The Oxford Research Encyclopedia of Asian History will be available via subscription on April 26. Visit About to learn more, meet the editorial board, learn about subscriber services.

Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, ASIAN HISTORY ( (c) Oxford University Press USA, 2019. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 23 April 2019

Ethnicity of Turkic Central Asia

Summary and Keywords

Ethnic identity is a fuzzy concept for several reasons. On the one hand, the very question of what is an ethnic group is not an easy one to answer. On the other hand, once this is established for a specific case, it is yet another task to define who belongs to it, and who does not, and how stable such assignments actually are. This is as true for Central Asia as for any other place in the world, and the fact that, for earlier periods of history, the records—both native ones and others—use a great variety of terms for human populations, does not make it any easier. Thus, it is largely unclear, which of the tribal groups or early statehoods correspond to a contemporary understanding of ethnicity.

Anthropological scholarship on Central Asia has, by contrast, stressed the rather vague and floating categories that people in the region used to define themselves and others. According to this view, the creation of ethnic groups was largely a product of more or less artificial engineering during Soviet times. Before, local communities and extended kin groups, regularly reshuffled and redefined in history, were of much greater importance for people’s identification and alliances than language or assumed genetic ties.

While there is some truth in that, the picture is more complex. Particularly among the Turkic-speaking groups in the region, a steady process of consolidation set in following the decline of the Mongol Empire, resulting in the emergence of contemporary ethnic groups out of earlier configurations. The underlying concepts of attachment and self-understanding vary, however, and can be distinguished in two different modes, roughly corresponding to the divide between nomadic and sedentary groups. Among the former, the idea of patrilineal descent, or a genealogical model, is at the bottom of internal divisions as well as external demarcation; in the oases, the prime criteria are proximity and shared culture, or a territorial model of ethnic identity. Kazaks and Uzbeks respectively represent examples of these two models.

Processes of ethnic demarcation have, however, been greatly accelerated during the Soviet period and its aftermath. Today, a hasty search for national identities can be observed across the region; while following lines of Soviet ethnicity concepts, these identities fundamentally change their understanding as well as inter-ethnic and majority-minority relations. This is still a very open and dynamic process leading to new (inter-)ethnic constellations and political power relations.

Keywords: Central Asia, Turkic, ethnicity, Uzbeks, Kazaks

Introduction: Defining Ethnic Identity

The formation and socio-political impacts of ethnic identities in Central Asia is a very complex and highly disputed topic. It is not only large scale migrations that led, over time, to frequently changing and overlapping units or boundaries of belonging that question a straightforward application of contemporary concepts, but also different modes of how to define what ethnic could mean in specific times and at specific places—or whether anything similar to our current understanding of the term has actually existed in earlier periods. While in the course of time, the Turkic elements became ever stronger in Central Asia, to clearly prevail among the contemporary groups in terms of population size and political dominance, the situation was much more shifting and complicated for most of the past. This is particularly true when it comes to regional variations, as the historical sources for many of the individual locations are often scarce.

This is, of course, not unique to Central Asia and the Turkic case at all. Ethnicity is both a fuzzy concept and a fluid reference in real life in most human societies.1 Central Asian historical sources, with all their ruptures and imprecision, simply go back much further into the past than, say, sources for Africa or the Americas. There exists more evidence for this ambiguity and fluidity, of similar or identical names with changing meanings, or referring to different populations of multi-lingual tribal confederations rather than clear-cut cultural units, of the uncertainty as to the overlap of linguistic, religious, and other markers without knowing what these categories actually meant for people of that time, or of parallel distinctions between aristocrats and common people.2 Many of the sources suggest that military alliances or religious beliefs were often more important for people’s identity than common language or ethnic origin; however, that may have been defined at a particular moment in time. They also show that external influences, such as cultural borrowings from neighboring empires and the political influence of colonial powers in more recent times, have always been of major impact on the significance of different identity markers.

The depiction of ethnic identities is further complicated by a shift in the academic literature and understanding of the concepts and their implication. This begins with the dual nature of the term identity, referring, on the one hand, to individuals and their self-understanding—or their being more or less consistent personalities over time—and on the other hand, to collective entities that experience themselves as being composed of similar human beings, sharing fundamental characteristics with each other.3 What makes any identity ethnic then is the reference to cultural closeness and/or perceived common origin, resulting in the former. As the relevant items that define culture may vary in each case, the term is equally broad and difficult to delimit from other collectivities, such as religious groups or social classes.4 It is also difficult to define because, as with any collective identity, it includes both a cognitive scheme that people have in their mind (consciously or not) and rules for conduct or mutual interaction.5

As a consequence of the ambivalent nature of the term ethnic identity and its applications, social scientists have for decades argued over the strength and solidity of the ties that develop. Within anthropology, this debate has been dominated by the contrasting views of primordialists and constructivists. Primordialists see ethnic identity as a virtually immutable feeling of belonging due to the strong impact of early socialization, which tells us how things are supposed to be done and which behaviors are acceptable or not. In this respect, it is not important if differences appear significant to outsiders in any objective way; only what is believed to be relevant by the people concerned is important.6 Constructivists, in contrast, think of ethnic identity as being fluid, responding to changing circumstances and opportunities. Thus people often have the chance to modify or manipulate their affiliation if this fits their interests.7 Most prominent in this regard is the concept of the ethnic boundary, as developed by Fredrik Barth.8 According to Barth, ethnicity should be understood as a mechanism of social differentiation, dividing people into “us” and “them,” that in each case picks up appropriate cultural distinctions to match the position of the boundary rather than the other way round. These boundaries, according to Barth, may survive any cultural change, as well as the switch of sizeable numbers of individuals across said boundaries. Analysis of historical configurations also demonstrates that, crucially, ethnic identities do not develop in isolation, where cultural purity can be preserved, but always in interaction with other equivalent phenomenon in opposition.

The Tribal Scenario

The early history of Central Asia is shrouded in mystery when it comes to the identification of specific populations in ethnic or linguistic terms. Many were simply labeled after the place of the archaeological finding, such as the Andronovo culture, because this is almost all that is known about them.9 Starting from around the third millennium bce, the names of some groups are transmitted, although it is usually unknown if they were used by the people themselves and whether they corresponded to clearly defined cultural entities with common languages or other features. In the anthropological and historical literature, these populations are traditionally referred to as tribes or tribal confederations, with the latter giving credit to their probably heterogeneous internal structure. The very term tribe, however, has come under attack in anthropology for some time because of its evolutionist connotations as well as imprecise usage in the literature. In particular, the conflation with other concepts of kinship and descent, such as clans and lineages, has turned it into a rarely used term in recent decades.10

It is indeed difficult to retrace the internal organization and identitary configuration of the earlier populations in Central Asia, even for those where a concrete name has survived. It is equally difficult to know if there existed a common pattern adopted throughout history or whether there were different types of social and political organization in place. For the first larger entities that populated the steppe region, namely the various Scythian and Sak groups, it is well established that they spoke Iranian languages, whether in their majority or at least among the ruling elites.11 The same applies for the oases further south and southeast, where the Sogdhians, Bactrians, Khotanese, and other akin groups settled. It has also been established that the later Yüe-chi, or Tokharians, spoke another Indo-European language. But for much of the earlier periods, little more than the names of rulers, dynasties, or confederations have survived, without allowing us to identify the people linguistically or ethnically. This is true in particular for the eastern steppe region where the Xiung-nu, Jou-jan, and other pastoral nomads occupied spaces, which would, a few centuries later, become the origin of the Turkic groups. While some authors claim a proto-Turkic or proto-Mongol origin, it is difficult to clarify the relationship with later configurations.12

It was legendary in and around the Altay Mountains that the label Türk emerged sometime during the 5th or 6th century, although the corresponding languages had probably existed for several centuries, if not millennia. The meaning of the term has been debated and so has the original usage as an ethnonym—a tribal denomination or, rather, a dynastic designation.13 The following centuries saw a rapid expansion during the first so-called Kök Turkic Qaghanate of the 6th to 8th centuries, which soon covered not only contemporary Mongolia but also territories up to the Aral Sea in the west. It was during this time that the gradual expansion of the Turkic languages began, first in the steppe region and later into the oases, although for a long time Iranian idioms would preserve a dominant position in most of sedentary Central Asia. The Qaghanate represented primarily a case of vassal overlordship, but it can be assumed that Turkic-speaking pastoralists slowly started to settle down near and around the oases, gradually assimilating with the local population.14 The situation did not change very much in the aftermath of the Kök Turks, when related groups or dynasties such as the Uygurs, Qarluqs, and Kyrgyz took over, although some of them restricted their area of influence to the eastern steppe zone. During this period, the Turkification especially of the eastern oases proceeded rapidly.15

