1.1 Defining What States ‘are’

In a recent comprehensive survey article on modern scholarship’s diverse engagements with the study of the premodern state classicist Walter Scheidel noted how there are being used two distinct types of definition of what ‘the state’ is, and how it can or should be studied. On the one hand, Scheidel explains, there are those who formulate exclusive definitions that prioritize modern Western statist experiences. On the other hand, there are those who pursue inclusive definitions that are in favor of universal heuristic applicability across time and space. ³ According to Scheidel, this analytical dichotomy is “emblematic of a more general rift between legalistic and political science approaches … [and] approaches of history, anthropology, and sociology”. ⁴ As we will argue below, this division has also characterized historiographical visions of the state and of its agency in late medieval Europe and fifteenth-century Islamic West-Asia. In fact, the exclusive determinism of ‘the modern state’ has for a long time had a substantial impact on these visions, and on their widely-shared organization around state/society binaries in particular. Especially in the study of fifteenth-century Islamic West-Asia this impact continues to leave many traces until today.

At the same time, from a comparative analytical perspective, it is also clear that across the board, the diverse definitions of ‘the state’ in modern scholarship all share a rootedness in Max Weber’s ideal type, with its emphasis on the combination of coercion, differentiation, a ruling apparatus and legitimate order:

A ‘ruling organization’ (Herrschaftsverband) shall be called a ‘political organization’ (politischer Verband) if and insofar as its existence and the effectiveness of its order (die Geltung seiner Ordnungen) within a specifiable geographical area are continuously safeguarded by the application and the threat of physical coercion (physischen Zwangs) on the part of the administrative staff (seitens des Verwaltungsstabes). A continuously operating compulsory political organization (ein politischer Anstaltsbetrieb) shall be called a state (Staat) if and insofar as its administrative staff successfully claims the monopoly of legitimate physical coercion (das Monopol legitimen physischen Zwanges) in the implementation of its order (die Durchführung der Ordnungen). ⁵

The diverse definitions in modern (Western) scholarship actually share not just Weberian roots, but also what Scheidel identifies as “a number of key features”. Reviewing seminal contributions to state formation debates by leading historians and sociologists such as Michael Mann, John Haldon and Charles Tilly, ⁶ Scheidel concludes that all of the most widely accepted definitions point at the usefulness of identifying a power constellation as a ‘state’ if in one way or another it combines these three factors: “centralized institutions that impose rules, and back them up by force, over a territorially circumscribed population; a distinction between the rulers and the ruled; and an element of autonomy, stability, and differentiation”. ⁷

To this triad of centralized coercive institutions, socio-political distinction and stabilizing political differentiation a fourth factor is very often added referring back to Weber’s insistence on the centrality of successful claims to order and legitimacy. Especially in the study of premodern or early states, this additional factor has also increasingly been acknowledged as an equally constitutive key feature for any definition of the ‘state’. This concerns the symbolic means that underscore the reality of the other three factors and that bind rulers and ruled into the shared imagination of an integrated, even natural, political whole. Byzantinist John Haldon and political scientist Jack Goldstone, in their own neo-Marxist exploration of a statist definition, stress the importance of this factor of “ideological integration”, which in their view often appears in the format of “the ‘ritual penetration’ of a society as represented by specific sets of social practices that express the legitimacy and belief system underpinning elite and central authority and that generally express and reinforce the structure of social relations of production”. Haldon and Goldstone explain that ideological integration and ritual penetration actually allow states to survive even when those key features of centralization, distinction or differentiation are under pressure. ⁸ Many years before Haldon and Goldstone, Joseph Strayer, one of the pioneers of European state formation studies, had made this point even more forcefully when he explained that “a state exists chiefly in the hearts and minds of its people; if they do not believe it is there, no logical exercise will bring it to life”. ⁹ The French sociologist Pierre Bourdieu, in one of the most compelling contributions to theories of the state and its formation, formulates similar insights, famously explaining and formulating the addition of this symbolic feature of ideological integration, ritual penetration and collective imagination as an explicit elaboration of the traditional Weberian definition.

I would say, using a variation around Max Weber’s famous formula, that the state is an X (to be determined) which successfully claims the monopoly of the legitimate use of physical and symbolic violence over a definitive territory and over the totality of the corresponding population. If the state is able to exert symbolic violence, it is because it incarnates itself simultaneously in objectivity, in the form of specific organizational structures and mechanisms, and in subjectivity in the form of mental structures and categories of perception and thought. By realizing itself in social structures and in the mental structures adapted to them, the instituted institution makes us forget that it issues out of a long series of acts of institution (in the active sense) and hence has all the appearances of the natural. ¹⁰

Following Bourdieu, it indeed seems important for any definition of the state to also acknowledge that states are the incarnation of a mutually constitutive combination of a hegemonic discourse and of an apparatus of coercion, distinction and differentiation. As such, states display a tendency to appear as natural or even meta-historical forms of political organization, which is however no more than a function of their existential claims to legitimacy and to any self-defined form of statehood. Rather than merely being as they thus appear, states are always in the process of becoming, in an endless “series of acts of institution”. “[S]tates are never ‘formed’ once and for all”, as George Steinmetz explains in his exploration of the study of the state-culture nexus. “It is more fruitful to view state-formation as the ongoing process of structural change and not as a one-time event”. ¹¹ In any definition or attempt at analysis, therefore, what matters most is this process of a state’s endless formation, and transformation, as that incarnation of a mutually constitutive combination of a hegemonic discourse and of an apparatus of coercion, distinction and differentiation.

