Internal Colonialism PDF

 Internal Colonialism Michael Hechter with a new introduction and a new appendix by the author  Chapter 2 Towards a Theory of Ethnic Change I have seen, in my time, Frenchmen, Italians, and Russians; I even know, thanks to...

 Internal Colonialism Michael Hechter with a new introduction and a new appendix by the author  Chapter 2 Towards a Theory of Ethnic Change I have seen, in my time, Frenchmen, Italians, and Russians; I even know, thanks to Montesquieu, that one may be a Persian; but as for man, I declare that I have never met him in my life; if he exists, it is without my knowledge. JOSEPH DE MAISTRE Every industrial and commercial center in England now possesses a working-class divided into two hostile camps, English proletarians and Irish proletarians. The ordinary English worker hates the Irish worker as a competitor who lowers his standard of life. In relation to the Irish worker he feels himself a member of the ruling nation and so turns himself into a tool of the aristocrats and capitalists of his country against Ireland, thus strengthening their domination over himself. He cherishes religious, social and national prejudices against the Irish worker. His attitude towards him is much the same as that of the 'poor whites' to the 'niggers' of the former slave states of the USA. The Irishman pays him back with interest in his own money. He sees in the English worker at once the accomplice and the stupid tool of the English domination in Ireland.... This antagonism is the secret of the impotence of the English working-class, despite their organization. KARL MARX THE persistence of ethnic attachments in complex societies is in certain respects as vexing today as it was to these two very different nineteenth- century thinkers. The current flourish of national consciousness among cultural minorities in advanced industrial societies poses something of a sociological dilemma, for these frequently territorial conflicts involving language and religion have been assumed to be endemic to the early stages of national development.1 Peripheral regions in developing countries have tended to resist the incursions of central authorities imposing their omnivorous bureaucracy and haughty culture. However, the advent of sustained economic and social development presumably serves to undercut the traditional bases of solidarity among extant groups. A familiar list of processes catalogued under the heading of modernization systematically increases the individual's dependence upon and loyalty to the central government. The major foundation of political cleavage in industrial society is then thought to become functional rather than segmental, thereby producing 'alliances of similarly oriented subjects and households over wide ranges of localities.'1 This provides the nucleus of a very generally held sociological theory of ethnic change. Such a theory predicts that the transformed conditions of industrial society alter the basis on which individuals form political associations. In consequence of modernization, the salience of cultural similarity as a social bond should give way to political alliances between individuals of similar market position, and thus, more generally, social class. When the Western industrial working class was judged to have been granted the full rights of citizenship following militant tradeunion and party organization, some observers relaxed enough to proclaim an end of ideology in the West. It now appears that the Flemings in Belgium, Celts in Britain and France, Slovaks in Czechoslovakia, French in Canada—to say nothing of various minorities in the United States—have not yet received the message.2 There is a new fear that these phenomena are the rumblings of the eventual dissolution of the Western nation state. In ten years the pendulum has swung far indeed.3 The specter of class warfare may have subsided only to be replaced by the hoarier aspect of 'racial' conflict. At any rate that state of social harmony sometimes referred to as national integration now seems to many an unexpectedly elusive goal. In the context of the resurgence of ethnicity as a major political factor in industrial society it may be useful to examine the concept of national development and evaluate some of the alternative ways in which it has been used. Accordingly this chapter will have four related foci. First, it will introduce a simplified model for the empirical study of national development. Second, it will review the main currents of previous theory from the perspective of this model, and evaluate its adequacy, in the light of current research. Third, it will define the concept of ethnicity and discuss the social basis of ethnic identification. Finally, it will attempt to sketch some tentative hypotheses about the conditions governing the prospects for ethnic change.  Dimensions of National Development: An Exploratory Model Among current concepts of social science, national development must rank high on a scale of ambiguity. One writer has detected five separate senses in which a related term has been used, varying not only in scope but in subject matter as well.1 Others have denied the utility of this kind of concept outright, particularly in its evolutionary and deterministic form.2 The general expansion of the size and scope of states since the late Middle Ages is justification enough for the utility of some concept of national development. Clearly the social organization of the Athenian polis would be inadequate to maintain a society as large and diverse as the United States, the factor of communications technology aside. The specification of this evolutionary development was the major focus of nineteenth-century social theory. It is by now apparent that the classical social theorists overestimated the extent to which industrialization would lead to a fully national society. A useful definition of national development would describe a process, the creation of a national society, rather than a particular state of affairs.3 Here national development will refer to those processes by which a state characterized by sectional, or otherwise competing economies, politics, and cultures, within a given territory, is transformed into a society composed of a single, all-pervasive, and in this sense 'national' economy, polity, and culture. This definition, it should be noted, does not include the analytically distinct problem of the incorporation of excluded social classes into the national society.1 It should be understood at once that not much is known about national development in these terms. Since so little systematic empirical work has been done in this area many writers have assumed much about these processes on a purely a priori basis. For instance, the relationship between regional economic specialization and interregional political solidarity has never been carefully studied on a comparative basis. Nevertheless, writers will often take sides on this question largely on the basis of theoretical arguments. The difficulty is that a good theoretical case can be made for either side in the dispute. In the following discussion I shall assume a simple model to illustrate processes of national development in industrial societies. In this model there are two collectivities or objectively distinct cultural groups: (1) the core, or dominant cultural group which occupies territory extending from the political center of the society (e.g. the locus of the central government) outward to those territories largely occupied by the subordinate, or (2) peripheral cultural group. The model therefore assumes that these respective cultural groups are to a large extent regionally concentrated.2 For the moment I will assume that these two collectivities are perfectly solidary, and, loosely speaking, possess a group consciousness. Hence, when I conceive of a collectivity as an acting unit, this of course refers to the action of the bulk of its members. Because the lacunae surrounding the problem of national development are so extensive, it may be wise to proceed initially as if the concept were a black box. Actually, enough is known to separate the problem into three analytically separate black boxes. The first might be labeled the problem of cultural integration. The extent to which groups are objectively differentiated by religion or language, for instance, is easily determinable. Such cultural differences are often invidious and at times contribute to intergroup conflict. Cultural integration includes those processes which lead to the gradual effacement of objective cultural differences between groups in contact. In turn, this would encourage the growth of a national identity, providing common access to national symbols and values to each collectivity. The sense of nationality, which entails a feeling of belonging to a corporate group, is likely to be pervasive only in culturally integrated settings. The strength and distribution of this attitude, and related attitudes, has frequently been gauged in survey research.  