Revised version [May 29, and Oct. 6, 1997] of paper presented on the panel on Insecurity: Migration (Refugees) and Ethnic Nationalism as Symptoms of World Systemic Crisis, at International Studies Association meeting, Toronto, March 1997. For publication in International Political Science Review, special issue edited by Fred Riggs.

Figures 1 and 2 Copyright (c) 1997 by Westview Press. Reprinted by permission of Westview Press. I thank the commentators and panelists for their insightful comments. Leslie S. Laczko also made useful comments. At DePauw University thanks to colleagues Eric Silverman and Thomas Ewing for inviting me to present an earlier version of this paper at the Faculty Research Colloquium, and to the John and Janice Fisher fund for faculty development for assistance in attending the International Studies Conference and in preparation of this paper. Last, but not least, to the considerable efforts of Fred Riggs in organizing and editing this collection. As usual, despite all this good will, I bear full responsibility for any errors.

Abstract: This paper discusses how incorporation into a world-system (ancient or contemporary) can create, transform, or destroy ethnic groups. It suggests that: (1) ethnically homogeneous states have never been common; (2) ethnicity has always been fluid with respect to identity, boundaries, cultural content, and membership; study; (3) ethnic processes cannot be understood without careful consideration of their interstate, or world-systemic, context; (4) contemporary ethnic conflicts have contemporary roots; (5) the differnces between the contemporary and ancient world need further study; (7) the origin of the ideal of the ethnically homogeneous state and shifts in ethnic processes in the 20th century lack adequate explanation.

For several years I have collaborated with Christopher Chase-Dunn in studying precapitalist, that is before the fifteenth century, world-systems. To do this we have had to modify extensively the world-system models originally proposed by Immanuel Wallerstein (Chase-Dunn and Hall 1997a). In particular I have elaborated the concept of incorporation into a world-system of either a region or a group, or both (Hall 1986, 1987, 1989a:Ch. 2; Chase-Dunn and Hall 1997a: Chap. 4). In this paper I focus on how world-systemic processes can create, transform, or destroy ethnic groups, and how those processes are the same, and differ between ancient and contemporary world-systems.

In order to do so, I begin with a brief discussion of definitional issues, then summarize the historic depth of ethnic-like relations. With the stage thus set I will summarize the comparative world-systems approach to these issues. After noting two remaining puzzles, I conclude with a few lessons from the study of ancient ethnic conflicts for the study of contemporary ones.

Definitional Issues
When scholars whose training and practice is rooted in different social science disciplines engage in discussions of common concerns it is all too easy to get bogged down in all sorts of conceptual and definitional issues. I prefer to extend Arthur Stinchcombe's (1978) dictum that counting is the last, not first step in theoretical and empirical investigation. Thus, definitions should emerge within theoretical and empirical discussions. In order to forestall such confusion I present a few garden variety definitions in order to begin the discussion.

I follow Wallerstein's definition of a world-system as an intersocietal system marked by a self-contained division of labor. It is largely self-contained and has some degree of internal coherence and forms a complete unit (Wallerstein 1974, 1979, 1980, 1989). Hence the hyphen in the term which has, itself, become a politicized (Thompson 1983a, 1983b; Wallerstein 1983). Only in the twentieth century has this "world" become truly global. Similarly, the term is now often used in the plural, as I use it here, again with some contention (Wallerstein 1993, 1995). While much world-systems work-- with or without the hyphen, with or without the plural--has roots in the work of Immanuel Wallerstein, his voluminous production is but a small fraction of it (Hall 1996a, 1996b).

The word "modern" and its derivatives is similarly vexing. In this essay I follow the usage of Fred Riggs and Anthony D. Smith (1994:392): using "contemporary" or "recent" rather than "modern," and using modernity for the constellation of recent, more-or-less democratic, industrialized, national states. Still, I must point out that in Wallerstein the "modern" in the "modern world-system" refers to approximately the last 500 years--most of which many scholars would not label "modern," or "recent," or "contemporary." Thus, in this essay I use "ancient" for the period before circa 1500 C. E. This, however, leaves the period from 1450 to ca 1800 problematic.

The most contentious terminological terrain is that relating to "ethnicity." To get started, I will begin with the following, typical, or garden variety meanings. I [and indeed all authors in this collection] refine and modify these definitions as we proceed. Ethnicity is a publicly recognized, shared feeling of belonging to the same group defined by kinship or biology. Race is the same thing, but typically marked by one or more phenotypical manifestations. To anticipate many comments, these markers are clearly socially constructed. Nationality is similar to ethnicity, but defined by citizenship in or subjection to a state. Clearly, a nationality may encompass many ethnicities. In this discussion I will use ethnicity as a cover term--not because I wish to gloss over important distinctions--but for the more prosaic purpose of simpler exposition.

I must note here, however, I have an important disagreement with the usage of Fred Riggs. For me ethnicity existed in the ancient world, albeit in substantially different forms. Here I follow the discussions of Smith (1994:381, 191) and Tamir (1995:494) on nationalism. To study the emergence of a new or changing social process we must be careful to not to build explanations of those changes into our concepts. Let me hasten to add that I fully concur with Fred Riggs's claim that "modern", or contemporary ethnicity is very different from ancient ethnicity. But precisely because I seek to understand how the modern version emerged from the ancient version, I do not wish to restrict the term "ethnicity" solely to the modern version.

These definitions are rather facile. In the late twentieth century due to increased race-ethnic-national mixing--the Tiger Woods phenomenon--these distinctions can blur significantly. However, in ancient times the distinctions sometimes collapsed to the same thing, especially for nonstate societies, where often--but not always--the there was no difference between "primordialist" and "situational" definitions of group identity.[1]

Furthermore, the tension between biology and politics is significant throughout the range of phenomena that these three terms address. Race is often seen as largely biological, whereas nationality is typically seen as largely political. This is far from accidental, but rather is a manifestation of the shift from kin- ordered societies to tributary or state-ordered societies (discussed below and in Chase-Dunn & Hall 1997a and Sanderson 1995a). Furthermore, imputing biological bases to political and social differences is an all-to-familiar strategy to manipulate perceptions and to influence intergroup relations.