1.2 Defining How States ‘become’

For Bourdieu, and for many like him, this endless act of state formation happens simultaneously in the “objectivity” of social structures and in the “subjectivity” of mental structures. Often, this is thought to happen in a dialectic interaction between power structures, or between varying combinations of such structures and all kinds of non-state phenomena. ¹² In many cases more specific moments in this process of formation are identified and defined, mostly as heuristic tools for analytical purposes rather than as actual stages in any teleological trajectory. In this respect, Goldstone and Haldon speak of a “line from local state to supra-local state to empire (and back again)”. ¹³ More specifically, they explain how

At one extreme of socio-political organization, the term “state” can refer to a relatively short-lived grouping of tribal or clan communities united under a warlord or chieftain who is endowed with both symbolic and military authority—in anthropological terms, a “Big-man” confederacy. […] At the other extreme we find more or less territorially unified political entities, with an organizational “center” (which may be peripatetic) from which a ruler or ruling group exercises political authority and that maintains its existence successfully over several generations; a key element in the formation and degree of permanence of such formations is that the authority of the ruler or ruling group is recognized as both legitimate and exclusive. ¹⁴

In Goldstone and Haldon’s very wide-ranging historical model, state formation between these extremes is not simply a contingent function of dialectic interaction between power structures, but “a longer-term evolutionary process in which social habits and institutions and state organizations respond to changing conditions through […] ‘competitive selection’ of practices”. ¹⁵ In this evolutionary process towards social transcendence and autonomy and towards a high degree of political integration and permanence, Goldstone and Haldon describe how “the potential for state formations to reproduce themselves” appears as a central feature. ¹⁶ In this respect, they reformulate Bourdieu’s idea of “the instituted institution” in historically and materially more concrete terms, which arguably represent some of the main heuristic parameters that are currently used in historical state formation studies:

The state becomes a specialized and dominant set of institutions, which may even undertake the creation ab initio of its own administrative personnel and that can survive only by maintaining control over the appropriation and distribution of surplus wealth that this specialized personnel administers. ¹⁷

Goldstone and Haldon’s analytical model of an evolutionary process shifting back and forth between two extremes appears here perhaps most clearly as deeply rooted in a Marxist interpretation of modes of production. ¹⁸ At the same time, they also explicitly acknowledge their indebtedness to other traditions. These included the ideas formulated by the fourteenth century North-African scholar ʿAbd al-Rahman Ibn Khaldun (d. 1406), ¹⁹ who famously described the process of social formation from a nomadic chieftaincy to a bureaucratic-administrative power constellation as a highly competitive iterative one, determined chiefly by social identities and relationships, economic structures, urbanization and cultural production and consumption. For Ibn Khaldun, nomadic formations are most powerful in social terms, but economically they remain relatively weak; their inevitable military empowerment is therefore bound to target economic empowerment as well, both of which materialize best in urban contexts of defense, accumulation, differentiation and distinction. However, the impact of the latter processes, including on social power, is transformative for the ruling constellation of the ‘state’ (al-dawla). Over time this ‘state’ becomes entirely dependent on new administrative personnel for its maintenance of control and authority, and at the same time is bound to collapse under the pressure of a new nomadic formation and its fresh social power. ²⁰

This tradition of imagining the endless act of state formation in a naturally occurring iterative succession of different forms of political organization actually has its equivalents in Mediterranean antiquity. It appears most explicitly in Book 6 of The Histories by the second century bce Greek historian and politician Polybius of Megalopolis (ca. 200–118 bce). Polybius explains how there are three forms of political organization (politeias)—kingship (basileias), aristocracy and democracy—and how time and again each of these three forms degenerates into its lesser equivalent—tyranny, oligarchy and mob-rule (okhlokratia)—, just as a living organism experiences birth, rise, decay and death. For Polybius the latter natural experience of life also marks the succession of these forms, with mob-rule eventually giving way to the chaos out of which a new strongman will rise as monarch, and the cycle of political organization (politeiōn anakyklosis) restarts. ²¹ Ibn Khaldun’s and Polybius’ thinking about state formation appears as very different, also demonstrating how they operated and embedded their theories in very different political contexts and concepts—fourteenth-century ce nomadic power and Islamic urban efflorescence in North-Africa and West-Asia for the former, Greek integration into the freshly won Roman domination of the second-century bce Mediterranean world for the latter. On a more abstract level Ibn Khaldun’s and Polybius’ modellings nevertheless also have many features in common, from their iterative logics and firm beliefs in the degenerating nature of power to their assumptions about the naturally increasing complexity of political organization and their conceptions of that organization’s active, even reproductive, integration of ever more stakeholders. ²²

Another seminal tradition to which Goldstone and Haldon explicitly refer as a source of inspiration—“Weber’s concept”—brings us back to the opening paragraphs of this chapter’s introduction, to the Weberian tradition. ²³ It is well known that Weber identified “three pure types of legitimate authority” (drei reiner Typen legitimer Herrschaft), which in many ways also remind us of Polybius’, and for that matter Ibn Khaldun’s, modellings. In line with Weber’s aforementioned definition of a state as “a political organization [that] […] successfully claims the monopoly of legitimate physical coercion in the implementation of its order”, these three analytical types differ along the fundamentally different rational, traditional or charismatic “grounds” (Charakters) on which they successfully make those claims, famously giving way to “legal authority” (legale Herrschaft), to “traditional authority” (traditionale Herrschaft) or to “charismatic authority” (charismatische Herrschaft) respectively. Weber usefully summarizes the socio-cultural essence of each of these three types as follows:

In the case of legal authority, obedience is owed to the legally established impersonal order (der legal gesatzten sachlichen unpersönlichen Ordnung). It extends to the persons exercising the authority of office under it (durch sie bestimmten Vorgesetzten) by virtue of the formal legality of their commands and only within the scope of authority of the office (kraft formaler Legalität seiner Anordnungen und in deren Umkreis). In the case of traditional authority, obedience is owed to the person of the chief who occupies the traditionally sanctioned position of authority and who is (within its sphere) bound by tradition (der Person des durch Tradition berufenen und an die Tradition [in deren Bereich] gebundenen Herrn). But here the obligation of obedience is a matter of personal loyalty within the area of accustomed obligations (kraft Pietät im Umkreis des Gewohnten). In the case of charismatic authority, it is the charismatically qualified leader (dem charismatisch qualifizierten Führer) as such who is obeyed by virtue of personal trust in his revelation, his heroism or his exemplary qualities so far as they fall within the scope of the individual’s belief in his charisma (im Umkreis der Geltung des Glaubens an dieses sein Charisma). ²⁴

Weber basically conceptualizes political transformation as a process of rationalization from personal to impersonal forms of political organization, towards the ideal type of legal authority “and its typical expression in bureaucracy” (seinen spezifischen Typus in der ‘Bürokratie’), and away from the charismatic and traditional types, the latter “typically represented by patriarchalism” (im ‘Patriarchalismus’ typisch repräsentiert). ²⁵ In premodern times, however, the charismatic and, especially, the traditional types always remained predominant for Weber, especially in patriarchalism’s more complex manifestation in patrimonialism (Patrimonialismus), representing an expansion and semi-bureaucratization of the ruler’s personal power, and at the same time “a decentralization of the household […] [which] leads inevitably to an attenuation of full patriarchal power (führt unvermeidlich zu einer inneren Abschwächung der vollen Hausgewalt)”. ²⁶ As will also transpire from the survey below, this typology of charismatic, traditional and legal power has arguably been as influential in modern, especially historical, studies of that endless act of state formation, as Weber’s aforementioned definition of the state has been. A case in point, especially for West-Asian historiography, is Karl Wittfogel’s much debated and largely outdated model of ‘oriental despotism’. In its alleged historical manifestation in the format of Asian ‘hydraulic empires’ this highly influential modelling was obviously informed by Marx’s historical materialism and his conceptualization of an ‘Asiatic mode of production’. ²⁷ In its imagination of the organization of discretionary personal power, however, it was rather more akin to the “extreme case” (im Höchtsmaß der Herrengewalt) of bureaucratic patrimonialism that Weber identified as ‘sultanism’ (Sultanismus). ²⁸

In speaking of ‘sultanism’ as an “extreme case” and identifying it as the outcome of a “transition” (Unterschied) that is “continuous” (fließend) and moves from tradition to discretion, Weber’s conceptualization of political transformation appears more complex (and Eurocentric) than any uniform and one-directional process of rationalization accounts for. ²⁹ In fact, his understanding of this relationship between patriarchal/patrimonial and sultanistic authority arguably displays surprising parallels with Ibn Khaldun’s and, especially, Polybius’ assumptions about the degenerating nature of (traditional) power. For Weber, indeed, premodern state formation must be analyzed against the patrimonial background of “the continuous struggle of the central power with the various centrifugal local powers”, an endless oscillation between the centralizing and decentralizing tendencies that are identified as patrimonialism’s “specific problem”. In fact, in many ways this incorporates another iterative logic of political organizations waxing and waning between ad hoc and more complex power constellations (at least until, for Weber, modern rationalism enabled Europe to escape from that logic). ³⁰

More generally, iterative models such as those of Ibn Khaldun, Polybius, and—at least for non-European political organization—Weber, and evolutionary models of state formation, which inform approaches such as Wittfogel’s or for that matter of modern legalistic and political sciences, both seem to be ideal types at the extremities of a rich continuum of interpretations of that endless historical act of a state’s formation. Most of these interpretations, however, including that of Goldstone and Haldon and more generally also that of Weber himself, situate themselves somewhere in between these evolutionary and iterative variables, trying both to avoid the pitfalls of determinism and also to allow for entropy. From a generalizing perspective, therefore, it appears more relevant to accept the reality of this analytical continuum in state formation studies than to identify where exactly on that continuum these and many more conceptualizations are to be situated. In fact, the most interesting general insight may well be that the variables that tend to be operationalized in this respect by most, if not all, relevant studies continue to relate directly to the shared essence of Ibn Khaldun’s and Polybius’ aforementioned conceptions and assumptions about the entropically increasing complexity of political organization and its active, even reproductive, integration of ever more stakeholders.