The second dimension of national development is clearly economic. Economic integration will refer to the evolution of substantially equal rates of social and economic development among the collectivities in a society. It can be measured by the extent to which cultural groups are differentiated by per capita income, infant mortality, literacy, and extent of political power. Such trends as the historical tendency towards increasing equality of regional per capita income1 and the extension of citizenship to formerly excluded social groups are included in this dimension.2 Last, it is possible to distinguish a separate political dimension to national development. Political integration may be said to occur to the extent that the social structural position of a collectivity determines its political behavior, whether electoral or extra-parliamentary. This may be illustrated by an example. If two groups sharing a common occupation (e.g. coal miners), but differing in objective cultural forms (e.g. of different religion), may be shown to have similar general political preferences, this indicates a high level of national political integration. Hence, political integration of this kind implies that objective cultural factors, such as language or religion, cease to have salience in the formation of a collectivity's political demands. The extent to which a collectivity is politically integrated depends ultimately on its definition of the political situation. At any given time each collectivity may either extend legitimacy to the central government or it may withdraw it.3 This grant of legitimacy is given to the government in return for what the collectivity defines as satisfactory societal membership. While this seems analogous to a contractual agreement, there are several key differences. Everything is not written down or specified. There is no way in which this agreement can be legally enforced since cancellation implies the denial of the legitimate authority of the state. Finally, in so far as the reneging collectivity is concerned there may be little or no non-contractual basis to the agreement. The underlying shared values of Durkheim's conscience collective may not be presumed to automatically cut across cultural lines in these circumstances. This is precisely the difficulty of the situation.  A collectivity may reassess its sense of societal membership, and hence may withdraw its grant of legitimacy, for essentially two types of reasons. On the one hand it might become aware of measurably unfavorable changes in its situation in the society—such as an outbreak of job discrimination directed against its members, in which case political integration ceases due to a change in the group's social structural position. On the other hand, the collectivity may, in effect, redefine the situation to demand greater rewards from the government in return for its continued support of the regime. In this case there is a subjective reassessment despite little objective social structural change. Tocqueville's notion of the socially disruptive effects of rising expectations among oppressed groups is one example of this phenomenon. Similarly, Marx's idea of the attainment of the true consciousness (Klasse für sich) is another attempt at interpretation. It need hardly be emphasized that the conditions under which such reassessments occur are central to the study of political mobilization, and will be briefly considered later in this chapter. One last caveat. At the risk of repetition, it should be emphasized that at any given time a collectivity's self-definition, or state of consciousness, may or may not be related to objective changes in its relative position within the society. Hence the realization of political integration is a precarious process and an easily reversible one. Since broad economic and cultural changes appear to be very gradual in the history of societies, except in very special circumstances, these types of integration are therefore assumed to be much more temporally stable than political integration. The key empirical question suggested by the model concerns the relationship between intercollectivity economic and cultural differences and the general prospects for political integration. It will be assumed that in the absence of such differences national political integration is a certainty. However, the lack of political integration in the face of economic and cultural differences cannot be presumed, because a collectivity may be resigned to its disadvantaged position in society. It is possible to arrange case studies in a two-by-two matrix bounded by relative levels of cultural and economic integration. At one end of the integration continuum the values assigned to economic and cultural integration are both low. Most societies of the Third World belong here. At the other end of the continuum are 'perfectly integrated' societies where the values given to economic and cultural integration are both high. It might be interesting to speculate about the politics of the perfectly integrated national society. This type of society would not be immune from political struggle between the members of different cultural groups, but the conflicts generated would be of a specific type, basically between constantly shifting interest groups, each mobilized by the specific political decision at hand more or less in a random order. There would, however, be no permanent constellations of culturally distinct groups such that regular 'ins' and 'outs' could be distinguished. Cultural status would therefore crosscut all other significant statuses in the social structure. Since both collectivities define the government as legitimate; since they both have equal access to the symbols of nationality, the salience of objective cultural distinctions would quickly fade from the political arena. It is evident that few existing societies approach this ideal to any appreciable extent.1 In order to examine such general propositions as whether cultural or economic integration is more significant for the realization of political integration, thereby seeking an elementary causal understanding of the evolution of national development, it must be made clear that little can be gained from the consideration of case studies clustered at the extreme ends of the integration continuum. This highlights much of the difficulty with research which has been completed on problems of national development. Case studies have been almost exclusively selected from a group of societies having no appreciable prospects for the realization of either economic or cultural integration in the near future.1 The critical case studies are those where an intermediate level of national development has occurred: either economically integrated societies composed of distinct cultural groups, or—if this is not a null set—culturally integrated societies composed of groups at different levels of economic and social development. By and large such examples may be found among contemporary industrial societies. Cross-national comparisons of societies at these two intermediate stages of integration will help determine the relative dominance of economic or cultural factors for political integration. Only then can the sequences of national development begin to be understood. In the absence of this systematic body of empirical research, I shall try to sketch and evaluate the relative merits of some alternative theoretical models concerning the processes of national development. Diffusion Models of National Development Diffusion models of national development ultimately assert that core- periphery malintegration can be maintained only in the absence of sustained mutual contact.2 Hence, the establishment of regular interaction between the core and the periphery is seen to be crucial for national development. Industrialization is usually conceived to be a necessary condition for intensifying contact between core and peripheral groups. Processes operating at various levels of social organization should reduce much of the diversity characteristic of pre-industrial societies. It may be useful to group diffusionist accounts of the integration of societies3 into those which emphasize the dominant importance of cultural elements in the realization of national development and those which essentially seek to explain national development by reference to change in the social structure. Cultural theories tend to presume a model in which the peripheral group is termed 'traditionally oriented' in contrast to the modernized core group. The most familiar elaboration of this dichotomy is Talcott Parsons's discussion of the pattern variables; they need not be discussed here. Once the peripheral group becomes exposed to the cultural modernity of the core, its values and normative orientations ought to undergo transformation. Frequently, however, this does not seem to occur. In the view of many of the cultural theorists, the maintenance of peripheral cultural forms and customs is an irrational reaction by groups which seek to preserve a backward life- style insulated from the rapid change of contemporary industrial society. The traditional life-style is a comfortable one. Groups tend to resist major change, in any case: there is a collective, if sometimes unstated, desire to optimize short-run interests while prospects for the long range are seriously compromised. The existence of a backward sloping labor supply curve, supposedly endemic to less developed societies, is commonly taken as evidence of a traditional normative orientation.1 The cultural theories thus seek to explain differences of values, norms, and life-styles among these collectivities as a function of the relative isolation of the peripheral group from the mainstream culture of the core. If only the choice between tradition and modernity could somehow be placed before individuals of the peripheral group, some have felt, modernity would easily win over. The solution to this problem of persisting cultural differences is, first, to stimulate a wide range of intercollectivity transactions, then, to let time work its inevitable course towards eventual cultural integration. Emile Durkheim presented an early statement of this theory when he discussed the effect of an increase in transactions between cultural groups as if it were analogous to the physical process of osmosis.1 Cultural differences between collectivities would in effect be leveled, presumably according to the relative volumes (magnitudes?) of the respective groups. This is one of the biological metaphors which Professor Parsons has so rightly disparaged in Durkheim's work. However, the anthropological version of this image, namely the process of