What these historical developments implied for the identity of people on the ground is hard to ascertain and probably forever unknown due to the lack of first-hand information. Certainly there were numerous mechanisms of internal distinctions at work, among which language and cultural similarity may have been variables. These were tribal confederations, or proto-statehoods, with a hereditary elite and thus some kind of rudimentary class differentiation. But this should probably not be taken too far as there existed a great amount of fluidity within and between different confederations, which resulted in a rather amorphous structure when it came to ethnic differentiations. Many of the alliances between groups were primarily strategic—as in most other parts of the world as well, before the rise of the modern nation state—and grounded in a multitude of variables people utilize to affiliate themselves with others. While the Turkic languages, already split into a number of dialects, seem to have dominated in the steppe regions by then, the information for specific regions is too scarce to establish this with any certainty, and many of the groups or confederations mentioned may have been multi-lingual in themselves.16

An often-referred to example of this general fluidity and the multiple sources of ethnic identity is the historical shifts among a category of people labeled as Uygurs. When they first showed up in history, it was as the successors of the Kök Turks, from whom they inherited overrule in the Mongolian steppes in 744 ce. As such, they were probably pastoral nomads in their majority, although it is held that the urban and sedentary element had gradually gained importance. At least the ruling elite of the Uygurs had adopted Nestorian Christianity as a kind of state religion. When they were expelled by the Kyrgyz a hundred years later, the Uygurs retreated toward the southwest, where they founded several petty kingdoms along the northern Silk Road in places such as Turfan and Kocho. Another hundred years later, after Kyrgyz power had vanished, they refused an invitation by the Chinese Emperor to return to Mongolia. Apparently, a sedentary lifestyle had by then become far more attractive to them. In the course of the following centuries, however, the name Uygurs gradually disappeared from the scene. This was, by all accounts, due to the fact that it became associated with Christianity, Manichaeism, and Buddhism, all popular in these oases at some point. With the gradual conversion to Islam, the term Uygur, as a reference to paganism, was dropped, yet no new common ethnonym emerged. For many centuries, people identified themselves by their locality, as Kashgarians, Khotanese, and the like, and it was only in the early 20th century that the old term Uygur was taken up again, this time as a Soviet, and later Chinese, “nationality,” which was not necessarily referring to the same group of people or their descendants.17

Historical Consolidations

The single most influential event in the history of Central Asia and in the formation of many of the contemporary Turkic-speaking groups was, without doubt, the rise of the Mongol Empire in the 13th century. It is not the place here to discuss its origins, which is similarly mysterious as that of the earlier Turkic ones. What seems clear is that, from the 10th century onward, Mongolian-speaking groups had gradually taken over the eastern steppes of Central Asia and, by the time of Chingis Khan, had transformed this region into a political power.18 Less certain is the linguistic situation among the various larger tribes and confederations that populated the eastern steppes by that time, such as the Nayman, Merkit, Kerait, and Tatar.19 At least some of these included, in all likelihood, substantial components of Turkic speakers. Within a few generations, the Mongolian armies swept across much of Asia and Europe, establishing themselves as the largest empire the world had seen so far.20 They mixed with other nomadic groups who were Turkic-speaking in their majority. While the period of glory held for only little over a century, it had a lasting impact on the political history and ethnic configuration in the region. The earlier tribes mentioned were divided among distant military units, to prevent any possibility of resistance, and thus were spread all over the empire’s territory. Consequently, they show up today among different ethnic and national groups, such as Kazaks, Kyrgyz, Tatars, or Bashkirs.

Even before the official end of the Mongol Empire, with the death of the last emperor of the Yuan dynasty in 1368, the various branches had achieved far-reaching autonomy. Of these, the so-called Golden Horde was of particular relevance for the further development of Turkic ethnicities in Central Asia. Being the realm of the successors of Chingis Khan’s eldest son Jochi, his descendants would soon switch to Islam as the dominant religion and to Turkic—in the Qipchaq variant—as the prime language. These two changes were apparently closely related, as the Mongol language, similar to what has been said about the case of the Uygurs, had become associated with paganism and was thus abandoned by the new converts.21 This coincided with the development of a new Turko-Muslim political ideology—to take full swing during the following Timurid period—that partly replaced the previous one inherited from the steppes and also gave rise to, and was fueled by, a new genre of historiographical writing such as the accounts of Rashid ad-Din and others.22 Within this amalgamation of different tribal units, such as the mentioned Nayman, Kerait, and Merkit, two new names emerged in the course of the 15th and 16th centuries. The first of these was the Özbeks, under the rule of Abul-Khayr, who would later invade the oasis belt to the south and establish themselves as the successors of the Timurid dynasty.23 Splitting off from them were the soon-to-become Kazaks, at first an alliance of dissatisfied followers of Abul-Khayr who would gradually form into a linguistically and culturally distinct group.24 Both were ruled for most of their history by descendants of the Jochid line of Chenggisids.25 In contrast, the Kyrgyz and Karakalpaks, although similar in linguistic and cultural terms to the Kazaks, lacked any connection with the royal lineage.

To the south and east, Turkic-speakers had intermingled for centuries with the local sedentary population in the oases, which were still dominantly Iranian by then. With the continuing influx of new groups of pastoralists settling down, things took a new turn, and one oasis after the other became more and more Turkic in character. They did so, however, in specific forms. First, the sedentary population in their great majority adopted East-Turkic dialects of the Qarluq branch, which had been prominent in the region since the rule of the Qarakhanid in the 10th and 11th centuries. The Qarakhanids had already converted to Islam and thus became an acceptable model and destination for assimilative processes. Things have not changed significantly since the later invasion of the Qipchaq-speaking Uzbeks in the year 1500.26 Second, the pastoralists, upon settling down, adopted the cultural patterns of the local population and became almost indistinguishable from those who remained speakers of Iranian languages. As Khazanov puts it, while the invading nomads established their rule and linguistic behavior on the oases, the latter ultimately imposed their modes of social organization and cultural practices on the conquerors.27 In some places, this resulted in highly mixed bilingual configurations, such as Bukhara and Samarqand as paradigmatic cases, while in others, Turkification was more thorough, as in Khorezm or Tashkent. Culturally, however, there was little difference between them and the population in the eastern oases of contemporary Xinjiang, who would later become Uygurs.28 During the Mongol period, this was the realm of the descendants of Chingis Khan’s second son, Chagatay, and his name would be used to identify both a kind of early ethnicity and the corresponding language, the predecessor of contemporary Uzbek.29 The following dynasty, the Timurids, greatly enhanced the spread of the Chagatay-Turkic language and at the same time strengthened the symbiosis with the Iranian-Islamic world.

On the southwestern edge of Central Asia another set of dialects, belonging to the Oghuz branch of the Turkic languages, were spoken. Here, the Mongol Empire had somewhat less of an impact, and the various tribes today subsumed under the label Turkmen can make reasonable claims to have existed as named groups before. The internal distinctions, however, as well as their relations with those Oghuz and Seljuks who left for the Iranian and Anatolian plateaus since the 10th century, are far from established. What is clear is a stronger political and cultural orientation of the Turkmens toward the Middle East. They had no legacy of Chenggisid descent and little of an indigenous aristocracy either. In contrast to the clans and lineages among Kazaks or Kyrgyz, the different Turkmen tribes, such as Yomut, Tekke, Ersari, and others, to some degree led existences of their own and differentiated each other according to dialect and cultural expressions.30 They represent a distinct type of ethnic identity and socio-political organization within the Turkic world of Central Asia.

Soviet and Independent Nation Building

With the rise of regional colonial powers, namely Russia and China, since the 17th century things began to change fundamentally. It would take its time, however, until the Turkic-speaking areas in Central Asia were effectively subdued under Tsarist and Manjurian suzerainty, respectively. The type of governance was indirect rule, for the most part, which left local elites and aristocracies in power for the time being. Toward the end of the 19th century, however, colonial interference gradually increased, in particular in the pastoral regions.31 Like other colonial empires, Russian and Chinese variants developed an interest in determining social and administrative categories, which often made use of cultural distinctions. A prime tool in this was the first All-Russian census, conducted in 1897, when a person’s language and cultural belonging, or ethnicity was determined and fixed on paper. This proved difficult, and boundaries were not at all that clear, but it was a first step of what was to come in the near future.32 It divided in particular the sedentary population into Turkic and Iranian speakers, with a somewhat ambiguous category of Sart—variously referring to either of these categories or both at the same time—in-between. Relations with neighboring Xinjiang were intensive and included a regular exchange of population.33

Socialist engineering had as its primary aim the creation of a new, classless and internationalist humankind that would guarantee the end of inequality and exploitative relationships. For early Soviet politics, the issue was how to deal with the colonized and suppressed populations, such as those in Central Asia, who could rightly define their national ambitions as liberation movements. The chosen path, decided on by the new man in power, Stalin, was to strengthen ethnic or national self-consciousness within an overarching model of the socialist state as an umbrella for their equitable unfolding. To achieve this, ethnographers, historians, and linguists were sent out to define those groups and categories deemed acceptable for future classifications. In this process, some groups were refused recognition, such as the Sart, while others were created more or less out of nothing, as in the Altayans in Southern Siberia. People had to assign themselves to each, and only one, ethnicity, natsionalnost’ in Russian usage, a decision largely based on descent and self-declaration. While there was some flexibility in this, only those categories officially recognized could be picked.34 This system was later adopted by the People’s Republic of China, which is officially made up of 56 “nationalities.”35 It has been argued that this splitting and consolidating was partly done to prevent the rise of any larger unit of identification, such as pan-Turkic or pan-Islamic ones in the region. It has also been argued that it helped downgrade ethnic distinctions and belonging to a matter of folklore.