1.3 Defining What States ‘do’

Another insight that follows from this generalizing perspective is that the state—as the integrative-cum-entropic process of the formation and transformation of an apparatus of coercion, distinction and differentiation in a mutually constituent combination with a hegemonic discourse—is, and always has been, experienced as a dominant mode of organization in history. “It appears that from very small beginnings some five thousand or more years ago”, Scheidel remarks, “the state soon became the demographically dominant type of human political organization”. ³¹ As a result, studies of the state have also always revolved around, and been inspired by, questions about states’ actual roles in, and impact on, history. Alongside questions about the state’s definition and formation, another big issue at stake in studies of the premodern state has thus been: What does the state actually do? Governing society, or social groups, communities and formations, seems the most obvious and most widely contemplated answer here. Even so, the empirical reality of the minimalist nature of government in premodern contexts often tends to add important caveats to this kind of answer. Sociologist Anthony Giddens importantly remarks in this respect that

[i]t is misleading to describe the forms of rule typically found in non-modern states as ‘government’, if ‘government’ means a concern of the state with the regularized administration of the overall territory claimed as its own. Traditional states did not ‘govern’ in this sense. Their ‘polities’ were mainly limited to the governance of conflicts within the dominant classes, and within the main urban centres. ³²

For most of human history, states simply did not have, aspire to, or consider acquiring, the power, resources and instruments to discipline a ‘society’ in the maximalist ways that the modern notion of ‘government’ suggests. ³³ At least, studies of the state have never stopped to grapple with the measure—whether minimalist or rather more maximalist—of this state-society relationship. An important voice in this debate undoubtedly is that of sociologist and historian Charles Tilly, whose contribution to state formation studies of late medieval and early modern Europe has been crucial (see below). For Tilly, states engage primarily in war-making against external enemies, in state-making by integrating or excluding internal competitors, in protecting their main supporters and allies against enemies or competitors, and in extracting resources from subject populations to enable war-making, state-making and protection. ³⁴ This focus on practices of exclusion, integration, reproduction and appropriation, however, at the same time implies that there are also many things that pre-modern states did not do. Many, if not most, voices in the field would certainly agree with the implication that Bourdieu’s aforementioned claim that the state exerts “violence over a definitive territory and over the totality of the corresponding population” seems too maximalist (and perhaps also too modern) a definition. ³⁵ Weber, again, appears to have been more nuanced in his understanding of this state-society relationship. As Scheidel explains, this is a nuance that easily tends to be forgotten, making Weber even more useful for studies of the premodern state than is generally acknowledged:

It is worth noting […] that Weber speaks essentially of a claim to legitimate force in the enforcement of state rules, and does not envision an effective monopoly on physical coercion per se. In this regard, his approach fits the situation of early states with their diffused coercive capabilities better than is sometimes realized. ³⁶

Among the multiple imaginations of the nature and impact of these “diffused coercive capabilities” Tilly’s interlocking practices of exclusion, integration, reproduction and appropriation certainly stand out as referential, and are represented at least partly in most studies of what premodern states do. At the same time, quite a few scholars wish to go several steps further with the minimalism implied, questioning in varying degrees the very reality of the state-society relation. One important voice in this debate was certainly social anthropologist Ernest Gellner. Gellner actually preferred to think of pre-modern state-society relations in terms of social segmentation, with a horizontally oriented elite constituting the state, and extracting taxes from and maintaining peace (and segregation) between otherwise largely disconnected social units of local communities. ³⁷ Other modern scholars tend to push the analytical balance in favor of what Gellner identified as communities, strictly avoiding over-structuralist or top-down approaches and incorporating into their thinking varying notions of the state’s social constructedness. Following Michel Foucault’s insistence on the ubiquitous, ‘capillary’ and productive nature of power, Middle-East historian Tim Mitchell questions the very notion of the state as a social actor, and suggests that any kind of state/non-state interaction should not be taken “as the starting point of the analysis, but as an uncertain outcome of the historical process”. ³⁸ For Mitchell, the state should be understood as “a structural effect” of that process, which although “appearing as an apparatus that stands apart from the rest of the social world” in actual fact should be studied as the product of diverse social practices and arrangements that, importantly, also “produce the apparent separateness of the state and create effects of agency and partial autonomy, with concrete consequences”. ³⁹ In this conception, coercion, distinction, differentiation and hegemony do not necessarily constitute any coherent apparatus of power, but rather a diverse and contingent set of social relations, which simultaneously create an effect of their own appearance as a coherent apparatus. ⁴⁰ In other words, states do not make history, but history makes states, as and when successful social practices of exclusion, integration, reproduction and appropriation start to appear and present themselves as a coherent apparatus of coercion, distinction, differentiation and hegemony.

In rethinking the state in this bottom-up manner, Mitchell actually takes direct issue with another, even more minimalist, trend in state studies. “The importance of the state as a common ideological and cultural construct”, Mitchell argues,

should be grounds not for dismissing the phenomenon in favor of some supposedly more neutral and accurate concept (such as political system), but for taking it seriously. Politics, after all, is a process built out of such shared constructs. ⁴¹

