MANZ Multi-Ethnic Empires

Ethnic and Racial Studies
ISSN: 0141-9870 (Print) 1466-4356 (Online) Journal homepage: http://www.tandfonline.com/loi/rers20
Multi-ethnic Empires and the formulation of

identity
Beatrice Manz
To cite this article: Beatrice Manz (2003) Multi-ethnic Empires and the formulation of identity,
Ethnic and Racial Studies, 26:1, 70-101, DOI: 10.1080/01419870022000025289
To link to this article: http://dx.doi.org/10.1080/01419870022000025289
Published online: 07 Dec 2010.
Submit your article to this journal
Article views: 598
View related articles
Citing articles: 2 View citing articles
Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=rers20
Download by: [Macquarie University]
Date: 02 January 2016, At: 06:27
Ethnic and Racial Studies
Vol. 26 No. 1 January 2003 pp. 70101
Multi-ethnic Empires and the

formulation of identity
Beatrice F. Manz
Downloaded by [Macquarie University] at 06:27 02 January 2016
Abstract
This article explores the effect of multi-ethnic empires on the formulation
of identity, examining particularly identities developed before the modern
period. Imperial state structures and legitimation influenced the understanding of ethnic identities; the resulting definitions and expectations often
outlived the empire. Modern European nationalism developed from the
group feeling and ideologies of medieval and early modern Europe, influenced by the Austro-Hungarian Empire. In western and central Asia largescale secular identities also existed in the pre-modern period within several
great empires: the Islamic caliphate, the Mongol Empire, and the Russian
Empire. In these states, the connections made between various markers of
identity language, origin, and territory were unlike those in Europe, and
the expectations connected with separate identity were also different.
Despite the spread of European nationalism and the creation of modern
nation-states throughout these regions, earlier systems of identity have
survived and influenced the form of modern national sentiment.
Keywords: Empires; identity; nationalism; Central Asia; caliphate; Mongol
Empire.
The fall of the Soviet Union and the collapse of Yugoslavia have refocused attention on two topics of interest to historians: the formulation
of secular identities and the legacy of multi-national empires. In this
article I shall trace the interaction between empire and identity over time
and space, looking particularly at identities developed outside of Europe
and before the modern period. In order to analyse the development and
meaning of such group loyalties, I move away from some of the concerns
which have dominated recent discussion of communal sentiment.
Scholars have most often explored national movements as they developed in Europe and spread to the rest of the world; for the modern world
these are seen as the primary form of political and ethnic solidarity. For
some time scholars were divided between those who viewed national
2003 Taylor & Francis Ltd

ISSN 0141-9870 print/1466-4356 online
DOI:10.1080/01419870022000025289
Multi-ethnic Empires and identity 71

sentiment as a modern and constructed identity, and those who considered it natural and primordial.
A few recent investigators have abandoned the modernist perspective
but continue to see the nation as a construct, and these scholars have
traced the history of national sentiment and imperial structures as they
developed in medieval Europe. (For example, Hastings 1997) Most comparative and theoretical writing still analyses communal loyalties of the
past primarily in relation to the modern nation of the European type,
with the goal of indicating how past identities pre-shadowed the nation
(Eley and Suny 1996, pp. 45, 1819; Smith 1996, pp. 10631). In studies
of empire and its connection to identity, the state that has particularly
influenced views of multi-ethnic empires is also European: the AustroHungarian empire, whose breakup created the map of modern Europe.
The most detailed examinations of early national identity have concentrated on Europe, and a number of scholars of the non-Western
world have protested against the imposition of Western or modern
categories on peoples distant in time and space. Writers reacting against
the historiography of nation-building argue that at least outside of
Europe no national or proto-national identities existed in the past, and
that to trace their development is a distortion of history.1 Many have
argued that identity centred around religion, tribe, clan or village, and
that intermediate communities held little meaning. However, for
historians of the Asian continent in the medieval period the denial of
large-scale secular identities does not offer a solution, because such communities are constantly referred to in our sources. In the Islamic world
for instance we read of Turks, Tajiks (Persian speakers), Arabs, Kurds,
and Kazakhs; these identities clearly had importance for the people of
the period. Since these groups were recognized entities, we must attempt
to find out what they meant and we must do so within the context of
their own time.
If we examine earlier identities primarily to determine their relation
to nationalism, to class them as either national or non-national, or as
sub-national, supra-national or proto-national, we risk missing the
political meaning that identities held for people whose loyalties were
quite simply unrelated to a national framework.2 In this article, I propose
to abandon issues of classification, and to step back from both the nationstate and nationalism, to examine the formation of long-lasting group
identities which had both political and communal importance. Many of
these had a number of the markers we are familiar with in modern
nationalism of the European type common history, language and perceived affinities. They differed sharply from modern national identities
not so much in size as in the connections perceived between various
ethnic markers and the expectation aroused by common communal
feeling.
One crucial element in the formation of pre-modern identities was
72
Beatrice F. Manz
their function within the multi-ethnic empires which used to control

much of the world. State structures and systems of imperial legitimation
influenced the formation and understanding of ethnic identities; the
resulting definitions and the expectations attached to them could outlive
the empire itself, and survive within another system. I would suggest that
the imprint of earlier imperial systems of thought still exists in many
parts of the world and continues to colour perceptions of identity, definitions of difference, and expectations of community. The importance of
history in the formation of nationalism helps to preserve earlier imperial
concepts within states of a far different structure.
Although state and imperial systems of Europe fostered the development of a nationalist sentiment which has now spread throughout much
of the world, the resulting understandings of nation may not be identical to those in Europe. The centrality of the Austro-Hungarian example
to European experience and the spread of nationalism in the modern
world should not blind us to the reality and importance of other types
of structure and of identity. The new states formed in western and central
Asia in the twentieth century emerged from several major imperial traditions which interacted to form a powerful set of identities and expectations. The formation of the Islamic caliphate in the seventh century
created a new framework for the understanding of difference and of
allegiance; in the thirteenth century the Mongol Empire first threatened
this system and then adopted much of its structure and ideology. The
Russian Empire, developing in the fifteenth and sixteenth centuries, fell
heir to much of the Mongol legacy, and thus indirectly to the Islamic.
Finally, when the Soviets reconstituted the Russian Empire, they preserved many of its systems and ideas. Just as the traces of the medieval
European states and the Austro-Hungarian empire still exist, so too do
the different traditions developed in the empires of western Asia.
The Habsburg Empire
Let me start with the familiar European experience, the one from which
we wish to move away. As nationalism intensified and more peoples
joined the ranks of conscious nationalities, Europe watched the drama
played out in its one local multi-national empire. When political thinkers
of the nineteenth and early twentieth century turned their attention to
national demands, they took Austria-Hungary as their example, and saw
its problems as the ones to be solved. The empires collapse during World
War I opened the way for the formation of the political map of twentieth-century Eastern and Southeastern Europe.3 We should recognize the
influence of the Austro-Hungarian example on thinking and policies at
the turn of the century, but we need not ourselves accept it as the
defining standard of a multi-national empire; in fact, if we compare it to
empires elsewhere, we find it interestingly atypical.

The Habsburg state originated from the Holy Roman Empire of the
middle ages, and combined an imperial ideology derived from Rome
with legal legitimation derived from numerous smaller polities. (Kann
1977, pp. 711; Ingrao 1994, pp. 26) The empire was built on the local
structures of Europe, where early medieval expansion had created
regions with strong local traditions of rulership, law and custom (Bartlett
1993, pp. 35, 12330, 16771, 197220). After 1438 the crown of the Holy
Roman Empire was held almost exclusively by the members of the
Habsburg house, who used it to enhance their prestige while continuing
to amass territory. At the time that the Habsburgs rose to pre-eminence,
much of Europe was organized in composite monarchies which
combined two or more separate regions, each possessing its own crown,
a landed nobility with inherited rights, and local laws. One monarch
could most easily achieve power over several territories not by abolishing their separate crowns which could provoke resistance but by
holding several crowns at once, serving simultaneously as king of one
region and perhaps archduke of another. The ruler remained dependent
on the loyalty of subordinate regional powers and was pledged to uphold
their laws. (Elliott 1992, pp. 505) The Habsburgs acquired some regions
through marriage, some through a combination of diplomacy and
pressure. Most lands came into the dynastys possession already formed,
retaining their political structures and individual legitimacy as duchies,
principalities, or kingdoms. Within each a local diet of hereditary nobles,
ecclesiastical officials and townsmen represented their estates, formally
recognized the power of the ruler, and together symbolized the status of
the region as a political and historical entity, with separate laws and
customs.4
The empire of the Habsburgs was exceptional in Europe not for its
structure, but for the length of time that it lasted as a composite state.
While most other European states centralized in the course of the seventeenth and eighteenth centuries, the difficulties and costs of the endeavour were readily apparent in rebellions and continued resentment from
regions stripped of what they considered their legal and historic rights
(Elliott 1992, pp. 638). The Habsburgs retained their composite structure through the nineteenth century, and continued to include peoples
of different languages, customs and history. Although the claim to the
title of Holy Roman Emperor was not abandoned until 1806, the most
effective legitimation of the empire remained regional and mosaic, based
on the acceptance of the emperor as ruler of separate provinces. Several
regions such as Bohemia and Hungary had originated as kingdoms, with
a founding myth centred around the conversion of a ruler, immortalized
as a patron saint; even after the suppression of many Bohemian rights
in the seventeenth century, the Habsburg ruler had himself crowned
King of Bohemia in the cathedral of Prague (Macartney 1962, pp. 1015;
Merinsky and Meznik 1998, pp. 3951; Slma 1998, 327).
74
Beatrice F. Manz
Even with greater centralization in the eighteenth century, the AustroHungarian emperors hesitated to deny the legitimacy of all local diets
and historic titles, preferring to use formal recognition by component
parts of the empire to enhance the legitimacy of the whole. (Taylor 1948,
pp. 1417; Kann 1977, pp. 5960) To placate nationalist sentiments and
bolster their central position, Habsburg rulers in the nineteenth century
revived some local ceremonials of rule (Taylor 1948, pp. 4751, 55). The
titles borne by the last Habsburg monarch, Franz Joseph, give a vivid
illustration of the structural basis of the empire:
Franz Joseph I, by Gods grace Emperor of Austria, King of Hungary
and Bohemia, king of Lombardy and Venice, of Dalmatia, Croatia,
Slavonia, Galizia, Lodomeria and Illyria; King of Jerusalem, etc.;
Archduke of Austria, Grand Duke of Tuscany and Krakow, Duke of
Lothringen and Salzburg, . . .
and so on for about twenty more regions, followed by lesser titles and
regions in equal number (Bezecny 1908, p. 11). By the twentieth century
most of these titles may have been anachronistic, but the fiction still
mattered.
The development of nation-states in western Europe presented a
model for the smaller nationalities of eastern Europe. In the declining
years of the Austro-Hungarian empire, peoples seeking autonomy and
local privileges based their demands on the sovereignty and sanctity of
the medieval kings who had defined their regions, or on the historic
privileges granted to local hereditary nobility (Taylor 1948, pp. 2830;
Sperber 1994, p. 98; Evans 2000, p. 188). For nobles and many emergent
nationalists the possession of local legitimacy served to bolster a
developing national identity in which language, historical community
and region were closely connected. The European formulation of nation
and identity gave high priority also to another trait the possession of
a written vernacular language. The political importance of language
appears to have begun in the Middle Ages, where we see the growing
prestige of conquest languages and occasionally an appeal to common
language to justify political alliances (Bartlett 1993, pp. 198202). The
possession of a written language, most notably English, German, or
French, came to be a defining attribute of a ruling nation and was sometimes imposed on related subject peoples as a sign of their subordination
to the centre (Hobsbawm 1990, p. 37; Elliott 1992, pp. 656; Hastings
1997, pp. 39, 457, 68, 72, 99).
Within the Habsburg monarchy, German was the language of government and most of the governing class. Two other developed languages,
Italian and Hungarian, were recognized as languages of high culture and
historical importance, while most of the numerous and less developed
Slavic languages within the empire, like Czech, Serbian, or Slovenian,

