Thesis Eleven http://the.sagepub.

com/

Between Chieftaincy and Knighthood: A Comparative Study of Ottoman and Safavid Origins
Babak Rahimi Thesis Eleven 2004 76: 85 DOI: 10.1177/0725513604040111 The online version of this article can be found at: http://the.sagepub.com/content/76/1/85

Published by:
http://www.sagepublications.com

Additional services and information for Thesis Eleven can be found at: Email Alerts: http://the.sagepub.com/cgi/alerts Subscriptions: http://the.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://the.sagepub.com/content/76/1/85.refs.html

>> Version of Record - Feb 1, 2004 What is This?

Downloaded from the.sagepub.com at City University Library on January 6, 2012

BETWEEN CHIEFTAINCY AND KNIGHTHOOD: A COMPARATIVE STUDY OF OTTOMAN AND SAFAVID ORIGINS1
Babak Rahimi

ABSTRACT Tracing the history of the Ottoman and Safavid empires back to the Middle Period of Islamic history, this article focuses on their origins in the chieftaincies and the hybrid cultural formations of the Anatolian regions. While considering the inter/intracivilizational historical context of their respective rise to power, it is argued that the structural makeup of the empires differed primarily in their disparate forms of Su-knightly cultures, identied here as knightly-heroic (Ottoman) and millenarian-populist (Safavid), which is essentially tied to two distinctive types of tribal political organizations: frontierchieftaincy (Ottoman) and sectarian-chieftaincy (Safavid). Although the original Ottoman and Safavid chieftaincies, based on the militant Ghazi and Qizilbash orders, dissolved once the roaming bands of warriors were replaced by more settled military formations in the course of long-term state-building processes, the inuence of their Su-knightly cultural heritages is still manifest in modern Iranian and Turkish societies. KEYWORDS chieftaincy civilization empire state tribe

INTRODUCTION The Iranian and Turkish revolutions of the early 20th century, and in particular the Islamic revolution of 19789 in Iran, reect basic tensions in the cultural framework of Islamic civilizational dynamics. Conicting trajectories
Thesis Eleven, Number 76, February 2004: 85102 SAGE Publications (London, Thousand Oaks, CA and New Delhi) Copyright 2004 SAGE Publications and Thesis Eleven Co-op Ltd [0725-5136(200402)76;85102;040111]DOI:10.1177/0725513604040111

Downloaded from the.sagepub.com at City University Library on January 6, 2012

86

Thesis Eleven (Number 76 2004)

of historical processes led one nation to establish a religious state (Iran) and the other a secular one (Turkey). In the case of Turkey, the so-called Young Turk Revolution of 1908 and the Kemalist movement of 191923 led to the proclamation of a Turkish secular republic in 1923. The 1908 revolution, sometimes described as the proto-type of Near Eastern military coups of the 20th century (Rustow, 1959: 13), began as a military elite revolt which overthrew Sultan Abdlhamid II (18791909). Its challenge to the Ottoman ancien rgime was half-hearted and had to be renewed by another movement after a lost war. In the case of Iran, the constitutional revolution of 190611 and the Islamic revolution of 19789 were predominantly political insurrections buttressed by popular mass support, mobilized on the basis of revivalist and utopian ideologies.2 In retrospect, and with reference to de Tocqueville, the contrasting versions of the esprit rvolutionnaire can only be understood in light of different historical backgrounds. The following brief discussion is an attempt to trace the historical processes and cultural sources that shaped the destinies of the two countries. As I will try to show, the history of these two divergent revolutions can be traced back to the complex political and religious culture of the Anatolian Suknightly tradition of the Turko-Persian ecumenical age beginning in the 11th century. But the long-term outcome was the creation of the Ottoman and the Safavid empires. The notion of Su-knightly culture indicates a social imaginary that brought together the Su mystical tradition (with its distinctive cosmology) and the knightly code of ethics, which came to prominence as a political current in the Middle Period of Islamic history.3 More precisely, I argue that the establishment of these empires represents the crystallization of two distinct types of Su-knightly cultures, dened here as knightly-heroic (Ottoman) and millenarian-populist (Safavid). The basic difference in the pristine cultures of Ottoman and Safavid Su-knightly orders which I propose to examine in this article lies in the distinct socio-organizational types of the respective political coalitions, dened here as frontier-chieftaincy (Ottoman) and sectarian-chieftaincy (Safavid). At the end of the article I will briey consider the implications of the two types for later developments, nothwithstanding the dissolution of the Su-warrior classes as a consequence of sedentarization processes unfolding in the Ottoman and the Safavid imperial formations from the 15th to the early 17th century. Attempting to engage in a comparative study of the Ottoman and the Safavid empires, both of which began as chieftaincies and went on to extensive conquests, ultimately means embarking upon a broader spectrum of tasks. Here I do not pretend to develop a systematic comparative treatment of these two empires, or of the subsequent paths of their divergent yet connected political histories. Nor do I intend to offer an analysis of the historicalinstitutional formation of the modern Iranian and Turkish nation-states, based on Ottoman and Safavid legacies. Given the constraints of space, this article will only outline basic parallels and contrasts between the two Islamic