This ‘cultural’ turn in state studies has indeed also given way to an analytical trend that involved even more radical questionings of the usefulness of the state as a heuristic concept to understand the social realities and impact of power. Seminal contributions to this include a posthumously published paper by historical sociologist Philip Abrams about “the difficulty of studying the state”, given what he described as “the secret of the non-existence of the state”. ⁴² Perhaps one even more significant contribution is a survey about the uses and abuses of the state concept in European medieval studies by medieval historian Rees Davies who concluded that “the state has been given far too privileged a role in the analyses of power in earlier societies”. ⁴³ This critical line of minimalist, if not nihilist, thought was informed by a growing unease with the dominant evolutionary paradigm of the medieval origins of European modern states (see below). It also displayed many belated echoes of similar debates that have been raging in the political sciences since the mid-twentieth century. ⁴⁴ This approach found its most explicit medievalist representative in a survey history of later medieval European politics first published in 2009 by John Watts, an historian of late medieval England (see also below). Watts even argued that “it is not necessary to frame—one might almost say burden—the structural history of politics with the notion of the state”. ⁴⁵ As a result, Watts decided not to speak of the state at all, but to adopt “a more open-ended perspective on the changing political structures of the period”. ⁴⁶

This dismissive attitude, however, is certainly not the dominant position in state formation studies at large. Echoing Mitchell’s above-mentioned criticism, John Haldon was perhaps one of those who warned most forcefully against throwing out the baby with the bath water when pushing a minimalist approach to these extreme ends.

It is important to stress that the state does have an identity as a field of action, as a role-constituting site of power and practices which can be independent, under certain preconditions, of the economic and political interests of those who dominate it. ⁴⁷

As in the case of Middle East historian Timothy Mitchell, Haldon entered this debate from a rather different background, not as a specialist of Western Europe but as a Byzantinist. In his thinking about the state as a tool in premodern historical research he was concerned with a far more theoretical “Marxist approach to the State”. Haldon asked questions on a Eurasian scale about “a historical materialist approach to the state, state elites, the relative or absolute autonomy of state structures and practices, and the role of the economic in Marxist historical interpretation”. ⁴⁸ From his Eurasian and comparative perspectives Haldon actually recasted these debates between minimalists and maximalists in interestingly processual terms. Ascribing to the aforementioned notion of state formation as an endless process, he stressed “that state formations differ qualitatively in the degree of their ‘stateness’”, forbearing any comparisons between their historical manifestations across time and space that do not take into account the specifics of “very different structural contexts”. ⁴⁹ Any study of states and of their impact on human history should therefore internalize the assumption that all statist manifestations are specific and can only be considered as contextually defined exceptions of any one analytical model that may be employed. Rather than trying to reduce that specificity and exceptionality to, prioritize their features within, or exclude them from a statist ideal, what should matter most in comparative, and, for that matter, non-comparative historical research is to use that ideal to acknowledge for, and to decipher, specificity and exceptionality. ⁵⁰ Scheidel believes that it is Tilly’s “model of state formation driven by interstate and class conflict”, through the interplay of war-making and state-making, that “holds promise for the study of any historical period, even as we must avoid the transfer of specifically European features to other environments”. ⁵¹ The latter critical distance may well be achieved by identifying where and how Tilly’s model resonates with or may be further refined by the thinking of others, including Max Weber, Ibn Khaldun and Polybius.

3.1 The Dominance of Legalistic Readings and the Sociological Turns of Lapidus, Hodgson and Chamberlain

As noted earlier in this chapter, the history of state studies in general has been marred by an analytical dichotomy between legalistic and sociological approaches. At least, this somewhat reductive legalistic versus sociological framing seems very useful to organize and understand how the state has been considered in the different dynastic historiographies of West-Asia. In fact, within these historiographies there is an interesting prevalence of the former, legalistic, frame. This is largely due to the fact that within these historiographical traditions there has always been a strong presence of Islamicists and modern specialists of Islam. These scholars have always been primarily interested in the formation of Islam as a normative cultural, social and political practice and have always tended to see the history of Sunni Islam’s legal system as one of the crucial vectors of this formative process. This Islamicist focus means that for a long time, understandings of political organization in Islamic West-Asia have been considered first and foremost from a legal perspective of formal procedures, structures and discourses that were somehow related to the implementation of God’s Law (shariʿa). Contemporary writings and related sources in Arabic and Persian certainly lend themselves easily to these legalistic interpretations. This is not just because of the legal training and employment shared by most of their authors, but also because of the prescriptive, normative and legitimating agendas that these texts all served. At the same time this legalistic frame was also informed by particular modern narratives of Asian otherness and of an Islamic state formation process that was considered fundamentally different from the secular and liberal destiny of the West. The notion of the absolutist rule of divine law and of legal conservatism offered a highly functional framework which many modern scholars have used to explain the post-Abbasid Islamic world’s absorption of Inner Asian nomadic conquerors and the maladroit condominium between violence-wielding Turko-Mongol ‘outsiders’ and that world’s sophisticated urban elites, especially its religious scholars and judges as well as the caliphs of Baghdad and, eventually, Cairo. ⁸⁵

In the early 1990s the Ottomanist Rifa’at ʿAli Abou-El-Haj heavily criticized this traditional reading of the state which had also for a long time permeated early modern Ottoman studies. In fact, in a programmatic publication on the study of Ottoman state formation he usefully summarized the basic—a-historical and absolutist—assumptions of this legalistic trend in scholarship. On the one hand, Abou-El-Haj lamented the fact that “in twentieth century scholarly writing on Ottoman affairs, the concept, the institution, and the nature of the state have been treated as if, regardless of the passage of time, the state had remained essentially the same”. On the other hand, he claims that modern West-Asian scholarship is predisposed “to regard the modern nation-state with its meritocratic bureaucracy as a paradigm”, so that “modern sociologically evolved standards as merit, public service, equity, and rationalized practices” tend to be used to “measure the early modern Ottoman state”. For Abou-El-Haj this perspective derives from “the misapprehension that prior to the seventeenth century the Ottoman state was a centralized, efficient, and rational public entity, unique in the period during which it flourished”. ⁸⁶ In Abou-El-Haj’s reading these are highly problematic assumptions of pre-seventeenth century Ottoman meritocratic maximalism and uniqueness. In fact, these assumptions have for a long time been constituent partners of the legalistic trend in general. In this dominant reading the idea, or the ideal, of the state as a continuous and autonomous constitutionalist bureaucracy actually served as an interpretative norm against which historians were supposed to weigh the course of state formation in post-Abbasid West-Asian history writ large, as if the whole was meant to culminate in Ottoman absolutism.