To begin to understand and analyze the tension and confusion in these terms, I will briefly recap what we can learn from world history on the matter.

Polyethnicity Is Normal, Not National States
William H. McNeill claims that world history shows "polyethnicity as normal in civilized societies, whereas the ideal of an ethnically unitary state was exceptional in theory and rarely approached in practice" (McNeill 1986:4). By polyethnicity McNeill means what sociologists conventionally call pluralist [2] or multicultural societies, without making any of the finer distinctions used by Fred Riggs (1994, 1997a, 1997c), and especially without reference to their "modernity" in Riggs's sense. McNeill couches his argument in terms of "civilization" and "barbarism," by which he means what Gerhard Lenski (Lenski, Nolan, Lenski 1995) would call agrarian states and horticultural, pastoral, or foraging societies, or what I would call tributary and kin ordered modes of accumulation (Chase-Dunn and Hall 1997a, espec. Chs. 2 and 3).[3] In short, the drive for nation- building, especially when focused on racial-ethnic-linguistic-cultural homogeneity is in aberration of the last few centuries in the over 5,000 years of history of states. The recent resurgence of states made up of multiple ethnic groups is a continuation of the historically natural condition of states.

In a review of McNeill's book Leslie Laczko (1990) questions McNeill's assertion that poly-ethnic hierarchy is on the rise everywhere. This does not gainsay the commonality of ethnic conflict. Laczko further notes that "the work does little to clear up the widespread conceptual confusion in the field when it uses the term national unity as a synonym for ethnic homogeneity, in much the same way as all those nationalist thinkers have often done!" (Laczko 1990, p. 428). Still, McNeill's account underscores that multi-ethnic states are typical, and ethnically homogeneous states have been very rare in the both the ancient and contemporary worlds.

This raises two questions. First, why and how did an ideal of an ethnically homogeneous state ever arise, especially in the face of overwhelming evidence that ethnic-racial-linguistic-cultural diversity remains the statistical norm in states (see Gurr 1993)? Indeed, why do some still hold on to this ideal? This question I will defer until later. Second, why is ethnic diversity so pervasive? I will begin with McNeill's explanation, then turn to world-systems elaborations of his explanation.

McNeill argues that three fundamental processes constantly generate polyethnicity in states.[4] Polyethnicity is common first and foremost because states, based on agriculture for over 95% of their existence, are far more productive, have higher population densities, and are much wealthier (even if very unequally distributed) than nonstate societies. Hence they often displace or conquer the others as they spread into new territories, absorbing new peoples.

Second, this expansion is fueled by differentials in the occurrence disease. As McNeill notes (1986:12; and 1976), until the mid 19th century cities were net population sinks. Their populations always had to be replaced. Where this did not occur voluntarily, it was accomplished through force. Slaves typically became lowest ranking members of society. Occasionally, though, they were palace guards. Once in a while these palace guards took over the government (e.g., Mamluks in Egypt). In either case, the importation of new populations tended to generate and increase polyethnicity.

Third, trade led to mixing. Some long distance traders took up residence in distance lands, creating trade diasporas. This facilitated trade because traders knew that at the end of a long exchange (whether direct, or indirect) they would be dealing with people who traded by the same rules. These communities maintained their differences from their host populations and were often recognized by host governments as having the right to govern their own affairs (Curtin 1984).[5] Later, when universal religions (Christianity, Islam, Buddhism, etc.) united large areas, ethnic enclaves were no longer as necessary as before. Trade diasporas gave way to what Curtin (1984) calls trade ecumenes--that is, large areas where values were shared. Where trade diasporas persisted, members often adhered to a different world religion as well as being ethnically distinct.

To sum up, McNeill's argument is that polyethnicity is due to "the political, commercial, and epidemiologic consequences of civilized social articulation" (McNeill 1986:25). Or put in more conventional language ethnic diversity is a necessary consequence of the normal functioning of states. To be sure this diversity is far from egalitarian. Typically it is hierarchical, often extremely so (Friedman makes the same point).

I now turn to an elaboration of this argument, or what might be called a "yes, but" critique. That is, I find McNeill's argument persuasive, but somewhat miss-specified in its emphasis on states. Rather, I argue that the generation of diversity also inheres in world-systems.

The Contribution of a World-Systems Perspective
A key insight and finding of a world-system perspective is that the world-system is a fundamental unit of analysis within which all other social processes and relations must be studies (Bach 1980). This does not mean that everything can be explained by or from a world-system perspective. It only means that a world-system perspective must be a part of any broad explanation. Thus, my critique of McNeill's analysis of polyethnicity, is not that it is wrong, but that it is incomplete.

Several scholars have modified and extended world-system theory into what is typically called "pre-modern ," herein called "ancient," that is pre 1500 C.E., settings. Though they disagree on a number of issues, they all agree that processes of long-term social change since at least 5,000 years ago require study from a world-system perspective, as well as from conventional, civilizational, state, and local perspectives.[6] At least two points from my work with Christopher Chase-Dunn are especially germane here. First, all world- systems expand and "pulsate" (Chase-Dunn and Hall 1997a, 1997b), that is expand and contract, or expand rapidly then more slowly. Second, we argue that all world-systems have four boundaries demarcated by sharp declines different types of interactions:

a) those shaped by bulk goods trade networks (BGN);
b) those shaped by political military interactions (PMN);
c) those shaped by prestige or luxury goods networks (PGN);
d) those shaped by information networks (IN).

Often these are nested, as shown in Figure 1. It is only in very small, isolated world-systems, islands, and in the late twentieth century when the world-system becomes truly global, these four boundaries coincide.