Downloaded from the.sagepub.com at City University Library on January 6, 2012

Rahimi: Between Chieftaincy and Knighthood

87

empires, with particular emphasis on the cultural formation of socioorganizational patterns. THE TURKO-PERSIAN ECUMENE OF THE MIDDLE PERIOD According to Marshall Hodgson, the Middle Period (9451503) of Islamicate history represents a civilizational shift of major importance. The period of genesis (c.600945) saw the replacement of Syriac and Pahlavi (IranoSemitic) traditions by an Arabic post-axial, agrarianate and citied civilization, underpinning an inclusive Muslim community between the Nile and the Oxus river developed on this basis. By contrast, the Middle Period was marked by a widening gap between state and society, the diffusion of Susm and the expansion of Persian as a literary language throughout a large part of the Afro-Eurasian landmass (Hodgson, [1958]1974).4 In a posthumous article published in 1970, Hodgson described this period as an age of great cosmopolitan creativity that reached its height by the 16th century, when the main region of Islamdom came under the control of empires (Mughal, Safavid and Ottoman) administered by military patronage states. This was the fth phase of Islamicate history, namely the era of Gunpowder Empires (Hodgson, 1970).5 In contrast to the conventional academic notion of a period of decline after the collapse of the High Caliphate, Hodgsons periodization of Islamic history offers us an alternative historical account, showing that Islamicate civilization underwent major transformations from the 10th to the 16th century. Although a critical study of Hodgsons system of periodization is beyond the scope of this article, it is reasonable to argue that the impact of the Turkish migration to Asia Minor played an integral role in shaping new patterns of socio-cultural crystallization in the Middle Period of Islamdom. The successive waves of Turkish migration from the steppe grasslands of Inner Asia to the settled regions of Anatolia and the Irano-Mesopotamian plateaus began in the ninth century, when Turkish slaves were recruited in order to create a new a military elite order, loyal to the Byzantine and the early Caliphate state. With the Ghaznavids (9771186), military slave elites of Turkish origin but Persianate culture emerged as heirs to the Caliphate political order. The Seljuq suzerainty in the 11th and the early 12th centuries, however, marked the establishment of the rst Turkish nomadic empire that led the way to the revival of Orthodox Sunnism. The establishment of a nonmilitary slave Turkish empire with vast expansionist aspirations represents the rst major nomadic conquest movement with religious revivalist dimensions. The gradual process of Turkish migration to the Anatolian regions, beginning in the 11th century, led to major demographic transformations. This occurred in two successive historical phases. The Seljuq victory over the Byzantine forces at the battle of Manzikert in 1071 inaugurated the rst

Downloaded from the.sagepub.com at City University Library on January 6, 2012

88

Thesis Eleven (Number 76 2004)

decisive stage, with major political consequences: the establishment of Turkish-speaking principalities in the western borderland marches as a way to challenge Byzantine control over Anatolia. This socio-demographic process, known as Turkicization, entered a second phase of development with the Mongol invasion of 1258, which intensied the Turkmen migration to the western regions of Anatolia, replacing the Greek-Christian peasant population with Turkish groups of nomadic origin. Though sporadic movements occurred throughout the 11th to 13th century, the 14th century highlights the nalization of a major demographic shift in Anatolia that involved radical changes of socio-cultural signicance. The transition from the early to the late Middle Period, as seen in the context of successive Turkish migrations to Anatolia, can be regarded as a revolutionary phase in two important ways. On one level, the complex process of hybridization of Arabic, Persian and Turkish cultural elements, from the end of the High Caliphate to the establishment of the Ilkhanate era in Iran in the 13th century, represents a new period of creative cosmopolitanism, and a new stage of the Turko-Persian ecumene (Caneld, 1991: xiv). The Turko-Persian Islamicate culture that had crystallized under the Samanids and the Qarakhanids on the eastern Iranian margin in the 11th century, and which was later exported to surrounding regions, was a product of intercivilizational encounters and open to further developments of that kind. In the particular case of the Turkicization of Anatolia from the 11th to the 14th century, the regional mixture of agrarian, nomadic tribal and urban settings was particularly favourable to cultural blending.6 The fusion of Arabic-scriptural, Byzantine-Greek, Turkish-nomadic and Persianiate-lettered traditions of the Middle Period paved the way for the creation of new cultural complexes. On another level, this mixed borderland civilization (Wittek, 1966: 20) also became a meeting ground of different religious traditions. In this sense, the blending of steppe (instrumental) religious practices of the Turkish nomads with the universal (soteriological) religions of Irano-Semitic and Byzantine-Greek societies represents the crystallization of new cultural milieus, where nomadic and settled civilizations had to some extent been amalgamated.7 From the 11th century onwards, the most original expression of this syncretistic process was the appearance of Anatolian Susm, in its distinct shamanistic form of Darvish Islam (or bb Islam), as a dominant aspect of the daily life of the Turkish nomadic population, and indeed the main factor for the conversion of rural Asia Minor to Islam. The development of Suesque heterodoxies, sectarian and millenarian movements throughout the Middle Period can in part be credited to this process of religious syncretism, in which Shii and Sunni ritual practices and creeds intermingled in close proximity and at times overlapped in the shifting spaces of everyday interaction.8 But it was the emergence of the Turkish Su-knightly brotherhood

Downloaded from the.sagepub.com at City University Library on January 6, 2012

Rahimi: Between Chieftaincy and Knighthood

89

orders the so-called Ghazi warriors that gave a political expression to this civilizational fusion. The origins of the Su-knightly associations could be traced back to the urban-based pure brethren of the Qarmati movement, which played a great role in the development of the Islamic guilds in the 10th century (Lewis, 1937), and the Futuwwa associations that were revived under the reign of Caliph An-Nasr in the late 12th and early 13th centuries, as a consequence of the expansion of trade and revival of towns under the Seljuq rule. In the early Middle Periods these relatively autonomous movements, as popular militia and volunteer Su-guild associations, played a crucial role in the local governance of Islamdom. The Futuwwa associations combined an ethical code of egalitarianism, in the form of Darvish fraternities, with non-egalitarian charismatic elitism, in the form of master (pr)disciple and patron-client relations. Ties of blood and kinship afliations were less important than competition for the sacred status of leadership in the clubs, manifested in the paradoxical notion of rst among equals, which reected the knightly-spiritual character of the associations. In the post-Mongol era of Islamic history, the late Middle Period, the Futuwwa associations began to merge with the Anatolian-Su orders (Lewis, 1937: 278), a process that during the 14th century spread further within Islamdom. The synthesis between the Futuwwa and the Anatolian-Su orders created the Akhyat al-Fityn or Akhis movements, which tended to fuse the horseback warrior culture of Inner Asia with the sedentary Irano-Semitic Messianic traditions.9 Built around the ethos of steppe heroism and the Quranic notion of justice, the brotherhoods lived by a strict code of honour, embedded in a culture of reverence for spiritual sacred persons (shaman) and belief in the potential to unite the mundane with the supernatural world through ritual, ceremony and, above all, war. FRONTIER CHIEFTAINCY: THE OTTOMAN CASE The so-called Ghazi thesis, proposed by the Austrian historian Paul Wittek, which described the original Ottomans as holy warriors, has come under serious attack in recent years. For Wittek (1966), the Ottomans were march-warriors inspired by the ideology of holy war who attacked and overran the frontier lands between the Byzantine and Seljuq empires.10 Their religious culture was rooted in the older Ghazi tradition of Anatolia. But part of the difculty with this thesis, as Rudi Lindner has argued, is that by placing too much emphasis on the Islamic dimension, Wittek obscures an Inner Asian legacy, in particular the shamanistic tendencies characteristic of the early Ottoman frontier principalities (Lindner, 1983: 10512). On this view, Wittek ignores certain heretical religious practices, like the rituals of human sacrice, which played an important role in early Ottoman society. In addition, Wittek does not do justice to the inclusive and tolerant policy of the early Ottoman conquerors. An adversary ideology, as postulated by the Ghazi thesis,

