Читать книгу Between Two Worlds - Cemal Kafadar - Страница 9

Introduction

Background and Overview

Osman is to the Ottomans what Romulus is to the Romans: the eponymous founding figure of a remarkably successful political community in a land where he was not, according to the testimony of family chronicles, one of the indigenous people. And if the Roman state evolved from a peripheral area to represent the center of the Graeco-Roman civilization, whose realm it vigorously expanded, so the Ottoman state rose from a small chieftainship at the edges of the abode of Islam eventually to become the supreme power within a much enlarged Islamdom. Once they came to rule, the Ottomans, like the Romans, gained a reputation as better administrators and warriors, even if less subtle minds, than the former representatives of their civilizations; they possessed less taste for philosophical finesse perhaps but had greater success in creating and deploying technologies of power. The “Romanesque” quality of the Ottoman political traltion has been noted before and was expressed recently by an eminent scholar of the Islamic Middle East: “The Ottoman empire…was a new and unique creation, but in a sense it also marked the culmination of the whole history of Muslim political societies. The Ottoman Turks may be called the Romans of the Muslim world.”¹

They were indeed called just that when they, like various other peoples of medieval Asia Minor, were referred to as Rml, that is, those of the lands of (Eastern) Rome.² This was a primarily geographic appellation, indicating basically where those people lived, but it did not escape the attention of geographers and travelers that the Turco-Muslim populations of Rm, a frontier region from the point of view of the central lands of Islam, had their own peculiar ways that distinguished them from both the rest of the Muslim world and from other Turks. Namely, being a Rm Turk also implied belonging to a newly emerging regional configuration of Islamic civilization that was on the one hand developing its own habitus in a new land and on the other engaged in a competition to establish its political hegemony over a rival religio-civilizational orientation. The proto-Ottomans, of whom we know nothing with certainty before the turn of the fourteenth century, were a tiny and insignificant part of this new configuration at first but their descendants and followers eventually came to dominate it and to shape it toward the creation of a new imperial order under their rule.

According to most historical traditions, the immediate ancestors of Osman arrived in Anatolia with the second great wave of Turkish migrations from central Asia, which took place in the wake of the Chingisid onslaught in the early thirteenth century. Once in Anatolia, they would have encountered a variety of Turkish-speaking communities—some in urban centers, some settled down to agriculture, but the majority engaged in pastoral nomadism like Osman's ancestors, most but not all of them speaking the Ouz dialect, most but not all of them Muslim, and even then divided into communities that understood different things about being Muslim—living in a complex ethnoreligious mosaic that included Christian and non-Turkish-speaking Muslim communities (especially Arab, Kurdish, and Persian).

The earlier wave, the tail end of the Vlkerwanderungen in a way, had occurred in the eleventh century when large numbers of Turkish tribes, belonging primarily to the Ouz dialect group and to the Ouzid idiom of Inner Asian political discourse, crossed the Oxus and moved toward western Asia. While the Seljuk family from among these tribes soon became involved in politics at the highest levels in Baghdad and ended up as a dynasty that held the sultanate, many tribes moved further west and piled along the eastern borders of the Byzantine Empire. Their incursions into Asia Minor were independent of and at least occasionally contradictory to the will of the Seljuk sultanate.

The Byzantine Empire had faced a similar and at first more threatening pressure from a more southerly direction in the seventh century with the appearance of Arab-Muslim armies. While raids and counterraids continued to rage in the next few centuries, however, these were relatively localized in a fluid frontier zone that developed in southeastern Anatolia with its own borderland institutions, heroes, traditions, and lore. The Turkish-speaking settlers and conquerors of the later medieval era were to inherit a good deal of those traditions from both the Muslim and the Christian sides.

In any case, ongoing friction in eastern Anatolia in the eleventh century led to the fateful encounter of the Seljuk and Byzantine armies in Mantzikert in 1071, the same year that the Eastern Roman Empire lost Bari, its last possession in the Italian peninsula, to other tribal warrior bands led by the Normans. The Byzantine defeat at Mantzikert was to be followed by deeper and more frequent raids or plain migration by Trkmen tribes into Asia Minor. The political landscape of the peninsula started to change immediately and was not to fully stabilize for four centuries, until the Ottomans established unitary rule over it in the latter part of the fifteenth century.³ Before the end of the eleventh century, most of Anatolia was divided up among petty potentates led by Turkish warriors, Armenian princes, Byzantine commanders, and Frankish knights arriving with the First Crusade (1096-99). The political configuration of the peninsula kept changing through mostly short-lived successes of different adventurers who were ready to enter into all sorts of holy and unholy alliances with others who were not necessarily of the same religious or ethnic background. Many an aspiring warrior seems to have enjoyed, to paraphrase Andy Warhol, fifteen days to fifteen years of glory before he disappeared or was sucked into the sphere of influence of a momentarily mightier one. The Byzantine Empire still held the coastline and some connected areas inland, especially after aka Beg, who was based in the Aegean, was murdered (in 1093, with the help of the Seljuk ruler) and the crusaders recaptured Nicaea from the Seljuks of Rm for the empire (1097).

Among Turco-Muslims, who were largely restricted to the inner plateau after some very early excursions to the coastal areas and who were replenished by continued migrations of Trkmen tribes, two powers were able to acquire prominence and enjoy some longevity. An offshoot of the Seljuk family and the House of Dnimend competed for the ultimate leadership of the Muslims of Anatolia for nearly a century. Melik Dnimend, whose gests were to be woven into the epic cycle of Anatolian Muslims, and his family seem to have cared less for state building than for what they did better than anyone else for a while: namely, capturing towns and undertaking daring raids that brought them tremendous prestige. The Seljuks of Rm, on the other hand, were keen to emulate more stable and structured modes of governing; they were particularly successful in that task after establishing Konya (ancient Iconium) as their capital during the reign of Mes‘d I (r. 1118-55). The often violent competition between the Dnimendids and the Seljuks of Rm, both of whom sought the alliance of the Byzantine emperor or local Christian or Muslim powers when it seemed expedient, was ultimately resolved in 1177 in favor of the latter, who captured their rivals' last major holding, Malatya, and decisively reduced them to vassalage.

This feat was accomplished only one year after another Seljuk victory, this one over Byzantine imperial armies in Myriokephalon (1176). This was, in the words of one of the most prominent scholars of medieval Anatolia, “after an interval of a century, a replica of Mantzikert, which showed that henceforward there existed a Turkey which could never be further assimilated.”⁴ Although the word “Turchia” indeed appeared in Latin geographic designations in the twelfth century, from the point of view of the Turkish-speaking populations and polities of the area, there was no Turkey, either as a geographical or as a political entity, until the end of World War I, when the European designation was finally accepted by the locals themselves. Instead, there was a changing set of competing political enterprises, many of which were led by Turkish-speaking warrior elites but which were never organized along ethnic lines or with an eye to eventual ethnic unity. The land was known as the land of Rm, and its people were divided into different communities of religious, linguistic, or political affiliation. The Ottoman ruling class eventually emerged as a combination of Muslims (some by conversion) who spoke Turkish (though not necessarily as a native tongue), affiliated (some voluntarily and some involuntarily) with the dynastic state under the rule of the House of Osman. And “Turk” was only one, and not necessarily a favored one, of the “ethnicities” ruled by that class.