acculturation, remains a rather current perspective in the study of intergroup relations. Acculturation is thought to occur when 'groups of individuals having different cultures come into continuous first-hand contact, with subsequent changes in the original culture patterns of either or both groups.'2 Like osmosis, it is often conceived to proceed automatically and irreversibly. Once begun, the diffusion of symbols and institutions from the core to the periphery should lead to a gradual rapprochement, to a stable cultural equilibrium.3 The proposition that the greater the frequency of interaction between collectivities, the greater the probability they will become more alike is sometimes held to follow from the experimental study of small groups.4 Since it is not difficult to find evidence that intergroup contact often leads to hostility rather than mutual accommodation, less inviolable explanations must be sought. Perhaps the most influential of these is the concept of social mobilization.1 The social mobilization perspective, as well, assumes that the initiation of cultural contact between collectivities is generally beneficent. But interaction per se is seen to be an insufficient condition for the realization of national development. Much emphasis is placed on the power of the central government to coax or coerce the recalcitrant collectivity into acceptance of the core culture. This is best done by the manipulation of cultural symbols and values, especially through the use of communications media.2 But the deployment of military force is also not ruled out. Social mobilization theorists such as Deutsch do not assume that any mere increase in the rate of intensity of interaction is sufficient to end culture maintenance in the periphery, since they recognize that it may be partially self-imposed. Hence, the active role of the central government is stressed, particularly in the establishment of what is often termed a national 'political culture.'3 Control of the national information network enables a régime to set national goals, create a national identity, teach needed skills, centralize its power, extend the effective market, confer status on certain groups at the expense of others, and generally manipulate large numbers of individuals through well- developed techniques of mass persuasion.4 However, with all these tools of behavioral management available, some well-established Western governments are facing ever stronger and more violent challenges to their authority along separatist lines. Another of the technical means by which a central régime may gain adherents is through an inspiring brand of leadership. The Weberian category of charismatic legitimation has been often invoked to account for specific integration successes.5 Recently, however, this explanation has come under criticism for certain theoretical inadequacies,1 as well as the fact that a host of charismatic leaders in the Third World did not survive as long as their academic observers had anticipated. Yet another factor which has been cited as having crucial importance for the process of national development is the encouragement of intercollectivity elite participation in shared, especially governmental, activities. It is thought that the experience gained in such activities encourages élite accommodation and mutual understanding, which then 'filters down' to the level of the masses. These so-called 'functional' theories of the political science literature2 presume, however, a whole host of assumptions about the ability of each collectivity's élite to influence their respective rank-and-file members with the same effectiveness. In fact, there is good reason to believe the results of these filtering down activities to be highly differentiated by collectivity. The theories of cultural diffusion have several obvious inadequacies. Since most of these hypotheses were derived from considering Third World failures of the integration process, they are not very applicable to the collectivities in developed societies. It is difficult to argue that peripheral groups in industrial societies are economically, politically, and culturally isolated from the core. This is evident from the most casual consideration of the history of Western European societies. The development of national economies in Western Europe began in the sixteenth and seventeenth centuries with the ascendance of mercantilism.3 Internal tariffs and other barriers to the free mobility of goods, labor, and capital were substantially eliminated. Weights, measures, and coinage were standardized throughout state territory. The central government succeeded in establishing control over foreign trade, especially in the realm of customs. Attempts to enforce central government control over techniques of production largely failed because of inadequate communications: the center was unable to carry out these policies because supervision was impossible in remote regions. National economies were basically realized after the Industrial Revolution made possible the proliferation of railway and canal systems throughout state borders. Similarly, the evolution of strong central administration, begun in the fifteenth and sixteenth centuries,1 was effectively completed in the nineteenth century, aided by means such as the standing army, which was capable of subduing internal threats to the central régime. Improved communications enabled the center to exert its control into the far reaches of the territory through its cultural influence and military might. Rural-urban migration and the creation of an urban proletariat forced an eventual extension of the suffrage to the working class, which resulted in the formation of mass national parties.2 Finally, the Industrial Revolution had direct and indirect effects on cultural isolation. In the late nineteenth century there were dramatic increases in mass literacy.3 Newspapers reached into the dark corners of the land. Later, radio and television—often operated by the regime in power— penetrated all territorial space, trumpeting the culture of the core. The establishment of national school systems significantly narrowed socialization differences among the youth of separate collectivities. As a consequence of the penetration of the core's cultural institutions into the hinterlands, the maintenance of distinctive languages, and other cultural forms, was severely threatened. These trends indicate that the persistence of a distinctive culture in peripheral areas cannot be explained by the periphery's isolation from the core culture, at least in Western societies. Instead, the persistence of peripheral culture suggests a pattern of resistance to assimilation, a resistance so virile that powerful behavioral management techniques cannot overcome it. Social structural diffusion theories claim that the malintegration of core and periphery arises from their essential differences of social organization. The 'modern' social organization of the core is characterized by a wide division of labor, high level of urbanization, capital-intensive production, small nuclear family, rationalistic bureaucratic structures, high per capita income, and those rational norms and values which naturally arise in such settings. On the contrary, the 'traditional' social organization of the periphery manifests a narrow division of labor, low level of urbanization, laborintensive production, large extended family, personalistic and diffuse structures, lower per capita income, and traditional norms and values. These are the stereotypical differences between advanced industrial societies and relatively backward agricultural societies and are the stuff of the literature of modernization. What is slightly different in this situation is that this conflict of social systems occurs within the framework of the advanced industrial society. The structural diffusion theories suggest that economic integration precedes, if it does not actually cause, cultural integration and subsequent national development. How can this survival of traditionalism within a sea of modernity be accounted for? The simplest, and most frequent explanation is that the peripheral collectivity is not, in fact, economically integrated into the society.1 How, then, can such integration be actively sought? One remedy is to incorporate the peripheral group into the modern industrial economy so that it becomes subject to the strains of structural differentiation. The widening of the division of labor loosens the hold of traditional authorities, creates new social needs and functions, and thus brings pressure for integration. 'Differentiation and upgrading processes may require the inclusion in a status of full membership in the relevant general community system of previously excluded groups which have developed legitimate capacities to "contribute" to the functioning of the system.'2 As the peripheral collectivity begins to participate in the national economic system, changes in its structural relations should lead to rational, performance-centered, and universalistic values. A more equivocal response to this problem of integration leans on the tendency towards unbalanced growth in capitalist societies. It is well known that some regional sectors develop faster and more completely than others. Hence the solution to the problem of economic integration is to promote development in the relatively backward regions.1 Neo-classical economic theory holds that the expansion of efficient, capital, labor and commodity markets into regions dominated by traditionally oriented groups should decrease regional economic inequalities in the society as a whole.2 Once the peripheral region is brought into the national network of commercial flows and transactions, inequality might temporarily increase; but in time an equilibrium will be reached and economic integration will be substantially achieved. This is another osmotic model, only in this case the causal factor is the exchange of material goods and services rather than the interaction of cultural elements per se.  