Parallel to this, more or less artificial administrative units of different scales were created, each built on one or more of the official ethnic units. At the highest level were the fifteen republics that constituted the Soviet Union, five of which were dedicated to the larger Central Asian ethnicities, namely Kazakstan, Kyrgyzstan, Tajikistan, Turkmenistan, and Uzbekistan. Below these were other levels, like the so-called autonomous socialist soviet republics, of which there was only one in Central Asia, Karakalpkistan within Uzbekistan, and autonomous regions, such as Gorno-Badakhshan in the eastern part of Tajikistan. In most cases, these new territorial units cut through the ethnic boundaries mentioned above, so that all of them contained substantial proportions of minorities as well. The different levels at the same time corresponded to an evolutionist hierarchy of Marxist provenance that divided humankind into different grades of development, from tribal groups to full-scale nations. Only the latter would be awarded status as a corresponding Socialist Soviet Republic.36 While there were some local disputes regarding the assignment of territories to different administrative units, this was by and large a top-down approach to define individual and collective belonging and identity.37 Because of this, scholars have been suspicious as to whether these new units of identification would make sense in everyday life. But it seems that this policy has been rather successful and turned people into Uzbeks, Tajiks, or Kyrgyz. At the same time, however, many of the previous smaller entities, such as tribal and regional affiliations, have survived until today.38

This was the situation when the five Soviet republics suddenly received their independencies in 1991. Maintaining the previous logic of ethnic and national hierarchies, it was exactly those units, based on categories that had been defined as most highly developed by the Stalin administration, that would become states of their own. Lower positioned units, such as Karakalpakistan—or Tatarstan and Chechnya for that matter—remained within the boundaries of their superordinate entities. Equipped with a state administration of their own and a Soviet-inherited understanding of ethnicity, all the newly independent republics were eager to pursue national ideologies on which to base internal and external politics for the future.39 Very often, it was not only the very concept of cultural unity and distinction that resembled the socialist period but also the concrete figures and emblems, such as national poets and musicians, historical monuments, and cultural symbols, which were preserved with few ruptures. In some cases, these led to disputes between different states in the region, in particular between Uzbekistan and Tajikistan, which share many of their historical figures and emblems.40 What differ from the Soviet legacy, however, are conceptualizations of inter-ethnic relations and the foundations of a national idea. These are taken up below with two distinct case studies as illustrations. Suffice to say that all states borrowed in varying forms from models of ethnic and civic nationalisms, as Brubaker and others have termed them.41

This change—or rather emergence—of state boundaries also implied the transformation of majority-minority relations. To some degree, all had been in a mutually equal yet subordinate position toward Russians during Soviet times, while it suddenly made a difference to be a Kyrgyz or an Uzbek in Kyrgyzstan, or to be a Tajik in Tajikistan or in Uzbekistan. These distinctions existed in the past, but their effect on everyday life was very limited. And with Russian as the ultimate state language, differences on the ground were of secondary importance. It is important to remember that these new minorities were, in most cases, not the result of migrations but were created by the sometimes-dubious drawing of boundaries by the authorities. As indicated, all new states developed distinct ideas regarding the status they were to provide to their respective minorities and the privileges granted to the titular group. The establishment of new national languages seemed a particular danger in this situation, and inevitably, some people were discriminated against.42 Different approaches toward those co-ethnics that remained outside the borders of the new states, the diasporas as they came to be called in recent years, caused additional issues. The fate of these are elaborated on in more detail within the two case studies below.

Some groups stayed outside of the new state building efforts. Among the Turkic-speaking groups, the Karakalpaks have already been mentioned. Several others live in the southern Siberian regions adjacent to Central Asia, such as Tuvinians, Altayans, or Bashkurts. More prominent is the fate of those populations under Chinese administration, notably the Uygurs in Xinjiang and some smaller groups, such as the Salar in Qinghai and the Yugur in Gansu. The floating history of the Uygurs, or earlier incarnations of groups by that name, has been explored. Most recently, Uygurs are probably best known for the suspicion the Chinese state harbors against their loyalty due to a series of violent incidents. Formally, the Uygurs have claim on an administrative unit that carries their name, the Xinjiang Uygur Autonomous Region, but the effective rights and opportunities this entails are limited.43 Due to a lack of data, it is difficult to know how deep is the unhappiness of the majority of the population. It is equally debated to what degree the ethnic name has by now prevailed over traditional local identities, such as Kashgarians, Turfanian, or Khotanese.44

Two Case Studies

These general remarks are illustrated by a detailed description of two cases of ethnic and national identity. Kazaks and Uzbeks, as well as the corresponding states, represent the two largest entities within Turkic Central Asia, and they stand for fundamentally different concepts of ethnicity, which I have elsewhere labeled as genealogical and territorial, respectively.45 Each of the two cases describes the historical evolution and changes of both categories as well as their contemporary understanding and modes of interaction across ethnic boundaries. The respective state politics and the national ideology they embrace play crucial roles. Of course, each represents an ideal type, which does not give credit to the complexities and fluidities of everyday life. But they do illustrate two models that stand for the range of possible formats ethnic and national identities may take in the region (and beyond). As such, they each resemble other categories, Kyrgyz, Karakalpak, or Uygur in particular. It may be said that, from the major Turkic groups in the region, only the Turkmens again represent a different mode of identity understanding. As there is too little ethnographic information available for their case, they are put aside for the purpose of this article.

Kazaks and Kazakstan

The origin of the Kazaks has been hinted at above. While the name, a generic term for “free riders” who break away from their overlords, has been reported for earlier periods in the Turkic world, the first mentioning for a specific group of people goes back to the 15th century.46 Unsatisfied with the rule of Abul-Khayr, some groups sought autonomy and followed Girey and Janibek Khan, both of Chenggisid background, in forming a distinct community in the eastern part of contemporary Kazakstan. Within a few decades they were able to accumulate larger numbers of people around them and gradually shifted from a social entity to one of an ethnic nature with a shared language, distinct from others around. It may be debated whether the pre-colonial organization resembled a type of statehood, similar to earlier steppe empires, or was rather a tribal confederation with an indigenous aristocracy that had little effective power beyond its immediate surrounding.47 But there seems to have been a general sense of cultural similarity across the steppe belt that showed signs of an ethnic identity.48

The territory of contemporary Kazakstan came into contact with the Russian Empire toward the middle of the 18th century when some of the nobles accepted nominal rule of the Tsar. This had little impact until the beginning of mass immigration of Russian settlers and a tightening of administrative structures during the second half of the 19th century. For many Kazaks, this meant a loss of traditional winter pastures now occupied by European peasants.49 Resistance was widespread but met with little success, and by then many Kazaks had decided to move away and seek new pasture areas in neighboring Xinjiang, which had been depopulated after the defeat of the last Oyrat Empire in 1757.50 This trend continued far into the Soviet period, in particular during the forced collectivization, which left an estimated 1.5 million, or 40 percent, of all Kazaks to die in famines, and caused hundreds of thousands to flee. When Kazakstan achieved independence in 1991, the new state thus saw itself confronted with a rather unfortunate demographic situation, from a national point of view, with only some 40 percent of the total population belonging to the titular group. To make things worse, many of those with Kazak ethnicity spoke only a rudimentary form of their alleged mother tongue, contributing to a fear of loss and weakness in the course of nation building.51

This does not, however, question the essence of Kazak identity to the degree it might do in other cases. At the core of Kazak self-understanding is the genealogical tree, the shezhire, and the positioning of each and every individual within that. At the top of the shezhire are the three hordes or zhüz, the Great (uli zhüz), Middle (orta zhüz), and Small Horde (kishi zhüz).52 Each of these hordes divides into a number of clans, which in turn consist of several levels of lineages. The smallest of these levels, in emic thinking conceptualized as the seven immediate ancestors, the zheti ata, remain key institutions today when it comes to marriage exogamy and mutual support, even if they play less of a role for economic cooperation and land entitlement than they did in pre-socialist times.53 Constitutive for the clan and lineage order is patrilineal descent, which implies that every Kazak, male and female, is clearly attached to one line of named groups, from the zheti ata to one of the hordes. It is important to note that this also implies a common origin of all Kazaks—except for the former aristocracy of Chenggisid origin and the Khoja lineages of alleged Arab background—a fact that has been increasingly stressed in national politics since independence.54 By reversal, each Kazak can, in theory at least, define exactly along which lines s/he is related with any other Kazak and how many generations they are apart by descent.