Downloaded from the.sagepub.com at City University Library on January 6, 2012

90

Thesis Eleven (Number 76 2004)

would have excluded Byzantines from joining the Ottoman forces. Using the anthropological literature on tribes, Lindner further argues that the early Ottomans were a tribal political group whose membership should be dened not in terms of exclusivist religious zeal but by wide-ranging shared interests of its heterogeneous members (Lindner, 1983). On the other hand, Cemal Kafadar has criticized both of these approaches. He argues that the simple dichotomy between Inner Asian and Islamic aspects misses the ambiguous cultural reality of early Ottoman society (Kafadar, 1995). Following Kafadar, we can reconstruct the early history of the Ottomans as a combination of the quasi-corporate organization of tribal nomadism with the Ghazi spirit of the heterodox frontier culture; as the Ottomans eventually established a centralized administration through military build-up, taxation and urbanization, this unstable mixture was replaced by more orthodox models. It is important to note that the term Ghazi used here does not refer to an exclusive adversary ideology of Holy War against an indel enemy, but rather to the Su-knightly hybrid culture of Anatolia in the Middle Period. In this sense, the early Ottomans should be dened as a frontier-chieftaincy entrenched in a Su-knightly culture of honoric ethos. This can be explained on two levels. First, the early Osmanli dynasty, although one of many petty and semiautonomous Turkoman principalities in western Anatolia, represented a distinct type built around the chief as the source of charismatic authority. The status of the early Ottoman chief was earned by victory in battle, and his authority depended upon accomplishments in the struggle for expansion. In a way, this type of associational organization dees the Ibn Khaldunian denition of tribal organization as based exclusively on mere kinship and religious feeling (asabiyya).11 As Ira M. Lapidus has argued, the founder of the Ottoman dynasty did not rely on lineage ideology (this was a later fabrication); rather, he relied on an agglomeration of diverse groups, including the Byzantine defectors, whom the chieftain united based on shared interest for military expansion (Lapidus, 1990: 334). What is the relationship between chieftaincy and tribe? Tribes, Lapidus denes, are not familial or ethnic groups but political and religious chieftaincies whose composition varies greatly (Lapidus, 1990: 27). This applies, more specically, to the tribal formations characteristic of Middle Eastern history. They are primarily characterized as heterogeneous culturally but politically united under a central authority, the chief (Tapper, 1983: 9). This model seems best suited to explain the success of Ottoman expansion, conquest and, ultimately, sedentarization, as it integrated diverse ethnic and religious groups into a exible, resilient and enduring world empire. Second, the structural aspect of the early Ottoman chieftaincy was related to the peculiar culture of an inclusive Su-knightly order, in the tradition of the Akhi association mentioned above. They combined mystical religiosity with the warrior ethos of bravery, independence and, above all,

Downloaded from the.sagepub.com at City University Library on January 6, 2012

Rahimi: Between Chieftaincy and Knighthood

91

honour. Deeply disciplined through the codes of honoric spiritualism, with a strong heroic antipathy to submission and defeat, the early Ottomans were male warriors who formed alliances and accepted the absolute authority of their spiritual guide (bb), conceptualized as a living embodiment of the supernatural world to protect the community. In more specic historical terms, the Su aspect of the Ottoman chieftaincy found expression in its close ties to the Bektsh order, which later played a central role in the formation of the religious culture of the slave soldiers, namely, the Janissary corps. Perhaps the most signicant aspect of the emerging frontier-chieftaincy was its hierarchical organization, a feature that continued to prevail in Ottoman society until the new property-holding class began to challenge the strict social codes of distinctions in the second half of the 17th century (Abou-El-Haj, 1991: 4951). Hierarchical order, unied at the top, divided the chief and his warriors from the subjects. This social-organizational trait appears to have become more prevalent with the consolidation of the dynastic lineage ideology (based on patrilineal descent), as a result of the increasing sedentarization of the Ottoman chieftaincy through expansion and conquest during the 14th century. Guided by a set of strict ceremonial and ritualistic codes of conduct that separated the knightly military caste from the rest of the population, the Ottoman elites, with the sultan at the head, governed their conquered territories as heroic protectors and guardians of the chieftain community. This caste-like system may of course appear to pose a problem for the egalitarian spirit of the Su-knightly order which had given birth to the Ottoman frontier-chieftaincy. But in fact the early Ottoman Su-knightly order was deeply implicated in the ethos of expansion, incorporating the culture of honoric competition to maintain division and rank among its confederate members and, above all, conquered subjects. The consolidation of hierarchy paved the way for centralization and empirebuilding in the 15th and 16th centuries. These developments pregured the later expansionism and militarism of the Ottoman empire, whose administration was essentially a military structure at least until the reign of Selim III (17891807) when reforms led to radical changes in the structure of the Ottoman military institutions. They also shed light on the cultural dimension of the Ottoman empire-building processes, especially after the conquest of Constantinople in 1453, and more particularly on the combination of expansionism and cosmopolitanism: whereas the former can be identied with the chieftain politics of warlike frontier culture, the latter can in turn be linked to a more inclusive culture of knightly-heroism, grounded in the ethos of bravery, independence, loyalty and self-abnegation. SECTARIAN CHIEFTAINCY: THE SAFAVID CASE The prehistory of the proclamation of Twelver Shiism (Ithna-ashari) as the state religion of Iran by Shah Ismail I (14871524) at the congregational