Expansion of world-systems involves gradually incorporating new regions and new peoples into the system as it expands. When one world-system expands into the territory of another, typically there is some type of merger, rather than full incorporation (this is discussed in detail in Chase-Dunn & Hall 1997a). A key point here is that incorporation can occur along any of the four boundaries, and these may not coincide.

I have shown (Hall 1986, 1989a) that incorporation is not simply of matter of being in or out of the system, but rather is a matter of degree (see Figure 2). I further argued that changes in a world- system itself can cause lessening of the degree of incorporation, despite a general trend toward tighter and tighter incorporation. The pulsation of world-systems is one of the causes of such reversals, as are changes in frontier policies and practices of states, and heightened resistance to incorporation by those being engulfed. As Friedman shows, cycles of hegemony also shape processes of incorporation and processes of ethnic change (Friedman 1994, this collection).

Indeed, these shifting zones of incorporation make up many of the world's frontier regions.[7] These are the places where the formation and transformation of ethnicity is most active as various peoples and regions are incorporated into world-systems. The process of incorporation varies with the type of world-system doing the incorporation--tributary or capitalist world-systems--and with the type of group being incorporated--state or nonstate group. The difference between states and nonstate groups is a salient distinction, but not the only one.[8]

Both tributary and capitalist world-systems have a range of subtypes. Tributary can range from highly decentralize (feudal systems) to highly centralized (something akin to Marx's oriental despotism--albeit not limited to or all that common in Eastern Eurasia). Capitalist systems range from early mercantile systems in 16th through 18th century Europe to fully productive capitalist systems, the one that has increasingly dominated the world since the late 18th century. This latter distinction overlaps somewhat with Riggs's (1997a, 1997b, 1997c) distinction between capitalism and industrialism, and shares his recognition that industrial states or production capitalism differs in its social consequences significantly from all early forms of states, and I would add, world-systems.

Of special concern here is how these processes differ in ancient settings, and what we may learn from them about ethnic relations and processes. In ancient settings, political considerations are as frequent as access to trade and resources as reasons for incorporation. These reasons remain common, if not the most frequent, reasons for incorporation in early merchant capitalism. Encounters with a large variety of nonstate societies were more common than in the modern world-system, although such encounters have been common in what Wallerstein calls the "modern world-system," (1450 to present). What is distinctive, as McNeill (1986) argues, is that ancient systems and state elites did not try to force ethnic uniformity. Rather, they tolerated difference as long as their goals were met. This does not mean that there were no forced changes in religion, governance, customs and so forth. Only that complete assimilation was almost never a goal. poly-ethnicity, or multi-ethnic states, were acceptable. Furthermore, since most states were multi-ethnic, there is no reason to expect that subjects would find it odious to be conquered by yet another ethnic elite, unless of course taxes increased or they were relocated forcibly or pressed into slavery. In that case their objections, and rare rebellions, would be couched in those terms, not ethnic terms.

Especially worthy of note are the transformative consequences of incorporation. Depending on a number of specific circumstances incorporation can fragment groups, destroying fragile states and thereby engendering a number of "tribes." Other circumstances favor formation of states by amalgamating nonstate societies. Both of these processes have occurred repeatedly among Central Asian pastoralists (Barfield 1989; Hall 1991; Frank 1992; Chase-Dunn and Hall 1997a :Ch. 8). Chiefdoms may be formed from formerly band organized groups (Comanches in the American Southwest [Hall 1989a], or Araucanians in La Plata [Hall in press a; Jones in press]. If there is prolonged interaction and interbreeding (and sometimes intermarriage) new hybrid groups may be formed, such as Met¡s descended from Native American women and male French fur traders; or gen¡zaros descended from Native American and Hispanic populations in southwestern United States (Meyer 1994; Hall 1989a, 1989b).

Frequently, incorporated groups lose former autonomy and become ethnic minorities. Native Americans in United States and Canada are familiar examples. This was also the fate of the Hispanic populations of New Mexico, California, and Texas after 1848 (see Hall 1989a). Entirely new identities can be created when groups are relocated and old identities stripped and replaced with new ones, as happened to African slaves imported into the United States. Where states, or portions of states are absorbed territorial minorities are created. This was especially common in colonial areas. These absorbed groups give rise many ethnic movements and civil wars in the contemporary world (Gurr and Harff 1994).

the contemporary world differs significantly from the ancient world. First, states, for the most part, have become much larger, encompassing more minority groups. Second, states have increasingly emphasized national unity, often interpreted to mean ethnic unity, in their nation-building efforts. Third, the right to a separate, ethnically homogeneous state is often used to justify claims of independence. Erstwhile native peoples have been especially savvy in using the doctrine of sovereignty to press such claims (Wilmer 1993). Fourth, the frequency with which claims of sovereignty are pressed directly contradicts all predictions of modernization theory. These occurrences appear to be much more common in the contemporary world.[9]

Finally, all of this suggests that there has been a fundamental shift in the nature of ethnicity in the contemporary world from what it was in ancient worlds. In ancient times a primordialist approach to ethnicity is at least plausible, and possibly more accurate. In the contemporary world, by contrast, situational, reactive, or interest-based ethnicity seems to be far more common, rendering the primordialist approach almost untenable (Hall 1984). Part and parcel of this change is the increasing politicization of ethnicity, and identity in general (see Friedman 1994 and this collection). Indeed, much of the contention that attends discussions of ethnicity and race derives from the new preponderance of situationally defined groups in confrontation with primordialist conceptions of what ethnicity and race mean.[10]

Some of the difference may be an artifact of differing time scales. For instance the formation of a new ethnic group, say Met¡s or gen¡zaros, takes many generations, whereas many contemporary conflicts focus around shorter-term issues. Similarly, it took decades or even centuries of interactions among nomadic peoples and with Spaniards and Anglos to create or transform the loosely connected "bands" of nomadic peoples inhabiting what is now southwestern United States into the Indian nations we know today.