The principle behind this is a genealogical one or, in the understanding of most Kazaks themselves, a genetic one. Kazakness is usually described as being transmitted by blood. Membership in this chain of groups is defined strictly by (patrilineal) descent and is conceptualized as being very difficult to either adopt or forfeit. Children of ethnically mixed origin are by definition in-between, although this does not necessarily translate into discrimination. Shared cultural practices are important, but their lack does not terminate one’s ethnic belonging. Those who do not speak their mother tongue or do not practice Islam may be bad representatives, but they are still Kazaks.55 Nevertheless, language is a key element of ethnic identity, and the fact that differences in dialect are marginal across a territory of several thousands of kilometers supports the idea of a common origin. When it comes to religion, the fact that nomads were allegedly always less pious is, by many Kazaks, turned into an aspect of positive differentiation toward the sedentary Uzbeks and Uygurs, who are described as fanatics and hypocrites.56

These basic principles are reflected also in national politics and in relationships with other ethnic groups and their own diasporas outside the state boundaries. Many authors consider Kazakstan a prime example of an ethnic nation state, which is built around a titular group with other minorities as different by definition. As a consequence, many Kazaks understand the state as theirs to some degree, while others are granted the status of permanent guests. Members of minorities complain about discrimination arising out of this. While discrimination is rarely observed in public life, the proportion of Kazaks in superior positions clearly indicates a certain truth to this.57 At the same time, this scenario does not promote any kind of systemic assimilation because all ethnic groups are perceived as profoundly distinct, and boundaries as more or less clear-cut and hard to permeate. In a similar vein, the diasporas, those descendants of Kazaks who left to settle in China, Mongolia, Turkey, and other places during the last two centuries, are considered kin, even if they behave very differently and even if one does not particularly appreciate their presence. As part of the common genealogical framework they are ultimately members of the titular group. Thus, despite much reservation, the majority has accepted the idea of the government inviting all Kazaks living abroad to join the native homeland (which approximately one million of them did by the mid-2010s). At the same time, however, a growing concern can be felt as to the ethnicity of the pre-Turkic inhabitants of the current state territory. While there was little debate in the early years of independence, the idea of the Scythians as ancestors of the contemporary population has been on the rise, and numerous exhibitions and excavations have been devoted to this. This might be termed a growing element of territorial thinking in the otherwise genealogically defined national concept, which is probably intrinsically linked to the idea of the state.

Uzbeks and Uzbekistan

While today the model of Uzbek identity is sharply contrasted with that of Kazaks, and the two respective states stand in clear opposition toward each other regarding regional hegemony, the origins of both are closely linked. It was the Uzbek, or better Özbek, confederation of Abul-Khayr from which the early Kazaks separated, in all likelihood a union of similar language, culture, and, presumably, identity at that time. Deriving their name from a direct descendant of Chingis Khan, Özbek Khan (1282–1341), a former ruler of the Golden Horde and instrumental in its conversion to Islam, the confederation consolidated again after this split and, at the turn of the 16th century, invaded the southern oases belt under the leadership of Mukhamad Shaybani, expelling the last Timurid rulers. This was the beginning of the Uzbek Khanate, which soon split into three political entities, named after the cities of Bukhara, Khiva, and Kokand. Each city was ruled by a distinct line of Chenggisids until the later 18th century, when other tribal dynasties took over.58

Colonial rule came to the oases during the second half of the 19th century. Within a few years, all the major towns were captured. The Khanate of Kokand was fully integrated into the Tsarist administration after a series of uprisings had been defeated.59 Khiva and Bukhara were forced to accept Russian suzerainty but continued as nominally independent states into the early Soviet period. The latter began less dramatically than had been the case for the Kazak steppes but would similarly result in a fundamental transformation of social structures and ethnic belongings. In the case of the oases, the national delimitation project proved particularly difficult, due to the intimate cohabitation of Turkic and Iranian speakers for centuries, which resulted in highly complex and usually bi- or multi-lingual local configurations with little or no internal boundaries. The Soviets had force people to define themselves as either Tajiks or Uzbeks, the latter having become—somewhat arbitrarily—the chosen label for the Turkic-speaking part of the population.60 Although the term was, until that time, hardly used among the oasis dwellers, it gained rapid acceptance and is today clearly the dominant reference of self-identification. Nevertheless, independence left the new state with a rather heterogeneous ethnic configuration, with the continuing meaning of sub-ethnicities such as Laqay or Qurama, and with regionally very different concepts of identity.61

Genealogical descent, as described for the Kazaks, plays hardly any role in defining Uzbek identity. This is intriguing given the fact that the original followers of Mukhamad Shaybaniy in all likelihood had a similar type of social organization and that for centuries the Uzbeks were described in the literature as consisting of 92 or 99 tribes.62 There are at least two complementary explanations for this. One is that the majority of the Uzbeks of today are not descendants of the 16th century invaders but of the population that had been settled in the oases for generations. This is attributed by the fact that, in contrast to the Qipchaq dialects of the Shaybanids, most Uzbeks speak a Qarluq variant of Turkic, a heritage of the Qarkhanid period, which also became the basis of the literary language.63 Indeed, many, if not most Uzbeks descended from a Turkified Iranian population, which presumably never had any lineages or clans. The other explanation is that even those legendary tribes apparently did not form elaborate systems of genealogical descent or different levels of named groups but were merely ascriptions of political belonging. Thus, the general mode of identification in the oases was traditionally, and so is today, by locality, as Bukharians, Khorezmians, and the like. Within these territorial clusters, linguistic preferences may be of secondary importance to shared cultural patterns and economic lifestyles.64

What this adds up to is, in contrast to the Kazak case, a concept of identity based on territorial belonging. One is Uzbek by living in a particular setting, growing up within specific social and cultural patterns, and speaking the corresponding local language(s). Members of other groups or categories, such as Tajiks, Arabs, or Turkmens, can—depending on their perceived similarity—be or become part of the local variant of Uzbekness. And Uzbeks in other sites are usually perceived as being less similar and less close than their fellow oasis men. Because inter-marriage is the norm in many settings, rather than the exception, hybridity becomes part of the very concept itself. More important, the transmission of identity and belonging is a matter of socialization, more than descent, and will thus depend on where and with whom people grow up. As a consequence, it is also much easier to change one’s ethnic affiliation, for example after marriage or a move to a different settlement. Over the centuries “Uzbekness” in the contemporary sense of a sedentary Turkic speaker, was a highly attractive model as it occupied an intermediate position between the local Iranian population and the Turkic nomads outside the oases.65 Being Muslim is a component of this cultural pattern, and though co-brethren from other parts of the Islamic world may look down upon the low level of knowledge regarding religious practices, this affiliation forms an important part of identity in contrast to the former nomads, believed to be less sincere about this.

As in the first explanation, this conceptualization of ethnic identity can be recognized in the corresponding design of the nation state. From the onset, independent Uzbekistan was thought of as a state that inherited an age-old tradition of a sedentary civilization. For the heirs to all previous populations, no matter which language they spoke, the focus is not so much on genealogies and purity but on inclusion and lifestyle. This may also explain the choice of Timur as the national hero, rather than Mukhamad Shaybaniy, who was considered an uncivilized nomad from the steppes.66 Minorities are welcome in this model and may even become part of the mainstream society if they behave appropriately. A particular issue in this regard is the large numbers of Tajik-speakers in places like Bukhara or Samarqand who are perceived as Uzbeks by the government and often have this as their official ethnicity mentioned in their passports. Rumor has it that the long-term president Islom Karimov, who died in 2016, was himself of half-Tajik origin. National politics thus have a stronger assimilation component than in Kazakstan. They also lack an idea of irredentism or the invitation of diasporas. As an extension of the territorial principle, Uzbeks in Kyrgyzstan, Kazakstan, or Tajikistan are different by definition, because they grow up in other circumstances and lack the same patterns of socialization. Sad proof for this was given to Uzbeks trying to flee violent conflict in Afghanistan or Kyrgyzstan who were stopped from entering their “ethnic homeland.”