Aside from the historical objections to the cultural diffusion theories discussed previously, there is a mounting body of evidence which seems to refute the thesis of peripheral isolation, particularly with respect to economic integration.3 Thus there are two major difficulties with the structural and cultural diffusion theories. First, the persistence of relative economic backwardness in the periphery cannot satisfactorily be explained by reference to its isolation from the national economy. In effect, peripheral economic development has occurred more slowly than the theory would predict. This has led to analyses which tend to blame peripheral economic sluggishness on the oppressive traditional culture which is maintained.4 But this points to the second problem: why is traditional culture so enduring in the periphery despite this substantial interaction with the core? Clearly the existence of a distinctive culture in the periphery cannot be taken as a given. The probability of successful acculturation, leading to the cultural homogenization of the two groups, should increase progressively with time. Here as well the diffusion theories present an overly optimistic assessment. An Alternative Model: The Periphery as an Internal Colony Common to both the structural and cultural diffusion theories is a unilateral conception of social and economic development. This type of development, as indicated by such measures as labor diversification indices and urbanization statistics, is assumed to spread from one locality to another though the mechanism of this diffusion is somewhat mysterious. However, an important distinction can be made between development which occurs as a result of factors endogenous to a specific society and that which is the result of basically exogenous forces. The second type of development—that usually associated with certain sectors of Third World societies—arose out of what Georges Balandier1 has termed the 'colonial situation.' Typically this involves domination by a 'racially' and culturally different foreign conquering group, imposed in the name of a dogmatically asserted racial, ethnic, or cultural superiority, on a materially inferior indigenous people. There is contact between the different cultures. The dominated society is condemned to an instrumental role by the metropolis. Finally, there is a recourse not only to force, to maintain political stability, but also to a complex of racial or cultural stereotypes, to legitimate metropolitan superordination. The pattern of development characterizing the colonial situation is markedly different in these respects from that which emerged from endogenous development in Western Europe and Japan. First, colonial development produces a cultural division of labor: a system of stratification where objective cultural distinctions are superimposed upon class lines. High status occupations tend to be reserved for those of metropolitan culture; while those of indigenous culture cluster at the bottom of the stratification system. The ecological pattern of development differs in the colonial situation, leading to what has been termed economic and social dualism. Since the colony's role is designed to be instrumental, development tends to be complementary to that of the metropolis. The colonial economy often specializes in the production of a narrow range of primary commodities or raw materials for export. Whereas cities arose to fulfill central place functions in societies having had endogenous development, the ecological distribution of cities looks very different in colonies, where they serve as way stations in the trade between colonial hinterlands and metropolitan ports. Hence cities tend to be located on coasts with direct access to the metropolis.1 Similarly, transportation systems arise not to spur colonial development—they are seldom built to interconnect the various regions of the colony—but to facilitate the movement of commodities from the hinterland to the coastal cities. Thus, the cultural contact engendered in the colonial situation did not lead to a type of social and economic development in the colony which was recognizably similar to that of the metropolis. Andre Gunder Frank has characterized the fruits of such contact as 'the development of underdevelopment.'2 It must not, however, be assumed that this colonial type of development is to be found only in those areas subjected to nineteenth- century overseas imperialism. Simultaneous to the overseas expansion of Western European states in the fifteenth and sixteenth centuries were similar thrusts into peripheral hinterlands:3 The prime aim of the new rulers in their expansionist efforts was to bring under sway all territory not already theirs within the 'natural frontiers' dimly coming to be perceived... small nationalities which had failed to develop as States were now swallowed up: Brittany (1491), Granada (1492), Navarre (1512), Ireland. Their languages and cultures persisted nevertheless and none of the governments succeeded fully in its program of unification. England strove in vain to absorb Scotland; Spain was only briefly able to absorb a reluctant Portugal. Frontiers thus surviving helped by mutual irritation to generate a corporate sentiment on both sides. By the seventeenth century an Englishman who did not look down on a Scotsman would have been only half an Englishman; a Scotsman who did not hate an Englishman would not have been a Scotsman at all. These internal campaigns were not in any sense coincidental to overseas colonization. There is reason to believe that both movements were the result of the same social forces in these states, among which the search for new sources of foodstuffs may have been of primary importance. Fernand Braudel has referred to this territorial expansion of the Western European states as a quest for 'internal Americas.'1 This bears a striking resemblance to the description of internal colonialism which has emerged from consideration of the situation of Amerindian regions in several Latin American societies. This conception focuses on political conflict between core and peripheral groups as mediated by the central government. From this perspective the 'backwardness' of peripheral groups can only be aggravated by a systematic increase in transactions with the core. The peripheral collectivity is seen to be already suffused with exploitative connections to the core, such that it can be deemed to be an internal colony. The core collectivity practices discrimination against the culturally distinct peoples who have been forced onto less accessible inferior lands. Some aspects of internal colonialism have been sketched, though not yet systematically demonstrated.2 These bear many similarities to descriptions of the overseas colonial situation. Commerce and trade among members of the periphery tend to be monopolized by members of the core. Credit is similarly monopolized. When commercial prospects emerge, bankers, managers, and entrepreneurs tend to be recruited from the core. The peripheral economy is forced into complementary development to the core, and thus becomes dependent on external markets. Generally, this economy rests on a single primary export, either agricultural or mineral. The movement of peripheral labor is determined largely by forces exogenous to the periphery. Typically there is great migration and mobility of peripheral workers in response to price fluctuations of exported primary products. Economic dependence is reinforced through juridical, political, and military measures. There is a relative lack of services, lower standard of living and higher level of frustration, measured by such indicators as alcoholism, among members of the peripheral group. There is national discrimination on the basis of language, religion or other cultural forms.1 Thus the aggregate economic differences between core and periphery are causally linked to their cultural differences. In this description national development has less to do with automatic social structural or economic processes, and more with the exercise of control over government policies concerning the allocation of resources. Since increased contact between core and periphery does not tend to narrow the economic gap between the groups, national development will best be served by strengthening the political power of the peripheral group so that it may change the distribution of resources to its greater advantage. Ultimately this power must be based on political organization. One of the foundations upon which such organization might rest is, of course, cultural similarity, or the perception of a distinctive ethnic identity in the peripheral group. The obstacle to national development suggested by the internal colonial model analogy, therefore, relates not to a failure of peripheral integration with the core but to a malintegration established on terms increasingly regarded as unjust and illegitimate. Thus the internal colonial model would appear to account for the persistence of backwardness in the midst of industrial society, as well as the apparent volatility of political integration. Further, by linking economic and occupational differences between groups to their cultural differences, this model has an additional advantage in that it suggests an explanation for the resiliency of peripheral culture.  