control their members, with whom he could negotiate (Bova, 1991). In these circumstances, controlled decompression was impossible. Democratization took the place of liberalization. For another, the national question can be seen to have occupied the new political space, and in such a way as to put the final nail in a social world presumed until very recently to be powerful and permanent. The reconstitution of the empire by 1921 and its expansion in 1939 and in the years from 1944 presented problems with which the Tsars would have been all too familiar. Several systems of rule were again contained within a single political umbrella, with the greatest difficulties again coming from the inclusion of advanced Western nations whose consciousness was so advanced as to make assimilation impossible. The situation was in fact worse than it had been for the Tsars: the Baltic states and Poland had tasted independence, the Czechs knew that socialism was taking away their industrial lead, whilst a united Ukraine, freed from fear of Poland and Germany, concentrated all its ire on Russia. But if the empire became an expensive burden, it is important to remember that the nationalities did less to cause the breakdown of the Soviet bloc than to make sure that reconstitution would be impossible. A political opening increases noise. Nerve is required to put up with new pressures, so that discontents take a normal form—from revolution to reform. The worst move in such circumstances is—what Gorbachev did—to step backwards, to make the newly vocal fear and thereby to confirm them in their suspicion of the continuity of an old regime. The interventions in Georgia and Lithuania were accordingly utterly disastrous. Yeltsin was given the cards by means of which he was able to destroy the Soviet Union. Rarely has a great power fallen so far, so fast.

The American Melting Pot

The discussion of nationalism in the abstract suggested that civic nationalism was not necessarily as liberal as its defenders imagined. This insight certainly helps us to understand the Leviathan of the contemporary world. Bluntly, the national experience of the United States is not as different from that of Europe as it would like to believe. Civic nationalism in America has encouraged a melting pot, homogenizing the many into a single unit. Differently put, the United States is not a social world favoring diversity. An initial consideration to that effect lies in the simple fact that white Anglo-Saxon settlers more or less exterminated the native population, thereby establishing their own hegemony. Further, the creation of the new state placed a very strong emphasis on uniformity. For one thing, a Constitution was formed, a singular set of ideals created, which thereafter was held to be sacred. For another, the United States was created by means of powerful acts, usually directed from below, of political cleansing. A significant section of the elite—in absolute numerical terms larger than those guillotined during the French Revolution, and from a smaller population at that—that had supported the Crown was forced to leave (Palmer, 1959: 188-202).

Perhaps the most striking general interpretation of American history and society, namely that proposed by Seymour Martin Lipset (1996), is that which insists on the power of these initial ideas, of continuity through continuing consensus. That is not quite right. If some alternatives were ruled out at the time of foundation, others were eliminated as the result of historical events. The two most important examples deserve at least minimal attention. First, we ought to remember that the United States remained unitary only as the result of a very brutal civil war. The Constitution had of course recognized the different interests of the slave-owning southern states, but the difference between North and South grew in the early years of the republic. War destroyed that diversity, with Lincoln trying at the end of the conflict to create unity by means of such new institutions as Thanksgiving. Of course, the South did not lose its cultural autonomy simply as the result of defeat in war, maintaining a key hold on federal politics well into the 1930s. None the less, over time the South has lost its uniqueness, especially in recent years as the result of political change and of population and industrial transfers from North to South. Since no one wants a second civil war of visceral intensity, there is no possibility of the United States becoming a multinational society. The second alternative vision was that of socialism, in one form or another. Revisionist history makes it equally clear that there was a genuine socialist stream of ideas and institutions in American history, represented most spectacularly in the militant unionism of the International Workers of the World. Further proof of the strength of working class activism can be found in the bitterness of labor disputes—whose end result was a very large number of deaths, second only to those at the hands of the late Tsarist empire (Mann, 1993). This is all to say that American ideals of individualism and enterprise were not so powerful or so widely shared as to rule out a challenge. Their ascendancy came about for two fundamental reasons. On the one hand, the fact that citizenship had been granted early on meant that worker dissatisfaction tended to be limited, to be directed against industrialists rather than against the state—thereby limiting its overall power. On the other hand, and crucially for this argument, socialism was literally destroyed—as is made apparent by that very large number of working class deaths. The recipe for social stability, which worked in the United States, is often the combination of political opening with absolute intolerance towards extremists.

It would be a mistake to leave matters at this point. For the rosier and milder face of the coin of American homogeneity can be seen at work in American ethnic relations. With the clear exception of African Americans, for the majority of Americans, ethnic identity is now, as Mary Waters (1990) makes clear, a choice rather than a destiny imposed from outside. Rates of intermarriage are extremely high, not least for the first generation of Cuban Americans in Florida, more than 50 per cent of whom marry outside their own group. Ethnic identity has little real content. It is permissible to graduate from kindergarten wearing a sari as long as one does not believe in caste—that is, as long as one is American. There are severe limits to difference, but similarity is now often achieved by much more civil means. The powers of homogenization in the United States, deriving as much from Hollywood and consumerism, of course, as from the factors examined here, remain intact. The melting pot still works, but it does so in a far more benign manner.