Downloaded from the.sagepub.com at City University Library on January 6, 2012

92

Thesis Eleven (Number 76 2004)

mosque of Tabriz in 1501 is complicated, and a detailed discussion is beyond the scope of this article. But a sketch of the historical background to the ascendancy of Ismail I will be necessary for a better understanding of the Safavid chieftaincy, and the distinct cultural and social organizational traits that set it apart from the Ottoman form of chieftaincy. The origins of the Safavids can be traced back to the Sunni-Su order of the mystic Shaykh Sa al-Din (d.1334), whose status as the spiritual head (pr) of the order gave rise to the label Safawiyya, or the Safavid.12 The Su order evolved into a messianic sectarian (gult) movement with the succession of Shaykh Junayd (145960), during which it underwent a momentous transformation from contemplative inner-worldly Susm to the openly outerworldly heterodox extreme Shiism (ghuluww). The evolution of the Safavids from a quietist to a militant revolutionary force in the 15th century, represents in a sense an escalation of the hybridizing process between the Inner Asian and Irano-Semitic civilizations. On one hand, old shamanistic beliefs continued to be blended with the extremist Shii conception in the notions of divine incarnation of God in man and the Christian belief in the trinity (Moosa, 1988: 40); on the other hand, though closely related, the fusion of certain Shii extremist practices with the Su belief in the mystical nature of reality saw the appearances of new heretical movements and militant millenarian orders, which developed throughout Islamdom in the 15th century.13 Two main factors explain these hybridizing civilizational processes. First, beginning in the 13th century, the laissez-faire approach of the Mongol rulers to religion led to the growth of religious heterodoxies that continued to spread well into the 15th century. The period produced a uid ambience of mixed religious practices, involving Christianity, Susm, extreme Shiism and Sunni Islam, and a veritable explosion of Su movements with pro-Shii tendencies. Second, the centralization of the Ottoman chieftaincy into an empire with the institutionalization of Sunni orthodoxy, beginning with the reign of Bayazid I (13891402), led to growing persecution of heretical religious movements throughout the Ottoman provinces. This process was a complement to the expansion of tax farming as a source of revenue for the centralizing Ottoman empire, applied both to the peasantry and to seminomadic groups. Even though uprisings by Darvish Turkoman groups, like the Bb shq movement in 1241, occurred in the rst half of the 13th century, the greater frequency of revolts in the 15th century reects stubborn resistance to the oppressive policies of Ottoman rulers, and to their efforts to subdue the nomadic heretical forces that roamed the peripheral regions of the empire. This also led to the emergence of revolutionary movements from diverse grass-root chieftain groups, their ideologies mixed with revivalist, utopian and eschatological themes. As I will try to show, it was at this crucial historical juncture that the chieftain and sectarian identities mixed to form the grass-roots religiouspolitical aspect of the Safavid revolutionary movement. Seen in this context,

Downloaded from the.sagepub.com at City University Library on January 6, 2012

Rahimi: Between Chieftaincy and Knighthood

93

the Safavid chieftaincy was composed of followers of a Su order recruited among the tribal groups (known in Inner Asia as Oymqs or tribes) located in diverse places like Anatolia, Syria and Inner Asia. Known as the Qizilbash or redheads, after their knightly red caps with 12 folds representing the 12 Shii Imams, the followers of the Safavid order also consisted of Turkoman nomads, descended from the Saljuq Ghuzz Turks, that faced persecution by the Ottomans as the latter transformed their frontier-chieftaincy into a sedentary empire in the early 15th century. In broad terms, the Safavid Qizilbashi movement during its early extremist and revolutionary phase was political in nature, representing the last bid for power by the Anatolian and Caspian regions, which had always resisted Islam in its centralized and orthodox form. On the explicitly political level the Qizilbash followed Shii traditions in recognizing the prophets cousin and son-in-law Ali and his descendants as the leaders of the Islamic community. But in line with the mystical tradition, they also attributed supernatural qualities to these hereditary leaders; they held esoteric keys to ultimate reality. This fusion of diverse sectarian elements served to articulate the opposition to Sunni power represented by the Ottomans and their allies in Inner Asia, the Uzbeks. The movement was thus also a radical religious one, proposing a heterodox alternative to the existing order. In the case of the Safavids, the eschatological element was particularly strong; they may therefore be described as a sectarian movement par excellence.14 In contrast to the Ottoman type of frontier chieftaincy, where religious belief was ultimately instrumental to the struggle against the Christian indels, the sectarian ideology of the Safavids involved belief in the heretical doctrine of reincarnation (tansukh) and the return of the deceased to the world in various forms (recalling certain practices of spirit-possession); it also included the belief in transmigration and the oneness of the sacred spirit incarnated in prophets and saints, and now embodied in the Su guide. On one hand, the chieftain dimension of the Safavid Qizilbash order highlights the Oymq, a type of political organization based on a territorially bounded collectivity of groups (Tapper, 1990: 68), inclusively heterogeneous and composite, where agricultural, pastoral and commercial (trade) economies coalesced under the charismatic leadership of an urban-based chief (Reid, 1978, 1979). On the other hand, the Safavid chieftaincy also involved the additional aspect of sectarian belief in the supreme charismatic leadership of a single chief, recognized as the divine source of sovereignty. The heretical notion of a transmigration of divine spirit through different human bodies (Babayan, 1994: 136) gave a specic avour to the Safavid version of charismatic leadership; this aspect was absent from the Ottoman model of frontier chieftaincy where spiritual authority was linked to a more heroic ethos of struggle. To sum up, the Safavid Qizilbash warriors differed from the Ottoman Ghazis primarily in regard to their religious-political structure as a dissident