On the Cause of Cultural Differences Between Collectivities The antecedent causes of cultural differentiation between groups sharing close geographical proximity have seldom been satisfactorily explored by social scientists from a theoretical perspective. Partly this may be due to the extraordinary slipperiness of the concept of culture, which at base must refer to the prism through which groups ascribe meaning to the physical and social world around them. While it is relatively easy to specify the content of a particular culture by observing such phenomena as socialization processes, it is a problem of a different order to ask why a child is nursed for five years in one society, or taught to mumble reverences to an inanimate flag in another. The participant's answer to this question will be of the form: 'But that is the way it is customarily done.' The sociologist of Durkheimian persuasion will hasten to illustrate the ways in which such practices simultaneously reflect and contribute to the solidarity of the group. But generally the specific form of the culture —whether it is animist or totemist, Catholic or Protestant—is regarded as of incidental significance when compared with the functions the culture performs towards the maintenance of the social order. It is much simpler to assume that a group's culture is an attribute of the social system rather than a variable to be explained. Thus the group may be defined as having a certain set of cultural traits which, when aggregated, make up its 'ethnicity.' This may be a valid assumption in those few societies which are largely self-enclosed and have only sporadic contacts with other societies. For instance it is clear that cultural differences between proximate groups may be a function of their mutual isolation. If these groups are suddenly brought into contact they will manifest these cultural differences to a greater or lesser extent. These initial differences may well result in different institutional arrangements in the two groups:1 What do we mean by a cultural explanation? When a people has lived together generation after generation, sharing a common history, answering often to a common name, it tends to develop distinctive institutions, a distinctive way of life, adapted of course to its physical environment and technology. When faced with new circumstances, the people may well adjust its institutions to meet them, but the adjustment will start from the old traditions, and a recognizable continuity will be maintained.... When we find two people practicing different sets of institutions—even though the differences may not be great, and even though they live in somewhat similar environments and employ a similar technology—we are apt, quite rightly, to explain the facts by saying that the two practice different cultures, the precipitates of different past histories. For the purpose of cross-sectional analysis at a single point in time the assumption of the givenness of a particular set of cultural traits may well be justified. But if the investigator is interested in the longitudinal process of social change, where the relevant groups are subject to frequent culture contact and exchange, then the maintenance of cultural differences over time may be an important clue as to the nature of their interrelationship, particularly in so far as dominance and subordination are concerned. The cultural theory is really no theory at all. It merely refers back to an antecedent period when the specific group in question fashioned its culture in an unspecified way. In the attempt to progress beyond 'the fruitless assumption that culture comes from culture,' the notion of cultural ecology was devised by Julian Steward,1 For Steward a group's culture emerges from particular characteristics of its social structure, which in turn is adapted to a specific physical environment. Culture, in this view, is largely responsive to changes in a core element of the social structure which is defined as:2 the constellation of features which are most closely related to subsistence activities and economic arrangements. The core includes such social, political, and religious patterns as are empirically determined to be closely connected with these arrangements. While accepting the logic of this search for explanatory variables in cultural differentiation, Clifford Geertz cautions that 'there is no a priori reason why the adaptive realities a given sociocultural system faces have greater or lesser control over its general pattern of development than various other realities with which it is also faced.'3 Geertz's objection seems true a fortiori in developed societies, where groups are much less constrained by the purely ecological considerations of how food and shelter are to be provided in the face of a hostile environment. For development typically offers several types of alternative life-styles to individuals. First, there are very many different ways to provide for sustenance through participation in the primary, secondary, or tertiary sectors of the economy. Thus the individual is liberated from crude ecological constraints. But secondarily, in developed societies some individuals are allowed choices about the kind of culture they will adhere to. Individuals may choose to be Protestant or Catholic, Ga or Ashanti, in certain settings.4 Frederik Barth has emphasized that ethnicity must be considered to be a boundary phenomenon. What ultimately separates one ethnic group from another is that it 'has a membership which identifies itself, and is identified by others, as constituting a category distinguishable from other categories of the same order.'5 The subjective element in ethnic identity in complex society has been illustrated in a different way by Ernest Gellner:6 If a man is not firmly set in a social niche, he is obliged to carry his identity with him, in his whole style of conduct and expression: in other words, his 'culture' becomes his identity. And the classification of men by 'culture is of course the classification by 'nationality.' It is for this reason that it now seems inherent in the very nature of things, that to be human means to have some nationality. In our particular social context, it is inherent in the nature of things. In an age of bureaucratic organization and mass literacy, cultural distinctions, particularly those involving language, assume great importance since education has an important place in allocating occupational and social status in the society at large. All of these considerations imply a different type of core element to account for cultural particularity than that which is tied to the ecological situation. What is problematic in complex society are the social conditions under which individuals band together as members of an ethnic group. It is clear that culture maintenance in the periphery can be regarded as a weapon in that it provides the possibility of socialization, as well as political mobilization, contrary to state ends. Max Weber termed those collectivities having distinctive life-styles and cultures status groups (Stände), each of which is allocated a different ranking on a hierarchy of social honor, or prestige.1 In effect, the difference between the type of social solidarity predominating in the core and that in the periphery is akin to the distinction Weber made between classes and status groups. Class solidarity is consonant with a high level of modernization and tends to be organized functionally in occupational groups. Trade unions, for instance, unite men in narrowly defined occupations at the primary, shop level, and confederations of unions tend to be much less solidary than constituent locals. Class solidarity assumes an individual orientation towards the marketplace, whereas status group solidarity involves a group, or collective orientation. The important question, of course, is to come to an understanding of the dynamics between class and status group solidarity. Why does status group solidarity appear to be stronger in the periphery than in the core? The persistence of objective cultural distinctiveness in the periphery must itself be the function of the maintenance of an unequal distribution of resources between core and peripheral groups. Initially, individuals in the disadvantaged peripheral group are not permitted to become acculturated to the core. For Barth,2 the persistence of ethnic groups in contact implies not only criteria and signals for identification, but also a structuring of interaction which allows the persistence of cultural differences... a set. of prescriptions governing situations of contact, and allowing for articulation in some sectors or domains of activity, and a set of proscriptions on social situations preventing inter-ethnic interactions in other sectors, and thus insulating parts of the cultures from confrontation and modification. Such boundaries are maintained by the differential allocation of social roles with the society:1 Common to all these systems is the principle that ethnic identity implies a series of constraints on the kinds of roles an individual is allowed to play, and the partners he may choose for different kinds of transaction.... The persistence of stratified polyethnic systems thus entails the presence of factors that generate and maintain a categorically different distribution of assets: state controls, as in some modern plural and racist systems; marked differences in evaluation that canalize the efforts of actors in different directions, as in systems with polluting occupations; or differences in  culture that generate marked differences in political organization, economic organization, or individual skills. This system of stratification is, in effect, a cultural division of labor. Gellner sees the initial advantage to one cultural group rather than another as an historical accident caused by the uneven spread of industrialization through territorial space. 