Splendours and Miseries of the South

It only takes a moment to think of issues in the South affecting the transformation of states. It may be that socialist China can manage to transform itself, both because it placed perestroika before glasnost and because it has very largely become a nation-state. More generally, however, the North has washed its hands of the South, much of which could drop off the face of the globe without the purportedly global economy even noticing (Hall, 2000). However, despite all the talk about globalization diminishing the significance of ethnic and national attachments, it seems that the opposite process is taking place. First, one of the consequences of more globalized economies is further expansion in migration from the South to the North. Nevertheless, the new cohorts of migrants differ significantly from their counterparts in the postcolonial era: in an environment of instability and insecurity within a globalizing world, assimilation and full citizenship in a host nation-state is often replaced with alternative forms of political loyalties such as dual citizenships, denizenships or living on the legal margins of the asylum system (non-documented immigrants, runaway deportees). Intensified by the development of modern means of transport and communication (Internet, mobile phones etc.) on one hand and economic stagnation in their home countries on the other, the new immigrants often opt for retention of strong ties with their countries of origin. These political, cultural and financial links are often fostered by the governments in the South, who view their transnational emigrants as key source of ‘remittances, investment capital and votes’ (Itzigsohn, 1999). Secondly, the changing nature of the globalizing economy coupled with the persistence of strong ethnic and national ties with the South creates a situation where new immigrants are less likely to develop a stronger sense of cultural and political membership in the country of immigration. Rather, they are more prone to transnationalism, identifying with ethnic group attachments that cross borders of a particular nation-state (Kearney, 1995). However, one should not overstate this largely economic-centered argument since the technical capacity of the states in the North to control their borders and the movement of people has also dramatically increased. In other words, politics matters as much as economics, if not more. One wonders whether politics can in the longer run be so subject to a new form of international apartheid as is economics. The spread of weapons of mass destruction, especially to states possessed of the fiscal advantages given by the possession of fossil fuel, first presented a crucial problem in the form of Saddam Hussein. It is hard to imagine that his will have been the last challenge, despite America’s much vaunted military revolution.

It is beyond our powers to do more than note the salience of these issues. But the perspective that has been argued does suggest the usefulness of considering the situation of multinational regimes in the South. Given that development seeks in its very essence to copy the advanced, it behooves us to ask whether the South’s twenty-first century will be as dark as that through which Europe has just passed. If there are obvious reasons to fear, there are—remarkably—reasons for optimism. It should be said immediately that the hope in question is not mere wishful thinking, not the placing of hope above analysis.

Some regimes in the developing world have managed multinationalism far better than did Europe. A general background condition was an initial realization in some quarters that imagination was needed so as to avoid disaster. It was precisely because African borders were absurd that it was, Julius Nyerere argued, essential to maintain them. Equally importantly, few states have an ethnic group of sufficient size that it is able to even imagine complete domination of the territory- there being, for example, perhaps 120 different ethnicities in Tanzania. Politics are therefore pushed towards multinationalism for structural reasons. These circumstances have bred a remarkable substantive achievement, that of the language repertoires of some African states and, above all, of India. David Laitin’s analysis of the Indian situation suggests that a fully capable Indian citizen needs a language repertoire of ‘3 plus or minus 1’ languages (Laitin, 1992). Two languages are needed to begin with because India has two official languages, English as well as Hindi—for Nehru’s desire to produce a unitary and monoglot society was stymied by the desire of civil servants to maintain their cultural capital, that is, the ability to function in English. A third language is that of one’s provincial state. But one only needs two languages when one’s provincial state is Hindi-speaking. In contrast, one needs four languages when one is in a minority in a non-Hindi-speaking provincial state. India is the most important exception to Gellner’s generalization that homogeneity is a functional prerequisite of modernity. This is a remarkable institutional success story, the creation of an Austro-Hungary that seems to work. And this sort of linguistic arrangement has been complemented in many parts of the developing world by a varied collection of agreements, habitually consociational and regional, which have allowed ethnic groups to survive within a single shell. The complex case of Malaya is a prime case in point (Horowitz, 1986).

Language is of course only one of the markers that can be used as the basis on which to homogenize peoples into a single nation, and one can always fear—though not, to this point, excessively—that religion could again serve as the basis for terrible ethnic cleansing in India. It is worth remembering in this context that the full impact of ethnic superstratification is felt during the process of modernization—which is by no means complete in most of the world’s polities. If hope has some descriptive base, the fact that there have been many failures of multinational federations—from Yugoslavia and the Soviet Union, to the Caribbean, sub- Saharan Africa and British Central Africa—should make us realize how very hard it is to make such arrangements work. Still more obvious are the genocidal horrors of Kampuchea and Rwanda—in which other peoples behave as did we Europeans in the very recent past. It is hard to imagine that such actions, now visible on our television screens, will not have any effect on the condition of those who inhabit the more comfortable zones of the world.


Our intent in this chapter has been to sound a cautionary note. When a society develops institutions to regulate conflict early on, then it is likely that emergent popular forces will be absorbed within a liberal mould flexible enough to tame and contain them. In contrast, democratization occurring before the advent of liberalism is likely to lead to social disruption, sometimes of the most repulsive sort. Establishing liberal institutions in the midst of fundamental social and political change is very difficult indeed. The world remains a very dangerous place—one in which nationalism may continue to cause disaster. For the characteristic political form of modernity remains that of the nation-state, whose character does indeed revolve around a good deal of homogeneity. When one remembers the amount of violence involved in creating such entities, one must fear for the future of the world. But this is an area in which a measure of hope is permissible, given the inventiveness of non-Europeans. Differently put, we can hope that they will not copy us. And it would be a terrible mistake to imagine that nationalism is now a problem for others, rather than for our own advanced countries. Brendan O’Leary (2000) has recently pointed out that federalism works best when it has at its core a demographically dominant Staatsvolk—the idea being that a ruling people, secure in its position, will be perfectly prepared to allow federal concessions. In the absence of such demographic dominance, federalism only works when consociational measures are added, so as to join different communities. Given that Europe, like Austro-Hungary, simply does not have enough Germans, the European Union would be well advised to retain all the consociational deals that reassure small states—as well as to find ways to give representation to such stateless nations as Catalonia and Scotland. Getting institutional design right even in an economically advanced and politically liberal Europe chastened by memories of its horrible past is going to be very difficult indeed.