An ethnic group is a distinct category of the population in a larger society whose culture is usually different from its own. The members of such a group are, or feel themselves, or are thought to be, bound together by common ties of race or nationality or culture. The nature of an ethnic group’s relationships with the society as a whole, and with other groups in it, constitutes one of the main problems in describing and analyzing such societies. As Ruth Benedict said of race conflict, it is not race that we need to understand, but conflict; so, for an understanding of ethnic groups in a social system, it is not on racial or cultural differences that we need to focus our attention, but on group relations.
The existence of distinct ethnic and cultural groups within societies is widespread and ancient and occurs at most levels of culture, ranging from the Bushmen of the Kalahari, who live within the framework of Tswana society, to modern Europe and America. Ethnic groups in the Near East were recorded by Herodotus almost 2,500 years ago and remained a persistent feature of the Byzantine, the Ottoman, and other Near Eastern empires. Similar situations also occurred in ancient India and in Chinese civilization at all stages of its expansion.
Although scholars in the past have often noted the existence of multiracial and multicultural societies, systematic examination of the sociological consequences of the phenomenon did not begin before the eighteenth century. And then it was principally in connection with the concepts of race and race relations as developed in the next century by writers such as Gobineau (1853–1855) and Chamberlain (1899). Linguistic scholars like Sir William Jones, the Grimm brothers, and Max Miiller not only examined the construction and development of Indo-European languages but also inadvertently encouraged the growth and elevation of the idea of race as an ideology and as the most significant index distinguishing culturally different groups from one another.
Earlier historians, including the writers of the Old Testament, had noted that ethnic groups might be found in a society as a result of the gradual migration of either whole populations or of segments, such as religious refugees, traders, craftsmen, or manual laborers. They also observed that military conquest might bring in its train soldiers and civilians, who either settled permanently in the area or administered their conquests for a period of years before retiring and being replaced from the homeland. Or, again, ethnic groups might be incorporated into a society by altered political boundaries. Sometimes a combination of processes was at work; but however a multiracial or multicultural system came into existence, the types of society in which ethnic groups could be found varied as widely as the processes that brought them into being.
Most investigations of ethnic groups have been made in connection with studies of race relations and stratified societies such as are found in Africa (MacCrone 1937; Patterson 1953), the southern states of the United States (Dollard 1937), parts of the Caribbean (Smith 1955; 1956), Central and South America (Freyre 1933), and in the plural societies of former colonial areas of Asia (Furnivall 1942). Ethnic groups that are not an integral part of a system of over-all social stratification are also found in countries like Switzerland and Nigeria, where they form units in the political system which, although perhaps internally stratified, are not ranked in relation to one another. Other types of multiracial and multicultural situations, as, for example, in northern Laos, Thailand, Burma, and India, have as yet hardly been examined. Frequently in these countries adjacent villages, or even sections of one village, may be linguistically and culturally different and yet be held together in a traditional system of social relations that is not part of the apparatus of a central government (Leach 1954). Similar conditions have been observed, although seldom analyzed, in the Indonesian archipelago, New Guinea, and parts of Africa.
Definitions. At this point it would be wise, for the sake of clarity, to make the distinction between a social group and a social category. By a group, sociologists usually mean an aggregation of people recruited on clear principles, who are bound to one another by formal, institutionalized rules and characteristic, informal behavior. Unless a group is to be no more than a temporary aggregation, it must in addition be organized for cohesion and persistence; that is to say, the rights and duties of membership must regulate internal order and relations with other groups. Members usually identify themselves with a group and give it a name. In practice social groups vary in the degree to which they are corporate; and in certain situations one of the principal difficulties of analysis may be to decide whether a particular social entity is in fact a social group or a mere category of the population, such as red-haired people, selected by a criterion that in the context is socially neutral and that does not prescribe uniform behavior. For any study of group relations this distinction is essential.