Inter-Ethnic Relations and Conflict

Ideas about oneself always imply difference about those who are “others.” Following Barth, these distinctions are at the core of any self-understanding.67 To think of ethnicity in terms of isolating specific elements within groups, that is to say to understand them out of themselves, is like trying to clap with one hand, as Eriksen puts it.68 Mutual differentiations are constitutive for any type of ethnic identity. These often come in the form of stereotypes, or standardized versions of how others are imagined and expected to be. In cognitive terms, they correspond to cultural schemes or associative models that people use—consciously or unconsciously—to screen real life situations for similar cases and experiences they have had in the past.69 Schemes are individual representations in the mind, in the sense that they derive from personal experiences, but they also show a degree of cultural commonality, built on socialization within larger social entities such as ethnic groups.

In Turkic Central Asia, as in other regions of the world, one major dividing line is that between (former) pastoral nomads and sedentary oases-dwellers. While data on the more distant past does not give a clear picture, it seems fair enough to assume that this is a rather old pattern of distinction. It refers to types of social organization and cultural patterns, as well as attitudes toward religion.70 One common feature is the idea among sedentary people that nomads are less civilized, dirty, and lazy. Conversely, pastoralists used to think of their counterparts as dishonest, faint-hearted, and prone to religious fanaticism. The term Sart epitomizes this distinction in many ways. For the nomads, and indeed for many contemporary Kazaks or Kyrgyz, it represents all sedentary people. Very much a derogatory label until the present, it may be used equally for Uygurs, Uzbeks, and Tajiks, but also, for example, by northern Kyrgyz for their southern co-ethnics to indicate their closeness to Uzbek culture.

Cultural schemes of “us” and “them” are not merely concepts that people have in their mind to position themselves within larger social settings. They are also guidelines for factual behavior and interaction. This does not say that people have rules of conduct for dealing only with their own kind. But those dealing with the “outside world” may be more or less elaborated, and thus more or less predictable. Inter-ethnic marriage is usually considered a prime indication of close ties. That this is not necessarily the case has been described already by Leach for the case of Burma, where marriage alliances systematically transcend ethnic boundaries without, however, questioning or blurring them.71 In most of Central Asia, mixed families of Uzbek and Tajik origin are very common and are usually considered fully appropriate. Those with other ethnic groups do exist but are less well looked upon.72 This applies even more to unions with non-Muslims. In the case of Western Mongolia, for example, there is hardly any case of Kazaks marrying with other local groups—that is to say Buddhists—although everyday relations are by and large peaceful.73 Another issue is economic specialization, very often associated with ethnicity. The traditional nomadic versus sedentary opposition is, of course, a case in point. This has been leveled during socialist times, but there were clear niches, such as Russians as industrial workers versus Uzbeks and Tajiks as primarily rural populations, making a living from agriculture. But it would probably be misleading to speak of a class-like hierarchy of ethnic relations in Central Asia (with the possible exception of Xinjiang).

As described, the traditional fluidity of ethnic boundaries, probably stronger among the sedentary than among the pastoralist populations, has been transformed to some degree by Soviet nationality policies that tried to fix people into pre-given categories. This has been successful, by and large, although there are still regions where boundaries are fuzzy, such as the Uzbek-Tajik settlements in the oases of Bukhara and Samarqand. The idea of ethnic and national entities as self-explaining units with clear-cut boundaries and individual assignments has been a very strong one. The new nation states follow this line, although with some differences, as demonstrated by the two case studies noted. Assimilation, or the unilineal shift of identitary belonging is by no means confined to the state of Uzbekistan. In Tajikistan, it is by contrast the local Uzbeks who come under increasing pressure to change their identity. In all states, the titular groups exert some attraction on others to join them for the structural advantages they provide. It is just that a move to become Kazak or Kyrgyz is more difficult to implement because it demands the fabrication of a genealogical slot in the overall social configuration, which is no precondition for becoming an Uzbek.74

Not all encounters between, or within, ethnic groups are peaceful. Public opinion has it that many conflicts in non-European regions have an ethnic cause at their beginning. This is misleading, and even if alliances and hostilities develop along such lines, the motivation to fight one another typically derives from other factors, be they economic resources or social and political discrimination.75 Luckily, the degree of violence in Turkic Central Asia since the days of independence has been less than expected. Still, the violent conflicts in southern Kyrgyzstan in 1990 and 2010, which each left several hundred people dead, are well remembered and have bedeviled Uzbek-Kyrgyz relationships profoundly.76 Less problematic in contrast to early expectations seem to be those relationships between indigenous groups and European immigrants. While Russians in Central Asia nowadays largely lead a life of their own, they are rarely despised or attacked. This is different in Xinjiang where many Uygurs experience the on-going Han-Chinese immigration as an immediate threat to their survival as a distinct ethnic group.77

A different situation, though somehow related, is tensions within ethnic groups that may be of similar explosiveness. One case in point is the Kazak repatriants, the oralman, who—as described above—have been officially invited to join their motherland but are deeply mistrusted and despised by many locals.78 Much more serious, of course, are the regional cleavages within Tajikistan, which are believed to have been a major source for the devastating civil war during the 1990s.79 In the case of Kazakstan, Kyrgyzstan, and Turkmenistan, internal conflicts along lines of descent were expected, and the threat of tribalism has become a popular feature both in academia and in local disputes.80 It has been pointed out that other authors see genealogical ties as rather uniting people.


Ethnic identity in Turkic Central Asia has gone through a long history of changes and variations. A crucial element in this has been the frequent and often overlapping migration movements that have brought people in new contacts and demanded new adaptations. One illustrative moment was the reshuffling of tribal units of various kinds and sizes during the period of the Mongol Empire, which resulted in such names as Nayman or Kerey showing up among different contemporary ethnic groups throughout Eurasia. We may only assume that this is based on the fact that members of the same clans and lineages ended up in various parts of the empire, contributing to the emergence of different new groups, as has been demonstrated for a number of East-African pastoralists as well.81

While shifts and migrations occurred in different directions, depending on circumstances, the overall trend since the early medieval has been a gradual process of Turkification, primarily at the cost of Iranian idioms once dominant in most of the region.82 Merging with the previous population different modes of identity have developed, based primarily on the respective economic lifestyle as either pastoral nomads or sedentary oases dwellers. For the former, mobility was a key issue, and wide-ranging genealogical connections secured access to distant pasture areas in times of need. In the oases, by contrast, the stress was on co-habitation and cultural similarity with a territorial concept of identity. Kazaks and Uzbeks respectively, taking the contemporary ethnic or national groups as points of reference, provide a kind of prototype for each of these models. The two corresponding states of today, it has been argued, resemble the “mechanisms behind” in regard to national ideology, minority treatment, and diaspora attitudes.

Ethnic identity, however, always contains ideas about others as well as rules for how to deal with them. These rules may be more precise or more restricting in some spheres, such as marriage, than in others. The nomadic/sedentary dichotomy has proved a strong dividing line until the present day because it stands for a larger idea of cultural practices, social organization, and individual behavior. It was also a main obstacle for an overarching pan-Turkic identity to develop. Even though livelihoods today do not represent as clear-cut oppositions anymore, the perception of difference is still alive and guides many of the mutual prejudices and constraints on social interaction. It has also fueled, to some degree, the conflict between Kyrgyz and Uzbeks, the most devastating Turkic Central Asia has seen in recent decades. But it has never prevented social interaction and the crossing of ethnic boundaries.

Discussion of the Literature

Historical sources are comparatively rich for Central Asia in general and the Turkic groups in particular. Most of these have been written by outsiders, often in hostile relations to the people concerned, so it is essential to maintain a certain precaution. And, obviously, in their majority they were not intended to capture the self-understanding of local communities and their shifting natures but were rather designed from a court or colonial perspective. The historical sources, indigenous as well as others, have been analyzed in great detail by a number of scholars, including Golden and more recently Paul.83

One could argue that the historical engagement with Central Asia goes back all the way to Herodot and his depiction of the Scythians, as the ultimate “others.” Travelers of later centuries include such famous figures as Marco Polo, Rubruck, Carpini, and Ibn Battuta, but none of these were particularly concerned with the topic at hand, namely people’s identity and mutual relationships. This changed to some degree during the colonial period when a number of insightful travel and academic reports, such as those by Levshine, Przhevalsky, Schuyler, or Radloff were produced.84 For the more eastern regions, such as Xinjiang, the works by Lattimore stand out as particularly important.85

More difficult again is the situation of the Soviet period, which was characterized by strong ideological ingredients of either Marxist provenance or anti-Bolshevism, which shaped their taking on the then famous “nationality question.” A lot has been written on the arbitrariness or malign influence of Soviet policies during this period.86 Soviet scholars developed an evolutionist model of ethnic hierarchies through time, which also provided legitimacy to the Stalinist creation of semi-autonomous regions and republics.87 Anthropological studies taking into account the way people felt about themselves and others were rare to non-existent. Works published during the Soviet era focused on topics of social and economic organization and were often based on secondary sources.88

Research has become much easier since then, although access to the now independent states of Kazakstan, Kyrgyzstan, Uzbekistan, and Turkmenistan, or to the Autonomous Region of Xinjiang in the People’s Republic of China, is not unhampered in every case. But recent decades have seen a flow of literature on various aspects of ethnic and national identity in the region.89 Much of this is within the realm of political science, which tends to look at identity formation from a top-down perspective, with states or governments defining who people are.90 Many of these studies tend to see the nation state as a kind of natural fact and try to explain why Central Asian republics still have a long way to go there. Others question the effect that imposed identities may have had on everyday relations or to what degree Soviet policies were able to suppress ethnic feelings (of animosity), which are now set free again.