Our understanding of these differences is not helped by the propensity of many such movements to impute deeply ancient roots to relatively recent events. Some of Vine Deloria's remarks (1995) about Lakota people always having been in what are now the Dakotas come to mind, and recent controversy's about the Bering Straights theory of the population of the Americas.[11]

According to Malcolm (1994) various partisans in Bosnia have sought to manipulate and use ancient conflicts to excuse or even obscure current motivations in the war there: t

The biggest obstacle to understanding the conflict is the assumption that what has happened in that country is the product--natural, spontaneous and at the same time necessary--of forces lying within Bosnia's own internal history. That is the myth which was carefully propagated by those who caused the conflict, who wanted the world to believe that what they and their gunmen were doing was done not by them, but by impersonal and inevitable historical forces beyond anyone's control (p. xix).

Malcolm makes two points that are germane here. First, that external factors shaped the conflict (Serbian manipulations and the misunderstandings of Western powers of the nature of the conflict). Second, that roots of the conflict are not ancient, but quintessentially modern. Rather, ancient conflicts have been manipulated as a rationalization for the seizure of property and power by any means whatsoever. Malcolm notes in support of this analysis the efforts to bomb archives, as specific targets, in order to destroy the history--all the better to revise it to suit current purposes.

This is precisely what Fred Riggs (1997a, 1997b, 1997c) has called a "para-modern" process: the harmful side of modernization. Contemporary ethnic conflicts have contemporary causes. As Riggs puts it, "If anyone living today could claim descent from the Philistines, they would have plenty to fight about--primordial wars mean nothing today except in the minds of those who can use the myths to buttress their contemporary struggles" (Riggs 1997b:9). Majid Tehranian (this collection) makes much the same argument when he claims that contemporary fundamentalist movements are not holdovers or reactionary, but post- traditionalist.

The western powers have, themselves, fallen victim to the ancient roots argument. However, some observers have noted that the conflicts in central Africa are not necessarily ethnic or tribal, but about scarce resources. Ethnic differences are the excuse for depriving some people of their resources and/or eliminating them as competitors. Thus, ethnicity has become the "rationalization of choice" for a wide variety of conflicts. The irony here is that such conflicts may, themselves, be ethnogenetic, creating the very groups whose existence is imputed to have been the root of the conflict.

Conflict as a cause of ethnogenesis, however, is not limited to the modern world. Conflicts between Burmese language groups and Thai language groups helped form Burma and Siam out of loose federations of what appear to have been chiefdoms. Still, the groups today are clearly distinct in culture and state structure. However, the constant taking of prisoners in the centuries of warfare insures that neither group is genetically pure, or even distinct. This, however, does not seem to hold as strongly for the numerous "hill tribes" in those two states.[12]

As Fred Riggs argues (1997a:7), cleavages, which occur when "a community rejects that identity [of the conqueror] and demands recognition of its sovereignty as an independent or autonomous people" often give rise to violence in the contemporary world. Under this definition, cleavages were extremely rare in the ancient world. Rather, diversity was common. As Riggs notes, the presence of democracy and nationalism in the modern world (along with industrialism) create conditions which not only foment cleavages, but promote violent ethnonationalist movements when democratic means can not achieve sufficient autonomy.

Before turning the conclusions that may be drawn from this discussion, I want to comment on two puzzles it raises, but does not solve.

Two Puzzles
As alluded to earlier, this analysis of polyethnicity or multi-ethnic states generates two puzzles: (1) why the ideal of ethnic homogeneity emerged and flourished in the 18th and 19th centuries; and (2) why ethnic conflict has become so common in the late 20th century?

Why Equate States With Ethnic Homogeneity?
In the 1950s and the early 1960s modernization theory dominated the social sciences and was more-or-less taken for granted by many historians. Those who accepted the claim by modernization theory that ethnic homogeneity was an inevitable consequence of modernization were dismayed by the proliferation of ethnic conflicts around the world. One of McNeill's goals in his 1985 lectures was to chide historians for accepting this view uncritically. While few people adhere to modernization theory uncritically, and it is close to dead (Wallerstein 1976), it still has considerable hold on the popular imagination. This is all too evident in the many claims on the need for ethnic unity in order to have a strong state. It is also a widely held view among contemporary beginning college students--as those of us who teach about ethnicity know too well.

McNeill's answer is again rooted in ancient history, and partially instructive. In their very early stages of formation Rome and Athens, were, indeed ethnically homogeneous. But as they grew they became more diverse. Writers from that transitional period saw this as a problem and wrote nostalgically about the glorious (presumably ethnically pure) past (Friedman's quotation from Seneca illustrates this nicely). In the late middle ages northern Europeans thinkers were reading these ancients and presumed this was the way to greatness.

John Bartlett in the Making of Europe (1993) argues that the Norman culture of knighthood spread throughout the western end of the Eurasian land mass, creating the idea, and the actuality of Europe as a distinctive region. The highly variegated terrain of western Eurasia favored small, and hence relatively homogeneous states. Thus, the intellectuals reading the ancients found themselves in much the same situation the ancient writers. They lived in recently homogeneous states which were becoming diverse.

Increasing communication and trade created pressure for developing standardized languages and uniform practices. The world- wide burst in population growth in the mid 18th century made it easier for cities to recruit workers from the same, rather than different, ethnic group. This facilitated the development of national languages, and made an ideal of ethnic homogeneity seem feasible.

The key in both McNeill's and my own view was the growing need for larger armies. As the need for infantry grew, state elites had to devise a way of recruiting more soldiers without raising the dangers inherent in arming peasants. The boon of citizenship proved to be the answer. Citizenship, however, carried with it a right to participation (democracy in some form) and thus further heightened pressures for cultural uniformity.