In east Africa, the African, the Arab, the European, and the Indian elements of society are closer to being categories of the population than social groups. Although, for example, a fully institutionalized Indian group, recruited from the general category of Indians, is likely to act in the Indian sphere of life, there is no certainty that it will; relations between the ethnic categories may therefore become blurred. The sections of an ethnically and culturally divided population may, according to circumstances, be institutionalized groups related to one another in a system of stratification, or they may be groups living side by side and related in other ways. Ethnic divisions may simply be categories of the population, as are Welshmen and Scotsmen living in England, or Indians, Chinese, and Creoles in Mauritius, who are beginning to lose a sense of ethnic separateness. It is, therefore, always important to be sure what is the exact sociological status of an ethnic or cultural division. Clarity in analysis depends upon it.
The division of society into broad strata, which form a hierarchy of prestige, wealth, and power, is a feature common to most societies and is one that has been used for classification. A few societies, mostly primitive or small in size, may not be stratified; social positions may not be sufficiently numerous or diverse to be easily grouped into strata or aggregates of individuals sharing an equivalent status that would differentiate them from members of other similar aggregates. This is not to say, of course, that statuses in such small societies may not be ranked but is merely to point out that they do not constitute groups. Sociologists traditionally classify the types of stratification as caste, estate, or class systems. As ethnic and subcultural groups may form the basis of a system of stratification, a closer examination of the matter is needed.
In any system of social stratification the following apply: (1) Individuals belong to strata that are groups in the sense that everybody in them shares some obligatory ways of acting that are typically and intentionally different from those in other strata. (2) Strata must be exclusive, so that nobody may belong to more than one at the same time. (3) Strata must be exhaustive, so that everybody in the society belongs to one. (4) Strata must be ranked. In using this criterion, differential access to political and economic resources is taken to be the most significant aspect of the ranking. These criteria do not distinguish different types of stratification, any one of which may be exemplified in a society where ethnic groups are a component element of the system.
Caste groups. One sociological definition of a caste system is that it is a hierarchy of endogamous groups in which status is rigidly ascribed by birth and in which mobility from one group to another is not possible. Correct relations between groups are maintained and validated by religious rules, especially the rule that improper contact between castes produces a state of impurity that entails ritual, legal, and other penalties.
In this definition no careful distinction is usually made between the four-fold division of Indian society into castes (varna) and subcastes (jati). In early literature the varna (priests, soldiers, businessmen, and laborers) are sometimes described in terms of what appear to be ethnic differences; but in historical times they have not constituted more than categories of value against which individuals and members of subcastes could measure their own and other people’s prestige. They were not groups, in the sense of imposing duties that were uniform throughout India. In short, the varna were ranked categories, not stratified groups. Not dissimilar arrangements are also found in some of the multiracial societies of Africa and the Caribbean.
Empirically, Indian society was made up of many small self-contained caste systems, each of which was a hierarchy of subcaste groups. Subcastes in turn were organized so that social labor was divided among them. Each subcaste traditionally held a monopoly of a particular service, so that all washermen, for example, although not bound to that occupation, could prevent others from practicing it. The essence of this division of labor was that it was cooperative and complementary, not competitive. In these small, closed caste systems relationships between individuals tended to be multiple in that two individuals could fill a number of roles in relation to each other. This “summation” or “involution” of roles is an attribute of small-scale and not large-scale social systems (Nadel 1957, pp. 64–72). The argument may be summarized thus: (1) Caste groups must be recruited by birth, that is to say, they must be closed. (2) Relationships between groups must be cooperative, not competitive.
In India subcastes are not usually separate ethnic or cultural groups, but an understanding of caste systems is essential in the analysis of society in Iran and parts of the Near East where a multiplicity of ethnic and cultural groups appears to be organized in small-scale caste systems (Barth 1960). It is possible, too, that society in certain parts of the southern states of the United States is similarly arranged (Dollard 1937).
If we are content to say, as are many students, that South Africa and India both exhibit a caste system, then no further distinctions need be sought. But writers on India do not usually agree that “color-bar” societies are ipso facto caste systems. Although the population of India was always very large, society there was characteristically composed of separate small-scale involute caste systems. South Africa, although less populous, is typical of large-scale Western society where relationships between roles are usually single-stranded and not the multiple ties of a small-scale society.