Further Reading

Abashin, Sergey N. “Post-Soviet Nationalism, Ethnos, Theory, and Constructivist Critique.” Anthropology and Archeology of Eurasia 44 (2006): 58–63.Find this resource:

Allworth, Edgar. The Modern Uzbeks: From the 14th Century to the Present: A Cultural History. Stanford, CA: Hoover Institution Press, 1990.Find this resource:

Babur. The Baburnama: Memoirs of Babur, Prince and Emperor. Edited by Wheeler Thackston. Oxford and New York: Oxford University Press, 1996.Find this resource:

Baldauf, Ingeborg. “Some Thoughts on the Making of the Uzbek Nation.” Cahiers du Monde Russe et Soviétique 32 (1991): 79–96.Find this resource:

Bellér-Hann, Ildiko. Community Matters in Xinjiang, 1880–1949: Towards a Historical Anthropology of the Uyghur. Leiden, The Netherlands: Brill, 2008.Find this resource:

Bromley, Yulian. Ethnos and Ethnography. Moscow: Nauca, 1973.Find this resource:

Critchlow, James. Nationalism in Uzbekistan: A Soviet Republic’s Road to Sovereignty. Boulder: Westview Press, 1991.Find this resource:

Cummings, Sally. N. Understanding Central Asia: Politics and Contested Transformations. London and New York: Routledge, 2012.Find this resource:

Edgar, Adrienne. Tribal Nation: the Making of Soviet Turkmenistan. Princeton, NJ: Princeton University Press, 2004.Find this resource:

Esenova, Saulesh. “‘Tribalism’ and Identity in Contemporary Circumstances: The Case of Kazakstan.” Central Asian Survey 17 (1998): 443–462.Find this resource:

Finke, Peter. “Competing Ideologies of Statehood and Governance in Central Asia: Turkic Dynasties in Transoxania and their Legacy in Contemporary Politics.” In States of Mind: Power, Place, and the Subject in Inner Asia, edited by David Sneath, 109–128. Bellingham, WA: Western Washington University Press, 2006.Find this resource:

Finke, Peter. Variations on Uzbek Identity: Strategic Choices, Cognitive Schemas, and Political Constraints in Identification Processes. New York, NY: Berghahn Books, 2014.Find this resource:

Finke, Peter, and Meltem, Sancak. “To Be an Uzbek or Not to Be a Tajik: Ethnicity and Locality in the Bukhara Oasis.” Zeitschrift für Ethnologie 13, no. 1 (2012): 47–70.Find this resource:

Frenkel, Yehoshua. The Turkic Peoples in Medieval Arabic Writings. London: Routledge, 2015.Find this resource:

Frye, Richard. N. The Heritage of Central Asia: From Antiquity to the Turkish Expansion. Princeton, NJ: Markus Wiener Publishers, 1996.Find this resource:

Gladney, Dru C. “The Ethnogenesis of the Uighur.” Central Asian Survey 9, no. 1 (1990): 1–28.Find this resource:

Golden, Peter. B. An Introduction to the History of the Turkic Peoples: Ethnogenesis and State Formation in Medieval and Early Modern Eurasia and the Middle East. Wiesbaden, Germany: Harrassowitz, 1992.Find this resource:

Hirsch, Francine. Empire of Nations: Ethnographic Knowledge and the Making of the Soviet Union. Ithaca, NY: Cornell University Press, 2005.Find this resource:

Irons, William. The Yomut Turkmen: A Study of Social Organization among a Central Asian Turkic-Speaking Population. Ann Arbor: University of Michigan Press, 1975.Find this resource:

Khazanov, Anatoly. M. After the USSR: Ethnicity, Nationalism, and Politics in the Commonwealth of Independent States. Madison: University of Wisconsin Press, 1995.Find this resource:

Krader, Lawrence. Peoples of Central Asia. Bloomington: Indiana University, 1966.Find this resource:

Martin, Terry. The Affirmative Action Empire: Nations and Nationalism in the Soviet Union, 1923–1939. Ithaca, NY: Cornell University Press, 2001.Find this resource:

Radloff, Wilhelm. Aus Sibirien: lose Blätter aus dem Tagebuche eines Reisenden Linguisten. Leipzig: Weigel, 1884.Find this resource:

Roy, Olivier. The New Central Asia: The Creation of Nations. London: Tauris, 2000.Find this resource:

Schatz, Edward. A. D. Modern Clan Politics: The Power of “Blood” in Kazakhstan and Beyond. Seattle: University of Washington Press, 2004.Find this resource:

Schoeberlein, John. “Identity in Central Asia: Construction and Contention in the Conceptions of ‘Ozbek,’ “Tajik,’ ‘Muslim,’ ‘Samarkandi,’ and Other Groups.” Unpublished PhD diss., Harvard University, 1994.Find this resource:

Tishkov, Valery. Ethnicity, Nationalism, and Conflict in and after the Soviet Union: The Mind Aflame. London: SAGE, 1997.Find this resource:


(1.) Thomas H. Eriksen, Ethnicity and Nationalism: Anthropological Perspectives, 3rd ed. (London: Pluto Press, 2010).

(2.) David Sneath, The Headless State. Aristocratic Orders, Kinship Society, and Misrepresentations of Nomadic Inner Asia (New York: Columbia University Press, 2007).

(3.) Peter Finke and Martin Sökefeld, “Identity in Anthropology,” in The International Encyclopedia of Anthropology (Wiley Online Library, 2018).

(4.) Richard Jenkins, Social Identity. Key Ideas, 3rd. ed. (London: Routledge, 2008); and Eriksen, Ethnicity and Nationalism.

(5.) Peter Finke, Variations on Uzbek Identity: Strategic Choices, Cognitive Schemas, and Political Constraints in Identification Processes (Oxford and New York: Berghahn Books, 2014).

(6.) Clifford Geertz, “The Integrative Revolution: Primordial Sentiments and Civil Politics in the New States,” in Old Societies and New States: The Quest for Modernity in Asia and Africa, ed. Clifford Geertz (New York: Free Press, 1963), 105–157; and Donald L. Horowitz, Ethnic Groups in Conflict (Berkeley: University of California Press, 1985).

(7.) Anthony Cohen, “Introduction: The Lesson of Ethnicity,” in Urban Ethnicity, ed. A. Cohen (London: Tavistock, 1969), ix–xxiv; Jack Eller and Reed Coughlan, “The Poverty of Primordialism: The Demystification of Ethnic Attachments,” Ethnic and Racial Studies 16 (1993): 183–202; and John Comaroff and Jean Comaroff, Ethnicity (Chicago: University of Chicago Press, 2009).

(8.) Fredrik Barth, “Introduction,” in Ethnic Groups and Boundaries: The Social Organization of Culture Difference, ed. F. Barth (Bergen, Oslo: Universitetsforlaget, 1969), 9–38.

(9.) James P. Mallory, “Andronovo Culture,” in Encyclopedia of Indo-European Culture, ed. James P. Mallory and Douglas Q. Adams (London: Fitzroy Dearborn, 1997); and Nicolo Di Cosmo, Cambridge History of Ancient China: From the Origins of Civilization to 221 B.C. (Cambridge, UK: Cambridge University Press, 1999).