'The wave creates acute cleavages of interest between sets of people hit by it at differing times in other words the more and less advanced.'2 This accounts for the ability of the superordinate group to impose the kinds of role restraints to which Barth makes reference. However, if the unequal distribution of resources is based on observable cultural differences there is always the possibility that the disadvantaged group in time will reactively assert its own culture as equal or superior to that of the advantaged group:3 This cleavage and hostility can express itself with particular sharpness if the more and the less advanced populations can easily distinguish each other, by genetic or rigid cultural traits. These aid discrimination and humiliation, and thus further exacerbate the conflict. If such differentiae are lacking, nothing happens: the 'backward' area becomes depopulated, or a depressed area within a larger unit, or an object of communal charity and assistance. If, however, the differentiating marks are available—whether through distance, 'race,' or cultural traits such as religion, they provide a strong incentive and a means for the backward region or population to start conceiving of itself as a separate 'nation' and to seek independence. Hence, if at some initial point acculturation did not occur because the advantaged group would not permit it, at a later time acculturation may be inhibited by the desires of the disadvantaged group for independence from a situation increasingly regarded as oppressive. This accounts for the cultural 'rebirths' so characteristic of societies undergoing nationalistic ferment. It is not that these groups actually uncover evidence of their ancient cultural past as an independent people; most often such culture is created contemporaneously to legitimate demands for the present-day goal of independence, or the achievement of economic equality.1  Conclusion: The Politics of Ethnic Change The internal colonial model would therefore seem to provide a more adequate explanation of the persistence of ethnic identity among peripheral groups in complex societies than do diffusion theories portraying the periphery as culturally and economically isolated from the core. In general, relations between core and peripheral groups may be schematized as follows: The uneven wave of industrialization over territorial space creates relatively advanced and less advanced groups, and therefore acute cleavages of interest arise between these groups. As a consequence of this initial fortuitous advantage there is a crystallization of the unequal distribution of resources and power between the two groups. The superordinate group, now ensconced as the core, seeks to stabilize and monopolize its advantages through policies aiming at the institutionalization and perpetuation of the existing stratification system. Ultimately, it seeks to regulate the allocation of social roles such that those roles commonly defined as having high status are generally reserved for its members. Conversely, individuals from the less advanced group tend to be denied access to these roles. Let this stratification system be termed the cultural division of labor: it assigns individuals to specific roles in the social structure on the basis of objective cultural distinctions. The cultural division of labor may be enforced de jure, when the individual from the disadvantaged collectivity is denied certain roles by the active intervention of the state. This is the racist solution to the maintenance of the status quo. The cultural division of labor may, alternatively, be preserved de facto, through policies providing differential access to institutions conferring status in the society at large, such as the educational, military, or ecclesiastical systems. This is the institutional racist solution to the maintenance of the status quo.1 Both types of policies insure that the character of the cultural division of labor remains intact. The existence of a cultural division of labor contributes to the development of distinctive ethnic identity in each of the two cultural groups. Actors come to categorize themselves and others according to the range of roles each may be expected to play. They are aided in this categorization by the presence of visible signs—distinctive lifestyles, language, or religious practices—which are seen to characterize both groups. Such visible signals allow for intergroup interaction, necessarily involving a certain commonality of definitions on the part of interacting partners, in the face of objective cultural differences between groups. Acculturation need not occur because each individual can adjust his behavior in accordance with the other's status (which can be perceived visually) even before interaction takes place.2 Regarded as a status, ethnic identity is superordinate to most other statuses, and defines the permissible constellations of statuses or social personalities, which an individual with that identity may assume. In this respect ethnic identity is similar to sex and rank, in that it constrains the incumbent in all his activities, not only in some defined social situations. The political position of the peripheral group within the society is likely to be feeble given this situation. This is so even in the most formally democratic politics since the peripheral collectivity is likely to be resource- poor relative to all other social groups. If the peripheral group is also a numerical minority its political situation is far worse. As a minority group it cannot independently force changes in central government policies, such as those which might provide a reallocation of income from the core group, on the strength of votes alone. This often results in politics of 'stable unrepresentation.'3 In most developed societies the above considerations hold only in a probabilistic sense. That is to say it is statistically possible for an individual of low ethnic status to achieve, for instance, high-status occupational roles, though of course it is very improbable. The realization of these conventionally forbidden roles makes it possible for statistically deviant individuals to reconsider their ethnic identity. They may have several types of options. By identifying with the advantaged group, these individuals may attempt to 'pass,' and thus undergo a subjective re-identification of their ethnic identity.1 Alternatively, they may attempt to maximize their individual power by acting as brokers between the two groups. As 'ethnic leaders' they may seek to incrementally narrow the material differences between the groups by appealing to the universalistic norms which many industrial societies aspire to. Or they may reactively assert the equal or superior value of their culture, claim the separateness of their nation and seek independence. The first option, basically one of selective co-optation, serves to remove potentially divisive leadership from the peripheral group and thereby ensures stabilization of the cultural division of labor. The ultimate consequences of the second choice are somewhat less clear, except for the probability that should any change occur it will be gradual. The slowness of actual economic integration in the face of larger expectations will most likely insure that a more militant group will form in the name of the ethnic nationalist position. In general, the probability of achieving economic integration within the society, as against other outcomes, such as actual secession or relative stasis, will be determined by factors such as the relative numbers of both groups, the in dispensability of the periphery's role in the national economy, and the kinds of policies adopted by the central government. The existence of ethnic solidarity in a given group should therefore be regarded as a special instance of the general phenomenon of political mobilization. Hence, ethnic change facilitating political integration cannot be expected to result in the periphery until there is widespread satisfaction that the cultural division of labor has largely been eliminated. Once placed in this framework, it is easy to see the reason for the frequency of political demands along ethnic grounds in industrial society. The Marxian discussion of political mobilization points to two fundamental conditions for the emergence of group solidarity. Substantial economic inequalities must exist between individuals, such that these individuals may come to see this inequality as part of a pattern of collective oppression. However, the aggregation of individual perceptions of economic inequality alone is insufficient for the development of collective solidarity. There must be an accompanying social awareness and definition of the situation as being unjust and illegitimate. Oppression by itself can, of course, if severe enough, precipitate random violence against the social system, as in the many instances of peasant jacqueries throughout history; but this is not the result of the attainment of group consciousness and hence lacks the organization and purposefulness to achieve effective ends. Thus another vital condition for the advent of collective solidarity is adequate communication among members of the oppressed group,1 Communication within collectivities generally occurs within the context of social institutions: neighborhoods, workplaces, schools, churches, social and recreational clubs, and the host of voluntary associations to which individuals may typically belong. To the extent that these contexts for social interaction are limited to members of a group sharing the social definition of an ethnic minority, the possibilities for intercommunication will be maximized. This is so because in the periphery there tends to be not only segregation in the workplace but also residential segregation as well. The concatenation of residential and occupational segregation gives a decisive advantage to the development of ethnic rather than class solidarity. Since the concept of social class seeks to deny the salience of cultural and residential differences among members of similar occupational groups, to the extent that such differences actually exist, class is ultimately more abstracted from the reality of everyday social life than is ethnicity. Finally, the very economic backwardness of the periphery contributes to the inevitability of such residential and occupational segregation. As an impoverished and culturally alien region there is little incentive for members of the core group to migrate there in force. Typically the periphery has a declining population, an overabundance of the elderly, and a disproportionate number of females, all of which reflect the lack of adequate employment opportunity which is both a result of peripheral backwardness and a cause of further economic disadvantages.1 From these general observations it is possible to make three propositions concerning the prospects for the political integration of peripheral collectivities into the society as a whole: 1 The greater the economic inequalities between collectivities, the greater the probability that the less advantaged collectivity will be status solidary, and hence, will resist political integration. 