In certain ways the system of closed groups in South Africa is nearer the model of an estate system than that of a caste or social-class system, but the sociological status of “color-bar” societies needs to be carefully re-examined. Studies of them have for the most part either used the concepts of stratification without careful consideration or have directed attention to economic functions (Boeke 1953) or to attitudes and other psychological concomitants of the existence of ethnic groups. [see CASTE.]
Social class groups. In some places, such as parts of the West Indies, ethnic groups are regarded as, and may in fact be, social classes. Sociologists generally consider a class to be an aggregate of people occupying roughly the same status, which is different from that of people in other classes and which, unlike status in a caste or estate system, allows movement from one stratum to another. It is never easy to decide to what degree a social class is an institutionalized group or exactly how it is related to economic and political status and prestige. When some of the qualifications for membership are also those for belonging to an ethnic category or group, the difficulties of analysis may become very great indeed. An aggregate of people is not a social class just because they think of themselves as one; it is a social class because some activities are obligatory to all or most members and act as a sign that the people form a group and are eligible for access (appropriately graded according to their class) to resources that are valued by the society. When these activities are also qualifications for membership of ethnic or cultural groups, then ethnic and class groups coincide.
The types of stratification that have been mentioned are, of course, models; and a particular system, whether its constituent elements are ethnic groups or not, may not correspond with the model. Racial differences used as insignia or badges to mark off groups from one another are not different in kind from clothing, speech, manners, property, or other cultural emblems that may serve the same ends. But since physical differences are permanent and may be strikingly visible and may also carry much emotion, the understanding of societies such as are found in Mexico, Nigeria, or Kenya has been made difficult by treating them as if they were altogether different from those more familiar to sociologists. The fact that signs of a special kind are used for distinguishing groups in multiracial societies does not mean that such societies are radically different from others.
In a study of early twentieth-century Burma and Netherlands Indonesia, Furnivall argued that countries in which “there is a plural society, with different sections of the community living side by side, but separately, within the same political unit” ( 1956, p. 304) were “a distinctive form of society with a characteristic political and economic constitution” (1942, p. 195). In such a situation, he believed, members of society are unable to develop the common values and demands generated by sharing common institutions. Another writer, M. G. Smith, regards Furnivall’s concepts as essential to comparative sociology, arguing that a plural society is composed of readily identifiable sections held together only by the fact that they are part of one central political system. Such sections, it should be noted, are not necessarily ethnic groups. Each is distinguished by having its own “core” of “basic” or “compulsory” institutions. Social systems may therefore be placed on a scale ranging from those that are fully plural with distinct sections fulfilling particular economic, political, religious, or other functions, to homogeneous systems in which one set of basic institutions is shared by all members (Smith 1960). Models of this kind have attracted anthropologists, historians, and economists, especially those working in multiracial or multicultural areas; but most sociologists have found the concept of a basic core of differentiating institutions even harder to define and handle than the concepts with which they were more familiar, and they have preferred to rely on older and better tested theories of social differentiation. [see STRATIFICATION, SOCIAL.]
Not all societies having ethnic groups within their boundaries incorporate them into a unified system of social stratification, and relations among ethnic groups may (to use a political metaphor) be more of a “federal” nature than one of ranked access to social resources.
Societies with a single administrative system. In Switzerland or Canada, for instance, cultural groupings are clearly differentiated and maintained, and each may be separately stratified; but access to power in the wider society is not limited by either ethnic or cultural origins, nor is it conditioned by a ranked evaluation of ethnic groupings within the society. In Malaya, too, although there are distinct ethnic groups, they are not stratified in relation to one another. The population is divided between Malays and immigrant Chinese and Indians. Constitutionally the machinery of government is in the hands of people selected by voting. The workings of electoral procedures and the staffing of the civil service and the armed forces have, in fact, placed most of the political machinery in the hands of Malays, leaving economic power largely with the Chinese. This distribution of the different kinds of power, with its open possibilities of real loss or gain, has tended to consolidate ethnic categories into political parties (Freedman 1960).