Downloaded from the.sagepub.com at City University Library on January 6, 2012

94

Thesis Eleven (Number 76 2004)

association with strong heretical tendencies.15 This popular, grass-roots aspect of the association is most apparent in the origins of the movement, as a Su order that was supported by the peasants, tribal orders and discontented groups in Islamdom. In the 15th century, when the movement was gaining momentum, it reconciled its followers from both the countryside and the towns, and they included a signicant number of nomadic Turkmans. The grass-roots origin of the Qizilbash were further reected in some egalitarian features of their power structure. As Kathryn Babayan has shown, the Qizilbash maintained during their revolutionary stage a conception of authority based on the notion of corporate sovereignty that recognized the sharing of political roles among the members regardless of gender differences (Babayan, 1998). This feature mainly represented steppe traditions of an egalitarian basis of authority, which tended to erode as the movement began to evolve from chieftaincy into a sedentary empire. The sharing of power among the tribal members, however, did not overshadow the sacred leadership of the chief; it only implied that this supreme authority operated in a more egalitarian manner, through a less stratied power structure than in the case of the Ottoman chieftaincy which operated as a frontier military order with strict hierarchical relations among its members. Although the gradual process of hierarchization also transformed the Safavid chieftaincy into an increasingly stratied imperial order, the movement was never based on the hierarchy of distinction. The tradition of love for the spiritual guide was a way to preserve the mystical idea of spiritual union on an equal basis, becoming one with the supernatural world in unconditional devotion for the Su pr. In short, certain distinct cultural patterns central to the identity of the Qizilbash and to their symbolic resources in the struggle for power set the Safavid chieftaincy apart as a specic kind of Su-knightly order. This cultural character of the Qizilbash warriors especially evident in their extremist stage of development when they entered the battleeld unarmed with the belief that Ismails supernatural power would protect them from the enemy (Babayan, 1994: 135) combined the Ghazi spirit of holy war with the shamanistic practices of exocannibalism, decapitating the enemys body for consumption (Arjomand, 1981a: 6). This fusion of hybrid Su and steppe cultures, grounded in sectarian revolutionary zeal, appears to have expanded under the leadership of Junayd when the emerging Qizilbash warriors began to combine the devotional practices of reverence for the Su master with the Shii belief in a messianic gure. In the 15th century, with the merging of the ethos of honour with the custom of devotion to the chief as a sacred gure in the 15th century, the knightly culture of the Safavid movement became more and more different from the Ottomans. In addition, apocalyptic beliefs in the end of the world, coupled with the populist yearning for redemption through self-sacrice, gave the Safavid culture a special avour, knighthood. The heretical Qizilbash version of Susm rejected the notion of resurrection, central to Shii orthodoxy, and replaced it with the belief that divine

Downloaded from the.sagepub.com at City University Library on January 6, 2012

Rahimi: Between Chieftaincy and Knighthood

95

spirit can migrate into different bodies through time. This notion translated into a strong model of sacred leadership, which served to channel populist protest against the established order and contain egalitarian and pluralist conceptions of authority. STATE FORMATION AND THE PACIFICATION OF THE SUFIKNIGHTS Despite the different structure of their chieftaincies, there were some remarkable similarities between the Ottomans and the Safavids. The two incipient states represent similar patterns of inclusive political organization, which enabled them to develop into empires with stable combinations of religious and political authority. In a broad comparative perspective, and in contrast to the Western knighthoods that emerged in the High Medieval times, the Ottomans and the Safavids did not evolve through the restructuring of an established nobility into a military elite (Arnason, 2002: 128). And in contrast to the Japanese samurai warriors, the Ottomans and the Safavids did not primarily rely on an honoric culture, as a symbolic system of stratication and the expression of collective elitist discipline (Ikegami, 1997: 11). As I tried to show in the previous two sections, the formation of these two imperial powers involved an intricate cultural mix of semi-nomadic politics and popular (or unofcial) cultures, together with a syncretic blending of steppe shamanistic, Shii, Su and Sunni religions; these processes give rise to two distinctive forms of chieftaincy. Furthermore, the development of the two models can only be understood in the context of the general transformation and diversication of Islamic political regimes during the Middle Period. Further similarities emerged during the complex process of sedentarization and the formation of centralized states. As the two chieftaincies grew into neighbouring empires, they applied parallel strategies of consolidation. In both cases, there were systematic and violent attempts to subdue the roaming bands of Su-knights, in their respective Ottoman Ghazi and Qilizbash forms, and replace them with state-organized professional armies. The Ottomans, especially under the reign of Bayazid I at the end of the 14th century, replaced the Ghazi warriors with slave soldiers, the Janissaries. This newly established military class went on to dominate the sultanate until 1826 when they were replaced by a more modern organized army which underwent further reforms during the Tanzimat era (183976). The development of scal centralization through a new tax system (timar) went hand in hand with the consolidation of a system of landholding in return for military service. Similarly, the Safavids, especially under the reign of Shah Abbas I (15871629), ruthlessly eliminated the original Qizilbash warriors and replaced them with a Georgian slave army (ghulms). The emergence of a military slave-state under the reign of Abbas I was the climax of Safavid power. At the same time, the imperial court brought