The field of Islamic West-Asian political history continues to be easily lured by the clear-cut categories and well-trodden narratives of legalistic readings. Starting in the middle of the twentieth century, however, there also emerged more sociologically inspired approaches to political organization and state formation in later medieval Islamic West-Asia. For a long time, such developments mostly happened in the margins of the field and its legalistic readings, and it has to be acknowledged that the pioneering beginnings of this alternative, sociological approach arguably lie in assumptions of the growing irrelevance of traditional statist structures in the face of Turko-Mongol irruption. This overture therefore engaged with a search for alternative conceptualizations of power and its organization in response to a presumed ineptness or even failure of the continuing legalistic state and its bureaucratic apparatus of power. The basic ingredients that were proposed to solve this analytical irrelevance of the legalistic state were Weberian traditional authority and the military household and family rule, patronage and mostly urban institutions for the local management and organization of resource flows. There are three pioneering contributions to this critical sociological turn in Islamic state formation studies that deserve to be expanded upon more explicitly. First and foremost, there are the writings of Ira Lapidus and Marshall Hodgson which are particular products of mid-twentieth century structuralism and systems theory. They fused the dominant legalistic reading of the state in Islamic West-Asian studies with a search for alternatives in highly creative and influential ways. Alongside these two authors we will also consider here the writings of Michael Chamberlain, as they represent an important translation of the ideas of Lapidus and Hodgson to late twentieth century post-structuralism and also as Chamberlain was arguably the first to truly think without the traditional legalistic paradigm.

Ira Lapidus, in his Muslim Cities in the Later Middle Ages, argued that the locus of power in the urban centers of late medieval Egypt and Syria no longer lay with the state. Indeed, here he still conceptualized the state in a legalistic fashion as the central bureaucratic apparatus that organized these centers’ protection and exploitation. Due to what Lapidus defined as “the privatization of power”, power instead lay with the capacities of the state’s agents to engage in mutually beneficial informal relations of patronage and service beyond tributary and military needs. These agents were military commanders (“the Mamluks”) with their expansive urban households and religious scholars (“the Ulama”), who were quasi-integrated in the state apparatus and simultaneously endowed with ambitious urban communal roles. Lapidus explained that he was “drawing on the work of Weber, Parsons, and other students of social process” and following the systemic readings by these theorists he proposed that “the Political System” was not confined to the state, but joined “the Mamluk State and the Urban Notables” in the balancing act of “a governing condominium” which boiled down to “a shared control over the society”. More specifically, this meant that

while the Mamluks took up the massive economic and military responsibilities, the notables lent their intimate grasp of local affairs to the service of the state and coordinated the government of society at more intricate levels. Their original social importance made them indispensable auxiliaries of the Mamluk state apparatus, and in turn their partial assimilation to the regime served to validate their local status and to assure their success in communal roles. ⁸⁷

Marshall Hodgson (1922–1968) is best known for his magisterial, posthumously published three-volume The Venture of Islam: Conscience and History in a World Civilisation. In this total history of Islamic societies and cultures Hodgson was certainly more explicitly concerned with processes of political transformation than his younger colleague Lapidus in Muslim Cities. At the same time, even though also indebted to the predominance of systems theory in post-World War ii times, Hodgson’s writings were less theoretically grounded than Lapidus’ work. Nevertheless, Hodgson’s particular conceptualizations of time, space and phenomena pertaining to the longue durée of Islamic history were no less compelling or influential. This is also apparent from Lapidus’ later magnum opus of similar ambition A History of Islamic Societies (1988). Lapidus’ discussion in that book of “the Post-ʿAbbasid Concept of the State” concludes, in a very Hodgsonian ‘historicization’ and ‘nomadization’ of his above-mentioned understanding of the state and wider ‘political system’, that

between 950 and 1500, a region-wide pattern of governmental institutions took shape. These institutions began with the late ʿAbbasid practice of the caliphate, the use of slave military forces and iqtaʿ forms of tax administration, and nomadic concepts of family and state authority. This new order first took shape in the eastern provinces of the former ʿAbbasid Empire under the aegis of the Buwayhids, the Ghaznavids, and the early Saljuqs. ⁸⁸

For Hodgson, as for Lapidus, the locus of power in what Hodgson termed ‘the Middle Periods’ (1000–1500 ce) had actually shifted towards specific social actors, in a double process defined as ‘militarization’ and ‘unitary contractualism’. In this process military amirs and urban notables (aʿyan) primarily pursued personal interests and responsibilities in mutual relationships that “amounted sometimes to a relation of personal patronage—a type of relation that played a major role in such a society”. In this egalitarian context, authority did not simply derive from any impersonal state apparatus, but was an achievement of personal relationships. These empowering bonds appeared in the format of contract-like arrangements that subscribed to a uniform communal framework of obligations and expectations, mainly though not exclusively informed by divine law (Shariʿa). Alluding to Weber’s aforementioned definitions of legitimate authority and its three ideal types as well as to Weber’s idea of the state as a monopolist of legitimate physical coercion, Hodgson formulated this ‘medieval’ empowerment of society, as a Muslim community, as follows:

Whether an independent position of authority was legitimized by appeal to personal charisma or to explicit law or to custom, it was conceived as established by mutual agreement and as assuming mutual obligations between one individual and others. […] always it was a contract-type arrangement which had to be renewed personally with each new holder of authority and was properly binding only on those who had personally accepted it. […] The religious community was […] almost—though not quite—liberated from dependence on an agrarian-based state; so that its communal law, built on its communal presuppositions, and not that of any territorial state, assumed the persisting primacy that accrues to whatever possesses exclusive legitimacy. It was not quite liberated: the ultimate sanction of force remained critical, and was left in the hands of the state. But the role of the state was as far reduced, especially in the basic sphere of law, as it ever has been in citied high culture. ⁸⁹

For Hodgson this ‘communal’ empowerment was part and parcel of a historical process of state-society polarization that involved “the collapse of the caliphal state and its bureaucracy”. This led to the subsequent difficulty “to develop any really integrated states”, and the formation—especially in the wake of the Mongol invasions of the thirteenth century—of new “military constructions”. Despite moments of brilliance, time and again these constructions “failed to provide sufficient strength to self-perpetuating bureaucracies which would have assured a continuity of authority despite the personality of the amir, and often failed to go far beyond a purely tribal, essentially irresponsible notion of power, in which the whole land became the proper prey of the prowess of a vigorous tribe”. ⁹⁰

Hodgson was trying to come to terms here with the Middle Periods’ historical context of short-lived military state formations constructed around political patterns informed by nomadism, Turko-Mongol ideas and ideals of power, and Inner Asian steppe life. For this context, Hodgson actually developed his own statist model parallel to Lapidus’ “governing condominium”, to understand how these patterns also included a “nomad-urban symbiosis” which eventually, in early modern times, helped to overcome the structural problem of failing political integration and insufficient bureaucratic power. This became the highly influential model of the ‘military patronage state’, in which “the steppe principles of nomad patronage of urban culture were generalized”, and which, for Hodgson, had a particular, historical effect on processes of state formation in late medieval and early modern Islamic West-Asia. In his reading of these patterns and processes, the state-society scales were again tipped in favor of the state, which thus gradually re-appeared as an absolutist legal structure and an institutional apparatus of centralizing power. Hodgson summarized this model and its evolutionary process as follows:

The Mongol states did, nevertheless, introduce the notions characterizing the ‘military patronage state’, which was to have a great future; and in doing so they modified the context of the Muslim institutions. Already in central Eurasia itself, the nomad-urban symbiosis had been raised above the level of simple plundering exploitation. In the agrarian societies of Islamdom, under the impulse of the tremendous successes of the Mongols and of the Turkic tribes that had shared their victories, the symbiosis was raised to a yet higher level. The Mongols from the first acted in a spirit of monumental achievement: they destroyed in the grand manner, they built in the grand manner too. All this had a relatively enduring institutional residue which we may pinpoint under three heads, recalling, however, that no one state is being described, but only features that frequently did occur under the Mongols and among their heirs, and that show mutual relevance. First, a legitimation of independent dynastic law; second, the conception of the whole state as a single military force; third, the attempt to exploit all economic and high-cultural resources as appanages of the chief military families. However, most of these institutional tendencies merely had their beginnings in Mongol times, which still displayed much continuity with the Earlier Middle Period; some were not developed fully till the sixteenth century, when the use of gunpowder weapons had given the central states (and the patterns they embodied) much more power. ⁹¹

Hodgson’s model of the ‘military patronage state’ ⁹² conceived of the state as a rudimentary military apparatus of violence and exploitation that pursued legitimate authority and local connectivity via its leading military families’ patronage of cultural elites and resources. It was applied as an interpretive tool in his detailed historical survey of the Middle Periods, where he built upon much extant historiographical scholarship to also engage with ‘medieval’ polities such as those of the fifteenth century. In the case of so-called Mamluk Cairo, Hodgson especially stressed the military element and the total exclusion of non-military society from an institutionalized type of absolutist power that he identified as oligarchic and as qualified by “chivalry” and “incessant intestine fights”. ⁹³ In the Timurid case, Hodgson preferred to highlight the factor of cultural patronage, explaining that “in the field of the arts the Timurids illustrate the patronage state at its best”. This was despite—or rather thanks to—the fact that “the chronicle of the Timurid reigns is even fuller of fratricide and power lust and low intrigue than is most Muslim history in the Middle Periods”. ⁹⁴ As for the Ottomans, they appear in Hodgson’s writings as combining the best of these two worlds: absolutist military states marred by internecine warfare combined with local societies blessed by a flowering of, especially, Persianate culture. The fact that the Ottoman polity was “founded in ghazi traditions rather than steppe traditions” meant according to Hodgson that it was organized more coherently around the Ottoman family, its ghazis or frontier warriors, and the many resources that the Muslim-Christian frontier had to offer. ⁹⁵ For Hodgson the mid-fifteenth-century conquest of Istanbul boosted this solid and integrated dynastic and military core to an imperial level, so that “from a frontier ghazi state, [it] became an absolutism assimilable to the military patronage type, and one of the cultural foci of Islamdom”. ⁹⁶ Eventually, in Hodgson’s reading, an absolutist centralizing state arose around the Ottoman court that integrated a wide diversity of social elites, practices and ideas into its stabilized military structures. In Hodgson’s own words, this culminated in “an absolutism in which the whole government—even the imâms of the mosques, as governmental appointees—were regarded as military (’askerî) even though not as ‘men of the sword’ (sayfî); and all that was valuable in society at large was regarded as in the dispensation of the royal family and its servants”. ⁹⁷