In Mexico, on the other hand, ethnic criteria have, on the whole, been abandoned in the formation of social groups because they no longer mark off differences considered significant. The people who control political, legal, and economic matters attain their positions without reference to race, although not without acquiring the dominant Spanish culture. In Thailand, too, the balance struck between demographic, political, economic, and cultural factors is such that the dominant Buddhist, Thai-speaking group is able to pursue a policy of assimilating ethnic minorities. The policy has had a measure of success with Chinese immigrants, but no attempt has been made to absorb the Muslim, Malay-speaking inhabitants of the southern provinces.
Where ethnic or cultural differences coincide with groups that tenaciously hold different religious opinions, the relationships of one group to another and to the central government may become very complex indeed and lead to serious conflict. But the problem is not one peculiar to ethnically diverse societies. In the sixteenth century, the division of France into Roman Catholic and Protestant Christians did not make it a plural society, even though some Protestants, especially in the south, were culturally and linguistically separate. Nor did it mean that groups that held these different opinions were different in kind from other groups that had been competing for political power before the religious differences arose.
The point here is an essentially simple one. In any society the immediately effective determinants of most social action lie in the political, legal, and economic spheres; and whether or not the main component groups of the society are stratified in relation to one another, an examination of the social system must be primarily concerned not only with relations between groups but also more specifically with those between rulers and ruled. The latter problem is an aspect of the former, but in a society with groups that differ ethnically, whose interests are in fact opposed to one another, the ensuing conflict may be phrased in racial terms and thus provoke more bitter hostility than in other struggles of the ruled against their rulers.
Societies without a single administrative system. In the examples considered so far, ethnic or cultural groups have formed part of an organized administrative and political system. Understanding the significance and range of such groups and their economic, political, religious, and cultural importance has revolved around the problem of ascertaining the exact position that they occupy in the system. In northern Thailand, Burma, and New Guinea, where small ethnic and cultural groups are dispersed and mingled over a wide area without traditionally being under the direct control of a single administrative system, it is also necessary to determine their exact relationship to one another.
A number of such “tribal” groups have been studied, but because anthropological fieldwork tends toward village studies and because the linguistic difficulties of examining such a heterogeneous system are very great, students have seldom made the study of the wider system the main focus of their attention. Groups of this kind appear to be linked to one another in a network of political, economic, ritual, and marriage alliances about which little information is available. But here, as in more politically unified systems, the balance struck in one place is seldom exactly the same as that in another, where the weighting of economic, political, cultural, and ideological forces may be different. The use of models of the kind that have been discussed in this article is essential in the analysis of any society; and where useful models do not exist, as is probable in the study of such tribally mixed areas as those just noticed, then they must be constructed. But their usefulness must not be misunderstood. It is highly unlikely that any model will correspond in detail with the complexity and variety of real life, especially that of multiracial and multicultural societies.
This article is concerned mainly with the theoretical problems of describing and analyzing the place of ethnic and cultural groups within social systems of different types. Most studies of societies of this kind have dealt less with theoretical problems than with the consequences of economic, political, or religious specialization of such groups within the wider society. Often these consequences are the result of the structural position of the group, but this in turn may be the result of the specialized tasks that it performs.
When Europeans first began to govern east Africa the difficulties of setting up an administration and of stimulating the trade needed to produce a revenue gave an opening to Indian immigrants, who were ethnically and culturally very different from the African and Arab inhabitants of the region. Even where common beliefs in Islam were held, this fact did not submerge ethnic or cultural differences or the hostility to, and suspicion and fear of, the immigrants, whose interests as middlemen and skilled workers brought them into conflict with all other ethnic categories of the population. Such a conflict tends to make a structural alignment even more rigid and to confirm and perpetuate associated attitudes.
The point can be illustrated in many parts of the world. Studies of Jewish ethnic groups have long been concerned with the political and other social results of economic specialization and the ways in which specialized minority groups, once established, are modified and maintained. Similarly, in all parts of southeast Asia, economic and political developments have produced ethnic specialization with a wide range of conflict.
The political consequences of the specialization of ethnic groups by occupation, and therefore of the kinds of power that they hold in society, is a problem of which all historians of colonial empires, from that of the Romans and the imperial Chinese to the sixteenth-century Spanish and the modern Europeans, are well aware. But it is also a problem that needs even closer attention in postcolonial societies, where, although the structural alignment of groups within them may have altered, the problems of cultural and ethnic diversity remain.