With their victories in Myriokephalon and Malatya behind them, the Seljuks looked like they had accomplished, “from the Byzantine territories in the West almost to the further limits of the East, the political unity of Asia Minor.”⁵ But to a student of the later and much more solid Ottoman state, like this author, the rule of the Seljuks of Rm in any period seems too fragile and ephemeral to be considered real political unity. All the major fault lines of those medieval Turkic states, built around the energies of tribal forces and ambitious warrior chieftains, were at work in the sultanate of the Anatolian Seljuks: there were many frontier zones of various sizes where the administrative apparatus hardly reached; there were many tribal groups that were not controlled; there were many ambitious warriors, some of them possibly made by the Seljuks, ready to imagine themselves independent of Seljuk authority; and when two or three of these came together, as they frequently did, they were able to shake, if not dissolve, state power. Finally, the Seljuks of Rm also continued the practice of dividing up their land among the heirs of the dynast; the same Seljuk sultan who won the two victories mentioned above carved his realm into eleven pieces for his nine sons, a brother, and a nephew. The realm could still remain united in principle, under the leadership of a “senior partner” recognized by the others, but it proved only a matter of time before some of the heirs found support among Trkmen tribes or warrior bands, and rival foci of power emerged. As we shall discuss in later chapters, the Ottomans, as if or perhaps because they were good students of history, and under different conditions no doubt, proved themselves much more successful in confronting these fault lines and eventually steering their course clear of them on the way to creating one of the most durable states in history.

In all fairness to the Anatolian Seljuks, it must be admitted that they were approaching a firm consolidation of their power in the first four decades of the thirteenth century, and that their ultimate failure is closely related to an unforeseen external factor: the invincible Mongol armies. Even before the Mongols, however, a Trkmen rebellion under the leadership of Baba Ilys and his followers presented a severe challenge to Seljuk authority between 1239 and 1241. It seems that the plight of the TIt seems that the plight of the Trkmen tribes was due, among other things, to the squeeze for land that arose with the second big wave of migrations, which is said, as was mentioned above, in most sources to have brought the tribe of Osman's grandparents into Anatolia.⁶ With the Seljuk defeat by the Mongol armies in Kseda (central Anatolia) in 1243, the tension-ridden pendulum of centripetal and centrifugal tendencies started to swing once again in favor of the latter. The political landscape was eventually, especially after the Mongols sent soldiers and horses to be fed in the name of establishing direct control over Anatolia (1277), thrown into turmoil with various forces waging a life-and-death struggle in a period of extreme violence and disarray. It is probably no coincidence that Ynus Emre, the classical poet of the newly forged Anatolian Turkish dialect, emerged in that context and produced a corpus of poems that are distinguished by the profundity with which they looked death right in the eye. In any case, continued political disarray and demographic pressure pushed many Turkish tribes and warriors further into western Anatolia, especially since the Byzantine capital was moved back to Constantinople in 1261 after having been seated in Nicaea since 1204 (the Fourth Crusade) and having brought heightened security and prosperity to the area for half a century or so. Before the end of the thirteenth century, endemic political fragmentation had led to the emergence of numerous small chiefdoms and relatively autonomous tribal domains in various parts of Anatolia.

The political turbulences and human catastrophes of the thirteenth century should not prevent us from observing the tremendous possibilities unleashed by an unprecedented “globalization” of the Eurasian economy thanks, in good part, to the Chingisid conquests and the pax mongolica. It is for good reason that it has become a commonplace to refer to the travels of Marco Polo when speaking of the Chingisids. There were signs even before Chingis that Asia Minor, once the jewel in Byzantium's crown and then having suffered a series of depredations, had regained sufficient stability to serve as a long-distance trade link (along a North-South as well as an East-West axis) and to benefit from the new commercial potential created by the mixed economies of urban, agrarian, and pastoralist populations. The outburst of caravanserai building activity, the primary area of architectural patronage by the Seljuk elite, was initiated in the late twelfth century and was to increase its tempo no matter what the nature of political turbulences. The early thirteenth-century acquisition of the port towns of Sinop (by the Black Sea) and Alanya (by the Mediterranean) brought the Seljuk system and the Turco-Muslim-dominated economies of Anatolia that it controlled at the time into direct touch with the Levantine sea trade. By the end of that century, more than one hundred caravanserais in the peninsula provided lodging and protection to merchants (and other travellers)⁷

It is revealing, for instance, that what is no more than a remote backwater in modern Turkey, an obscure plateau between Kayseri and Mara once entertained a lively international fair where Middle Eastern, Asian, and European merchants exchanged commodities like silk textiles, furs, and horses. True, the fair does not seem to have survived long into the era of Mongol Ilkhanid direct rule, but trade is not known to have suffered in general. The chiefdoms that emerged in western Anatolia, where even Mongol power could hardly reach, to some extent built their power on raids and pillaging, but the western Anatolian coastline was integrated into a brisk Levantine trade around 1300, and the chiefs were signing commercial treaties with the likes of Venice in the early fourteenth century.⁸ In fact, fragmentation and the emergence of small local powers may well have increased the possibility for a more local redistribution of resources that would otherwise have been siphoned off to distant imperial capitals.

One of those small chieftains, situated in the northwest in what was still partly Byzantine Bithynia, belonged to the clan of a certain Osman. He belonged to an exceptional generation (or two) of creative minds and social organizers who, either personally through their deeds or through their legacy as it was constructed and acted upon by followers, became the pivotal figures, the magnets, around whom the vibrant yet chaotic social and cultural energies of the Turco-Muslims of medieval Asia Minor ultimately found more-regular paths to flow. Since then, these figures, as embodied in the rich lore that has been built around them (whatever the relationship of such lore to their “real” or “historical” life), have represented the “classics” of western, or one might also say Roman, Turkish culture. Mevln Celleddn Rm and Hc Bekta Vel, for example, are the spiritual sources, respectively, of the two largest and most influential dervish orders in Ottoman lands. But their influence reaches far beyond any particular set of institutional arrangements, however large these may have been, and cuts across orders, social classes, and formal institutions. They have rather been fountainheads of broad cultural currents and sensibilities over the centuries. Ynus Emre's appeal has been even more ecumenical, with his poetry considered by successive generations to be the most moving and unadulterated expression of piety in Anatolian Turkish; dozens of imitators tried to pass their own works off as those of Ynus, and dozens of villages claimed to have his shrine. A certain Nareddn of thirteenth-century Anatolia seems to have been at least the excuse for the creation of the lore of Nasreddin Hoca, the central figure of a corpus of proverbial jokes that now circulate, with many later embellishments of course, from the Balkans to central Asia. Ah Evren may be the least well known of these figures in the modern era, but his cult once played the most central role in the now defunct corporations of artisans and tradesmen, providing the basic structures and moral codes of urban economic and social life, at least for Muslims. Widely popular legends of a certain an alu, who also is honored at numerous burial sites, portrayed him as the most pivotal character in spreading Islam in the Balkans.