(10.) While the literature on Central Asia has been less affected by these shifts, partly due to the fact that it was outside the main interest of anthropology for most of the 20th century, this critique has been adopted recently by David Sneath. He questions not only the concept of tribe as an invention of European evolutionist thinking but also the underlying idea of (patrilineal) kinship as the basic organizing principle. According to his reading of the historical sources on the Mongols, genealogical lines and named descent groups existed only among the aristocracy, and he believes this to have been true for other groups in Central Asia as well. The social history of the region should thus be conceived more as one of class distinction rather than kin-based; see Sneath, The Headless State. It should be noted, however, that his position has come under strong critique for his interpretation of sources, for example, in Thomas Barfield, “Review: David Sneath, The Headless State: Aristocratic Orders, Kinship Society, and the Misrepresentation of Nomadic Inner Asia,” Comparative Studies in Society and History 51, no. 4 (2009): 942–943; and Peter B. Golden, “Review of The Headless State: Aristocratic Orders, Kinship Society, and Misrepesentations of Nomadic Inner Asia, by David Sneath,” Journal of Asian Studies 68, no. 1 (2009): 293–296. Also, his disregard of kinship can hardly be sustained for the Turkic groups as well (cf. Finke, Variations on Uzbek Identity.

(11.) Anatoly M. Khazanov, “The Dawn of Scythian History. Physical Description,” Iranica Antiqua 17 (1982): 49–63; and Richard N. Frye, The Heritage of Central Asia: From Antiquity to the Turkish Expansion (Princeton, NJ: Markus Wiener, 1996).

(12.) Thomas Barfield, The Perilous Frontier: Nomadic Empires and China, Studies in Social Discontinuity (Cambridge, MA: Blackwell, 1989); and Peter B. Golden, An Introduction to the History of the Turkic Peoples: Ethnogenesis and State Formation in Medieval and Early Modern Eurasia and the Middle East (Wiesbaden, Germany: Harrassowitz, 1992).

(13.) Golden, An Introduction to the History of the Turkic Peoples.

(14.) Frye, The Heritage of Central Asia.

(15.) Golden, An Introduction to the History of the Turkic Peoples.

(16.) Golden, An Introduction to the History of the Turkic Peoples.

(17.) Dru C. Gladney, “The Ethnogenesis of the Uighur,” Central Asian Survey 9 (1990): 1–28; and Ildiko Bellér-Hann, Community Matters in Xinjiang, 1880–1949: Towards a Historical Anthropology of the Uyghur (Leiden, The Netherlands: Brill, 2008).

(18.) David Morgan, The Mongols: The Peoples of Europe (Oxford: Blackwell, 1986).

(19.) Golden, An Introduction to the History of the Turkic Peoples.

(20.) Joseph Fletcher, “The Mongols: Ecological and Social Perspectives,” Harvard Journal of Asiatic Studies 46 (1986): 11–50; and Morgan. The Mongols.

(21.) Golden, An Introduction to the History of the Turkic Peoples.

(22.) Beatrice F. Manz, The Rise and Rule of Tamerlane (Cambridge, UK: Cambridge University Press, 1989); Peter Finke, “Competing Ideologies of Statehood and Governance in Central Asia: Turkic Dynasties in Transoxania and their Legacy in Contemporary Politics” in States of Mind: Power, Place, and the Subject in Inner Asia, ed. David Sneath (Bellingham, WA: Western Washington University Press, 2006), 109–128; and Yeloshua Frenkel, The Turkic Peoples in Medieval Arabic Writings (London: Routledge, 2015).

(23.) Manz, The Rise and Rule of Tamerlane; and Aurdrey Burton, The Bukharans: A Dynastic, Diplomatic and Commercial History, 1550–1702 (Richmond, VA: Curzon, 1997).

(24.) Martha B. Olcott, The Kazakhs (Stanford, CA: Hoover Institution Press, 1987).

(25.) Golden, An Introduction to the History of the Turkic Peoples.

(26.) Finke, Variations on Uzbek Identity; and Manz, The Rise and Rule of Tamerlane.

(27.) Anatoly M. Khazanov, “Nomads and Oases in Central Asia,” in Transition to Modernity ed. J. A. Hall, and I. C. Jarvie (Cambridge, UK: Cambridge University Press, 1992), 69–89.

(28.) Bellér-Hann, Community Matters in Xinjiang.

(29.) Beatrice F. Manz, “The Development and Meaning of Chaghatay Identity,” in Muslims in Central Asia: Expressions of Identity and Change, ed. Jo-Ann Gross (Durham, London: Duke University Press, 1992), 27–45.

(30.) William Irons, The Yomut Turkmen: A Study of Social Organization among a Central Asian Turkic-Speaking Population (Ann Arbor: University of Michigan Press, 1975); and Adrienne Edgar, Tribal Nation: the Making of Soviet Turkmenistan (Princeton, NJ: Princeton University Press, 2004).

(31.) Seymour Becker, Russia’s Protectorates in Central Asia: Bukhara and Khiva, 1865–1924 (Cambridge, MA: Harvard University Press, 1968); Edgar Allworth, The Modern Uzbeks: From the 14th Century to the Present: A Cultural History (Stanford, CA: Hoover Institution Press, 1990); and Steven Sabol, Russian Colonization and the Genesis of Kazak National Consciousness (Basingstoke, UK: Palgrave, 2003).

(32.) Terry Martin, The Affirmative Action Empire: Nations and Nationalism in the Soviet Union, 1923–1939 (Ithaca, NY: Cornell University Press, 2001).

(33.) Bellér-Hann, Community Matters in Xinjiang.

(34.) Yuri Slezkine, “The USSR as a Communal Apartment, or How a Socialist State Promoted Ethnic Particularism,” Slavic Review 53, no. 2 (1994): 414–452; and Martin, The Affirmative Action Empire; and Francine Hirsch, Empire of Nations: Ethnographic Knowledge and the Making of the Soviet Union (Ithaca, NY: Cornell University Press, 2005).

(35.) Dru C. Gladney, “Representing Nationality in China: Refiguring Majority/Minority Identities,” The Journal of Asian Studies 53, no. 1 (1994): 92–123.

(36.) Bromley, Yulian, Ethnos and Ethnography (Moscow: Nauca, 1973).

(37.) Martin, The Affirmative Action Empire; Hirsch, Empire of Nations; and Finke, Variations on Uzbek Identity.

(38.) Finke, Variations on Uzbek Identity.

(39.) Martha B. Olcott, The New States of Central Asia (Washington, DC: United States Institute of Peace, 1996); and Sally N. Cummings, Understanding Central Asia: Politics and Contested Transformations (London: Routledge, 2012).

(40.) Stuart Horsman, “The Tajik Minority in Contemporary Uzbekistani Politics,” in Ethnicity and Democratisation in the New Europe, ed. Karl Cordell (London and New York: Routledge, 1999), 197–212.

(41.) Roger Brubaker, Mara Loveman, and Peter Stamatov, “Ethnicity as Cognition,” Theory and Society 33, no. 1 (2004): 31–64.

(42.) William Fierman, Language Planning and National Development: The Uzbek Experience (Berlin: Mouton de Gruyter, 1991); and Jacob M. Landau and Barbara Kellner-Heinkele, Politics of Language in the Ex-Soviet Muslim States: Azerbayjan, Uzbekistan, Kazakhstan, Kyrgyzstan, Turkmenistan, and Tajikistan (London: Hurst, 2001).

(43.) James Millward, Eurasian Crossroads: A History of Xinjiang (New York: Columbia University Press, 2007); and Gardner Bovingdon, The Uyghurs: Strangers in Their Own Land (New York: Columbia University Press, 2010).

(44.) Bellér-Hann, Community Matters in Xinjiang.

(45.) Finke, Variations on Uzbek Identity.

(46.) Lawrence Krader, Social Organization of the Mongol-Turkic Pastoral Nomads (The Hague: Mouton, 1963).

(47.) Olcott, The Kazakhs.

(48.) Aleksandr I. Levshin, Opisanie Kirgiz-Kazakh’ikh, ili Kirgiz-Kaisatskikh, ord i stepei (Almaty: Sanat, 1996) originally published in 1832; and Wilhelm Radloff, Aus Sibirien: lose Blätter aus dem Tagebuche eines Reisenden Linguisten (Leipzig: Weigel, 1884).

(49.) George J. Demko, The Russian Colonization of Kazakhstan, 1896–1916 (Bloomington: Indiana University, 1969); and Sabol, Russian Colonization and the Genesis of Kazak National Consciousness.

(50.) Linda Benson and Ingvar Svanberg, China’s Last Nomads: The History and Culture of China’s Kazaks (New York: M.E. Sharpe, 1998).

(51.) Bhavna Dave, Kazakhstan: Ethnicity, Language and Power (London and New York: Routledge, 2007).