held lower status. Within much of eastern Europe, the master languages were those of the landed aristocracy and the urban middle class,
who formed the elite over a largely Slavic agricultural population. With
the rise of nationalism among the peoples of the Habsburg monarchy in
the nineteenth century, the right to the development of local literary languages took a prominent place. Education and the publication of texts
in the vernacular became a common demand from the Revolutions of
1848 onwards, and the existence of historical rights, laws and corporate
identities served to justify such demands (Taylor 1948, pp. 23, 517, 626;
Havrnek 2000, pp. 124130).
As I have shown, the political development of Europe fostered a
particular set of connections among social, political and ethnic traits
which became central to modern European nationalism. Formal written
language came to be associated with political power, and ancient kingships with modern political aspirations. Within the heterogeneous lands
of the Habsburgs the accepted markers of identity language, territory,
corporate history and common customary practices very often failed
to match. In the Kingdom of Bohemia both German and Czech traditions were deeply rooted, while the Hungarian crown of St. Stephen
included lands inhabited by Magyars and numerous Slavic populations.
Nonetheless, the components of national identity were too widely
accepted to be discarded. Discrepancies in the territory inhabited by the
speakers of a language and the historical sovereignty that they claimed
led to conflict and repression or to imaginative administrative proposals,
but not to a rethinking of the formulation of nationality or of the legitimacy of the idea. The regionally based structure which I have described
was not like that of most empires; indeed for those of us used to reading
non-European history, the Habsburg empire looks peculiar, almost nonimperial. If we are to understand the legacy of identity left by more fully
universalistic empires, we should leave the Austro-Hungarian experience behind us and investigate different structures and ideologies.
The Islamic caliphate
The development of the Islamic caliphate produced a less regional ideological structure, in which language and local identities played a very
different role. When in 634 the Arabs emerged from the Arabian peninsula to conquer half of the Roman empire and all of the Persian, they
were a provincial people, but they had a new revelation and a mission
to make Gods will manifest on earth. The formation of the caliphal
empire was strikingly different from the development of the Habsburgs,
and more common in the history of empires. The Arab conquest was a
victory of the periphery over the centre; it was achieved by force of arms
and in a remarkably short time. Almost all the regions that the Muslims
took over were already part of an imperial system. As rulers of the
76
Beatrice F. Manz
Middle East, the Arabs fell heir to two traditions more sophisticated,
more complex and richer than they. They made little attempt at first to
change the societies they took over but they did formulate for themselves
a new legitimation. Their rulers were called Caliphs deputies of God,
or of His Prophet and the stunning success of their conquests provided
proof for their claims. Both the Islamic religion and the empire were universalistic and absolute, meant for all peoples alike.
The caliphate was similar to the Habsburg empire in that it preserved
differences among its subjects which served to demarcate the structure
of society and its relation to the government. At the beginning the fundamental division of society was a simple one, that between ruler and ruled.
The Muslim Arab ruling class was to remain separate and to live off the
taxes of its subjects. Most Arab immigrants were settled in new garrison
cities, and the regime taxed only the lands held by non-Muslims, on the
assumption that conquered peoples would retain their beliefs and
continue to hold most of the land. The new ruling class was defined at
once by faith and by ethnicity. The first distinction of the ruling class was
its religion Islam and the major legal distinction between populations
in the caliphate was drawn on religious lines; Muslims were taxed at a
lower rate and formed a legally privileged group. Each religious community retained its own laws and leadership over internal affairs, while
its members were subject to administrative and Islamic law in their
dealings with people outside the community. The autonomy enjoyed by
religious groups had little territorial expression; with the conversion of
the majority of the population over several centuries, Christians, Jews
and members of other religions remained as communities scattered
throughout the Islamic realm.
As a small minority ruling over more developed peoples, the Arabs
could well have become assimilated to their subjects, and their survival
as a separate people was the result of deliberate policies. Ruling in
Damascus, employing sophisticated Greek and Persian bureaucrats,
early Arab rulers cultivated a separate identity based explicitly on their
peripheral origins. They justified their status as a ruling class, first
because they were proponents of the true religion, and second because
they were Arabs: pure, tough soldiers, uncontaminated by city life. While
the Muslim ruling class might be less sophisticated than its servitors, its
members claimed the superior virtues of the soldier and nomad (Pellat
1953, pp. 1256, 13541; Goldziher 1967, pp. 11415; Beeston et al. 1983,
pp. 389, 3936). There was a deliberate attempt to retain and even reconstitute elements of past lifestyle. Although the army of conquest had not
been organized primarily along tribal lines, the caliphs revived Arab
tribal structures and formally enshrined them as military and administrative units. Tribal genealogies were preserved and elaborated as
important sources of privilege (Crone 1980, pp. 2932, 418; Hawting
1986, pp. 3642; Hinds 1971, pp. 3469, 3567). The Arabs further used

their tribal structure to block access to the ruling stratum and preserve
the distinction between groups. A new convert to Islam had to become
the formal client (mawla), of an established tribe, and this act entailed
accepting inferior status (Goldziher 1967, pp. 99104).
Arab purity and the monopoly of Islam could not be preserved
without effort. While maintaining the theory of Arab exclusivity, the
caliphs relied heavily on the personnel of the dynasties they had taken
over to administer their new lands and they lived off the fruit of nonArab labours. Arabs married, or kept, non-Arab women acquired in the
course of conquests, and these women bore quite a few of their children.
The distinction between the Arabs and their subject peoples became a
willed one, probably on both sides. The revival and collection of preIslamic poetry and the study of pure Arab language served to codify and
enshrine Bedouin Arab ways, while the development of tribal genealogy
as a science helped to limit the increasingly frequent invention of tribal
descent lines (Pellat 1953 pp. 512, 345; Goldziher 1967, pp. 12631,
16775; Morony 1984, pp. 195, 20810, 2378, 2548). The fear that
Arabs were becoming too strongly assimilated to a new, imperial,
sophisticated lifestyle was sometimes openly expressed, as Arab commanders called for a return to the starker lifestyle and military virtues
of a former time (Pellat 1953, p 5, note 2; Morony 1984, pp. 2623).
After about a century of Muslim rule, the concepts of Arab and
Muslim could no longer be considered coterminous. The classification of
populations took on a different character, now serving to distinguish one
type of Muslim from another. The way in which this happened and the
traits emphasized were a reflection of the social and political concerns
of society. At the beginning of the caliphate both the Roman and the
Persian populations influenced the Arab ruling class, but with the
triumph of the Abbasid dynasty in 750 A.D., the capital of the Islamic
state moved to Baghdad, close to the capital of the former Sasanian
dynasty, and Iranian culture became paramount in court circles. As the
number of converts increased, the requirement of attachment to an Arab
tribe and the disadvantages attached to client, or mawla, status had to
be abandoned. With the development of a significant class of educated,
bilingual, Muslim Iranians, Iranian tradition became central to Islamic
culture and learning. Although philosophical and scientific works were
translated into Arabic from Greek, works on history and statecraft came
primarily from the Persian, and the court of the caliph adopted
numerous aspects of Iranian imperial ceremonial and legitimation.5 Two
trends, however, discouraged the creation of a uniform culture and
society: first, an increased competition between Arabs and Persians in
the cultural sphere, which led to a sharp literary controversy over the
merits of different populations and, second, the importation of new
nomad manpower from the Eurasian steppes to serve in the army of the
caliphate.
78
Beatrice F. Manz
The early Abbasid period, with its active cultural borrowing, brought
renewed defensive action to protect Arab culture. The Arabic sciences
of philology, poetry and genealogy continued to develop, and Bedouin
informants became so sought after that scholars had to protect themselves against impostors. The famous ninth-century author al-Jahiz
complained of the assumed Bedouin accent of affected people, and
particularly of the nomads who pitched their tents in the neighbourhood
of main roads and busy markets, and pretended still to be true, uncontaminated Bedouin (Pellat 1969, p. 105). By this time converts to Islam,
more secure in numbers and influence, could think of aspiring to equality
with the Arabs. In the ninth century a literary movement known as the
Shuubiyya arose, which was made up of Iranian writers who complained
about continuing claims of Arab ethnic superiority and asserted the
equality if not superiority of their own traditions. The writers identified
with this party were Muslim and wrote in Arabic; the movement neither
promoted the Persian language nor disparaged Islam. What the Shuubi
party did do was to ridicule a number of Arab ideals, most particularly
those connected with desert Bedouin origins. Its adherents pointed out
that while the Persians were creating a great and sophisticated civilization, the Arabs had been in the desert, uneducated and uncouth, living
in misery and surviving on dried leather and lizards. Pro-Arab writers
replied that while the effete Persians prided themselves on elaborate
manners and luxurious fashions in hair and dress, the noble, unspoilt
Arab lived in the desert, uncontaminated by city life, delighted with his
diet of dried leather and lizards, hedgehogs and snakes (Lecomte 1965,
pp. 34951; Hutchins 1988, pp. 12832).
What is striking about the Shuubiyya is its concentration on ideal
cultural types and appropriate characterization rather than political traditions or goals. The aim of the pro-Iranian party was to enhance the
status of Iranians and Iranian culture within the Islamic world, not to
promote cultural or political autonomy. One major result of the
Shuubiyya was a set of well-defined characteristics for Iranians and
Arabs, contrasting the courtly, urban, sophisticated Iranian with the
military, tribal Arab of nomad origin. There was little disagreement on
the characteristics to be ascribed to the Arabs and the Iranians; what was
at issue was the relative worth of peoples and of the character traits
agreed on.6 These characterizations remained important, enshrined for
centuries in literature of lasting popularity, most notably in the essays of
the enormously influential author al-Jahiz.
By the ninth century, when the traits of Arabs and Iranians were
epitomized, they were already becoming archaic. The Arabs in particular were now a largely settled and civilian people, and their place in the
central armies of the caliphate was being filled by a new source of
manpower: Turkic slave soldiers imported from the steppes of Inner
Asia. In this way the Turks first arrived in the Middle East. Military