It is much more coincidental but still worthy of note that Osman was also a near contemporary of two figures, very remote from his sphere of action in thirteenth-century terms, whose descendants were to share with his house the limelight of international politics in the sixteenth century. One of these was Rudolf of Habsburg, who acquired his Erblande in 1278. The Habsburg dynasty was to become the main competitor of the Ottomans on the central European and Mediterranean scenes, and the two states were to follow more or less the same rhythms until both disappeared as ruling houses in the aftermath of the First World War. And the other relevant member of the cohort, a much less likely candidate at that time for inclusion in our comparison here, was Sheikh af al-Dn (1252-1334) of Ardabil, whose own political role as a renowned Sufi may have been considerable but was primarily indirect. His legacy and the huge following of his order would eventually be shaped by late-fifteenth-century scions into the building blocks of the Safavid Empire, the main competitor of the Ottomans in the Muslim world in the early modern era.

Osman, a near contemporary of theirs, is the founder of a polity that rose over and above all its Anatolian and Balkan rivals to be eventually recognized, whether willingly or reluctantly, as the ultimate resolution of the political instability that beset Eastern Roman lands since the arrival of Turkish tribes in the eleventh century. He is a much more historical (i.e., much less legendary) character than Romulus of course. Nevertheless, he is equally emblematic of the polity that was created after his name and legacy. As Marshall Sahlins points out in his study of the stranger-king motif in Hawaiian and Indo-European political imagination, “it is not significant that the exploit may be ‘merely symbolic’ since it is symbolic even when it is ‘real.’ ”⁹

One of the most influential legends concerning Osman is the one that depicts his whole conquering and state-building enterprise as having started with an auspicious dream. Variants of this legend were retold in dozens of sources until the modern era, when the dream was dismissed in terms of its historicity but still, by some vengeful intervention of ancestral spirits perhaps, did not fail to occupy a central place in much of the debate among historians, as we shall see in the next chapter. According to one of the better-known versions, Osman was a guest in the home of a respected and well-to-do Sufi sheikh when he dreamt that

a moon arose from the holy man's breast and came to sink in Osman Ghazi's breast. A tree then sprouted from his navel, and its shade compassed the world. Beneath this shade there were mountains, and streams flowed forth from the foot of each mountain. Some people drank from these running waters, others watered gardens, while yet others caused fountains to flow. [When Osman awoke] he went and told the story to the sheykh, who said, “Osman, my son, congratulations for the imperial office [bestowed by God] to you and your descendants, and my daughter Mlhn shall be your wife.” He married them forthwith.¹⁰

Thereafter, it is a story of success that culminated in the phenomenal expansion of the territories controlled by the House of Osman. In the early twentieth century, after various parts of the empire had been gobbled up or seceded throughout the eighteenth and nineteenth centuries, Ottoman forces were still defending what they considered to be their own territory in as diverse parts of the world as Macedonia, Libya, Yemen, and the Caucasus.

Phenomenal as it was, however, the Ottoman expansion was slow when compared to the empire-building conquests of some other Inner Asian/Turco-Mongol tribal formations, such as those led by Chingis and Timur, or when compared to the swift rise of the House of Seljuk to the sultanate in Baghdad, the fabled seat of the Islamic caliphate. That is probably why it was so much more durable. Relatively speaking, the Ottomans took their time building their state and it paid off. They took their time in constructing a coalition of forces and reconstructing it as it changed shape, while they were also keen on institutionalizing their political apparatus. It was a gradual and conflictual process of state building that took more than a century and a half from Osman's earliest ventures to the conquest of the Byzantine capital by his great-great-great-great-grandson, Memed II (r. 1451-81), when the Ottomans can finally be said to have graduated to an imperial stage.

When Memed the Conqueror visited Troy later in his reign as sultan, khan, and caesar, he seems to have been aware of the explanation of Ottoman successes by the theory, upheld by some in Europe, that Turks were, like the Romans before them, vengeful Trojans paying back the Greeks.¹¹ Standing at the fabled site, the sultan is reported to have inquired about “Achilles and Ajax and the rest” and then, “shaking his head a little,” to have said: “It was the Greeks and Macedonians and Thesalians and Peloponnesians who ravaged this place in the past, and whose descendants have now through my efforts paid the right penalty, after a long period of years, for their injustice to us Asiatics at that time and so often in subsequent times.”¹²

HISTORIOGRAPHY

Modern historiography, of course, has had little patience with dreams and legends as explanation. Still, both the dream story mentioned above and, even more so, the “true origins” of the proto-Ottomans have functioned as pivotal issues in twentieth-century discussions of Ottoman state building.

The first study devoted to the rise of the Ottoman state, published in 1916 by H. A. Gibbons, held that that successful enterprise could not have been built by “Asiatics.” The dream story, contended Gibbons, though not to be taken at face value, implied that Osman and his tribe, who must have been pagan nomads of Inner Asian background, converted to Islam at some point and set out to Islamize their Christian neighbors in Byzantine Bithynia. Converts among the latter made up the majority of the proto-Ottomans and provided the expertise for setting up an administration.

In the charged environment of the early twentieth century, where nationalism was linked with racialism even more explicitly than it is today, this argument was obviously loaded. The emerging Turkish nationalism of the republican era (1923-), busily occupied with redefining the role of Turks in world history, was not entirely sympathetic to the later and “corrupt” phase of the Ottoman Empire that the Republic replaced; however, the same nationalists could not but proudly appropriate the earlier history of invasions, settlement, and state building, including the most successful case, represented by the Ottomans, that established the Turkish presence in the region.

In the formulation of M. F. Kprl, the leading Turkish historian of his generation, who elaborated his views in the 1930s and framed them partly as a response to Gibbons, the military-political expansion of medieval Anatolian frontiers was primarily due to the demographic pressure of Turkish tribes fleeing the Chingisid armies. According to Kprl, Osman's immediate entourage consisted of members of his tribe, who must have been of common descent as later Ottoman rhetoric claimed. As they set out to carve themselves a body politic, their numbers were replenished, on the one hand, by other Turkish elements of the same region and, on the other, by experienced representatives of the hinterland's sophisticated Turco-Muslim political-administrative culture. A number of conversions took place, but the Ottoman state was essentially a Turkish state; it was built by Turks, and almost all elements of Ottoman political culture can be explained by reference to their Turco-Muslim heritage deriving from central Asia and the Middle East. Tribal and ethnic cohesion as well as a sophisticated institutional legacy enabled the building of a state out of demographic pressure in a relative political void. Kprl's vision was hailed and continues to serve as a building block of Turkish national historiography.

It was Paul Wittek's theory, however, formulated in the 1930s and partly as a response to Kprl, that was to gain international recognition as the most convincing account of Ottoman success. Wittek found the rhetoric of tribalism unconvincing; nor did he dwell on the question of ethnicity as such, though he underlined continuities in Turco-Muslim culture without failing to note the occurrence of conversions and Christian-Muslim cooperation in some passages. For him, what fueled the energies of the early Ottoman conquerors was essentially their commitment to gaza, an “ideology of Holy War” in the name of Islam. Ottoman power was built on that commitment, as expressed in an inscription erected in Bursa in 1337, which referred to Osman's son as a “gazi, son of gazi.” Wittek found the same spirit in the earliest Ottoman histories, none of which, however, dates from before the fifteenth century. While a shared ethos provided warriors who banded together with cohesion and drive, the realm they brought under their control was organized according to the administrative experiences of scholars and bureaucrats from the Islamic cultural centers. There was some tension between the two elements because the gazis belonged to a heterodox frontier culture; over time, orthodoxy prevailed as the Ottomans established a stable administration. Wittek's formulation, which has much in common with Köprülü's but avoids the ethnicist controversy and seems to place singular emphasis on religious motivation as the root cause of Ottoman power, was widely accepted and recycled in the pithy formula of “the gaza thesis.”