(52.) In the literature, especially among historians, these three entities are often translated as Senior, Middle, and Junior Horde, respectively, because it is argued that the difference between them is based not on size but on seniority. It is true that as of today, the Uli Zhüz is in fact the smallest of the three in terms of membership, but this may have been different during earlier periods, as they were also particularly affected by violent intrusions of the west Mongolian Oyrats. Anthropologically, there is little evidence for such a claim. First, the Kazak genealogical system does not include an aspect of seniority. Inheritance is by ultimogeniture, and elder descendants do not occupy superior positions. The fact that the current president is from the Uli Zhüz is not accounted for by the latter’s seniority, and relations between the three categories is not conceptualized as hierarchical by any means. As Hudson was told by his informants back in the 1930s, it may well be that some members of the Uli Zhüz considered themselves to be superior to others; certainly no one else did so. See Alfred Hudson, Kazak Social Structure (New Haven: Yale University Press, 1938).

(53.) Hudson, Kazak Social Structure; Krader, Social Organization of the Mongol-Turkic Pastoral Nomads; and Peter Finke, Nomaden im Transformationsprozess: Kasachen in der post-sozialistischen Mongolei (Münster: Lit, 2004).

(54.) Saulesh Esenova, “‘Tribalism’ and Identity in Contemporary Circumstances: The Case of Kazakstan,” Central Asian Survey 17 (1998): 443–462; and Edward A. Schatz, Modern Clan Politics: the Power of “Blood” in Kazakhstan and Beyond (Seattle: University of Washington Press, 2004).

(55.) Peter Finke, “Historical Homelands and Transnational Ties: The Case of the Kazak Oralman,” in “Mobility and Identity in Central Asia,” special issue, Zeitschrift für Ethnologie 138, no. 2 (2013): 175–193.

(56.) Bruce Privratsky, “Turkistan: Muslim Landscape and Kazak Identity,” Journal of Central Asian Studies 2, no. 1 (1997): 46–61.

(57.) Dave, Kazakhstan: Ethnicity, Language, and Power.

(58.) Allworth, The Modern Uzbeks; Burton, The Bukharans; and Svat Soucek, A History of Inner Asia (Cambridge, UK: Cambridge University Press, 2000).

(59.) Becker, Russia’s Protectorates in Central Asia.

(60.) Allworth, The Modern Uzbeks; Ingeborg Baldauf, “Some Thoughts on the Making of the Uzbek Nation,” Cahiers du Monde Russe et Soviétique 32 (1991): 79–96; and Maria E. Subtelny, “Symbiosis of Turk and Tajik,” in Central Asia in Historical Perspective, ed. Beatrice F. Manz (Boulder: Westview Press, 1994), 45–61.

(61.) Finke, Variations on Uzbek Identity.

(62.) Allworth, The Modern Uzbeks, 1990.

(63.) Interestingly, those groups of Uzbeks who still speak a Qipchaq dialect and until recently have led a semi-nomadic lifestyle in the steppe areas, are today rather marginal to the ethnic or national mainstream culture. While they could legitimately claim a direct line to the Shaybanid invaders who gave their name to the present population, they are perceived as culturally backward, even by themselves. In everyday life, they are most commonly named as “joqchi,” referring to an initial sound they pronounce differently from other Uzbeks and more similar to Kazaks or Kyrgyz (see Finke, Variations on Uzbek Identity).

(64.) Finke, Variations on Uzbek Identity.

(65.) Finke, Variations on Uzbek Identity.

(66.) Finke, “Competing Ideologies of Statehood and Governance.”

(67.) Barth, “Introduction,” in Ethnic Groups and Boundaries.

(68.) Eriksen, Ethnicity and Nationalism.

(69.) Roger Brubaker, Mara Loveman, and Peter Stamatov, “Ethnicity as Cognition”; and Finke, Variations on Uzbek Identity.

(70.) Khazanov, “Nomads and Oases in Central Asia.”

(71.) Edmund R. Leach, Political Systems of Highland Burma: A Study of Kachin Social Structure (Cambridge, MA: Harvard University Press, 1954).

(72.) Leach, Political Systems of Highland Burma: A Study of Kachin Social Structure.

(73.) Finke, Nomaden im Transformationsprozess.

(74.) Finke, Variations on Uzbek Identity.

(75.) Günther Schlee, “Taking Sides and Constructing Identities: Reflections on Conflict Theory,” Journal of the Royal Anthropological Institute 10 (2004): 135–156.

(76.) Aksana Ismailbekova and Baktygul Karimova, “Ethnic differentiation and conflict dynamics: Uzbeks' marginalisation and non-marginalisation in southern Kyrgyzstan,” in Understanding the city through its margins: pluridisciplinary perspectives from case studies in Africa, Asia and the Middle East, ed. André Chappatte, Ulrike Freitag, and Nora Lafi (London and New York: Routledge, 2018), 161–184.

(77.) Bovingdon, The Uyghurs. Strangers in Their Own Land; and Agnieszka Joniak-Lüthi, “Han Migration to Xinjiang Uyghur Autonomous Region: Between State Schemes and Migrants’ Strategies,” Zeitschrift für Ethnologie 138 (2013): 155–174.

(78.) Finke, “Historical Homelands and Transnational Ties.”

(79.) Stephane A. Dudoignon, Communal Solidarity and Social Conflicts in Late 20th Century Central Asia: The Case of the Tadjik Civil War (Tokyo: Islamic Area Studies Project, 1998).

(80.) Erlend H. Hvoslef, “Tribalism and Modernity in Kirgizia,” in Ethnic Encounter and Culture Change, ed. M’hammed Sabour and Knut Vikør (Bergen, London: Hurst, 1997), 96–108; Schatz, Modern Clan Politics; and Kathleen Collins, Clan Politics and Regime Transition in Central Asia (Cambridge, UK: Cambridge University Press, 2006).

(81.) Günther Schlee, Identities on the Move: Clanship and Pastoralism in Northern Kenya (Manchester, UK: Manchester University Press for the International African Institute, 1989).

(82.) Golden, An Introduction to the History of the Turkic Peoples.

(83.) Golden, An Introduction to the History of the Turkic Peoples; and Paul Jürgen, “Zentralasien,” Neue Fischer Weltgeschichte 10 (Frankfurt: Fischer, 2012).

(84.) Levshin, Opisanie Kirgiz-Kazakh’ikh, ili Kirgiz-Kaisatskikh; Nikolay M. Przhevalsky, Mongolia, the Tangut Country, and the Solitudes of Northern Tibet, Being a Narrative of Three Year’s Travel in Eastern High Asia (London: Sampson Low, Marston, Searle and Rivington, 1876); Eugene Schuyler, Turkistan: Notes of a Journey in Russian Turkistan, Kokand, Bukhara, and Kuldja (New York: Praeger, 1966); and Radloff, Aus Sibirien.

(85.) Owen Lattimore, Inner Asian Frontiers of China (Boston: Beacon Press, 1940).

(86.) Allworth, The Modern Uzbeks; Martin, The Affirmative Action Empire; Edgar, Tribal Nation: The Making of Soviet Turkmenistan; Alisher A. Ilkhamov, “Archaeology of Uzbek Identity,” Central Asian Survey 23, nos. 3–4 (2004): 289–326(38); Hirsch, Empire of Nations; and Sergey N. Abashin, “Post-Soviet Nationalism, Ethnos, Theory, and Constructivist Critique,” Anthropology and Archeology of Eurasia 44 (2006): 58–63.

(87.) Bromley, Ethnos and Ethnography.

(88.) Hudson, Kazak Social Structure; Lattimore, Inner Asian Frontiers of China; and Krader, Social Organization of the Mongol-Turkic Pastoral Nomads.

(89.) John Schoeberlein, “Identity in Central Asia: Construction and Contention in the Conceptions of ‘Ozbek,’ ‘Tajik,’ ‘Muslim,’ ‘Samarkandi,’ and Other Groups” (Unpublished PhD diss., Harvard University, 1994); Esenova, “‘Tribalism’ and Identity in Contemporary Circumstances”; Madeleine Reeves, Border Work: An Ethnography of the State at its Limits in the Ferghana Valley (Unpublished PhD diss., University of Cambridge, 2007); Bellér-Hann, Community Matters in Xinjiang, 1880–1949; Jay Dautcher, Down a Narrow Road: Identity and Masculinity in a Uyghur Community in a Uyghur Community in Xinjiang China, Harvard East Asian Monographs 312 (Cambridge, MA: Harvard University Asia Center, 2009); Rita Sanders, Why Did They Stay Behind? Identities, Memories, and Social Networks of Kazakhstani Germans (Unpublished PhD diss., Martin-Luther-Universität Halle-Wittenberg, 2010); Morgan Liu, Under Solomon’s Throne: Uzbek Visions of Renewal in Osh (Pittsburgh, PA: University of Pittsburgh Press, 2012); and Finke, Variations on Uzbek Identity.

(90.) Olcott, The New States of Central Asia; Schatz, Modern Clan Politics; Collins, Clan Politics and Regime Transition in Central Asia; and Cummings, Understanding Central Asia.