With the growth of good communications and the spread of travel, ethnically and culturally diverse societies are likely, in the short term, to increase in number rather than diminish. As the sociological study of society ceases to be solely a Western discipline, the need to find appropriate conceptual tools for analyzing ethnic and cultural variation will undoubtedly become a major preoccupation of the discipline.
H. S. MORRIS
[see alsoMINORITIES; PEASANTRY; RACE RELATIONS; SOCIAL STRUCTURE; TRIBAL SOCIETY. Other relevant material may be found inASIAN SOCIETY, article onSOUTH ASIA; CARIBBEAN SOCIETY; ’LINGUISTICS, article OnTHE SPEECH COMMUNITY.]
BARTH, FREDRIK 1960 The System of Social Stratification in Swat, North Pakistan. Pages 113–146 in Edmund R. Leach (editor), Aspects of Caste in South India, Ceylon and North-west Pakistan. Cambridge Papers in Social Anthropology, No. 2. Cambridge Univ. Press.
BENEDICT, BURTON 1962 Stratification in Plural Societies. American Anthropologist New Series 64:1235–1246.
BOEKE, JULIUS H. 1953 Economics and Economic Policy of Dual Societies as Exemplified by Indonesia. New York: Institute of Pacific Relations.
CHAMBERLAIN, HOUSTON STEWART (1899) 1910 Foundations of the Nineteenth Century. New York and London: John Lane. -→ First published in German.
Cox, OLIVER C. 1948 Caste, Class and Race: A Study in Social Dynamics. Garden City, N.Y.: Doubleday.
DOLLARD, JOHN (1937) 1957 Caste and Class in a Southern Town. 3d ed. Garden City, N.Y.: Doubleday.
FREEDMAN, MAURICE 1960 Growth of a Plural Society in Malaya. Pacific Affairs 33:158–168.
FREYRE, GILBERTO (1933) 1956 The Masters and the Slaves: A Study in the Development of Brazilian Civilization. 2d ed., rev. New York: Knopf. → First published in Portuguese.
FURNIVALL, JOHN S. 1942 The Political Economy of the Tropical Far East. Journal of the Royal Central Asian Society 29:195–210.
FURNIVALL, JOHN S. (1948) 1956 Colonial Policy and Practice: A Comparative Study of Burma and Netherlands India. Issued in cooperation with the International Secretariat, Institute of Pacific Relations. Cambridge Univ. Press; New York Univ. Press.
GOBINEAU, JOSEPH ARTHUR DE (1853–1855) 1933 Essai sur I’inégalité des races humaines. 6th ed., 2 vols. Paris: Firmin-Didot. → Partially translated into English in 1915 as The Inequality of Human Races.
JAYAWARDENA, CHANDRA 1963 Conflict and Solidarity in a Guianese Plantation. London School of Economics and Political Science, Monograph on Social Anthropology, No. 25. London: Athlone.
LEACH, EDMUND R. 1954 Political Systems of Highland Burma: A Study of Kachin Social Structure. London School of Economics and Political Science; Cambridge, Mass.: Harvard Univ. Press.
MACCRONE, IAN D. 1937 Race Attitudes in South Africa: Historical, Experimental and Psychological Studies. Oxford Univ. Press.
MORRIS, H. S. 1957 The Plural Society. Man 57:124–125.
NADEL, SIEGFRIED F. 1957 The Theory of Social Structure. London: Cohen & West; Glencoe, Ill.: Free Press.
PATTERSON, SHEILA 1953 Colour and Culture in South Africa. London: Routledge; New York: Humanities.
REX, JOHN A. 1959 The Plural Society in Sociological Theory. British Journal of Sociology 10:114–124.
SMITH, M. G. 1955 Framework for Caribbean Studies. Mona (Jamaica): University College of the West Indies.
SMITH, M. G. 1956 Community Organization in Rural Jamaica. Sociai Economic Studies 5:295–312.
SMITH, M. G. 1960 Social and Cultural Pluralism. New York Academy of Sciences, Annals 83, no. 5:763–785.