2 The greater the frequency of intra-collectivity communication, the greater the status solidarity of the peripheral collectivity. 3 The greater the intergroup differences of culture, particularly in so far as identifiability is concerned, the greater the probabilty that the culturally distinct peripheral collectivity will be status solidary. Identifiable cultural differences include: language (accent), distinctive religious practices, and life-style. It should be underlined that when objective cultural differences are superimposed2 upon economic inequalities, forming a cultural division of labor, and when adequate communications exist as a facilitating factor, the chances for successful political integration of the peripheral collectivity into the national society are minimized. The internal colonial model predicts, and to some extent explains, the emergence of just such a cultural division in labor. 1 Clifford Geertz, 'The Integrative Revolution: Primordial Sentiments and Civic Politics in the New States' in C. Geertz, ed., Old Societies and New States (New York: 1963). 1 Seymour M. Lipset and Stein Rokkan, 'Introduction' in S. M. Lipset and S. Rokkan, eds, Party Systems and Voter Alignments (New York: 167), p. 10. 2 E. Allardt and Y. Littunen, eds, Cleavages, Ideologies, and Party Systems (Helsinki: 1964); Robert A. Dahl, ed., Political Oppositions in Western Democracies (New Haven: 1966); and Stein Rokkan, Citizens, Elections, and Parties (New York and Oslo: 1970). 3 For a graphic demonstration compare two essays written at opposite poles of the past decade by Daniel Bell, 'The End of Ideology in the West' in The End of Ideology (New York: 1962), pp. 393– 407, and 'Unstable America,' Encounter, 34 (1970), pp. 11–26. 1 Myron Weiner, 'Political integration and political development,' Annals of the American Academy of Political Science, 358 (1965), pp. 52–64. 2 Randall Collins, 'A Comparative Approach to Political Sociology' in Reinhard Bendix et al., eds, State and Society (Boston: 1968), pp. 42–67. 3 Gunnar Myrdal, An International Economy (New York: 1956), pp. 9–11. 1 This problem has received substantially greater attention. See T. H. Marshall, 'Citizenship and Social Class' in Class, Citizenship, and Social Development (New York: 1964), pp. 71–134, and Reinhard Bendix, Nation-Building and Citizenship (New York: 1964). 2 This restriction limits the applicability of the model to a subset of the universe of settings for intergroup conflict. However, my central concern in this study is with industrialized societies where this model has wide relevance, except for those societies, like the United States, formed largely from voluntary migration. As R. A. Schermerhorn has argued, 'Studies of ethnic relations based chiefly on data from voluntary immigrations cannot serve as the model or foundation for ethnic relations as a whole.' Comparative Ethnic Relations (New York: 1970), p. 156. 1 Jeffrey Williamson, 'Regional inequality and the process of national development: a description of the patterns,' Economic Development and Cultural Change, 13, 4 (1965), pp. 3–45. 2 Marshall, op. cit. 3 Obviously this is an extreme simplification. It will be objected that there is an intermediate situation where legitimacy is in fact neither granted nor withdrawn, and the state of affairs is passively accepted as inevitable, though it is negatively evaluated. I will consider this sense of resignation to be support for the state of affairs, since there is no expression of withdrawal of legitimacy, such as through changed voting patterns, or sporadic outbreaks of anti-régime violence. 1 Switzerland is often cited as a developed poly-ethnic society which is politically integrated, but no sufficient qualitative and quantitative evidence exists to support this claim. 1 Thus a sample of the most noted case studies includes such societies as Indonesia (Clifford Geertz, Agricultural Involution, Berkeley: 1963); Ceylon (W. H. Wriggins, 'Impediments to unity in new nations: the case of Ceylon,' American Political Science Review, 55, 2 (1961), pp. 313–20); and Ghana (David Apter, Ghana in Transition, Princeton: 1963). 2 A useful definition of diffusion is A. L. Kroeber, 'Diffusionism' in E. R. A. Seligman and A. Johnson, eds, The Encyclopedia of the Social Sciences, III (1930), reprinted in A. and E. Etzioni, eds, Social Change: Sources, Patterns and Consequences (New York: 1964). 3 This discussion does not pretend to offer a review of the literature bearing on these problems, but only on those general theoretical images of the mechanism of national development, i.e., a particular problem for the theory of social change. There is a vast anthropological literature on Third World societies which categorizes a variety of types of internal ethnic stratification as being more or less 'pluralistic,' without, however, attempting to theoretically account for changes in these situations. 1 Max Weber is among the many writers who have assumed this to be a defining characteristic of the traditional life-style, The Protestant Ethic and the Spirit of Capitalism (New York: 1958), p. 59. See also Wilbert E. Moore and Arnold Feldman, Labor Commitment and Social Change in Developing Areas (New York: 1960). For a somewhat skeptical view see Elliot J. Berg, 'Backward sloping labor supply functions in dual economies—the Africa case,' Quarterly Journal of Economics, 75, 3 (1961), pp. 468–92. 1 'Some have seen [the increase in national homogeneity] to be a simple consequence of the law of imitation. But it is rather a leveling analogous to that which is produced between liquid masses put into communication. The partitions which separate the various cells of social life, being less thick, are more often broken through.... Territorial divisions are thus less and less grounded in the nature of things, and, consequently, lose their significance. We can almost say that a people is as much more advanced as territorial divisions are more superficial.' The Division of Labor in Society (New York: 1964), p. 187. 2 This is the classical definition of acculturation from Robert Redfield et al., 'Outline for the study of acculturation,' American Anthropologist, 38, 1 (1936). Since then there have been attempts to specify the types of situations in which contact lessens intergroup conflict. See Gordon W. Allport, The Nature of Prejudice (Cambridge, Mass.: 1954); Robin M. Williams, Strangers Next Door (New York: 1964); and Thomas F. Pettigrew, 'Racially separate or together?,' Journal of Social Issues, 25, 1 (1969). 3 This is not to deny that acculturative processes occur with some regularity as a result of interaction between collectivities. What is questioned is the simple assertion that acculturation necessarily leads to cultural integration under all circumstances. Acculturation studies have  failed to describe the nature of the conditions under which assimilation can be expected to occur. 'Although Kroeber's characterization of acculturation as a passing fad was belied it was still true by 1960 that generally accepted propositions about the nature of change under contact conditions were lacking.... [Subsequently] the main line of development of acculturation studies continued to be descriptive.' Edward H. Spicer, 'Acculturation,' International Encyclopedia of the Social Sciences (New York: 1968) vol. 1, p. 23. It might be added that the large literature on nationalist movements has similarly not yielded a theoretical perspective. 4 George C. Homans, The Human Group (New York: 1950), chapter 5. 1 Karl W. Deutsch, 'Social mobilization and political development,' American Political Science Review, 55, 3 (1961), pp. 493–514. Actually, Deutsch's precise position on this issue shifts from article to article and defies pinning down: see Walker Conner, 'Nation-building or nation- destroying?,' World Politics, 24, 3 (1972), pp. 319–55. I feel that, over all, his sympathies lie with the cultural diffusion model. 2 Lucian Pye, 'Introduction' in Communications and Political Development (Princeton: 1963), pp. 3–23. 3 Gabriel Almond and Sidney Verba, The Civic Culture (Princeton: 1963). 4 Wilbur Schramm, 'Communication Development and the Development Process' in Pye, op. cit., pp. 30–57. 5 Apter, op. cit., and Seymour M. Lipset, The First New Nation (New York: 1963). 1 See Claude Ake, A Theory of Political Integration (Homewood, Ill.: 1967). It should be pointed out that these are less Weber's difficulties than those of many of his subsequent commentators. 2 Ernst Haas, Beyond the Nation State (Stanford: 1964) and Ernst Haas and Philippe Schmitter, 'Economies and differential pattern of political integration: projections about unity in Latin America,' International Organization, 18 (Autumn, 1964), pp. 705–37. 