slaves proved useful and became a central part of most major armies.
These new soldiers were at once disliked and admired. They were seen
as uncouth, foreign and destructive but, nonetheless, perfect for soldiers
because of their unspoilt character and their nomadic upbringing. As alJahiz characterized them they were similar in many ways to the Bedouin
indeed, he called them the Bedouins of the non-Arabs.7
In the middle of the eleventh century the Turkic presence in the
Middle East was strengthened by the arrival of nomad Turks as conquerors and rulers. In 1035 several Muslim Turkic tribes under the
leadership of the Seljukid dynasty crossed the Oxus looking for pasture,
and soon turned to war. The Seljukids quickly conquered the central
lands of the caliphate, but they did not bring about a fundamental change
in traditions of rule because they fitted into a pattern already partly
established. Leaving the Arab caliph nominally in control, the Seljukid
leaders took the lower title of Sultan and ruled by virtue of a patent from
the caliph, and through the strength of their armies, divided now
between imported slave soldiers and the tribal Turkic soldiers they had
brought with them.
From the ninth century onwards, Arabs, Iranians and Turks coexisted
in the Islamic world. Their relationship was not without strife, but it was
based on the assumption that all should live in the same realm, in which
each had a place and function. The most salient differences among the
three peoples related to their perceived lifestyles. Arabs, whatever their
actual situation and habits, long remained in some way associated with
the image of the Arabian peninsula at the birth of Islam whether with
the Bedouin of the desert or with the modest merchant lifestyle of the
Prophet and his companions. The image of the Persian with his sophisticated and imperial past remained attached to court bureaucracy and
landed aristocratic life. Finally, the Turks were firmly characterized by
their nomad origins, combining cultural inferiority with military superiority.
Each of these groups spoke a different language, and other segments
of the populations spoke yet others. Like the Habsburgs, the caliphs
ruled over a polyglot realm, but the political importance of languages
and the relation among them developed very differently. Despite their
small numbers, the Arabs were extraordinarily successful in imposing
their language. They were helped by the central religious position of
Arabic, embodied in the Quran, and probably also by the fact that many
of their western subjects spoke Semitic languages related to Arabic. As
speakers of related languages converted, most eventually became Arabic
speakers. Starting in the beginning of the eighth century, Arabic was the
language of caliphal administration and it soon became likewise the
language of religious studies, scholarship and literature. With the dominance of Arabic, written Persian began to languish and, as we have seen,
the Iranians who argued for their cultural superiority did so in Arabic.
80
Beatrice F. Manz
For a while Arabic held a position analogous to that of both German

and Latin in the early Habsburg realm.
In the late tenth century, when a significant proportion of the Iranian
population had converted to Islam, Persian re-emerged as a written
language and a vehicle for high culture, now in the Arabic script and
with a significant percentage of Arabic vocabulary. The Persian literary
rebirth began in the eastern regions of the caliphate, promoted first by
the Iranian Samanid dynasty (8191005) which patronized literature in
new Persian and adopted the language for government administration.
However, there were few dynasties of Iranian origin in the medieval
Middle East, and what allowed New Persian its success was its use by
the numerous dynasties of Turkic origin. The next dynasty to adopt it
was the Ghaznavid dynasty (9771187), of Turkish slave origin, which
took over both the territory and the bureaucrats of the Samanids. When
the Seljukid Turks invaded the Middle East, they came in through
eastern Iran, and after they defeated the Ghaznavids, they hired their
bureaucrats and took them west to conquer Baghdad, thus returning the
written Persian language to the earlier centres of its culture. The
Samanids, Ghaznavids and Seljukids all continued to patronize literature and scholarship in Arabic as well.
The Seljukids and their nomad followers remained self-consciously
Turkic but they made no attempt to impose Turkic on their subjects or
to develop it as a literary medium. Instead they patronized scholarship
and literature in the languages of their subjects; religious studies, philosophy and scientific literature remained largely in Arabic, while history,
poetry and belles-lettres were more often written in Persian. The
Seljukid pattern of promoting cultural production in the high culture of
the subject peoples while preserving a separate identity became a
standard practice among the dynasties of the Middle East, many of
whom were of Turkic origin. The symbiotic relationship of Turkic rulers
with Iranian bureaucrats and scholars became a pattern throughout the
eastern and northern Islamic world, and a series of dynasties of steppe
origin carried Persian literature into Anatolia in the west, to India in the
Southeast. In the later middle ages Turks began to write their language
in the Arabic alphabet, with the addition of Arabic and Persian vocabulary, but only in the Ottoman Empire did written Turkic supplant Persian
or Arabic. Whereas in Europe the language of the ruling class became
the language of high culture, in the caliphate, from the time of the
Seljukid invasion, the prestigious languages were those of the subjects,
while the vernacular of the rulers remained less developed.
One particularly striking difference between the Habsburg monarchy
and the Arab caliphate lay in the relative unimportance of territorial
legitimation within the Middle East. We should not take such a trait to
imply the absence of meaningful local boundaries, loyalties and politics.
Large territorial entities created under the pre-Islamic Iranian Sasanian

Empire remained visible and important throughout the Islamic period.
(Fragner 1999, pp. 1316) The caliphate was divided into provinces
whose approximate boundaries remained stable over many centuries of
increasingly decentralized government and within provinces cities served
as reference points, providing a centre for market towns and a strong
focus for local loyalty and identity. As the caliphal government declined
in power, it was not infrequent for a governor to make himself essentially independent in a provincial capital and win the loyalty of the elite
of the provincial cities. A few regions, such as Sistan and the Caspian
provinces, appear to have harboured traditions of local rule, resulting in
a long series of dynasties based in the region. While these areas repeatedly made up parts of larger empires, they retained active and apparently relatively independent political systems.
What differentiated Islamic regionality from that of Europe was its
lack of connection with legal rights and political legitimation. Although
there appear to have been local aristocracies holding land over long
periods, city and regional elites did not have legally inherited positions
and rights, and the reciprocal ties between them and local rulers were
not formalized as they were in the European diets. Rulers might base
their power and wealth on the population of the region, but when they
had to justify their rule they turned to the centre. As long as the caliphate
existed, local rulers could legally claim only contingent sovereignty, confirmed by a patent from the caliph, though they might extort their patent
by force. If leaders needed further legitimation, they sought it in relation
to imperial traditions, notably the Iranian. In the regions which had
made up the Sassanian empire we often find rulers reviving the old titles
and claims of the pre-Islamic Persian monarchy. These pretensions
however were largely divorced from territorial claims and from actual
dynastic descent. No single dynasty was the exclusive owner of Iranian
legitimation; it could be used by any ruler, and in fact was a significant
element of the court culture of Arab caliphs and later of Mongol khans
in the Middle East.8
We find therefore that in Islamic imperial traditions, the most strongly
expressed identities brought with them no specific territorial claims. The
accepted principles of rule were based on the traditions and recognized
functions of the three major peoples of the Islamic Middle East: Arabs,
Persians and Turks. The delimitation of regional legitimacy, however, was
not connected with identity, depending instead largely on the ability to
control the area in question. Power was justified less through specific than
through general principles, which varied little throughout the Islamic
world. Claims of independent, sovereign power were limited to the caliph,
who had to be descended from the tribe of Muhammad, and thus be of
Arab origin. Other Muslim leaders acquired legitimacy through a patent
from the caliph, and proclaimed as a major goal of their rule the protection of Islam, which as a religion was still associated with a scripture and
82
Beatrice F. Manz
with the tradition of scholarship in the Arabic language. Further legitimation was available from the Iranian tradition, and was accessible to any
ruler within the lands of the former Iranian empire, or even beyond.
After the arrival of the Seljukid Turks in the eleventh century, Turkic
dynastic principles were at least informally accepted; membership in the
Seljukid house and Turkic provenance were major advantages for
aspiring rulers. Turkic legitimation, although limited to members of the
ruling tribe, could be applied in any area a member of the clan could
conquer and hold. Whatever the origin of the ruling dynasty, the standing
army was made up of troops of steppe origin, either slave troops or newly
arrived Turkic nomads. Such steppe troops had the twin advantages of
nomad skill in warfare and separation from the population as a whole.
Within the caliphal empire, it was not expected that the ruler would
be of the same ethnic group as the bulk of the population indeed, the
contrary expectation held sway. It came to be an understood principle
of politics that the ruler and military should be above and separate from
society, able to balance the needs of competing groups within society,
rather than furthering the interests of the one to which they belonged
(Mottahedeh 1980, pp. 1759). The early Arabs had emphasized their
distance and difference from the major populations they controlled, and
later caliphs, though less different from their increasingly Muslim and
Arabized subjects, ruled through a foreign military force. After the
arrival of the Seljukids, most dynasties in the Middle East and Central
Asia were of Turkic or other tribal origin.
The Mongol Empire
The Arab caliphate served as a system of power for about three hundred
years, and as a system of legitimation and social order for a further three
hundred. What destroyed it was the rise of another great universal
empire: that of the Mongols. The dynasty founded by Chinggis Khan was
far from short; it lasted from 1206 to the fall of the Kazakh khanates in
the nineteenth century. Mongol rule left a strong mark on the regions it
covered and remains a politically important issue even today. Like the
Arabs, the Mongols came from the periphery and quickly conquered a
series of regions far richer and more developed than they. Although they
brought no new religion, they claimed to rule through Gods favour,
demonstrated by the success of their conquests. The world belonged by
right to them; peoples were either submitted or rebellious, not neutral.
(Rachewiltz 1973, pp. 245) One would think the Islamic and the Mongol
empires inimical and incompatible. Both were universal, one based on
the settled and agricultural regions, the other in the nomad steppes, one
founded on strict monotheism, and the other on a combination of
paganism and prudent respect for other religions. Yet as it happened,
the two empires combined to a large extent. The Mongols conquered the