This is far from a sufficient explanation for the rise of the ideal of the ethnically homogenous state. There is much more to it. As Yael Tamir (1995:438) observes, "It is truly ironic that the power of nationalism depends precisely on those aspects of the nation's image least supported by research evidence", that is, its antiquity. What remains a puzzle is how such a clearly empirically incorrect view ever took hold, and why its hold has persisted so long despite overwhelmingly contradictory evidence.

Why the Rise in Ethnic Conflict?
McNeill claims the European-based ideal of the ethnically pure state began to erode around the time of World War I. He attributes this to the waves of decolonization and the realization by imperial powers that ethnic homogeneity was neither possible nor necessary.

This account of the rise and demise of the ideal of ethnically homogeneous nationalism is far from satisfactory. But to explicate it fully would take far more time and space than is appropriate here. Clearly though, as the Bosnian example shows, ethnonationalism remains a convenient ideology for rationalizing seizure of power and property. That is, the ideal of ethnically homogeneous states can be marshaled for other political causes. Be that as it may, we can conclude with McNeill (1986:36) that: "The accelerated mingling of diverse peoples within state boundaries that we everywhere witness in our own time, and specifically since World War I, is therefore a return to normal as far as western European nations are concerned." Though I agree that recent increases in ethnic conflict are a return to "normal," this does not fully explain the timing of the rise.

I will leave aside the puzzles of why the chimera of ethnically homogeneous states persists, despite its obvious everyday empirical contradictions, and the timing of the resurgence of ethic conflict. I turn to the conclusions that emerge from the analysis of ancient states and world-systems.

Conclusions
The first, and possibly most important conclusion is that ethnically homogeneous states have been rare since the first formation of states. When they do occur they are peculiar accidents of history, or brief phases on the way to ethnic diversity. This not only reframes the problem from why so much ethnic unrest today, to why should anyone have seriously have expected otherwise, but also suggests radical changes in policies regarding ethnicity. Chief is that we should continue to criticize the drive for ethnic purity as fundamentally unnecessary, in Bosnia, Rwanda, Canada, the United States, and everywhere else. Furthermore, there is no real need for assimilationist policies. Multi-ethnicity, or in sociological terms, pluralism, is not only possible, but preferable. However, this does not give any guidelines for making it workable.

Second, there are many criteria by which groups have been distinguished. These criteria rise and fall in salience with world- systemic changes. Key differences follow the four world-system boundaries: bulk goods networks, political-military networks, prestige goods networks, and information networks. Hence there are different incorporation processes, or trajectories of incorporation, as well as varying degrees of incorporation along each of these boundaries. All give rise to different kinds of ethnic minorities. In the modern world-system, as these four boundaries have come to coincide, the mechanisms which generate ethnic groups have changed. Whether they have become fewer remains unclear. That they are different, however, is inescapable. Similarly, as Jonathan Friedman shows, cycles of hegemony in the world-system also transform identity politics in important ways.

Thus, third, ethnic phenomena and relations can not be understood solely by reference to local processes, but must always studied within their broader national and international contexts. This is true for both ancient and modern instances. Yet those contexts, are themselves, significantly different in the ancient and modern worlds. In short, lessons from the ancient world must be used with caution.

Fourth, world-systems, as well as states, create diversity. More often they transform it, primarily, but not exclusively by incorporation. The processes are part and parcel of the normal functioning of states and world-systems. Thus, ethnic diversity is normal. Similarly, cycles in the world-system (e.g., hegemony, as argued by Friedman) shape and transform identity politics in systematic ways.

Fifth, ethnicity has always been somewhat fluid with respect to identity, boundaries, cultural content, and membership. It changes through time. It is impermanent. It is this fluidity, as Tamir (1995) argues, that generates a need to impute deep historical roots--precisely to solidify that which is inherently ephemeral.

Sixth, a major difference between the modern world and the ancient world is volume and velocity of ethnic processes and the pervasiveness of states. In short, while a primordialist approach may have made sense in many ancient contexts, it very seldom does in the contemporary world. Phrased alternatively, the very nature of ethnicity has changed in the contemporary world. This is precisely why there is so much contention about what it is and means, and why it is subject to so much postmodern deconstruction and reconstruction. This difference is the basis of Fred Riggs's argument that ethnicity is a (para-)modern social phenomenon.

Seventh, in the case of nonstate societies, and even some states, incorporation into a world-system, or even a state, transforms formerly autonomous social organizations into sub-parts of a larger whole. That is, "tribes" become minority groups.

Eighth, the drive for sovereignty by such groups is an attempt to regain that autonomy. Formerly nonstate societies have been especially adept at manipulating the concept of sovereignty to their own advantage. As Franke Wilmer (1993) observes, to deny their claims to sovereignty is to deny validity of the concept of sovereignty itself.

Ninth, the processes creating ethnic conflicts are part and parcel of modernity. That is, not only did modernization theory err in predicting the demise of ethnicity (and status markers generally), if failed utterly to understand that modernity creates such ethnic conflicts. As Fred Riggs has put it (1997a, 1997b, 1997c) modernity has its downsides, ethnic conflict being a major one. I would amend his argument only slightly to note (as in the third point) this downside of modernity is a property of the world-system, not of states. Thus, explanations of its workings must attend the forces driving world-system processes.

Finally, the pursuit of ethnic homogeneity is chimera, often used to justify pursuit of other goals, like wealth and power. It is a rationalization. Like all rationalizations, it works better if there is some plausible factual basis upon which it might draw. Thus, the policy recommendations listed under the first item above. In terms of praxis, it is high time that the concept of the ethnically pure state be constantly and harshly criticized. This, however, does not mean that claims for equal treatment should be ignored. Rather, it means that the task at hand for social scientist, politicians, and citizens is to devise ways in which states and societies may function while differences are supported and even celebrated. This, to be sure, is a daunting task.

[1] For typical definitions of these approaches see, for instance, Gurr & Harff (1994:78ff).