Downloaded from the.sagepub.com at City University Library on January 6, 2012

96

Thesis Eleven (Number 76 2004)

the former Qizilbashi efdoms under control and used their revenues to pay the salaries of the new military. In both cases, the institutionalization of slave armies laid the foundations for a new political order and led to the disestablishment of preimperial warrior elites. In close connection with the policy of detribalization, the Ghazis and the Qizilbashs lost their inuence at the courts of the Ottoman sultan and the Safavid shah. These developments strengthened the imperial centres and marginalized the dissenting elements of warriors that had continued to challenge the centralized empires during their formative stages. As a result of progressive bureaucratization, the original charismatic force of the Ghazi and Qizilbash movements was routinized and subordinated to the unfolding imperial projects. Imperial competition and military conicts on an ever-increasing scale shaped the course of institutional change on both sides. In brief, the accumulation of military power and the structural changes to military organization were at the centre of more comprehensive political transformations. On the level of religious culture, the institutionalization of a centralized orthodoxy led to the disintegration of Ottoman and Safavid heretical religiosity. In the Ottoman case, the process of establishing a Sunni orthodox legal order unfolded from the early 15th to the late 16th centuries. Especially under the rule of Mehmet I (141321) and Mehmet II (142151), the creation and the consolidation of the ofce of Shaykh-al-Islam led to the suppression of frontier heterodoxies and the strengthening of central religious institutions under the supervision of the sultan (Pixley, 1976). The Safavids also pursued the institutionalization of orthodoxy beginning with the ofalization of Shiism in 1501 and culminating under Abbas I in 1591. They supported the migration of the Shii ulama from Arabic Iraq and Lebanon to centres of learning under imperial control (Arjomand, 1981b, 1984). The increasing power of the shar-minded ulamas at the court, a process that reached its highest point under the reign of Shah Sulayman (166694), shifted the focus of religious life away from Qizilbashi sectarianism towards a Twelver Shii puritanical and hierarchical religious order. In short, the bureaucratization of the religious sphere not only signied the consolidation of a centralized state under the rule of the shah and the sultan, but also indicated the systematic eradication of non-ofcial religions that had been characteristic of the original Su-knightly culture of the early Ottoman and the Safavid chieftaincies. Despite the domestication of Ghazi and Qizilbash warriors, in the course of long and bloody struggles involving civil wars and periodic rebellions, the original Su-knightly cultures left signicant traces in the symbolic frameworks of the two sedentarized empires. The civilizing effort to domesticate the Ghazi and the Qizilbash orders was accompanied by a systematic dissociation of Su-knightly cultures from their original social bearers. This mainly occurred by means of state-led construction of new urban spaces and the institutionalization of rituals, ceremonies and customs in connection with

Downloaded from the.sagepub.com at City University Library on January 6, 2012

Rahimi: Between Chieftaincy and Knighthood

97

crystallization of new collective identities. In this regard, the Su-knightly cultures were reshaped in the imperial contexts of knightly initiation rites (Ottoman) and millenarian rituals of death and redemption (Safavid) that demarcated the symbolic domains of the integrative and normative imperial collectivities.16 After the collapse of the empires in the early 18th (Safavid) and the early 20th centuries (Ottoman), new political institutions and social movements drew heavily upon the symbolic resources of the Su-knightly traditions in order to legitimate their projects and self-understandings. Knightly-heroic culture played a central role in the establishment of the Turkish republic, in connection with the struggle to protect the nation from external threats. The tradition of millenarian populism in post-Safavid Iranian society surfaced in mass-based political movements which used eschatological language to articulate their struggle against oppression. But in a more general sense, the legacy of the two cultures was crucial to the formation of modern Iran and Turkey: the sustained struggle for independence and selfdetermination owed much to traditions inherited from the warrior elites of the formative phase. CONCLUSION The task of historical sociology is to link theoretical arguments to concrete analyses of events, processes and transformations. Comparative analysis is an essential part of this project and the best way to avoid an overgeneralization from particular cases. In this spirit, the present article suggests ways of comparing the Ottoman and Safavid empires. Popular stereotypes of the two empires have identied them, respectively, with Sunni (Ottoman) and Shii (Safavid) Islam. As I have argued, such views miss the very complicated inter/intracivilizational macro-historical trajectories that were crucial to the creation of the two empires: they also ignore the divergent patterns of cultural idioms embedded in their respective political orders. By emphasizing the Su-knightly cultural dimensions of the early Ottoman and the Safavid chieftaincies, I tried to show how the gradual domestication of the chieftain warrior elites was closely tied to the state-building processes; at the same time, their cultural heritages were important to the construction of new collective identities. It should be noted that the divergent versions of the original Suknightly ethos were not the only cultural paradigms that enabled the Ottoman and the Safavid empires to prevail over other competitors in the eld at the dawn of early modern Islamic history. They played prominent roles in the processes of state and nation formation. A more broadly based comparative analysis would have to account for their interaction with other models and traditions. But the legacy of the Ottoman and the Safavid Su-knightly cultures was alive and active in recent history. In the early 20th century, the

Downloaded from the.sagepub.com at City University Library on January 6, 2012

98

Thesis Eleven (Number 76 2004)

rise of Atatrk to power in 1923, as the father (bb) of the newly established Turkish (secular) republic, is a fascinating reminder of the Su-knightly culture of reverence for the heroic knight as the guide and protector of the community. Similarly, the rise of Ayatollah Khomeini to power in 1979 as a popular quasi-messianic leader of the new Islamic republic, with the title of Imam, showed that the ethos of millenarian populism was still present in modern Iranian politics. And to conclude, contemporary trends of the same kind should be noted. The military version of the knightly-heroic ethos, with its secular Jacobin tendencies, remains a powerful factor in Turkish politics and an obstacle to the consolidation of democracy. On the other hand, millenarian populism, in its modern Shii nationalistic form, remains a potent symbolic resource for movements struggling against the hierocratic regime in post-revolutionary Iran, as can be seen from recent popular demands for a more democratic government.

Babak Rahimi is currently writing his PhD thesis on the history of the Safavid state and society relations, titled Between Carnival and Mourning: The Muharram Rituals and the Emergence of the Early Modern Iranian Public Sphere in the Safavid Period, 15901641, at the European University Institute, Florence, Italy. Address: European University Institute, Department of Political and Social Sciences, Via dei Roccettini 9, 1-50016 San Domenico di Fiesole, Italy. [email: brahimi77@hotmail.com].