eastern, Iranian, regions of the Middle East, and this area remained
under Mongol influence for a long time. Starting at the end of the thirteenth century, the ruling class of the western Mongol Empire adopted
Islam and began to speak Turkic, the language of most of their nomad
subjects; from this time on they are best characterized as Turco-Mongolian. They formed a relatively uniform ruling class stretching over a vast
territory, from the Altai and Turkestan to the Black Sea steppe, through
Iran, eventually reaching into India.
By the early fourteenth century, the institution of the great khan had
disappeared, but the Turco-Mongolian ruling class retained loyalty to the
Mongol political system and accepted, at least in theory, the exclusive
right of descendants of Chinggis Khan to the sovereign titles of Khan and
Khaghan. Though tribal chiefs might rule independent polities, they contented themselves with modest titles and often ruled through puppet
khans. The primary legitimation for Mongol rule lay in two things: first
the charisma and universalist claims of the descendants of Chinggis Khan,
and second, the preservation of Mongol law and custom, referred to as
yasa and tre. Mongol Khans presented themselves as guardians of world
order and enhanced their prestige through large-scale cultural patronage.
Within Muslim regions, an additional principle was added, namely the
role of the ruler as protector and promoter of Islamic religion and law.
We find under the Mongols, as under the caliphate, a deliberately plural
society in which different cultures and languages were intertwined. The
ruling stratum of court and army jealously preserved its separate identity.
Turco-Mongolian courts patronized high culture primarily in the languages of their subjects, celebrating the history and culture of these people
along with their own. In the Mongol Middle East and Central Asia,
Iranians made up the peasant subject class and they also provided the
educated elite. Like earlier Turkic rulers, the Mongols hired Persian
bureaucrats and brought them into new regions now often far beyond
the borders of the caliphate. The Mongols also made free use of Iranian
legitimation, and it was under their rule that the Iranian verse epic, the
Shahnama, achieved its status as a royal book, the subject of elaborate
illustration (Fragner 1999, pp. 5161). As in the caliphate, peoples and
cultures lived together, ruled by a heterogeneous and interdependent elite
which preserved disparate cultural, historical and linguistic traditions.
We can see in the Mongol Empire a re-enactment of some of the processes found in the early Islamic period. Once again, subjects and rulers
influenced each other and tried to resist mutual assimilation. Each group
saw itself as superior and did not hesitate to criticize the other. Persians,
guardians of high culture and now also of religion, considered the TurcoMongolian elites uncultured, overbearing, and threatening even to the
religion which they claimed to espouse, since they remained faithful to
the un-Islamic customs of the Mongol Empire.9 The nomad elites in their
turn thought Persians inferior in military skill and bravery, effete,
84
Beatrice F. Manz
duplicitous and untrustworthy. These opinions did not prevent the

different groups from working together and learning each others languages and customs, but it produced a kind of ritualized conflict
expressed in literature and in administrative in-fighting. Mongol commanders in Iran used Iranian troops but resented Iranian pretensions to
military prowess; the Mongol Ilkhans, Tamerlane and his descendants,
and the Uzbek khanates all reserved the highest military ranks for their
Turco-Mongolian followers. The histories written for the court usually
played down the role that Iranian soldiers and commanders played in
campaigns.10 Formal ethnic separation was further reinforced by the
revivalist streak found in both Islamic and steppe societies.
All students of medieval and modern Islamic history are familiar with
periodic Islamic revival, and with criticism of ruler and society for failing
to adhere to appropriate standards of Islamic behaviour. On the nomad
side there was a similar tradition, attested in the seventh- and eighthcentury Turkic inscriptions of the Orkhon river, in which the decline of
Turk power was blamed on assimilation to settled societies. The Turk
Mahmud al-Kashghari, writing in the Middle East during the Seljukid
period, repeated the need to maintain purity within Turkic traditions.
Such calls to revival, Islamic and nomad, are constantly repeated in the
Mongol period.11
As I have shown, the characterization of separate ethnic groups within
the Islamic caliphate had mirrored the concerns of early Islamic society:
Arabs, Persians and Turks were characterized according to origin, lifestyles, and equally importantly, in relation to the functions they filled in
a multi-ethnic empire. Under the Mongol empire ethnic distinctions
remained useful and were reworked to form a new condominium of
nomad Mongols and Turks with settled Iranians. The need to preserve
difference arose not simply from personal, or communal, dislike, but also
because accepted ideas of legitimation, on both sides, made sharp distinctions desirable. The Mongols believed that supreme rule belonged
to the house of Chinggis Khan, and that the military should remain in
the hands of the descendants of his armies, distributed among his sons.
The ruling class should remain at least partly nomadic and, above all,
faithful to the traditions of the Mongol Empire. The Iranians, on their
side, were used to having foreign rulers and military, and sought efficient
protection for the practice of agriculture, trade, and religion, best
achieved through strong armies. Through long experience and tradition,
they believed that the best armies were those of nomadic origin. On both
sides therefore, the difference between nomad rulers and settled subjects
was not only tolerated but desired.
The creation of new identities in the fifteenth century
The fifteenth century was a transitional period in the Mongol empire
when several new political and ethnic identities appeared, some of which

have lasted to the present. The new groups came into being as a result
of strife within Mongol society; we can see, both in their creation and in
their characterization, the concerns which animated Mongol politics.
The empire was regionally divided almost from its inception. Regions
were allocated to members of Chinggis Khans family, who were to
acknowledge the supreme power of one great khan descended from
Chinggis Khan. Over time four major Mongol successor states came into
existence, ruled by descendants of Chinggis Khan: the Ilkhanid dynasty
in Iran, the Ulus Jochi in the steppe lands of Russia and Kazakhstan, the
Yan dynasty in China and Mongolia, and the Chaghatayid khanate in
Central Asia.
The fifteenth century witnessed a further subdivision of the Mongol
world into new states and confederations, many of which had lasting
ethnographic impact. It is at this time that we find the appearance of new
names attached not simply to dynasties, but to the people making up
different polities: the names Chaghatay and either Moghul or Chete for
the populations of separate sections of the Chaghatayid Khanate, controlled by the descendants of Chinggis Khans second son Chaghatay;
the names Uzbek and Kazakh for adherents to different confederations
in the eastern section of the Russian steppe, inherited by the family of
Chinggis Khans eldest son, Jochi. The Chaghatay and Moghul identities
proved ephemeral, but Uzbek and Kazakh identification have lasted to
the present, along with the older Tajik and Turkmen identities. The
major ideological motivation for new splits in the ruling class appears to
have been the issue of adaptation to settled society. In order to exploit
their sedentary subjects efficiently, the Mongol ruling class had come
into increasingly close symbiosis with urban and agricultural populations. This brought economic and political gains, but opened the door
to accusations of unfaithfulness to Mongol tradition.
Chaghatay identity: We can find an illustration of the controversy over
assimilation in the creation of the new identity known as Chaghatay.
The descendants of Chaghatay held Central Asia, including both western
and eastern Turkestan up to the Altai. Their state, known as the
Chaghatayid Khanate, broke up in the fourteenth century. As medieval
histories relate the story, the split resulted from the actions of its ruler
Tarmashirin Khan (132634) who converted to Islam, adapted to settled
ways, and abandoned many of the old Mongolian customs. Some of his
subjects, angry at the betrayal of Mongol tradition, rebelled and deposed
him; in the resulting disorders, the khanate divided in two. The eastern
section remained under Chinggisid khans, pagan for another generation,
and continued a traditional nomadic lifestyle. The western section, now
known as the Ulus Chaghatay, was centred in the agricultural sections
of the former khanate and while the Mongol aristocracy and their followers remained largely nomadic, they lived in close contact with settled
people. They also appear to have been Muslim from this time on, and
many were bilingual in Persian and Turkish.
86
Beatrice F. Manz
In the Ulus Chaghatay power was held by tribal leaders, ruling

through a Chinggisid puppet khan. Here, in 1370, a new leader arose:
Temr or Tamerlane, who conquered Iran, Khorezm and the Jaxartes
region, and though himself not a descendant of Chinggis Khan, ruled a
state based on Mongol political traditions and intimately involved in the
politics of both the Mongol and the Central Islamic worlds (Manz 1989,
pp. 227, 4165). Throughout Temrs reign and to a large extent
through that of his successors, the Turco-Mongolians descended from
the nomads of the Ulus Chaghatay retained a privileged status and
formed the backbone of the standing army. These people came to be
known as Chaghatay, a name applied to the Turco-Mongolian population first of the Ulus Chaghatay (13341370), and then of the Timurid
state through the fifteenth century. While the term Chaghatay derived
from the name of Chinggis Khans son Chaghatay, it did not serve in this
form to designate only his descendants, nor was it applied particularly
to the Timurid dynasty, rather it applied to those of their subjects and
followers who originated within the Mongol tradition. As the name was
used occasionally in sources written specifically for the dynasty, it was
clearly a positive term.12
To understand the content of Chaghatay identity, we must examine
the groups from which the Chaghatay differentiated themselves. The
most basic distinction was between nomad and settled: between the
Chaghatay and the Iranian population, referred to consistently and
sometimes pejoratively as Tajik. The most common name applied to
the Chaghatay was the generic term Turk, which served to distinguish
them as people of Turkic speech and nomad extraction. In some cases,
we find the term Turk and Tajik used to designate the population of
the realm as a whole; in others, Tajik bears a more specific cultural
connotation, as ethnic Iranian (Hasani 1994, pp. 80, 92; Subtelny 1994,
pp. 489). In relation to Tajiks the Chaghatay might identify themselves
simply as Turks, but within the nomad world they made further distinctions. The most important dividing line was that between those who
belonged within the Mongol tradition and those who remained outside
it. The Chaghatay were clearly separated from the Turkmen, a term
applied to the western (Oghuz) Turks, the descendants of the tribesmen
who had come into the Middle East with the Seljukid dynasty, before
the Mongol invasion, and had remained distinct from the eastern Turks.
There were large Turkmen confederations in the Middle East at this
time, some allied and some opposed to the Timurids, but all clearly differentiated from the Chaghatay and other Turco-Mongolian nomads.13
The differentiation of Turco-Mongolians from the settled and from
non-Mongol nomads was an old one; what was new to the fourteenth
and fifteenth centuries was the development of separate identities among
the Turco-Mongolians themselves. Here the Chaghatay were distinguished in particular from two neighbouring groups: the Uzbeks and the