Lawrence Stone, historian of early modern England, has described the fate of historical theses in terms of a tongue-in-cheek quasi-Hegelian spiral of generational cycles.¹³ The dominant view of one generation is turned on its head by the next but is then reclaimed, hopefully in an improved version, by the next. It may be due to the ambiguous nature of “generation” as an analytical concept or the backwardness of Ottoman studies that the beat seems to have skipped a generation or two with respect to the gaza thesis, which was formulated in the 1930s. Alternative or supplementary explanations were occasionally aired in more-general studies, but there was no lively debate producing new research and ideas—until, that is, the 1980s, when many voices were raised, independently of each other, against the Wittek thesis.

The main tenor of those voices reflected a dissatisfaction with an explanation that put so much emphasis on “Holy War ideology” when early Ottoman behavior, it was claimed, displayed heterodoxy vis-à-vis Islam and accommodation vis-à-vis Christian neighbors. The early Ottomans could not have been driven by the spirit of gaza, because they were neither good orthodox Muslims nor zealous exclusivist ones. Ottoman sources that speak of gaza might as well be read as representatives of a later ideology, addressing Islamized audiences by putting a respectable religious veneer on earlier actions that had been driven by pragmatic considerations such as plunder and power.

No matter how radical a departure they claimed to represent from the consensus of the former generations, however, all of these critical voices still subscribed to the essentialism of earlier historiography. Its legacy, in other words, has proven sufficiently powerful even in the hands of critics to perpetuate a dichotomous analysis that wishes to see the early Ottoman conquerors and state builders as fundamentally Turkish, tribal, and driven by pragmatism and plunder or fundamentally Muslim and driven by zeal for holy war.

It is argued in this book that early Ottoman history and the statebuilding process cannot be properly appreciated within the framework of such dichotomous analyses. While the identities, beliefs, values, and actions of the early Ottomans are naturally bound to constitute the basic material for analysis and explanation, they do not need to be framed in terms of ahistorical either /or propositions. Human beings display many complex and even contradictory behaviors, and it is in that very complexity that explanations for historical phenomena must be sought. To be more specific, it is argued that the recent debate over the normative “Muslimness” of the gazis obscures the historical reality of the distinctive culture and ethos of the march environment within which the Ottoman state was born. Beyond reassessing the historiography, it is my aim to reconstruct that distinctive ethos as well as the social and political environment of the marches in late medieval Anatolia in order to reach a better understanding of the rise of the Ottoman state.

PLAN AND APPROACH

This book analyzes its problem and elaborates its perspective in three layers. Chapter 1 is a discussion of modern scholarship on the rise of the Ottoman state. It introduces, in a much more detailed canvas than the sketch given above, the specific issues that have been raised and the main perspectives developed with respect to that particular theme. It is not a survey of scholarship on medieval Anatolia, the way Norman Cantor, for instance, has recently examined the history of medieval European studies as a field.¹⁴ Rather, this chapter is only a narrowly focused treatment of history-writing on the problem of the Ottoman state's emergence. It maps the wiring, as it were, of modern historiography on that problem in order to highlight the currents of tension and the nodes that are charged; any new conceptualization or reconstruction, including that of this author, will need to be assessed in terms of such a map.

Chapter 2 presents first a survey and analysis of the sources emanating from the Turco-Muslim frontier milieux of Anatolia, legendary accounts of the lives of warriors and dervishes, in order to illuminate how the people of the frontiers conceptualized their own actions and assigned meaning to them. This is the first attempt to reach a historicized understanding of the complex of values and attitudes embodied in or related to the notion of gaza, which both Wittek and his critics were more or less content to treat in terms of its dictionary definitions. This first part of the chapter demonstrates that the frontier ethos was intricately bound up with the gaza spirit, ubiquitous in the relevant sources, but nonetheless incorporated latitudinarianism and inclusiveness.

The second part of chapter 2 turns to a close reading and comparison of certain passages in a particularly relevant body of interrelated sources: the chronicles of the House of Osman, which were at least partly based on earlier oral narratives but were not rendered into writing before the fifteenth century, the more substantive compositions not emerging until the latter decades of that century. While enmeshed in frontier legends and myths, these sources at the same time present themselves as straight histories. As such, they have by and large suffered from either an uncritical adoption as factual accounts—a naive empiricism—or a nearly wholesale dismissal as myths—a hyperempiricism. The latter attitude or an outright neglect has severely limited the use of relevant hagiographical works that also develop their own historical arguments, in terms of the parameters of that genre of course, with respect to the early Ottomans.

Methodologically, my discussion is an attempt to transcend the positivistic attitude, still dominant in Ottoman studies, that every bit of information in the sources can and must be categorized as either pure fact or fiction. More specifically, my reading of the sources reveals that representatives of different political tendencies tried to appropriate the symbolic capital embedded in claims to success as gazis in their own ways through differing historical accounts. It is established through this discussion that the pertinent hagiographies and “anonymous” calendars and chronicles are far from being the inert products of accretion of oral tradition or chance coagulation of narrative fragments but rather represent internally coherent ideological positions articulated by authorial or editorial hands. By drawing out these several historiographical strands embedded in variants, the chapter enables an understanding of the gazi milieu as a social and cultural reality that sustained political and ideological debate. It identifies the major point of tension in the early Ottoman polity between centralizing and centrifugal tendencies, which shaped its trajectory until the conquest of Constantinople, when the triumph of centralized absolutism was sealed.

Chapter 3 proceeds to the old-fashioned task of reconstruction and is intended to be neither exhaustive nor definitive. It deals selectively with aspects of the process whereby the political enterprise headed by a certain Osman in the western Anatolian marches of the late thirteenth century was shaped into a centralized state under the House of Osman in a few generations. The discussion here moves from the gaza ethos to the gazis and other social agents in that scene and aims to re-present the preimperial Ottoman polity as the historically contingent product of a culturally complex, socially differentiated, and politically competitive environment rather than as the necessary result of a unitary line of developmental logic. It focuses on the sociopolitical plane, with particular emphasis on locating gazi warriors and dervishes, as well as their neighbors—tribal or settled, Christian or Muslim, rural or urban—within a matrix of shifting alliances and conflicts in late medieval Anatolia.

Insofar as it is a narrative of early Ottoman history, it is a highly selective treatment, intended only to highlight the process of coalition formation and dissolution and some of the most significant steps in the institutionalization of Ottoman power along a contested path that succeeded in circumventing the fault lines of medieval Turco-Muslim polities mentioned above.¹⁵ A brief overview is provided here for the reader who may need an introduction to the orientation of the author and to the discussion of specifics before the third chapter. It might also be worthwhile to consult the chronology of events (pp. xvii-xix) after reading this introduction.