Notes
1. An earlier version of this article was originally presented to the World Sociology Congress in Brisbane, Australia, in 2002. I am grateful to Dr Jari Eloranta, who has critically reviewed sections of this work, and also to Professor S. N. Eisenstadt for his encouraging comments. 2. Together with the Russian, Mexican and Chinese revolutions, the Iranian Constitutional Revolution of 190611 was a signicant part of a global popular revolutionary movement of the early 20th century. As Janet Afary argues, the Constitutional Revolution was not merely a transformation of the elite structure, but rather a grass-roots based political and socio-cultural revolution, which reduced the powers of the shah and his ministers (Afary, 1996), marking the rst popular constitutional revolution in Asia (Mottahedeh, [1985]2002: 6). The Islamic revolution of 19789 was also a popular insurrection against arbitrary power, but the context was very different. As Ervand Abrahamian and Sad Amir Arjomand have shown in their seminal studies, the 1979 revolution was an outcome of socio-economic changes on the political structure, especially under the rule of the Pahlavi regime (Abrahamian, 1982), and of changes to the dual system of ulama and royal authority that had prevailed before the centralization of the Pahlavi state (Arjomand, 1988). 3. The term knightly does not signify the Western form of knighthood, i.e. the military elite of the High Middle Ages that maintained a privileged social status by having control over land. Though similarities can be detected between Western and Islamic knighthoods, the idea of knightly order here represents

Downloaded from the.sagepub.com at City University Library on January 6, 2012

Rahimi: Between Chieftaincy and Knighthood

99

4.

5.

6.

7.

8.

9.

10. 11.

12. 13.

distinctive Islamic traits in reference to the history of the Ottoman and the Safavid chieftaincies. Hodgson (1974) lists six periods of Islamic history: the formative (to 692 CE), the High Caliphate (to 945), the International Civilization (to 1258), the Age of Mongol Prestige (to 1503), the era of the Gunpowder Empires (to c.1800) and Modern Times, with the emergence of nation-states. The Middle Period lumps together the third and the fourth period in this categorization. In this article, Hodgson argues that egalitarian and cosmopolitan elements in Islam, incorporated and institutionalized in the civilization of the Irano-Semitic societies, have made a lasting impact on interregional developments on a hemispheric-wide basis. Although Claude Cahen has argued that it is obviously impossible to give any gure for the Turkish immigration into Asia Minor (Cahen, 1968: 143; Cahen, 1969), evidence indicates a long-term process of conversion of the natives to Islam with the migration of Turkish Muslims to the region from the 11th to the 13th century (Mnage, 1979; see also Vryonis, Jr, 1971). As David Gellner explains, the difference between soteriological and instrumental religions is primarily based on their experiential orientation towards the supernatural: whereas the former type represents the belief in salvation with practices directed towards appeasing the supernatural with the aim of redemption, the latter is directed towards making specic things happen in the world through magical practices of shamanism and spirit-possession. In this sense, instrumental religions are not based on the faith, but rather on notions of efcacy of spiritual experience to control the supernatural (see Gellner, 1992). For the best exposition of Su history in this period, see Arjomand (1984: 6684). It is important to note, however, that Susm and Islamic messianism (especially in its Shii form of Mahdism) existed in earlier periods of Islamic history. The histories of the Abbasid and the Ismaili (Fatimid) revolutions in the eighth and tenth centuries, for instance, are replete with apocalyptic and messianic beliefs in the Mahdi that spread widely beyond other extremist Shiite groups (Arjomand, 2002: 114). It is important to note that the name Akhis is not derived from the Arabic word for brethren but is purely a Turkish word, meaning knightly or noble (Blake and Langer, 1932: 500). See also Wittek (1982: 285319). It should be noted in passing that I do not accept the functionalist notion of segmentary organization, especially as it is used by Ernest Gellner, dened exclusively in terms of lineal descent, with tribal members at each segment of the association equally balanced by others. As has been noted by a number of anthropologists, segmentary lineage theory ignores divergent cultural aspects that involve complex ambiguous relations between the actual groups of a tribe. For an interesting critique of this theory, see Hammoudi (1980). For a general history of the Safavids, see Mazzaoui (1972) and Savory (1980). I base my argument here on Annemarie Schimmels assertion that this apparent fusion between Susm and Shiism (both in its orthodox and heretical form) occurred in the centuries prior to the rise of the Safavid to power, which facilitated the advent of the dynasty as a result of this hybrid process (see Schimmel, 1974).

Downloaded from the.sagepub.com at City University Library on January 6, 2012

100 Thesis Eleven (Number 76 2004)

14. It should be noted that Ottoman politics, from chieftaincy to the classical imperial period, also manifested features of millenarian religiosity. But such millenarian elements were limited to the legitimization of power and the monopolization of authority, especially in the 16th century when the Safavid movement represented the greatest threat to the Ottoman control of eastern Anatolia. Unlike the Safavids, millenarianism never played a central symbolic role in the formation of the Ottoman empire. For a study of Mahdism and the messianic politics of the Ottoman state see Fleischer (1992: 15778). 15. In a generic sense, the early Qizilbash also regarded themselves as Ghazi, holy warriors for the faith. The specic name of Ghazi is used in this work in a nominal sense, and is to address the early Ottoman Turks so to differentiate them from the Qizilbash. 16. Here, I am mainly referring to the Ottoman Imperial Circumcision, celebrated for the circumcision of a prince, and the Safavid Muharram ceremonies, performed in commemoration of the martyrdom of the Prophets grandson, Hussain, on the plains of Karbala, 680. In a later article I shall discuss the importance of these two rituals in the formation of the Iranian and Turkish collective identities.