Moghuls. Moghul was the name most often applied to the people of the
eastern Chaghatayid khanate, the section which had separated at the
time of Tarmashirin Khan and had retained a more conservative nomad
lifestyle and ideology. The people of the eastern Chaghatayid khanate
were distinguished clearly from the Chaghatay in Timurid sources, called
either by the neutral term Moghul, or by the more pejorative term
Chete meaning robber, a name which referred clearly to their lack of
discipline and order, stemming perhaps from a looser system of rule and
frequent succession struggles (Manz 1992, p. 38). We find Moghuls
among Temrs commanders, identified by their ethnic name.
The Chaghatay were further differentiated from the Uzbeks, also of
Turco-Mongolian provenance.14 The use of the term Uzbek was much
broader than that of Chaghatay, and is harder to characterize precisely.
When it was used by settled people, or by more assimilated, bilingual
nomads, the name could have a general pejorative meaning, suggesting
uncouth nomads, inferior in respect to sedentary culture (Subtelny 1983,
p. 133). In this sense, it could be applied to any nomad or semi-nomadic
people, including the Timurid population. We find for instance that a
local ruler in Azarbaijan unfriendly to the Timurids referred to their
army as Uzbek although not, obviously, when writing to the powerful
Timurid ruler himself.15 However, the most frequent use of the name
Uzbek in contemporary sources was to denote the tribesmen of the
eastern Ulus Jochi, the section of the Mongol Empire controlled by the
descendants of Chinggis Khans oldest son Jochi.16 If, then, we look at
the boundaries of Chaghatay identity, we can see them defined first of
all against the world of settled Iranians and of Turks outside the Mongol
enterprise. Among the Turco-Mongolian population the Chaghatay
were further differentiated by their origin within a particular section of
the Mongol Empire and by their willingness to adapt to settled populations and civilization.
The Uzbeks and Kazakhs: Two other major new groups arose during
the fifteenth and early sixteenth centuries which have remained active
up to the present: the Uzbeks and the Kazakhs. The story of their separation is reminiscent of the history of the crisis within the Chaghatayid
Khanate which gave rise to the split between the Moghul and Chaghatay.
In the early fifteenth-century a new Chinggisid leader appeared in the
Ulus Jochi. This was Abul Khayr Khan, a descendant of Jochi, who
gathered a large number of tribes in the eastern regions around himself,
proclaimed himself Khan in 1428 and for a period controlled the region
from the Urals to the northern Syr Darya. It is here that we see the beginning of the political entity which later formed the Uzbek khanates. In
the 1450s, Abul Khayr suffered a serious defeat at the hands of the
Oirats, a Mongol confederation from the east. As a result, a number of
his tribal followers deserted him, led by two Chinggisids of a different
Jochid lineage, named Girey and Janibeg. The desertion of tribes from
88
Beatrice F. Manz
Abul Khayr seems to have originated partly in protest against the level
of authority he exerted, and from a desire to retain tribal autonomy
under the leadership of a different Chinggisid line.17 The deserters came
to be known as renegades in Turkic Qazaqs (or Qazaq-Uzbeks). Abul
Khayr attacked them but was defeated and died in 1468. Girey and
Janibeg then returned with their followers to the north and took over
much of the region of Abul Khayr Khan, bringing with them the new
name they had acquired. The original confederation, however, did not
disappear. Abul Khayr Khans nominal position was later inherited by
his grandson, Muhammad Shaybani, who spent his early career with a
small number of followers under the protection of the Mughals and
Timurids on the borders of Transoxiana, and for two years within the
Timurid realm, in the city of Bukhara.
At the beginning of the sixteenth century Shaybani Khan gathered a
nomad force, including about 5060,000 tribesmen of the original confederation, many of whom had in the meantime lived under Kazakh rule,
but now rejoined the Abul Khayrid family, whose fortunes appeared to
be rising. With these troops Muhammad Shaybani took Transoxiana
from the descendants of Tamerlane, proclaiming a new and more legitimate Chinggisid state (Sultanov 1982, p. 20). Here begin the Uzbek
Khanates which lasted up and into the Russian conquest of Central
Asia.18 While Muhammad Shaybani and his followers took over Transoxiana, the family of Girey and Janibeg with their nomad followers
remained in the northern steppe. This confederation, also under Chinggisid khans, was designated not by a separate dynastic name, but as
Kazakhs. The order they upheld was both Chinggisid and deliberately,
conservatively, nomadic. The term Uzbek from this time applied not to
the nomads of the steppe, but to those in and around Transoxiana, under
Shaybanid leadership. The border between the Kazakh and Uzbek
khanates usually lay a little to the north of the Jaxartes, or Syr Darya,
River (Kliashtorny and Sultanov 1992, pp. 2536).
When we look for a pattern in the splits creating new polities and identities within the Mongol lands we see similar patterns within the
Chaghatayid and Jochid sections of the empire. The Chaghatayid
Khanate split apparently over the question of adherence to conservative
Mongol custom; the eastern sections adhered to older ways, to the
Chaghatayid khans, and, for one generation, to Mongol shamanism,
while the western section chose Islam and coexistence with the agricultural and urban populations. The split between Uzbeks and Kazakhs
began probably both as a conflict of loyalties to different dynastic lines
and as a protest against the imposition of strong central control, but as
the Uzbeks became increasingly involved with the politics of Transoxiana, the split turned also into a distinction between more and less conservative nomadic lifestyles.
The political importance of lifestyle is indicated by the care with which

distinctions were articulated. We have seen that the more assimilated
Turco-Mongolians within the Middle East carefully separated their
sphere from that of the settled population, while at the same time maintaining a difference between themselves and the Uzbek nomads of the
steppe, considered less adept in high civilization. The Chaghatay used the
pejorative term Chete for their eastern Chaghatayid neighbours, close in
origin, but different in lifestyle. On their side, the more conservative
nomads emphasized their loyalty to older traditions, seeing it as a virtue.
The Moghuls used the term qaraunas, or half-breed, for the Chaghatay,
suggesting that they intermarried with the settled population (Manz 1992,
p. 38). In the case of the Kazakhs there seems to have been a deliberate
effort to emphasize their nomadism in relation to other TurcoMongolian peoples. We find in the Tarikh-i Rashidi, a history written in
Persian in the early sixteenth century by a Moghul prince of the eastern
Chaghatayid state, a quotation from the Kazakh Khan Kasim:
We are men of the steppe. Little good is to be found here other than
horses. Our food is horse flesh and in our region there are no gardens
or buildings. Our recreation is to inspect our herds.19
This was addressed not to the ruler of a sedentary state, but to another
Turco-Mongolian ruler, Said Khan of the Moghuls.
The accepted markers of modern European national identity
language, common culture, origin and place played a subordinate role
in the late Mongol world. It is notable that the tribal origins of several
different groups were similar, and the histories written for and about
them make no attempt to obscure the fact. Chaghatays and Uzbeks had
a number of tribal names in common, as did Chaghatays and Moghuls.
Among the Kazakhs and Uzbeks, the confluence of tribal origins is
striking; these two people derived from a common place and background; when they split, it was not only along tribal lines, but also
through divisions within tribes (Sultanov 1982, pp. 7, 203). When
members of the same original tribes belonged to separate political
organizations they developed different identities. Territory and language
also played a secondary role in divisions among Turco-Mongolian
nomads. Groups such as the Chaghatay and Kazakh expanded their
holdings and moved out of their original lands without losing their
separate name and identity. The written Turkic language was common
among many states, and dialects were not taken into consideration.
Political allegiance, whether to a branch of the Chinggisid house or to
a dynasty of tribal origin, held greater importance in the formation of
identity. It is in the coalitions centring around a ruler that we find the
genesis of most independent groups. These confederations, consisting of
nomadic tribes able to move and to switch allegiance at will, could exist
only with the continued consent of all members, at least of the tribal
90
Beatrice F. Manz
aristocracy. For this they required both loyalty to a successful dynasty

and a common political culture an area of agreement on style of rule
and relationship to settled culture (Manz 1992, pp. 2936). It is above all
in the central questions of Mongol political culture that we find the
markers of separate Mongol identities. The issues of central versus dispersed power and of separation from the settled population, as opposed
to close exploitation, lay at the base of political splits along the steppe
frontier. When we look at identities formed by such divisions we find
them defined by differences along this spectrum.
Despite new group identities, states within the Islamic Mongol world
long continued to use systems of legitimation worked out in the
Caliphate and the Mongol Empire, which were based on broad imperial
ideals. If we compare the titles accorded to rulers in the Mongol and
Islamic worlds to those of the Habsburgs, whom we discussed earlier, we
shall get an idea of the differences we are discussing. The Habsburgs, in
collecting titles, concentrated on regions with separate historical traditions, preserved within the empire. In the late Mongol world we find
rulers also using a combination of titles from different traditions, but
each of them suggests universal rule: the Arab titles Protector of the
Caliphate, Sultan of Sultans, the Iranian title Padshah, meaning
sovereign, and the Mongol title Khaghan, meaning supreme Khan.
Even in regional states, legitimation remained imperial.
The political system developed within the Islamic Mongol regions had
an exceptionally long life, particularly in Central Asia. Within the Uzbek
khanates and their successor states, the formal division between a military
and court of nomad origin, and a civilian, city administration from the
settled population lasted up to the Russian takeover in the nineteenth
century. Although by this time many Uzbeks were no longer nomads, and
a large part of the settled population had become either bilingual or
Turkic-speaking, the old distinction between nomad and settled traditions
remained important to the structure of government and society. Even at
the beginning of the nineteenth century, connection to the Turkic and
Mongol dynasties of the past remained central to legitimation. Despite
the increasing number of Turkic speakers, Persian remained the primary
language of high culture. Central Asian society remained consciously
heterogeneous, and the markers distinguishing different groups remained
similar to those developed in the Mongol period. Just as the Habsburg
imperial titles of the early twentieth century preserved the regional basis
of the imperial structure, the Central Asian khanates of the nineteenth
century still mirrored the Chinggisid and Islamic traditions.
Imperial legacies within the Russian and Soviet states
The lands of the Russian Empire stretched from Europe to Asia, and
were heir to both Asian and European traditions. The Muscovite state

which formed the base of the empire rose to prominence within the
Golden Horde, the westernmost section of the Mongol Empire, and it
owed many elements of its political structure and ideology to its Mongol
heritage. The Muscovite rulers gradually gathered the Russian lands,
partly by manoeuvre and negotiation, like the Habsburgs, and partly by
force of arms. When Constantinople fell to the Ottomans in 1453,
Muscovy was able to lay claim to the Roman and Christian imperial
inheritance. The adoption of Byzantine and Orthodox legitimation did
not lead Russian rulers to abandon Mongol legitimation; early rulers
continued to use insignia and symbols of the Mongol empire. As late as
the time of troubles in the early seventeenth century, the enthronement
of a Chinggisid tsar appeared as a possibility (Cherniavsky 1975, p. 134).
When the Muscovite rulers began to expand their territory beyond the
original Russian lands, the first regions they took were the successor
states of the Golden Horde, with which they had remained intimately
involved. As the Tsars pushed their territories east, they took over the
northern remnants of the Mongol Empire, replacing the highest levels
of government and incorporating the military classes into the Muscovite
army. Many elites of the conquered regions retained their land, noble
status and local positions, though now as part of the Muscovite state
(Kappeler 1992, pp. 304, 545). In the first periods of Muscovite expansion, Chinggisid prestige continued as an element both in state legitimation and in the ranking of newly incorporated elites.
Russias turn towards Europe in the late seventeenth century brought
in new ideas of empire and legitimacy, but also reinforced some earlier
distinctions between peoples. The non-Russian regions of the empire
came to be seen in a new light, as regions either to be assimilated or to
be identified as foreign. Peter I (r. 16891725) instituted Christianizing
policies; the Mongol, or Tatar, nobles who became Christian retained
their status, while those who refused forfeited their titles. Catharine II
(r. 176296), while proclaiming religious tolerance, sharply differentiated the civilized agricultural populations from nomadic and huntergatherer societies, which were classed as fundamentally different
peoples, called inorodsty. In this way, the Mongol and Islamic emphasis
on nomadism and lifestyle continued to exist, but was reversed. The
military abilities and resistance to assimilation shown by nomads
changed from a mark of superiority to one of inferiority. In Catharines
administration nomadic societies were considered particularly backward
and dangerous.
European ideas of empire also affected the administration of newly
acquired territories. Western regions inhabited largely by Slavs were
influenced by the nationalistic ideologies of central and western Europe.
We find here the demands for cultural development seen in the Habsburg
empire and, in response, pressures from the government towards assimilation and Russification. Outside of Europe, the experience of empire was
92
Beatrice F. Manz
a very different one. The eastern, and especially the non-Christian,