OVERVIEW

The scene is set in terms of the political wilderness and competition that characterized western Anatolia at the end of the thirteenth century. Byzantine, Mongol-Ilkhanid, and Seljuk powers still had some control over the region, but a number of chieftains or community leaders engaged relatively freely in acts that would determine their political future. A Turco-Muslim leader with a following and a recognized realm was called a beg or emir, and his competitive, expansion-oriented enterprise was called a beglik or emirate. Some local Christian lords, called tekvur in Turkish sources, controlled fortified or naturally protected settlements and surrounding agricultural areas that constituted raiding territory for the forces of the begs. This frontier environment also witnessed a high degree of symbiosis, physical mobility, and religious conversions that facilitated the sharing of lore (legends about earlier heroes, for instance), ideas, institutional practices, and even warriors among inclusive political formations that were at the same time steeped in the ethos of championship of the faith.

Like his competitors, Osman Beg not only undertook raids with forces under his command and carried off booty (mostly slaves and precious objects) but also constructed a set of alliances with some of his neighbors with an eye to increasing his sphere of influence. Bonds of solidarity would be formed in joint raids or through neighborly relations that included trade and intermarriage. One of Osman's wives was the daughter of a rich and respected sheikh of a dervish community; one of Osman's sons married the daughter of a tekvur. The chieftain of a Christian village near Osman's base was a scout and an ally in some early expeditions. It cannot be imagined that other begs of the frontiers failed to appreciate the value of such ties and to forge similar alliances. Cliental relations with wonder-working dervishes, who indeed worked wonders in capturing the hearts and minds of tribesfolk as well as of Christian or ex-Christian peasants through syncretism, were not the monopoly of the Ottomans. Nor were the Ottomans the only ones who could claim to be “raiders in the name of the faith.” Moreover, a policy of fiscal leniency (relative to late Byzantine practices), which worked well toward gaining Christian producers as subjects, was followed not only by the Ottomans but also by their rivals. They all benefited from the military potential in the restless energies and martial skills of the nomads and adventurers who had been “going West” in search of pastureland and other opportunities since around the mid thirteenth century.

Though it is difficult to assess whether the Ottomans had any comparative advantages in the use of any of these means toward expanding their sphere of control, it must at least be noted that Osman and his followers made effective use of them. Osman's political career seems to have started during the last years of the thirteenth century and to have carried him from the leadership of a community of nomadic pastoralists to the chieftainship of a beglik after he seized a few Bithynian fortresses. One particular advantage of his beglik was its location, since this base provided its forces with relatively easy access to poorly defended Byzantine territory. Successful military expeditions brought fame and riches, which were essential in attracting more warriors and dervishes as well as scholar-bureaucrats from the centers of Islamic culture. By the time of Osman's death (1323 or 1324), his small polity had the material and organizational means to strike coins, issue endowment deeds, and use siege tactics that required much more than competency in nomadic warfare. Particularly with the conquests of Bursa (1326) and Iznik (Nicaea) (1331) under Orn, Osman's son, the Ottomans controlled all the major towns of Bithynia and were in a position to build lasting institutions. Some recognized scholars, such as andarli ara all (d. 1387), seem to have arrived in Ottoman lands around this time, occupied top administrative positions (judicial and vezirial), and initiated new institutional mechanisms. The first Ottoman medrese, or college, was established in Iznik in 13 31 and started to train scholar-scribes and judges. But the Ottomans were not the only ones who undertook lucrative expeditions, and at first they were not the most renowned of the emirates. Even their location right next to Byzantine territory to the southeast of the Marmara Sea was matched by that of the Karasi emirs, who controlled the southwestern half of the region. Insofar as the extension of gazi activity into southeastern Europe constituted the next significant step in the imaginations of the begs of western Anatolia, the Karasi were in fact more favored in terms of their location and more knowledgeable in military-strategic terms. Factional strife in Byzantine imperial politics in the 1340s invited both emirates to Thracian ventures, where the Ottomans and the Karasi were called to aid the factions, respectively, of Kantakouzenos and his rivals (including Batatzes, the father-in-law of a Karasi beg). The former won, and the Ottomans overran the Karasi emirate and incorporated its experienced gazis. By 1354, Gelibolu (Kallipolis), the strongest Byzantine fort across the Dardanelles, was captured, and the gazis could now hope that their engagements in Thrace would be more than temporary. The goal of permanent control in this new territory was buttressed by the colonization of kindred populations there.

In taking charge of the raiding and settlement in Thrace, which eventually paved the way for conquests in southeastern Europe, the Ottomans gained a decisive advantage over other emirates, since this role brought not only immense prestige but also access to substantial material resources in the form of booty and tax revenues. But they soon faced the quintessential Ibn Khaldunian predicament of tribal war-band leadersturned-state builders: namely, the loosening of the bonds of solidarity among members of the war band as the administrative mechanisms and stately pomp of imperial polities are adopted by the leaders of the successful enterprise. In other words, as the House of Osman was being transformed into a dynasty at the head of an emerging administrative network of controls, the relatively egalitarian community of gazi commanders was giving way to a widening hierarchical space between central power and subordinate begs; not all of the latter were content with this role.

The disruption of communications between Asia Minor and Thrace for a decade after 1366, when Orn's son Murd I (1362-89) lost control of Gelibolu, encouraged some of the gazi warlords in Thrace to imagine themselves at the head of autonomous enterprises, especially because they could claim that many of the Thracian conquests were their own achievements. Many former states in the Muslim world, as Ibn Khaldn observed, had begun a process of disintegration at a similar stage; the nature of the challenge might differ somewhat from case to case, but the basic problem was one of dissolving cohesiveness. The Ottomans rose to the challenge, however, not only by eliminating the challengers after Gelibolu was recaptured ca. 1377 by also by creating an institution of artificial kinship, the Janissary standing army, that functioned as an extension of the royal household. This institution was the centerpiece of what hereafter would be a self-consciously centralizing administrative apparatus under “sultans” from the Ottoman family.

The goal of the sultanate thereafter was to enlarge its territory on the one hand and to control fissiparous dynamics on the other. Deftly making use of divisions among feudal polities, either by carrying out fierce raids or by gaining the loyalty of local populations through fiscal concessions and/or religious propaganda and of some of the local lords through incorporation, the Ottomans rapidly extended their power in the Balkans in the final decades of the fourteenth century. Of their major rivals, the Serbian kingdom was reduced to vassalage after the Battle of Kosovo (1389), and the Bulgarian one eliminated by 1394. Keeping almost a geometric sense of centralism, the Ottomans pursued a symmetrical expansion in Anatolia and reduced some of the weaker begs—first to unequal partnership, then to vassalage, and then to incorporation as removable appointees—just as they had done to many early Bithynian allies, gazi commanders, and Balkan local lords. It should be added that the forcefulness of Ottoman expansionism was also due to their extrapolitical logic in targeting important routes of commerce and sites of production.

In both the subjugation of gazi emirates and the building of bureaucratic mechanisms to buttress central government control over resources, Byezd I (r. 1389-1402) was widely perceived, especially among gazi circles and the dervishes close to them, to have gone too far and to have relied too heavily on the help of scholar-bureaucrats like members of the andarh family. The latter were in fact seen as the archenemies of a nostalgically reconstructed frontier ethos devoid of all imperial trappings. The last great conquering army of Inner Asia, led by Timur (Tamerlane), was drawn into Anatolia through the pleas of the leading families of subdued emirates. In the ensuing Battle of Ankara (1402), Byezd was routed when his Anatolian Muslim vassals defected to the other side and his gazi commanders abandoned the battlefield, where the sultan was left with only his Janissaries, still a small force, and Balkan Christian vassals.