References
Abou-El-Haj, Rifaat (1991) Formation of the Modern State: The Ottoman Empire. Sixteenth to Eighteenth Centuries. Albany: New York State University Press. Abrahamian, Ervand (1982) Iran: Between Two Revolutions. Princeton, NJ: Princeton University Press. Afary, Janet (1996) The Iranian Constitutional Revolution, 19061911. New York: Columbia University Press. Arjomand, Sad Amir (1981a) Religious Extremism (Ghuluww), Susm and Sunnism in Safavid Iran: 15011722, Journal of Asian History 15(1): 135. Arjomand, Sad Amir (1981b) The Shiite Hieroracy and the State in Pre-modern Iran: 17851890, Archives Europennese de Sociologie 22(1): 4078. Arjomand, Sad Amir (1984) The Shadow of God and the Hidden Imam: Religion, Political Order and Societal Change in Shiite Iran from Beginning to 1890. Chicago, IL: Chicago University Press. Arjomand, Sad Amir (1988) The Turban for the Crown: The Islamic Revolution in Iran. Oxford: Oxford University Press. Arjomand, Sad Amir (2002) Messianism, Millennialism and Revolution in Early Islamic History, in Abbas Amanat and Magnus T. Bernhardson (eds) Imagining the End: Visions of Apocalypse from the Ancient Middle East to Modern America, 10628. London and New York: I. B. Tauris. Arnason, Johann P. (2002) The Peripheral Centre. Melbourne: Trans Pacic Press. Babayan, Kathryn (1994) The Safavid Synthesis: From Qizilbash Islam to Imamite Shiism, Iranian Studies 27(14): 13561. Babayan, Kathryn (1998) The Aqd Al-Nisd: A Glimpse at Safavid Women in Local Isfahn Culture, in R. G. Gavin (ed.) Hambly Women in the Medieval Islamic World. New York: St. Martins Press. Blake, Robert P. and Langer, William L. (1932) The Rise of the Ottoman Turks and its Historical Background, The American Historical Review 37(3): 468505.

Downloaded from the.sagepub.com at City University Library on January 6, 2012

Rahimi: Between Chieftaincy and Knighthood

101

Cahen, Claude (1968) Pre-Ottoman Turkey: A General Survey of the Material and Spiritual Culture and History c. 10711330. New York: Taplinger. Cahen, Claude (1969) The Turkish Invasion: The Selchkids, in Kenneth M. Setton (ed.) A History of the Crusades, pp. 13576. Madison: University of Wisconsin Press. Caneld, Robert L. (1991) Introduction: The Turko-Persian Tradition, in Robert L. Caneld (ed.) Turko-Persia in Historical Perspective, pp. 134. Cambridge: Cambridge University Press. Fleischer, Cornell H. (1992) The Lawgiver as Messiah: The Making of the Imperial Image in the Reign of Sulleyman, Soliman le Magnique et son temps, Acts du Colloque de Paris, Galeries Nationales du Grand Palais, 710 March 1990. Paris: La Documentation Francaise. Gellner, David (1992) Monk, Household and Tantric Priest: Newar Buddhism and its Hierarchy of Ritual. Cambridge: Cambridge University Press. Hammoudi, Abdullah (1980) Segmentarity, Social Stratication, Political Power and Sainthood: Reections on Gellners Thesis, Economy and Society 9(3): 279303. Hodgson, Marshall G. S. ([1958]1974) The Venture of Islam: Conscience and History in a World Civilization, Vol. 2. Chicago, IL: Chicago University Press. Hodgson, Marshall G. S (1970) The Role of Islam in World History, International Journal of Middle Eastern Studies 1(2): 99123. Ikegami, Eiko (1997) The Taming of the Samurai: Honoric Individualism and the Making of Modern Japan. Cambridge, MA: Harvard University Press. Kafadar, Cemal (1995) Between Two Worlds: The Construction of the Ottoman State. Berkeley: University of California Press. Lapidus, Ira M. (1990) Tribes and State Formation in Islamic History, in Philip S. Khoury and Joseph Kostiner (eds) Tribes and State Formation in the Middle East, pp. 2547. Berkeley: University of California Press. Lewis, Bernard (1937) The Islamic Guilds, The Economic History Review 8(1): 2037. Lindner, Rudi Paul (1983) Nomads and Ottomans in Medieval Anatolia. Bloomington: University of Indiana Press. Mazzaoui, Michael M. (1972) The Origins of the Safavids: Shiism, Susm and the Ghult. Wiesbaden: F. Steiner. Mnage, V. L. (1979) The Islamization of Anatolia, in Nehemia Levtzion (ed.) Conversion to Islam, pp. 5267. London and New York: Holmes & Meier. Moosa, Matti (1988) Extremist Shiites: The Ghulat Sects. New York: Syracuse University Press. Mottahedeh, Roy ([1985] 2002) The Mantle of the Prophet: Religion and Politics in Iran. Oxford: Oneworld. Pixley, Michael M. (1976) The Development and Role of the Seyhlislam in Early Ottoman History, Journal of the American Oriental Society 96(1): 8996. Reid, James J. (1978) The Qajar Uymaq in the Safavid Period, 15001722, Iranian Studies 11: 11743. Reid, James J. (1979) Comments on Tribalism as a Socioeconomic Formation, Iranian Studies 12(34): 27581. Rustow, Dankwart A. (1959) The Army and the Founding of the Turkish Republic, World Politics 11(4): 51352. Savory, R. M. (1980) Iran under the Safavids. Cambridge: Cambridge University Press.

Downloaded from the.sagepub.com at City University Library on January 6, 2012

102 Thesis Eleven (Number 76 2004)

Schimmel, Annemarie (1974) The Ornament of the Saints, Iranian Studies 7(1): 88111. Tapper, Richard (1983) Introduction, in Richard Tapper (ed.) The Conict of Tribe and State in Iran and Afghanistan, pp. 182. New York: St. Martins Press. Tapper, Richard (1990) Anthropologists, Historians, and Tribespeople on Tribe and State Formation in the Middle East, in Philip S. Khoury and Joseph Kostiner (eds) Tribes and State Formation in the Middle East, pp. 4873. Berkeley: University of California Press. Vryonis, Speros Jr. (1971) The Decline of Medieval Helenism in Asia Minor and the Process of Islamization from the Eleventh through the Fifteenth Century. Berkeley: University of California Press. Wittek, Paul ([1936]1982) Deux chapters de lhistorie des Turks de Roum 1. Les traits essentials de la periode seldjoucide en Asie Mineure, in V. L. Menage (ed.) La Formation de lEmpire Ottoman. London: Variorium Reprints. Wittek, Paul (1966) The Rise of the Ottoman Empire. London: Royal Asiatic Society of Great Britain and Ireland.

Downloaded from the.sagepub.com at City University Library on January 6, 2012