regions that the Russians conquered in the eighteenth and nineteenth
century were less closely incorporated into the imperial administration.
They were considered inferior; the Russians came in as a civilizing race,
ruling from outside. In the modern empire the Muslim regions, in particular Central Asia and Kazakhstan, held low status. There was no question
of their peoples assimilating to the Russian way of life or of their upper
classes entering the elite of the empire. (Kappeler 1992, pp. 35, 13940,
159, 166) The eastern regions therefore neither felt the full pressure of
Russification, nor became exposed to eastern European national movements.
In the twentieth century the lands of the Russian Empire became
those of the Soviet Union, and within this new framework the legacy of
past empires continued. In the development of policy the most direct
influence was the one from outside: from the Austro-Hungarian empire.
When Lenin and Stalin developed a nationalities policy for the Bolshevik party, it was natural for them to look to their fellow socialists in
eastern Europe, who were deeply concerned with issues of nationality.
The ideas they took from them clearly mirrored the experience of
Austria-Hungary. Peoples were to be defined as nations on the assumption that language, local custom, national history and ethnic loyalties
would shape the political aspirations of autonomous national units.
Above all, it was assumed that separate ethnic or national groups were
likely to seek at least cultural autonomy, if not independence. The Bolsheviks disagreed with Austro-Hungarian Marxist thinkers like Otto
Bauer and Karl Renner on several important issues; for instance, while
Austro-Hungarian socialists attempted to solve the problem of dispersed
populations through the programme of non-territorial cultural
autonomy, the Bolsheviks insisted on the convergence of territory with
other national characteristics (Stalin 1936, pp. 513; Pipes 1964, pp.
2349). The differences in Bolshevik policy should not, however, blind
us to the similarity of their concepts of nation and nationality.
In the European regions of the USSR, Soviet policy was applied to
peoples who had been involved in European nationality politics and had
formed their identities within the same mould as the peoples of central
and southern Europe. Further east, the Soviets met populations with
totally different ideas of identity, regionalism and legitimation. Throughout the Soviet period, nonetheless, nationality policy adhered to the
criteria and concerns with which it began. National republics were
created with the help of ethnographers, on the assumption that language
and ethnic characteristics were the primary factors to be evaluated.
While the justification for boundaries concentrated on ethnicity, it seems
likely that the Soviets, perhaps unconsciously, also resembled the Habsburgs in their acceptance of legitimacy based on historical statehood.
This trait is illustrated in the formation of the Soviet Central Asian

republics. In choosing an ethnic name for the central republic of Transoxiana, the planners rejected that of Sart, applied mostly to Turkic
settled peoples, because they regarded it as pejorative. Instead they
chose Uzbek, which denoted the population descended from the Turkic
tribes which had entered with Shaybani khan, and was attached to the
former ruling class.
The Uzbek republic then received the capital regions of all three of
the Uzbek khanates, inhabited not surprisingly by many people calling
themselves Uzbek. The Tajiks and Turkmens who had formed part of
the populations of these khanates got only the territories earlier dominated by leaders of their own ethnic groups, regions now separated from
their former agricultural and administrative centres. Once the republics
had been formed, they were encouraged to create identities which would
fit at once their languages, their histories and, if possible, their borders.
Past rulers and luminaries of the several nations had to be assigned to
individual republics. Language policy followed the Central European
model; where no written language existed it was quickly developed, and
much of nationality politics revolved, as in the Austro-Hungarian
empire, around questions of education and administration in republican
languages (Simon 1991, pp. 3163; Slezkine 1996, pp. 20230).
In Soviet policy, the Austro-Hungarian influence dominated, but in the
realm of culture and practice, we see other legacies equally active. Like
the Tsars and the Mongols before them, the Soviet regime considered
nomadism an important factor in the classification of peoples. For a brief
period in the 1920s, nomads were idealized as a classless society, but in
the early thirties nomad societies came to be characterized as feudal
thus premodern and regressive (Edgar 1999, pp. 1327). The drive
towards collectivization in the late twenties and early thirties was used
to attack nomadism in the regions where it still survived. In Central Asia,
the Soviets preserved the cultural distance with the belief in their superiority and civilizing mission that the Russians of the nineteenth century
had held. Thus, the distinction between different lifestyles and between
eastern and western non-Russian regions, begun under the Tsars,
remained intact through the Soviet period.
The Soviet regime exerted more direct pressure in Central Asia than
had the nineteenth-century Russian regime, but despite its intransigence,
it left intact a number of structures and ideas from the pre-Russian
period. Indeed, in certain ways, the Soviets fit into the pattern set by the
Islamic Caliphate and the Mongol Empire. Once again a foreign ruling
class held power, basing its legitimacy on its outside origin, its separate
culture and language. Where earlier the administrative and cultural elites
had been bilingual in Turkic and Persian, now they were bilingual in
Russian and either Turkic or Tajik. The Soviet government promoted
culture in both Russian and republican languages. The settled population had often expressed its irritation with the Turco-Mongolian ruling
94
Beatrice F. Manz
class or its foreign armies, now Turks and Tajiks resented Russians. Like
nomad rulers of an earlier period, Russians felt themselves superior and
made no attempt to hide the fact.
Like all creators of national identity, Soviet functionaries used the past
to create the present. While trying to promote new republican loyalties,
the Soviet regime paradoxically kept alive many old ideas of identity.
The emphasis that the government placed on ideology and its belief in
the political importance of historical writing made it impossible for the
Central Asians to forget their past. Although the writing of history was
monitored by the state, the prestige of the field attracted serious scholars,
especially for the medieval period, and the basic textual and factual
research was solid. The results of this research, presented in the appropriate ideological form, formed a central part of public education. The
need to create separate republican characters and histories served as a
reminder of the commonality of historical experience, as scholars of
different republics tried to parcel out the luminaries of the past. Struggling to define what had belonged to each republic, Central Asians could
not forget that Tajiks and Uzbeks had lived close to each other and had
shared both rulers and scholars, while Uzbeks and Kazakhs reading of
their ethnogenesis were reminded of historical and linguistic bonds.
The differentiation of identity through the distinction between nomad
and non-nomad peoples also remained alive. In writing their history, the
peoples of Central Asia had to deal with the question of their nomad
and Chinggisid origins, now a brand of shame rather than a source of
glory. The Uzbeks, inhabiting an intensively agricultural region and
descended from nomads who had lived close to the settled, chose to play
down their nomadism and stressed the local, Central Asian aspect of
their heritage. For the Kazakhs, this was a less workable solution, and
we find some attempts to defend the nomad tradition (Pishchulina 1977,
pp. 78; Subtelny 1994, pp. 523). Thus we find in the Soviet period an
echo of the contests of earlier times.
By the time the USSR collapsed the Central Asian republics had
achieved modern identities showing the elements of European nationalism emphasis on region, common historical experience, culture and
language. The meaning of these elements and the way they are interconnected, however, need not be the same as what we find in Europe.
Within the new identities of Central Asia, older concepts have remained
alive and active. It is in the connections made between different aspects
of cultural and political identity that the heritage of earlier empires
remains important: the relationship between language use and political
loyalty, between ethnic differentiation and movements towards separatism. When we examine events since the breakup of the USSR, we can
see the influence of earlier understandings of identity. While the expression of republican patriotism might be similar to that of Western
nationalism, the expectations attached to it can be quite different.

In the 1980s many observers regarded Soviet Central Asia as the soft
underbelly of the Soviet beast. The fact that Central Asia was strikingly
different from Russia brought an assumption of greater separatist
feeling. It was clear that the Russian Soviet presence was resented, and
with reason, since forced cotton specialization had led to widespread
ecological damage. Anti-Russian feeling was well documented. Both
Soviet and Western scholars noted the persistence of Islamic observance
and the reluctance of many Central Asians either to intermarry with
Russians or to work outside their region. Nonetheless, when the Soviet
Union broke up, the Muslim Central Asian republics were among the
last regions to declare independence. If our expectations are geared to
European models, this behaviour seems odd, but I suggest that in the
context of the Islamic and Mongol legacies it is logical. The Russians
who ruled Central Asia had directly replaced earlier rulers and filled a
similar function. Earlier foreign ruling classes had been resented, often
vocally, but their presence was, nevertheless, accepted because they
provided security and military power. Within the Soviet Union, Central
Asians could count on the protection of the huge Soviet army, on continuing cultural patronage, and on the prestige of belonging to a major
power. Russians could be criticized as foreign but the expression of
feeling against them did not lead to a movement towards separatism.
One subject of interest to Western scholars was the republic boundaries drawn in the 1920s, portrayed by some writers as arbitrary,
designed specifically to divide the peoples of Central Asia. The denial
of national identities in pre-modern Central Asia led some writers to
state that identity was based primarily on religion and below that on
tribe, village or confessional community, and that intermediate identities, comparable in size to the modern nation, did not exist (Khalid 1998,
pp. 184190). The distinctions made between different Turkic peoples
and the creation of separate written languages for Uzbeks, Turkmens,
Kazakhs and Kirghiz, has been seen as an imposition of artificial boundaries (Bennigsen 1971, pp. 1745). With the development of free speech,
the revival of Islam, and the resumption of pan-Turkic ideals after the
fall of the Soviet Union, scholars and journalists speculated about the
creation of a Muslim Turkic coalition or even state, and the possibility
of serious disagreement over borders. So far, despite rhetoric, little has
happened on this front, and here again, one may see the survival of older
ideas about the relationship of territory and identity. As we have seen,
language was not an important factor in the formation of earlier Turkic
ethnic groups. The Turks, sharing closely related languages, historic
origin and consciousness, have usually lived in separate and often
warring states. While the distinctions between modern republican languages are in part a twentieth-century construction, the borders between
the Turkic states of Uzbekistan, Kazakhstan and Turkmenistan are not
very different from earlier borders between the Kirghiz and Kazakh
96
Beatrice F. Manz
nomads of the steppe and the settled populations in the Uzbek khanates
of Bukhara and Khiva (Manz 1994, p. 16). Except for the inclusion of
the Khivan oasis in Uzbekistan, they do not represent a major break with
earlier tradition. We must remember further that in the Mongol and
Islamic worlds regional identification was not necessarily connected to
political loyalty or state legitimation. Thus while some activists suggested
a unitary Turkic state, both the past and the present were against them.
The one current Central Asian republic whose territory is a lasting
source of dispute is the one most solidly founded on western national
markers of historical precedent, ethnic identity and linguistic difference,
namely Tajikistan. Having chosen for the Tajiks the one region of
Central Asia which had not only been inhabited, but even largely ruled
by Iranians, the Soviet state assigned to them the mountainous borderlands which Turkic nomads never settled or ruled directly. The mountain
populations, however, had been known as Ghalcha, and were considered different from the agricultural and urban Tajiks. Tajik identity
had long been defined not by territory or monarchy, but by function
within a large, rich and powerful society. The Tajiks, under Islamic,
Mongol and Uzbek rulers, were the bearers of high culture throughout
Central Asia. The other strongholds of Tajik language and population
and those most important in their history were at the opposite end of
the spectrum of civilization, namely the two great cities of the region:
Samarqand and Bukhara, from which many Tajik intellectuals originate.
It was in these cities that the cultural heroes of Persian civilization,
assigned to the Tajiks, wrote their masterpieces. The exclusion of their
centres of high culture continues to be a sore point for the Tajiks, whose
actual territory has little connection either with their past identity or with
the cultural history formulated in the Soviet period.20
Conclusion
The group identities of the modern world have a long and varied history,
and this history continues to matter. We can see in medieval and early
modern Europe the origins of ideas and structures that combined to
form the ideal of the nation-state. Regions were formally linked to a
complex of laws, customs and legitimation. Language was an important
marker of identity, a tool for the imposition of central rule, and if it was
also a literary language, a badge of superiority. It was on these structures
that the Austro-Hungarian Empire was built, and by preserving them,
the empire determined the lines along which it would be divided into
nations.
In the two great empires which dominated Eurasia the Islamic and
the Mongol identity was differently structured and understood. Here the
nomadic and sedentary lifestyles formed the most basic and unchanging
markers of identity, strong enough to survive even the sedentarization of