The Timurid debacle did not end the Ottoman empire-building project, especially since Timur soon left Asia Minor to pursue higher ambitions as a conqueror in Asia itself, but only led to temporary confusion until the whole realm was reunited under one heir after eleven years of internecine strife among Byezd's sons. The gazi warlords, most of whom were based in the Balkans, were no longer powerful enough to challenge the House of Osman, but they did play a major role in the unfolding and outcome of events as the different princes negotiated for their support. For a few decades after the Interregnum, under Memed I (r. 1413-21) and Murd II (r. 1421-44, 1446-51), state-controlled expansionism and raider activity coexisted with relatively little tension, partly due to pragmatism on both sides and partly due to the great success of the expeditions in enlarging the pie of redistribution. The new modus vivendi with the gazi warlords (and with formerly independent begs of gazi emirates who had been rendered Ottoman appointees) entailed their subordination but did not totally undermine their ability to take independent action or their access to booty and glory. Centralizing and fiscal policies were not abandoned, but neither were they pursued as aggressively as under Byezd I.

Starting around 1440, strains emerged again between the level of Ottoman centralization and the empire-building project. A number of Hungarian-led “anti-Turkish” leagues undertook incursions that made the Ottomans feel on the brink of losing the Balkans; internal factionalism rose to the fore between a “war party,” headed by gazi warlords, and a “peace party” headed by representatives of a central administration that was still led by the andarli family. With the conquest of Constantinople, however, the young sultan Memed II (r. 1444-46,1451-81), who rode on the tremendous prestige of that feat, eliminated the leaders of both factions as well as the significance of both types of forces in Ottoman politics. A newly conceived imperial project was set in motion that spelled the ultimate victory, in terms of the internal dynamics of Ottoman state building, of the centralist vision.

Identity and Influence in the History of Nations

The larger story of medieval Asia Minor, within which the early history of the Ottomans would have been but one of many analogous episodes had they not turned out to be the ones who wrote the concluding chapter, had its counterpart in Iberia. From the eleventh to the fifteenth century, in the two peninsulas at the two ends of the Mediterranean, there raged a long series of confrontations that were fought between people who considered themselves, or found their means of legitimation as, representatives of their respective religio-civilizational orientations, Islam and Christianity.

It should be pointed out that this grand clash of two world religions did not determine each and every action of each and every actor on this scene. Nor were Muslims and Christians constantly engaged, in their actions or thoughts, in a struggle against each other. Coexistence and symbiosis were possible and probably more common. Besides, even these provisos set the scene in terms of a match between two teams, that is, in terms of two clearly designated different people who lived either at peace or at war with each other. This overlooks the fact that many individuals or groups changed sides and identities. Through conversion or enslavement, one could over time “become a Turk,” within limits set by social and ideological structures, as in the case of “becoming an American.” Furthermore, the sides were at any given moment divided within themselves into hostile camps or polities that did not think twice about establishing alliances with camps or polities from the other side.

Nonetheless, against these complex and shifting loyalties, a larger pattern over those four centuries can be reconstructed as a competition for political hegemony between powers that saw themselves as members of different religio-civilizational orientations. The periods of fragmentation in the political life of both peninsulas in the medieval era are in fact referred to by the same term, mulk al-tavif,16 by Muslim historians (Spanish, taifas), who clearly saw the structural affinities. In Asia Minor, the Muslims, whose own competition was won by the Ottomans, ended that struggle with the conquest of the last remnants of sovereign Christian power, Constantinople (1453) and Trebizond (1461), only slightly before the Catholic king of Spain captured the last stronghold of Muslim power in Iberia, namely, Granada (1492). The two victors, the Ottoman Empire and Spain (after passing into the hands of the Habsburgs), were soon to lock horns over what they would deem world supremacy, basically the old Roman world and its suburbs. Perhaps it was part of the same synchrony that Sleymn the Magnificent abandoned the use of luxury items and adopted a more pious orthodox image for the sultanate at around the same time that Charles V withdrew into a monastery. Maybe the emperors were just feeling the signs of strain that would become much more obvious in the public consciousness of their empires around the end of the sixteenth century when a decline and reform discourse appeared in both.

Like the Anatolian case, the Iberian one has been an ideological quagmire of modern historiography. The best example of this can be seen in the trials and tribulations of Américo Castro (1885-1972), the Spanish historian, who had to suffer disfavor and unpopularity in his homeland for writing of the Arabo-Muslim past of the peninsula as part of the Spanish heritage.¹⁷ This is not the place, nor the author, to discuss the nature of the Spanish heritage, but it is impossible to overlook the parallels in the historiographies that have been forged in the age of national consciousness and nation-state building. In that context, the meaning of medieval Muslim invasions has been a particularly problematic one to deal with among many nations of Eurasia.¹⁸ It does not seem so farfetched to imagine that long periods of coexistence, with both warlike and peaceful encounters (eight centuries in Spain, four centuries of instability in Anatolia followed by five centuries under relatively stable Ottoman rule), would shape peoples and cultures on either side in profound ways, and not just in the sense of developing a sense of enmity or in the mechanical sense of “influences“ as cultural commodities taken from one side to the other. Taking one's commingling with the “other”seriously in the historical reconstruction of heritages, however, seems to demand too much of national historiographies.

National historiographies (indeed modern historiography in general, to the extent that it functions as the history of nations) have tended to assume more or less sealed cultural identities of peoples (Turks, Greeks, Spaniards, Arabs, etc.) who have come into contact within the framework of a larger bipolar division of equally sealed civilizational identities (East/West, Muslim/Christian, etc.). Spain was Spanish before the Muslims and reverted to its “Spanishness” after expelling the invaders (after eight centuries!). Greece, Bulgaria, or any other post-Ottoman state did the same in terms of national history-writing, though the situation is not always as clear-cut as in Iberia, where systematic expulsions and forced conversions followed the reconquest. In the Balkans, as in India, it remains complex because of the intertwined existence of peoples who identify with different layers of a given country's past; so long as each layer remains exclusive of other layers in historical consciousness, such identification also has the potential of turning to exclusivism (“cleansing,” to invoke the most notorious recent example) in reality. In terms of history-writing, all this implies that historicizing the identities of those peoples, and thus underlining their plasticity and multiplicity over time, is taken, as was the case for Castro with respect to “Spanish-ness” as questioning the essence of nationhood. That, I would argue, is precisely what needs to be done in understanding the Turkish invasions of and migrations into Asia Minor and in reconstructing the formation of the Ottoman state.

Most current historiography, however, tends to operate on the basis of a “lid model” whereby at least some empires (the oriental ones?) are conceived as lids closing upon a set of ingredients (peoples) that are kept under but intact until the lid is toppled and those peoples, unchanged (unspoilt, as nationalists would like to see it), simply reenter the grand flow of history as what they once were. They may have experienced changes in terms of numbers and material realities but not in essence. Readers may also be familiar with this view from the recent example of Soviet dissolution, which was widely analyzed in terms of history beginning again for the peoples of the former USSR. But can one see the expression of Kirghiz or Belarus national identities, for instance, in terms of a reassertion? Were they not constructed to a large degree, in terms of identifying with a particularly delineated territory as homeland, for instance, during the Soviet era, which was a formative historical experience for all of them?