[2] The term "pluralism" as used by sociologists and anthropologists, refers to societies composed of more than one ethnic group who share some, but not all, institutions. Sometimes a further distinction is drawn between cultural pluralism (a situation in which only some, but not all cultural attitudes, beliefs, and lifestyles are shared) and structural pluralism (where groups share only some, but not all institutions) (See Farley 1995:166-167 for a typical sociological discussion). This "pluralism" is very different from J. S. Furnivall infelicitous concept of "plural society" by which he means a society "comprising two or more elements or social orders which live side by side, yet without mingling, in one political unit" (Furnivall 1939:446). Furnivall's concept, at best would be an extreme form of pluralism. Indeed, if one pushed the sociological analysis further and noted that assimilation, pluralism, and separatism delineate a continuum of adaptation of one group to another ranging from complete to minimal, Furnivall's "plural society" would be placed very close the separatism pole of the continuum. To be fair, Furnivall developed his concept for dealing the peculiarities of the colonial situation. Indeed, I argue that his plural society is a characteristic of a certain degree and type of incorporation which cannot be explained solely from colonial society.

[3] This terminological difference reflects roots in history as a profession and a civilizational approach to history as opposed to various social science approaches. The differences are more than terminological. For discussion see Sanderson (1995b) especially chapters by McNeill and Melko, and Sanderson's introductory essays. For purposes of this paper, however, these equivalencies are adequate.

[4] I use polyethnicity, in accord with McNeill's usage, to mean any sort of racial-ethnic-cultural-linguistic mixing. Note, in my view religion is subsumed under culture.

[5] Kotkin (1993) makes much of the trading "tribes" in the contemporary world. To be sure, his point that they remain important today is well taken. He is mistaken, however, in seeing them as something new. They are as ancient as long distance trade itself.

[6] For a reviews of this literature see Chase-Dunn and Hall (1997: Chs 1-3), Frank and Gills (1993), Peregrine (1996a; 1996b); Hall (1996a, 1996b); and Sanderson (1995b). An entire subsection of the section on International Political Economy, the World Historical Systems subsection, has been formed around this subject.

[7] I have addressed the issue of frontier dynamics elsewhere. See Hall (in press a, in press b, 1996a, 1996b).

[8] I want to point out the range of types and variation within types is far greater among nonstate societies than state societies. While vitally important to understanding both long-term social change and incorporation processes, discussion of these differences would detract from the discussion at hand. These issue are discussed in detail in Hall (1989a) and Chase-Dunn and Hall (1997a).

[9] This, however, could be an artifact of our lack of detailed knowledge of ancient history with respect to ethnic movements. At this time, the preponderance of evidence strongly suggests that this is a real difference and not an artifact of selective knowledge. Still, further research on this could substantiate this claim more robustly.

[10] Fred Riggs (1994, 1997a, 1997b, 1997c) comes to a similar conclusion, that contemporary ethnicity differs significantly from ancient forms of ethnicity. He also draws a number of useful distinctions among types of ethnicity and movements that warrant closer examination than I give them here.

[11] For instance, the virtual discussion group, Native-L carried a great deal of debate about this issue. With some participants attributing anthropological/archaeological explanations of the peopling of the Americas via the Bering land bridge as an attempt to undermine and degrade indigenous religions.

[12] This discussion of Southeast Asia merely glosses very complex processes. These remarks are meant to be suggestive, not definitive. It is based on the following sources: Aung-Thwin 1985; Cady 1966; Chutintaranond 1985; Chutintaranond and Tun 1995; Coedes 1966, 1968; Hall 1985; Kulke 1986; Marr and Milner 1986; Penth 1994; Vickery 1986; Wang 1986; Wicks 1992; Wyatt 1984, 1994. Wang and Chutintaranond especially argue that the concepts of state and empire need to be rethought in the Southeast Asian context. Their discussions of the "mandala" of statelets resonates quite well with the discussion of marcher states, and state formation in Chase-Dunn and Hall 1997a, especially Chap. 5.

References

Aung-Thwin, Michael. 1985. Pagan: The Origins of Modern Burma. Honolulu: University of Hawaii Press.

Bach, Robert L. 1980. "On the Holism of a World-System Perspective." Pp. 289-318 in Processes of the World-System, edited by T.K. Hopkins and I. Wallerstein. Beverly Hills: Sage.

Barfield, Thomas J. 1989. The Perilous Frontier. London: Blackwell.

Bartlett, Robert. 1993. The Making of Europe: Conquest, Colonization and Cultural Change 950-1350. Princeton: Princeton University Press.

Cady, John F. 1966. Thailand, Burma, Laos, and Cambodia. Englewood Cliffs: Prentice-Hall. Photocopy.

Chase-Dunn, Christopher and Thomas D. Hall. 1997a. Rise and Demise: Comparing World-Systems. Boulder: Westview Press.
______. 1997b. "The Comparative Study of Pulsations." Paper presented at the International Studies Association meeting, March, 1997, Toronto.

Chutintaranond, Sunait. 1995. "'Mandala,' 'Segmentary State' and politics of Centralization in Medieval Ayudhya." Pp. 67-93 in On Both Sides of the Tenasserim Range: History of Siamese Burmese Relations, edited by Chutintaranond, Sunait and Than Tun. Bangkok: Chulalongkorn University. [first published Journal of the Siam Society 78:1(1990)].

Chutintaranond, Sunait and Than Tun. 1995. On Both Sides of the Tenasserim Range: History of Siamese Burmese Relations. Asian Studies Monographs No 050. Bangkok: Chulalongkorn University.

Coedes, Georges. 1966. The Making of South East Asia, Trans. H.M. Wright. London: Routldge & Kegan Paul. [Original Les Peuples de la Peninsule Indoshinoise, Paris, Dumond, 1962].
_____. 1968. The Indianized States of Southeast Asia, edited by Walter F. Vella, Trans. Susan Brown Cowing. Honolulu: East-West Center Press. [French 3rd ed 1964].