3 A very useful summary may be found in Eli Heckscher, Mercantilism (London: 1955), vol. I. 1 For England, the first state to undergo such political centralization, see G. R. Elton, The Tudor Revolution in Government (Cambridge: 1966). 2 Moisei Ostrogorski, Democracy and the Organization of Political Parties (New York: 1964). 3 Lawrence Stone, 'Literacy and education in England, 1640–1900,' Past and Present, 42 (1969), pp. 69–139. 1 Erik Allardt, 'A Theory of Solidarity and Legitimacy Conflicts,' in E. Allardt and Y. Littunen, eds, op. cit. 2 Talcott Parsons, Societies: Evolutionary and Comparative Perspectives (Englewood Cliffs: 1966), pp. 22–3. This general position has long been held, despite its evident empirical inadequacy. Thus, Max Weber goes to some length to discredit it with respect to the Indian caste system: 'One might believe, for instance, that the ritual caste antagonisms had made impossible the development of 'large-scale enterprises' with a division of labor in the same workshop and might consider this to be decisive. But such is not the case. The law of caste has proved just as elastic in the face of the necessities of the concentration of labor in workshops as it did in the face of a need for concentration of labor and service in the noble household.' Hans Gerth and C. Wright Mills, eds, From Max Weber: Essays in Sociology (London: 1948), p. 412. 1 For a survey, see John Friedmann and William Alonso, eds, Regional Development and Planning (Cambridge, Mass.: 1964). 2 M. Tachi offers an example in 'Regional income disparity and internal migration of population in Japan,' Economic Development and Cultural Change, 12, 2 (1964), pp. 186–204. 3 For example, see Werner Baer, 'Regional inequality and economic growth in Brazil,' Economic Development and Cultural Change, 12, 3 (1964); J. R. Lasuen, 'Regional income inequalities and the problems of growth in Spain,' Regional Science Association: Papers, no. 8; and J. F. Riegelhaupt and Shepard Forman, 'Bodo was never Brazilian: economic integration and rural development among a contemporary peasantry,' Journal of Economic History 30, 1 (1970), pp. 100–16. If the economic isolation of peripheral groups cannot be demonstrated in these societies, then in industrial societies such groups are, a fortiori, fully incorporated into the national economy. 4 Edward C. Banfield, The Moral Basis of a Backward Society (New York: 1958). 1 Georges Balandier, Sociologie actuelle de l'Afrique noire (Paris: 1963). 1 Hence Christaller's theory predicting the location of cities in central places (cf. Edward Ullman, 'A theory of location for cities,' American Journal of Sociology, 46, 3 (1941), pp. 853–64), and devised from South German data, seems to apply best in Western Europe, or exactly that area  which had endogenous development. After the sixteenth century, Eastern Europe experienced extensive refeudalization and came to serve as a major source of primary products for the Western maritime states. Hence these Eastern European societies exhibit many of the characteristics of an area of exogenous development. It has been claimed that this is a significant distinction in the development of feudalism as well. 'A rough general distinction may be drawn between feudalism (in an institutional sense) growing up "naturally" from below, or planted from above. West European feudalism seems to belong in the main to the former type. Peering with due caution into the mists of time and expert witness, one may associate this fact with the long Dark-Age struggle of its region of origin against the pressure of worse barbarism from outside.' V. G. Kiernan, 'State and nation in Western Europe,' Past and Present, 31 (1965), pp. 21–2. The defensive nature of Western European feudalism, for Kiernan, led to conditions favoring social solidarity. 2 Andre Gunder Frank, Capitalism and Underdevelopment in Latin America (New York: 1969). 3 Kieman, op. cit., p. 33. 1 'Mediterranean man has always had to fight against the swamps. Far more demanding than the problem of forest and scrubland, this colonization is the distinguishing feature of his rural history. In the same way that northern Europe established itself or at any rate expanded to the detriment of its forest marches so the Mediterranean found its New World, its own Americas in the plains.' The Mediterranean and the Mediterranean World in the Age of Phillip II (New York: 1972), p. 67. 2 Pablo Gonzales-Casanova, 'Internal colonialism and national development,' Studies in Comparative International Development, 1, 4 (1965), pp. 27–37; Rodolpho Stavenhagen, 'Classes, colonialism, and acculturation,' Studies in Comparative International Development, 1, 6 (1965). Internal colonialism, or the political incorporation of culturally distinct groups by the core, must be distinguished from internal colonization, or the settlement of previously unoccupied territories within state borders. An example of the latter process may be cited from twelfth-century France: 'The wild unoccupied spaces belonged almost without exception to the highest nobility, and the attitude of the latter towards these open spaces underwent a change: they now decided to organize their colonization. This choice was frequently dictated by political considerations. It might be a question of securing the safety of a road by settling the forests through which it passed, or else of asserting control over the frontiers of a principality by establishing strong peasant communities obliged to perform armed service in  wooded and deserted marches which had hitherto formed a protective ring round it.' Georges Duby, Rural Economy and Country Life in the Medieval West (London: 1968), p. 76. 1 There does not seem to be a general consensus on a small number of essential defining features of internal colonialism. Since the concept evolved from the study of ethnic conflict in Latin American societies, the above list is particularly applicable to societies with a similar history, especially with regard to Spanish and Portuguese patterns of colonialism. However, with certain modifications, the notion of internal colonialism may be much more general in scope. What if all but one or two conditions seem to be met? The danger, of course, is to so relax the meaning of internal colonialism that almost any instance of stratification may fall somewhere within its boundaries. Let me give an example. A strict case study of internal colonialism should probably include administrative differentiation, such that there are both citizens and subjects, as dictated by the colonial analogy. This qualification is easily met by many Third World societies, but probably by only one developed society, the Republic of South Africa. Are we therefore to conclude that internal colonialism is an inappropriate concept in modern European history? Ireland is a perfect example of an internal colony under the old United Kingdom until 1829, when Catholics were nearly granted full civic and political rights. But it would be folly to consider that this legislation ended the essentially colonial status of the island. Are we then to refer to post-1829 Ireland as an instance of internal neocolonialism? Similarly, is territoriality a necessary condition, or can American Blacks be considered an internal colony as Robert Blauner ('Internal colonization and ghetto revolt,' Social Problems, 16, 4 (1969), pp. 393–408) and others have suggested? These are some of the problems that remain to be worked out if internal colonialism is indeed to become a useful concept. 1 George C. Homans, 'The explanation of English regional differences,' Past and Present, 42 (1969), p. 29. 1 Julian Steward, Theory of Culture Change (Urbana: 1963), p. 36. 2 Ibid., p. 37. 3 Geertz, Agricultural Involution, op. cit., pp. 10–11. 4 Such subjective 'tribal' re-identification has been described by Immanuel Wallerstein in 'Ethnicity and national integration in West Africa,' Cahiers d'Études Africaines, 1, 3 (1960), pp. 129–39. 5 Frederik Barth, 'Introduction' in F. Barth, ed., Ethnic Groups and Boundaries (Boston: 1969), p. 11. 6 Ernest Gellner, 'Nationalism' in Thought and Change (London: 1969), p. 157. 1 Max Weber, Economy and Society, eds Guenther Roth and Claus Wittich (New York: 1968), vol. II, pp. 926–40. 2 Barth, op. cit., p. 16. 1 Barth, op. cit., p. 17. 2 Gellner, op. cit., p. 171. 3 Ibid., pp. 171–2. 1 David C. Gordon, History and Self-Determination in the Third World (Princeton: 1971). 1 Stokely Carmichael and Charles V. Hamilton, Black Power (New York: 1967), pp. 3–4. The failure of national educational institutions to provide equal training for members of all status groups is now well recognized, and has generated an exhaustive literature. 2 Barth, op. cit., p. 17. 3 This may be said to occur when the 'political system normally operates to prevent incipient competitors from achieving full entry into the political arena. Far from there being built-in mechanisms which keep the system responsive, such groups win entry only through the breakdown of the normal operation of the system or through demonstration on the part of challenging groups of a willingness to violate the "rules of the game" by resorting to illegitimate means for carrying on political conflict.' William A. Gamson, 'Stable unrepresentation in American society,' American Behavioral Scientist, 12, 2 (1968), p. 18. 1 It need hardly be added that to the extent that differential ethnicity is symbolized by phenotypical differences between groups such an option is correspondingly limited. 1 Karl Marx, The Eighteenth Brumaire of Louis Bonaparte (New York: 1926), p. 109. 1 This vicious circle of regional underdevelopment is lucidly discussed in Gunnar Myrdal, Rich Lands and Poor (New York: 1957), pp. 23–38. 2 For a parallel discussion of the effects of such superimposition on the intensity of class conflict, see Ralf Dahrendorf, Class and Class Conflict in Industrial Society (Stanford: 1959), pp. 213–18.