A recent publication that appeared in the most authoritative encyclopedia in the field of oriental studies takes us closer to our specific subject matter. In the lead essay to the entry on “Othmanli” (= Ottoman), “the subject peoples of the Balkans” are described as “for centuries peoples without history” until the nineteenth century.¹⁹ Where, in this depiction, could a historian fit the Muslim Slavs and Albanians, for instance? Or, how does one deal with the movement, under Ottoman rule, of Orthodox Slavs to areas now contested in Bosnia?

The Ottoman state/identity was not a lid that closed upon already formed national identities (of Arabs, Bulgarians, Turks, etc.) only to be toppled after a few centuries when those identities reasserted themselves. Some of these identities were formed to some extent, but they were reshaped (some might say, de-formed) under the aegis of, through the structures of, in response or reaction to, the Ottoman Empire. This is not a question of Ottoman influence but of a long and formative historical experience that shaped various communities and peoples under Ottoman rule through their interaction with each other as well as with peoples and ideas from neighboring civilizations. So the establishment of Ottoman rule in southwestern Asia and southeastern Europe, even if one sees it in black-and-white terms—namely, as either a yoke or a blessing—did mean much more than a lapse in what would otherwise have been the natural flow of the history of a given set of nations. Ottoman rule is part of the history of various communities, some of whom were able (and some unable) to shape and imagine themselves into a nation in the modern era thanks to a “historical consciousness” of their own (real or imagined) pre-Ottoman identity on the one hand and to that long and formative historical experience mentioned above on the other.

Specific issues of policymaking often do not require recourse to a historical argument, but the deeper matrix of orientations in which policy is made is inextricably linked with issues of national, that is, historical, identity in the political culture of nation-states. “Who we are,” at least in political discourse, is taken to be a major determinant of “how we should act.” Such linkages ultimately bring history-writing to the political sphere since national identity is defined and redefined through a historical discourse. “Who we are” is a culmination of “who we have been.” While this is valid for all nation-states to some degree or other, it is particularly pointed in some which have not resolved their identity questions as successfully as others; and this is valid for most, if not all, of the Ottoman successor states.

This is not to say that any such resolution is ever final; I do not mean that France, for instance, has defined its national identity in a decisive manner, that its historiography has comfortably removed itself from the sphere of politics, and that younger nation-states will eventually do the same. There is nothing to warrant such optimism. There are similar problems in the “mature” nation-states as well, and we are constantly reminded by the reactions in Europe and North America to the growth of the numbers of Muslims or blacks or freedom-or opportunity-seeking refugees from the Third and Second Worlds that these problems can easily become more acute in the West. The assertion of regional identities across Europe is another reminder of the fact that the relative homogeneity of modern European nation-states, which arguably served as models in much of world politics and historiography, disguises a multilayered history. It now seems that that homogeneity is in part a cultural construction, built through not only historical exigencies and certain forms of exclusivism but also a linear narrative of the story of “our true nation, one united people across time.”²⁰

There is no doubt, however, that the question of identity is particularly acute in Ottoman successor states, including Turkey, a relatively young nation-state, the historiographical (and thus also political) discourse of which has been the major ingredient of Ottoman studies in this century. The thrust of political ideologies runs just as deep in the historical consciousness of the other post-Ottoman nation-states of the Balkans and the Middle East. But the Turkish case is the more significant one for our purposes because of its obvious, but not necessarily to-be-taken-for-granted, centrality in Ottoman studies.

Three major issues still make up the underlying currents of tension in different national interpretations of Ottoman state building, though versions of the near consensus reached among Turkish historians in the first half of this century are accepted by international scholarship at large.²¹These issues are not necessarily discussed any more—at any rate they do not inspire many original research projects—but one can still feel the tension generated by differences of opinion on them. First is the rather racially conceived question of numbers: how many Turks came to Asia Minor, how many Anatolian Christians converted, what is the ratio of “real” Turks to converts in the composition of the later “Turkish” society under the Ottomans? Although there was heated debate on this question in the earlier part of this century, as we shall see in the next chapter, it was resolved in favor of “real” Turks.²² National historiographic discourses could hardly accommodate a different answer, whether one considers the Ottomans to be one's own or one's enemy. Thus, a nationalist Turkish and a nationalist Greek historian might easily agree that the Ottoman state was built by Turks, while the ethnic origins of a particularly favorable character such as an artist or a “good vezir” may be disputed.

Second is the issue of dislocation and violence caused by the migrations and the invasions; like the next question, it is one of respectability. There is a tendency on the one hand to portray them as sheer violence and, on the other, to see the migratory process as rather pacific. Here, most of the Turkish historians took the position that such disruption was minimal, while the nationalisms that defined themselves as liberation movements from the yoke of an alien people tended to emphasize the violence in the process that led to the rise of Ottoman power.²³ It is not only the desire to cater to the pacific values of the modern world that compels historians to magnify the level of violent power displayed by the other side or to reduce that displayed by their own. Some macho bravado is also involved here. Namely, it is also to portray defeat as due to the numerousness and violence of the enemy's military forces or to protect victory from being attributed to sheer numbers and brute power (rather than bravery, values, faith, tactics, etc.).

Whatever the motives, there are two main strategies here that can be mixed in varying degrees: to argue that the expansion was very or only minimally violent, or to argue that the rule of the conquerors or of the former regime was more tyrannical than the other and made violence legitimate. Thus Turkish historians, for instance, have gladly borrowed and chosen to focus on the modern European image of a “degenerate” Byzantium and of a rapacious feudalism before the Ottoman order: some violence may have been exerted but only for a good cause in the end.²⁴

Third is the issue of influence whereby an otherwise hostile national tradition might recognize the good things in its foe only to then “demonstrate” that the good thing is actually of one's own but taken over by the other as an “influence.” Here the burning issue was, and to some extent still is, whether and how much the Ottomans (read Turks) were influenced by the Byzantines (read Greeks), though on this issue the “Turkish consensus” is most deeply respected in the international scholarly community. As we shall see in the next chapter, the creation of the Ottoman administrative apparatus has been particularly controversial in this regard, with some historians arguing that it was all based on Byzantine models and others that the Ottomans could find all they needed in their Turco-Muslim heritage. In terms of broader cultural exchange or “lifestyles,” too, various sides of nationalist polemics have tended to see the influence of their side in, say, shared musical or culinary practices. The problem with both sides of this debate stems partly from their adherence to a static notion of cultural “goods,” whether one conceives of them in the realm of state building or cooking. In other words, “influence” is understood as a creative party giving one of its own “goods” to an imitating, uncreative other—a notion that needs to be recast now that historians realize influence is not possible without interaction, without a choice by the allegedly passive receiver. And even then, common cultural traits are not necessarily reducible to influence.

The issue of influence in cultural history is also skewed by its being understood in terms of a sexual act conceived as an unequal relationship. The influencer is like the one who penetrates and is proud, and the influenced is like the one who is penetrated and thus put to shame. A superior culture is naturally one that has more to be proud of in this manner. If “we” were the first ones to come up with the discovery of this (say, the dessert called baklava) and the invention of that (say, the shadow puppet theater called Karagz) and “our” cultural possessions then infiltrated other cultures and gave birth to offspring, then “our” culture must have been superior and dominant the way a male is over a female. If you have been influenced by others, on the other hand, you have acted like a “passive” partner in intercourse; you have been “inseminated.”