Curtin, Philip D. 1984. Cross-Cultural Trade in World History. Cambridge: Cambridge University Press.

Deloria, Vine Jr. 1995. Red Earth, White Lies: Native Americans and the Myth of Scientic Fact. New York: Scribner.

Farley, John E. 1995. Majority-Minority Relations, third ed. Englewood Cliffs, NJ: Prentice-Hall.

Frank, Andre Gunder. 1992. The Centrality of Central Asia. Amsterdam: VU University Press for Center for Asian Studies Amsterdam (CASA), Comparative Asian Studies No.8.

Frank, Andre Gunder and Barry K. Gills, eds. 1993. The World System: Five Hundred Years of Five Thousand? London: Routledge.

Friedman, Jonathan. 1994. Cultural Identity & Global Process. Thousand Oaks: Sage.
_____. 1997. "Transnationalization, Socio-political Disorder and Ethnification as Expressions of Declining Global Hegemony." (this collection)

Furnivall, J. S. 1939. Netherlands India. Cambridge: Cambridge University Press.

Gurr, Ted Robert. 1993. Minorities at Risk: A Global View of Ethnopolitical Conflicts. Washington, D.C.: United States Institute for Peace Press.

Gurr, Ted Robert and Barbara Harff. 1994. Ethnic Conflict in World Politics. Boulder, CO: Westview Press.

Hall, Kenneth R. 1985. Maritime Trade and State Development in Early Southeast Asia. Honolulu: University of Hawaii Press.

Hall, Thomas D. 1984. "Lessons of Long-term Social Change for Comparative and Historical Study of Ethnicity." Current Perspectives in Social Theory 5:121-144.
_____. 1986. "Incorporation in the World-System: Toward A Critique." American Sociological Review 51 (June):390-402.
_____. 1987. "Native Americans and Incorporation: Patterns and Problems." American Indian Culture and Research Journal 11:2:1-30.
_____. 1989a. Social Change in the Southwest, 1350-1880. Lawrence, KS: University Press of Kansas.
_____. 1989b. "Is Historical Sociology of Peripheral Regions Peripheral?" Pp. 349-372 in Studies of Development and Change in the Modern World, edited by Michael T. Martin and Terry R. Kandal. New York: Oxford University Press.
_____. 1991. "The Role of Nomads in Core/Periphery Relations." Pp. 212-239 in Core/Periphery Relations In Precapitalist Worlds, edited by Christopher Chase-Dunn and Thomas D. Hall. Boulder, CO: Westview Press. [see link http://www.depauw.edu/~thall/hp1.htm/books]
_____. 1996a. "The World-System Perspective: A Small Sample from a Large Universe." Sociological Inquiry 66:4(November):440-454.
_____. 1996b. "World-Systems and Evolution: An Appraisal." Journal of World-System Research, 2, No. 5 (1996): 1-109. [electronic journal http://csf.colorado.edu/wsystems/jwsr.html]
_____. in press a. "La Plata and Las Provincias Internas: Comparisons, Conclusions, Conundra." To be published in Contested Ground, edited by Donna Guy and Thomas Sheridan, University of Arizona Press.
_____. in press b. "Frontiers and Incorporation into the Modern World-System: Northern New Spain and Southwestern United States, 1598-1880." Forthcoming in Geographic Perspectives on Social Change, edited by Carville Earle and Leonard Hochberg, Stanford: Stanford University Press.

Jones, Kristine L. in press. "Comparative Raiding Economies: North and South." To be published in Contested Ground, edited by Donna Guy and Thomas Sheridan, University of Arizona Press.

Kotkin, Joel. 1993. Tribes: How Race, Religion and Identity Determine Success in the New Global Economy. New York: Random House.

Kulke, Hermann. 1986. "The Early and the Imperial Kingdom in Southeast Asian History." Pp. 1-22, in Southeast Asia in the 9th to 14th Centuries, edited by David G. Marr and A. C. Milner. Singapore: Institute of Southeast Asian Studies & Research School of Pacific Studies, ANU.

Laczko, Leslie S. 1990. Review of McNeill's Polyethnicity and National Unity in World History. Canadian Review of Sociology and Anthropology 27:3(Aug.):426-428.

Lenski, Gerhard, Patrick Nolan, and Lenski Jean. 1995. Human Societies: An Introduction to Macrosociology, 7th edition. New York: McGraw-Hill.

Malcolm, Noel. 1994. Bosnia: A Short History. New York: New York University Press.

Marr, David G. and A. C. Milner, eds. 1986. Southeast Asia in the 9th to 14th Centuries. Singapore: Institute of Southeast Asian Studies & Research School of Pacific Studies, ANU.

McNeill, William H. 1976. Plagues and Peoples. Garden City, New York: Doubleday.
_____. 1986. Polyethnicity and National Unity in World History. Toronto: University Toronto Press.
_____. 1995. "The Rise of the West after Twenty-Five Years." Pp. 289-306 in Civilizations and World-Systems: Two Approaches to the Study of World-Historical Change, edited by Stephen K. Sanderson. Walnut Creek, CA: Altamira Press.

Melko, Matthew. 1995. "The Nature of Civilizations." Pp. 11-31 in Civilizations and World-Systems: Two Approaches to the Study of World-Historical Change, edited by Stephen K. Sanderson. Walnut Creek, CA: Altamira Press.

Meyer, Melissa L. 1994. The White Earth Tragedy: Ethnicity and Dispossession at a Minnesota Anishinaabe Reservation, 1889-1920. Lincoln: University of Nebraska.

Penth, Hans. 1994. A Brief History of Lan Na: Civilizations of North Thailand. Chiang Mai: Silkworm Books.