most nomads. Language and religion, while important, were expandable; one state or region could be expected to contain populations of
several religions and languages, and such diversity was desirable.
Finally, regional identity, while strong, had no connection to imperial
legitimation. By the late medieval period, in both the Mongol and the
Islamic worlds an elite of nomad provenance ruled over a multi-ethnic
settled population which provided the scholars, bureaucrats, and the
language of high culture for the rulers. Like the Austro-Hungarian state,
these were multi-ethnic empires, but in contrast to the European
peoples, their major ethnic groups were scattered throughout the
empire, and thought to belong not in an original homeland, but where
they currently lived.
The Russian Empire provided a bridge between these two systems,
both physically and temporally. The core of the empire lay firmly in the
Russian lands but it began its development within the Mongol system
and showed Mongol influence in its early expectation of difference and
respect for nomad origins. Russian expansion into European territories
and the rise of western European prestige brought European ideas into
Russia, but as I have shown, earlier systems of thought remained alive
in the eastern regions. The Soviet state, despite its attempt to transform
society, kept alive much of what it inherited from earlier empires. I have
illustrated in this article how such influences have shaped modern ideas
in Central Asia.
The identities of Central Asia Uzbek, Tajik, Turkmen, Kazakh
are real, powerful and based in history, but they have not all fostered
the expectations and actions Western nationalism would demand of
them. As national, republican identities, all have changed and evolved
considerably during this century, but they are not new creations, formed
entirely by Western forces. They are rather identities with a long history
and with traits developed in a variety of situations. If we are to understand the post-Soviet societies of Central Asia, or the post-colonial
societies of other regions, we should look back beyond the history of
the last century, dominated by European ideas and by European
nationalism, and be prepared to see the legacy of other imperial structures and of the ideas which they imposed on the peoples within them.
I have tried to show in the example of post-Soviet Central Asia the
variety of ideologies and loyalties still active, and to caution against the
imposition of European nationalist expectations on identities formed in
other moulds.
Acknowledgements
I wish to thank the many colleagues and students who have read and
commented on one or other draft of this article, particularly Robert
Krikorian and Ronald Suny.
98
Beatrice F. Manz
Notes
1.
For an example of this argument see Mustafa Vaziri, Iran as Imagined Nation, the
Construction of National Identity, New York: Paragon House 1993, pp. 111, 6196.
2.
Prasenjid Duara has focused on this problem, particularly for China and India, and
proposed broadening the definition of nation and nationalism to include a variety of formulations of identity. (Prasenjid Duara, Rescuing History from the Nation, Chicago:
University of Chicago Press, 1995, pp. 5169).
3.
See, for example, the articles in Richard L. Rudolph and David F. Good, Nationalism and Empire: the Habsburg Monarchy and the Soviet Union, New York: St. Martins
Press, 1992.
4.
Kann 1977, pp. 123. For the importance of the landed nobility and their institutions in the formation of pre-modern European kingships and the identities attached to
them, see J. Strayer, The laicization of French and English society in the 13th century,
pp. 25363, and The statute of York and the community of the Realm pp. 26983, in
Joseph R. Strayer (ed.), Medieval Statecraft and the Perspective of History, Princeton, NJ:
Princeton University Press, 1971. Also, for early kingship and nobility in Europe, Jarich
G. Oosten, Ideology and the development of European kingdoms, in Henri J.M. Claessen
and Jarich G. Oosten (eds), Ideology and the Formation of Early States, Leiden: E.J. Brill,
1996, pp. 220241.
5.
Since the term Iranian refers to a number of related languages, I use the term
Persian in this article for the central literary language.
6.
Most of what we know about the Shuubiyya comes from its rebutters, and much
about it, including the identity of its promoters, remains unclear. (Susannah Enderwitz,
Shuubiyya, EI, 2d. edn, Goldziher, The Shuubiyya, and The Shuubiyya and its manifestations in scholarship, Muslim Studies, vol. I, pp. 13898, H.A.R. Gibb, The social
significance of the Shuubiyya, Studies in the Civilization of Islam, Princeton, NJ: Princeton
University Press, 1982, pp. 6272, Roy P. Mottahedeh, The Shubyah Controversy and
the social history of early islamic Iran, International Journal of Middle East Studies, vol.
7, 1976, pp. 16182, Lecomte 1965 pp. 34359).
7.
Pellat 1969, pp. 927, Ulrich W. Haarmann, Ideology and history, identity and
alterity: the Arab image of the Turk from the Abbasids to modern Egypt, International
Journal of Middle East Studies, vol. 20, no. 2, 1988, pp. 17581. The description given by
al-Jahiz was unusually positive, but his assessment of their martial and nomad character
was typical.
8.
The use of Iranian legitimation seems to have occurred most often in peripheral
regions such as Sistan and the Caspian provinces, and it may be that the imperial traditions
survived most fully in areas where the rule of empires, both past and contemporary, had
rested lightly. See G.C. Miles, Numismatics, in Cambridge History of Iran, vol. IV,
Cambridge: Cambridge University Press, 1975, p. 367, S.M. Stern, Yaqub the Coppersmith and Persian National Sentiment, in C.E. Bosworth, (ed.), Iran and Islam, in Memory
of the Late Vladimir Minorsky, Edinburgh: Edinburgh University Press, 1971, pp. 53845,
Wilferd Madelung, The assumption of the title Shahanshah by the Buyids and The reign
of the Daylam (Dawlat al-Daylam), Journal of Near Eastern Studies vol. XXVIII (1969)
pp. 859, 1057.
9.
See, for example, Ata al-Malik Juwayni, trans. J.A. Boyle, The History of the
World-conqueror, Manchester: Manchester University Press, 1958, vol. I, pp. 78, 2022.
10.
Beatrice F. Manz, Military manpower in late Mongol and Timurid Iran, LHritage
timouride, Iran-Asie Centrale-Inde, XVe-XVIIIe sicles, Cahiers dAsie Centrale, 34
(1997), pp. 4356. For examples of the way in which Mongols, and indeed Iranians classed
Iranians as men without military talent, see references to Tajiks in Hasani 1987, pp. 80, 92,
and in Mun al-Dn Zamj Isfizar, Rawd.at al-jannat f aws.af madnat Harat, ed. Sayyid
Muh.ammad Kaz.im Imam, Tehran, 1338/1959., pp. 14, 108, 115.
11.
Beatrice F. Manz, Temr and the problem of a conquerors legacy, Journal of the