These are admittedly the more extreme versions of positions that are held much more moderately, if made at all explicit, by scholars who have developed a remarkable sensitivity to such issues and prefer to leave them alone. I feel, however, that we must recognize these issues as underlying currents of much of the historiographies (and of the historical consciousnesses that shape the scholars who create the historiographies) on the four centuries of turmoil and construction in Anatolia that led to the building of a world empire. I think these issues cannot be resolved as they are conceived; if there is any hope of transcending them, it is possible only through reformulating them, not through burying them.

Compared to the historiography on the rise of the Ottoman state, discussed in detail in the next chapter, that on the emergence of Norman power in medieval France does not run into the same sorts of complications, though there are striking parallels in the basic issues raised by their story. Did the Normans constitute a blood-tie society? Did they continue Carolingian practices and institutions or did they create their own?²⁵ These too are serious questions for the historian, but they lack the imbroglio of nationalist polemics, at least in a directly perceptible way. Had Normandy been a nation-state or a province with self-assertive Normans, the situation might have been different. Whereas, if in the questions posed above you were to replace Normans with early Ottomans, and Carolingian with Byzantine, you could have a whole range of settings for a politically charged discussion, from the corridors of a university to a coffeehouse in Istanbul or Thessaloniki. Charming as this may sound, there are drawbacks, as illustrated by the case of Petropoulos, who was persecuted under the Greek junta for, among other things, writing a book called Ho tourkikos kaphes (“Turkish coffee,” which the junta ordered to be called “Greek coffee”). Similarly, under the Turkish junta of the early 1980s, an academician could suffer for writing about the presence of Armenian and other Christian fief-holders in the early Ottoman armies, whose existence is an undeniable fact. Such tensions may get more serious under military regimes but are not limited to them; even in “normal times,” one can feel the heat generated by arguing whether Cyril and Methodius were Greek or Slavic, or by dealing with the ethnic origins of Sinn, that most accomplished of Ottoman architects who was recruited in the sixteenth century through a levy that was applied to non-Muslim subjects.

True, the majority of historians have scoffed at this sort of thing, but without directly tackling the assumption of a continuous national identity, a linear nationhood or national essence that underlies even their own nonchauvinistic historiography. If “we” (from the point of view of modern Turkish consciousness) or “the Turks” (from the point of view of non-Turkish historiography) have come on horseback from Inner Asia and established a state that replaced the Byzantine Empire of “the Greeks,” it is only human that to be “one of us” (or “one of the Turks”) one needs to assume, or in some cases feels compelled to “prove,” that one's ancestry derives from the steppe nomads. And if one is of the Greeks, then one “knows” that one's ancestors have been oppressed by the Turks. Fortunately, such assumptions or presumptions can usually be made relatively easily at the individual level, where one slips into the role of citizen-as-member-of-the-nation (unless one is from a self-conscious minority, in which case one is a member of another “we” in a similarly linear story). It must have been rather traumatic, however, for a republican descendant of Kse Mial (Mikhalis the Beardless), one of the founding fathers of the Ottoman state, a Bithynian Christian who joined forces with Osman, converted, and started a minidynasty of raiders in the service of the House of Osman. The twentieth-century Turk, proud enough to take Gazimihal as his last name, at the same time felt compelled to write an article “proving” his glorious ancestor was also a Turk. In this fanciful account, Mial is one of those Christianized Turks employed by the Byzantines who eventually chooses to join his brethren on the right side.²⁶

The essentialist trap cannot be avoided unless we, the historians, problematize the use of “the Turks” (or any other ethnonym for that matter), systematically historicize it and confront its plasticity, and study its different meanings over time and place. I certainly do not wish to imply that being a Turk or a Muslim or a Christian did not matter; for many, it was all that mattered in the grand scheme of things in the long run of the thirteenth to nineteenth centuries. In fact, individuals changed their identities presumably not only because they could but also because it mattered. Nor do I wish to imply that it is meaningless to speak of ethnic or national identities (Turks or others), but these should not be—and this is particularly clear in the case of medieval Anatolia, where religio-political affiliations and thus identities were in rapid flux—conceptualized as an original stock and descendants going through a linear series of adventures in time and, along the way, clashing with other original stocks and descendants going through similarly linear series of adventures. Historians tend to overlook the fact that (America is not the only case where) one is not necessarily born into a people; one may also become of a people, within a socially constructed dialectic of inclusions and exclusions.

It is not the purpose of this book to disentangle these questions of identity and influence embedded in varying degrees in any historical writing on the Ottomans. My purpose here is basically to problematize them before dealing with one specific era when identities were in particularly rapid flux.

We must note, however, that the issue is not an intellectual abstraction for those who consider Asia Minor their homeland and grapple with its heritage in terms of their own identity. Having grown up in a village which seems to have enjoyed continuous settlement for five millennia and where sprang up the cult of one of the Turkish colonizerdervishes in the thirteenth century, a republican Turkish archaeologist muses that studying that village “would be a contribution to our national history since it is one of the earliest Turkish villages established in Anatolia in the 13th century,” but he adds that “the cultural past of a village which has lived intertwined with the ruins of the prehistorical and the antique ages, which has not rejected those, which has appropriated and used the cave cemeteries, which has carved on its tombstones figures inspired by the wall paintings of those caves and used such figures in its embroideries, starts undoubtedly with the earliest human settlements on these lands.”²⁷ Despite this openness to continuity at one level, the distinction between the “we who came from yonder” and “they who were here” is maintained with surgical precision. Turkish colonizers are immediately taken as the main actors when the archeologist moves to the analysis of some tombstone carvings that replicate designs seen in antique objects in the same region: “When the Turkish colonizers encountered such tombstones, they acquired and maintained the tradition.” There is not even a doubt expressed that the tombstones of the “Turkish era” may have been carved by the very same people who had been producing the pre-Turkish stones, namely, by the locals who “became Turks.” The liquidity and fluidity of identities in those centuries is hard to imagine in the national age.

It was not difficult for a sixteenth-century Ottoman intellectual, however, to appreciate the plasticity of identities that had gone into the making of the neo-Rms:

Those varied peoples and different types of Rumis living in the glorious days of the Ottoman dynasty, who are not [genetically] separate from those tribes of Turks and Tatars…, are a select community and pure, pleasing people who, just as they are distinguished in the origins of their state, are singled out for their piety, cleanliness, and faith. Apart from this, most of the inhabitants of Rum are of confused ethnic origins. Among its notables there are few whose lineage does not go back to a convert to Islam…either on their father or their mother's side, the genealogy is traced to a filthy infidel. It is as if two different species of fruitbearing tree mingled and mated, with leaves and fruits; and the fruit of this union was large and filled with liquid, like a princely pearl. The best qualities of the progenitors were then manifested and gave distinction, either in physical beauty, or in spiritual wisdom.²⁸

In that grand tableau, huge amounts of material—poetic, hagiographic, epigraphic, archeological—still await gathering and sifting. This is not the place to attempt such a tableau. It is the more modest ambition of this book to problematize the origins of the Ottoman state by engaging the historiography and thus hopefully opening early Ottoman history to wider debate. It is also my hope that some of the related questions will be taken up by future researchers to improve, alter, or disprove my suggestions.