Peregrine, Peter N. 1996a. "Archaeology and World-Systems Theory." Sociological Inquiry 66:4(Fall):486-495.
_____. 1996b. "Introduction: World-Systems Theory and Archaeology." Pp. 1-10 in Pre-Columbian World-Systems, edited by Peter N. Peregrine and Gary M. Feinman. (Monographs in World Archaeology No. 26) Madison, WI: Prehistory Press, 1996.

Riggs, Fred W. 1994. "Ethnonationalism, Industrialism and the Modern State." Third World Quarterly 13:4:583-611.
_____. 1997a. "The Para-Modern Context of Ethnic Nationalism." Unpublished Ms, available from http://www2.hawaii.edu/~fredr.
_____. 1997b. "The Malody of Modernity: Some Remedies." Available from http://www2.hawaii.edu/~fredr
_____. 1997c. "The Modernity of Ethnic Identity and Conflict." (this collection) Available from http://www2.hawaii.edu/~fredr

Sanderson, Stephen K. 1995a. Social Transformations: A General Theory of Historical Development. London: Basil Blackwell.
_____, ed. 1995b. Civilizations and World-Systems: Two Approaches to the Study of World-Historical Change. Walnut Creek, CA: Altamira Press.
_____. 1995c. "Part I: Civilizational Approaches to World- Historical Change, Introduction." Pp. 1-10 in Civilizations and World-Systems: Two Approaches to the Study of World-Historical Change, edited by Stephen K. Sanderson. Walnut Creek, CA: Altamira Press.

Sanderson. Stephen K. and Thomas D. Hall. 1995a. "Part II: World System Approaches to World-Historical Change, Introduction." Pp. 81- 94 in Civilizations and World-Systems: Two Approaches to the Study of World-Historical Change, edited by Stephen K. Sanderson. Walnut Creek, CA: Altamira Press.
_____. 1995b. "Part III: Civilizations and World Systems: Dialogue and Interplay, Introduction." Pp. 215-224 in Civilizations and World- Systems: Two Approaches to the Study of World-Historical Change, edited by Stephen K. Sanderson. Walnut Creek, CA: Altamira Press.

Smith, Anthony D. 1994. "The Problem of National Identity: Ancient, Medieval and Modern?" Ethnic and Racial Studies 17:3(July):375-399.

Stinchcombe, Arthur. 1978. Theoretical Methods in Social History. New York: Academic Press.

Tamir, Yael. 1995. "The Enigma of Nationalism." World Politics 47(April):418-440.

Tehranian, Majid. 1997. "Globalism, Localism and Islamism: The Dialectics of Migration and Identity." (this collection).

Thompson, William R., ed. 1983a. Contending Approaches to World System Analysis. Beverly Hills: Sage.
_____. 1983b. "Introduction: World System Analysis With and Without the Hyphen." Pp. 7-24 in Thompson (1983a).

Vickery, Michael. 1986. "Some Remarks on Early State Formation in Cambodia." Pp. 95-115 in Southeast Asia in the 9th to 14th Centuries, edited by David G. Marr and A. C. Milner. Singapore: Institute of Southeast Asian Studies & Research School of Pacific Studies, ANU.

Wallerstein, Immanuel. 1974. The Modern World-System: Capitalist Agriculture and the Origins of European World-Economy in the Sixteenth Century. New York: Academic Press.
_____. 1976. "Modernization: Requiescat in Pace." Pp. 131-135 in Lewis Coser and Otto Larsen, eds., The Uses of Controversy in Sociology. New York: The Free Press (reprinted as chapter 7 in Wallerstein 1979).
_____. 1979. The Capitalist World-Economy. Cambridge: Cambridge University Press.
_____. 1980. The Modern World-System II: Mercantilism and the Consolidation of the European World-Economy, 1600-1750. New York: Academic Press.
_____. 1983. "An Agenda for World-Systems Analysis," Pp. 299-308 in Thompson (1983a).
_____. 1989. The Modern World-System III: The Second Era of Great Expansion of the Capitalist World-Economy, 1730-1840s. New York: Academic Press.
_____. 1993. "World System vs. World-Systems." Pp. 291-296 in Frank and Gills (1993).
_____. 1995. "Hold the Tiller Firm: On Method and the Unit of Analysis." Pp. 225-233 in Civilizations and World-Systems: Two Approaches to the Study of World-Historical Change, edited by Stephen K. Sanderson. Walnut Creek, CA: Altamira Press.

Wang, Gungwu. 1986. "Introduction." Pp. xi-xviii, in Southeast Asia in the 9th to 14th Centuries, edited by David G. Marr and A. C. Milner. Singapore: Institute of Southeast Asian Studies & Research School of Pacific Studies, ANU.

Wicks, Robert S. 1992. Money, Markets, and Trade in Early Southeast Asia: The Development of Indigenous Monetary Systems to AD 1400. Ithaca, NY. Southeast Asia Program, 180 Uris Hall, Cornell University.

Wilmer, Franke. 1993. The Indigenous Voice in World Politics: Since Time Immemorial. Newbury Park: Sage.

Wyatt, David. K. 1984. Thailand: A Short History. New Haven: Yale University Press.
_____. 1994. Studies in Thai History. Chiang Mai, TH: Silkworm Books.

Go Tom Hall's Home Page

Thomas D. Hall is Lester M. Jones Professor of Sociology and Director of Conflict Studies at DePauw University. He is author of "Social Change in the Southwest, 1350-1880" (1989, Kansas) and along with Christopher Chase-Dunn, he is author of Rise and Demise: Comparing World-Systems (1997 Westview). email: thall@depauw.edu;

Department of Sociology & Anthropology
DePauw University
Greencastle, IN 46135-0037
765-658-4519
Web Page: http://www.depauw.edu/~thall/HP1.HTM

For other papers in the IPSR symposium see: [] Friedman || Tehranian || Riggs || Bigo ||Teune || Glossary []