Royal Asiatic Society, 3d series, vol. 8, part 1, 1998, pp. 368, Jean Aubin, mirs mongols
et vizirs persans dans les remous de lacculturation, Studia Islamica, Cahier 15, Paris, 1995,
pp. 14, 325, 3843. For the earlier period see Vilhelm Thomsen, Alttrkische Inschriften
aus der Mongolei, Zeitschrift der deutschen morgenlndischen Gesellschaft, vol. 78, no. 2,
1924, pp. 1417.
12.
Dawlatshah Samarqand, The Tadhkiratush-Shuara (Memoirs of the Poets) (ed.)
E.G. Browne, London, 1901, pp. 384, 519; also Manz 1992, p. 37.
13.
See, for example: Abd al-Razzaq Samarqand, Mat. la al-sa dayn wa majma albah.rayn, Muh.ammad Shafi (ed.), Lahore, 136068/194149, pt. II, pp. 644, 684, H
. afiz.-i
Abru, Majma al-tawa rkh, ms. Istanbul, Fatih 4371/1, f. 8a, H
. afiz.-i Abru, Zubdat altawarkh, ed., Sayyid Kamal H
. ajj Sayyid Jawad, Tehran, sh. 1372, I, pp. 332, 357, II, 609.
14.
See, for example: Niz.am al-Dn Sham, Histoire des conqutes de Tamerlan intitule
Z.afarnama, par Niz.a muddn a m (ed.) F. Tauer, Prague, vol. I, 1937, vol. II, 1956, vol. I,
p. 140, Muizz al-ansab f shajarat al-ansab, ms. Paris, Bibliotque Nationale, #67, ff. 129b,
133a, 134a, 142b.
15.
Abd al-H.usayn Nawa, ed., Asnad wa makatiba t-i ta rkh-i Iran, Tehran,
2536/1977, p. 215.
16.
This usage begins before the rise of Abul Khayr Khan, who formed the confederation later known as Uzbek. It is used, for instance, in the Zubdat al-tawarkh of H
. afiz.i Abru, written at Shahrukhs court in the 1420s, to refer to the followers of Edigey. (H.afiz.-i
Abru, Zubdat, vol. I, pp. 387, 394, 422.
17.
Rene Grousset, The Empire of the Steppes: a History of Central Asia, trans., Naomi
Walford, New Brunswick, 1970, pp. 47980, T. I. Sultanov, Osnovnye voprosy istorii kazakhskogo naroda v xv-xvii vv. po persidsko-tadzhikskim i tiurkskim istochnikam,
aftoreferat, Leningrad, 1971. The brief account given in the Tarikh-i Rashidi suggests
pressure from Abul Khayr against potential rivals (Dughlat 1996, text, pp. 2267, trans.,
p. 176).
18.
It does not appear that the followers of either Abul Khayr Khan or Muhammad
Shaybani used the name Uzbek for themselves. The name was used liberally by their neighbours. At the time of Muhammad Shaybanis conquest at the turn of the 1516th centuries,
Timurid and Moghul observers characterized Muhammad Shaybanis steppe followers as
Uzbek, and distinguished them sharply from Moghuls, Chaghatays and Kazakhs. In this
usage, the followers who came with Muhammad Shaybani from the northern steppes are
distinguished from those he acquired on the borders of Transoxiana where he spent some
time under Moghul protection. At a later date, in the late seventeenth century, it appears
that the Turco-Mongolian population of the Uzbek khanates did use the name Uzbek for
the tribal population, though not for the dynasty. (R. McChesney, Waqf in Central Asia,
Princeton, NJ: Princeton University Press, 1991, pp. 4950.)
19.
Dughlat 1997, text p. 229, trans. p. 178.
20.
See for example: Reinhard Eisener, Auf den Spuren des tadschikischen Nationalismus, Berlin: Das Arabishche Buch, Occasional Papers, #30, for expressions of this
sentiment during perestroika, and for a fuller exposition at the time of the breakup of the
USSR, Rakhim Masov, Istoriia topornogo razdeleniia, Dushanbe: Irfon 1991.
References
BARTLETT, R. 1993 The Making of Europe: Conquest, Colonization and Cultural
Change, 9501350, Princeton University Press
BEESTON, A. F. L., JONSTONE, T. M., SERGENT, R. B., SMITH, G. R. (eds) 1983
Arabic Literature to the End of the Umayyad Period, Cambridge History of Arabic Literature, Cambridge: Cambridge University Press
BENNIGSEN, A. 1971 Islamic and local consciousness among Soviet nationalities, in E.
Allworth (ed.), Soviet Nationality Problems, New York: Columbia University Press, pp. 16882
100
Beatrice F. Manz
BEZECNY, A. 1908 Die Thronreden Seiner Majestt des Kaisers Franz Josef I. bei der
feierlichen Erffnung und Schliessung des sterreichischen Reichsrates, Vienna: Manzsche
k.u.k. Hof-Verlags und Universitts-Buchhandlung
CHERNIAVSKY, M. 1975 Russia, in O. Ranum (ed.), National Consciousness, History
and Political Culture in Early-modern Europe, Baltimore, MD: Johns Hopkins University
Press, pp. 11843
CRONE, P. 1980 Slaves on Horses, Cambridge: Cambridge University Press
DUGHLAT, Muhammad Haydar 1996 Wheeler M. Thackston (ed.) and trans., Mirza
Haydar Dughlats Tarikh-i Rashidi, A History of the Khans of Moghulistan, (Sources of
Oriental Languages and Literatures, Central Asian Sources II), Cambridge, MA
EDGAR, A. 1999 The creation of Soviet Turkmenistan, 19241938 PhD dissertation,
Berkeley, CA: University of California at Berkeley
ELEY, G. and SUNY, R.N. (eds) 1996 Becoming National, Oxford: Oxford University
Press
ELLIOTT, J. H. 1992 A Europe of composite monarchies, Past and Present, No. 137, pp.
4871
EVANS, R. J. W. 2000 18481849 in the Habsburg monarchy, in R. J. W. Evans and H.
von Strandmann (eds), The Revolutions in Europe 18481849: from Reform to Reaction,
Oxford: Oxford University Press, pp. 181206
FRAGNER, B. 1999 Die Persophonie: Regionalitt, Identitt und Sprachkontakt in der
Geschichte Asiens, Berlin: Das Arabische Buch (ANOR 5)
GOLDZIHER, I. 1967 trans., C. R. Barber and S. M. Stern, Muslim Studies, London: Allen
and Unwin
J AL-DI N H.ASAN B. SHIHA
B YAZDI 1987 Jami al-tawarkh-i H
H.ASANI, TA
. asan
(ed.), H. usayn Mudarris T. abat. aba and Iraj Afshar, Karachi: Danmishgah-i Karachi
HASTINGS, A. 1997 The Construction of Nationhood: Ethnicity, Religion and Nationhood, Cambridge: Cambridge University Press
HAVRNEK, J. 2000 Bohemian spring 1848 conflict of loyalties and its picture in
historiography, in A. Krner (ed.) 1848 A European Revolution? International Ideas and
National Memories of 1848, New York, NY: St. Martins Press, pp. 12439
HAWTING, G. R. 1986 The First Dynasty of Islam, London: Croom Helm
HINDS, M. 1971 Kufan political alignments and their background in the mid-seventh
century A.D, International Journal of Middle East Studies, vol. 2, pp. 34667
HOBSBAWM, E. 1990 Nations and Nationalism since 1780, Cambridge: Cambridge
University Press
HUTCHINS, W. M. 1988 Nine Essays of al-Jahiz, New York: P. Lang
INGRAO, C. 1994 The Habsburg Monarchy, 16181815, Cambridge: Cambridge
University Press
KANN, R. 1977 A History of the Habsburg Empire, 15161918, Berkeley, CA: University
of California Press
KAPPELER, A. 1992 Russland als Vielvlkerreich, Munich: C. H. Beck
KHALID, A. 1998 The Politics of Muslim Cultural Reform: Jadidism in Central Asia,
Berkeley, CA: University of California Press
KLIASHTORNY, S. T. and SULTANOV, T. I. 1992 Kazakhstan, letopis trex tysiacheletii,
Alma Ata: Rauan
LECOMTE, G. 1965 Ibn Qutayba (mort en 276/889). Lhomme, son uvre, ses ides,
Damascus: Institut Franais de Damas
MACARTNEY, C. A. 1962 Hungary, A Short History, Edinburgh: Edinburgh University
Press
MANZ, B. 1989 The Rise and Rule of Tamerlane, Cambridge: Cambridge University Press
1992 The development and meaning of Chaghatay identity, in Jo-Ann Gross (ed.),
Muslims in Central Asia: Expressions of Identity and Change, Durham, NC: Duke
University Press, pp. 2745

1994 Historical Background, in B. Manz (ed.), Central Asia in Historical Perspective, Boulder CO: Westview Press, pp. 424
MNKSK and MEZNK 1998 The making of the Czech state: Bohemia and Moravia
from the tenth to the fourteenth centuries, in M. Teich (ed.), Bohemia in History,
Cambridge: Cambridge University Press, pp. 3958
MORONY, M. 1984 Iraq after the Muslim Conquest, Princeton, NJ: Princeton University
Press
MOTTAHEDEH, R. 1980 Loyalty and Leadership in an Early Islamic Society, Princeton:
Princeton University Press
ahiz, Paris: Librairie d Amerique
PELLAT, C. 1953 Le milieu basrien et la formation de G
et dOrient, Adrien-Maisonneuve
1969 The Life and Works of Jhiz, Translations of Selected Texts, Berkeley, CA:
University of California Press
PIPES, R. 1964 The Formation of the Soviet Union, Cambridge, MA: Harvard University
Press
PISHCHULINA, K. A. 1977 Iugo-vostochnyKazakhstan v seredine xiv-nachale xvi vekov,
Alma-Ata: Nauka
RACHEWILTZ, I. 1973 Some remarks on the ideological foundations of Chingis Khans
empire, Papers on Far Eastern History, vol. 7 (March) pp. 2136
SIMON, G. 1991 Nationalism and Policy towards the Nationalities in the Soviet Union,
Boulder, CO: Westview Press
SLMA, J. 1998 Boiohaemum-echy, in M. Teich (ed.), Bohemia in History, Cambridge:
Cambridge University Press, pp. 2338
SLEZKINE, Y. 1996 The USSR as a communal apartment or how a socialist state
promoted ethnic particularism, in G. Eley and R. G. Suny (eds), Becoming National,
SMITH, A. 1996 The origins of nations, in G. Eley and R. Suny (eds), Becoming National,
SPERBER, J. 1994 The European Revolutions, 184851, Cambridge: Cambridge
University Press
STALIN, J. 1936 Marxism and the national question, in Marxism and the National
Colonial Question, London: Lawrence and Wishart, pp. 361
SUBTELNY, M. 1983 Art and politics in early 16th century Central Asia, Central Asiatic
Journal, volume 27, no. 12, pp. 12148
1994 The symbiosis of Turk and Tajik, in B. Manz (ed.), Central Asia in Historical
Perspective, Boulder, CO: Westview Press, pp. 4561
SULTANOV, T. I. 1982, Kochevye plemena Priaralia v xv-xvii vv., Moscow: Nauka
TAYLOR, A. J. P. 1948 The Habsburg Monarchy, 18091918, A History of the Austrian
Empire and Austria-Hungary, London: Hamish Hamilton
BEATRICE F. MANZ is Associate Professor of History at Tufts University.

ADDRESS: Department of History, Tufts University, Medford, MA
02155, USA.

MANZ Multi-Ethnic Empires

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

MANZ Multi-Ethnic Empires

Transféré par

Droits d'auteur :

Formats disponibles

Ethnic and Racial Studies

ISSN: 0141-9870 (Print) 1466-4356 (Online) Journal homepage: http://www.tandfonline.com/loi/rers20

Multi-ethnic Empires and the formulation of

Published online: 07 Dec 2010.

Submit your article to this journal

Article views: 598

View related articles

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at

Date: 02 January 2016, At: 06:27

Ethnic and Racial Studies

Vol. 26 No. 1 January 2003 pp. 70101

Multi-ethnic Empires and the

Downloaded by [Macquarie University] at 06:27 02 January 2016

2003 Taylor & Francis Ltd

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 71

Downloaded by [Macquarie University] at 06:27 02 January 2016

their function within the multi-ethnic empires which used to control

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 73

Downloaded by [Macquarie University] at 06:27 02 January 2016

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 75

Downloaded by [Macquarie University] at 06:27 02 January 2016

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 77

Downloaded by [Macquarie University] at 06:27 02 January 2016

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 79

Downloaded by [Macquarie University] at 06:27 02 January 2016

For a while Arabic held a position analogous to that of both German

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 81

Downloaded by [Macquarie University] at 06:27 02 January 2016

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 83

Downloaded by [Macquarie University] at 06:27 02 January 2016

duplicitous and untrustworthy. These opinions did not prevent the

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 85

Downloaded by [Macquarie University] at 06:27 02 January 2016

In the Ulus Chaghatay power was held by tribal leaders, ruling

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 87

Downloaded by [Macquarie University] at 06:27 02 January 2016

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 89

Downloaded by [Macquarie University] at 06:27 02 January 2016

aristocracy. For this they required both loyalty to a successful dynasty

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 91

Downloaded by [Macquarie University] at 06:27 02 January 2016

a very different one. The eastern, and especially the non-Christian,

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 93

Downloaded by [Macquarie University] at 06:27 02 January 2016

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 95

Downloaded by [Macquarie University] at 06:27 02 January 2016

Downloaded by [Macquarie University] at 06:27 02 January 2016

Multi-ethnic Empires and identity 97

Downloaded by [Macquarie University] at 06:27 02 January 2016