Michael Chamberlain is another historian of Islamic West-Asia who deserves to be mentioned in this overview of the sociological turn in West-Asian studies, despite the fact that he did not engage with the fifteenth century. In fact, Chamberlain is one of the very few scholars who have engaged with the work of both Lapidus and Hodgson and who also thought about the concept of the state. This was done in his work on the organization of elite life in thirteenth- and fourteenth-century Damascus and in two later surveys of West-Asian power politics between the eleventh and thirteenth centuries. ⁹⁸ Chamberlain’s thinking about political organization followed that of his teacher Ira Lapidus in particular in being very consciously and critically informed by contemporary sociological theory. In Chamberlain’s case, however, this involved late twentieth-century post-structuralism rather than systems theory and functionalism. ⁹⁹ As a result, Chamberlain arguably emerges as the most successful representative of this sociologically inspired trend in West-Asian historiography, since unlike Lapidus and Hodgson he was not seduced by the structuralist categories of the dominant legalistic tradition. Chamberlain actually managed to entirely break with that tradition’s evolutionary Hegelianism, freeing himself from the recurrent maximalist interpretations of the state in legalistic readings. He was able to think beyond the strict Orientalist categories of a state-(civil) society relationship that should be measured in bipolar terms of strength or weakness, or of the state’s failure, or not, to integrate society. ¹⁰⁰ Basically, Chamberlain rethought the latter issue that had also haunted Lapidus and Hodgson, namely the relationship between the legalistic appearances of the state and the actual locus of power in ‘the Middle Periods’: military and other, especially urban strongmen and their kin groups or households. Whereas the two other scholars thought about the changing quality of that relationship in terms of “privatization” (Lapidus) or of “militarization” (Hodgson), Chamberlain radically questioned the validity of this entire relational construct. He actually moved away from the idea of society’s ‘subalternity’ and disruptive ‘medieval’ empowerment to that of the priority of social practice and the state as a fluid and messy by-product of that practice. This analytical reversal followed the ideas of Norbert Elias and, especially, Pierre Bourdieu, to favor social practice and power as a social phenomenon and moved against the idea of the state as a metahistorical bureaucratic actor. Chamberlain develops this approach most forcefully, and as a tacit correction to Lapidus, in Knowledge and Social Practice. Here, he is very explicit about his highly critical and minimalist thinking about the state as a historical concept, explaining that

The state in this period was not an impersonal entity, possessing specialized agencies, capable of formulating long-term strategies to pursue political goals. The politics of the city consisted of continuously renegotiated relationships among the ruling household, the important amīrs and their households, and civilian elites with specialized knowledge or religious prestige. If we are to speak of the state at all, it is as an abstraction of the personal ties of alliance, dependence, and dominance among these three groups. Rather than look for the mechanisms by which the state, as the primary embodiment and agent of power, diffused power from the top down, we need to understand a more complex situation. Studies of the bureaucracy, of such entities as the sultanate and the caliphate, of the legal and ‘public’ aspects of power, are undeniably useful in themselves. These, however, do not cover the entirety, or even perhaps the most important part, of relations among power, cultural practices, and the social strategies of groups. Such approaches have been useful in medieval Islamic history only with so much qualification that they lose the very precision they are intended to introduce. ¹⁰¹

In Knowledge and Social Practice Chamberlain defined the pattern of power relationships that emerges from this analytical turn to urban practice as “a kind of maladroit patrimonialism”, and it was through this pattern of personal ties rather than through any state structure that “Ayyūbid and Mamlūk warrior households […] made political use of existing social, cultural and administrative practices”. ¹⁰² In two subsequent studies, Chamberlain further qualified this correction of Lapidus’ functionalist universe through the notion of “maladroit patrimonialism” by bringing in the Hodgsonian concept of “the military patronage state”. This happened as Chamberlain was expanding his historical horizon beyond the narrow confines of thirteenth- and fourteenth-century Damascus. In so doing he was summarily following in the footsteps that had already led his teacher Lapidus from the Syro-Egyptian focus of his Muslim Cities to the global one of his Islamic Civilizations and its critical engagements with Hodgson’s conceptualizations of “the Middle Periods”. Chamberlain’s own critical engagements focused especially on how competition, conflict and related social practices helped to better understand Turko-Saljuq and post-Saljuq West-Asia as a meaningfully connected political space. In making this argument, Chamberlain simultaneously expanded the horizon of the concept of a ‘military patronage state’ beyond the spatial and temporal (and structuralist) confines of Hodgson’s Mongol and post-Mongol uses. In his contribution to the Cambridge History of Egypt, published in 1998, Chamberlain described how at the time of the late twelfth-century Ayyubids, Egypt also came to share “a number of characteristics with […] the post-Saljuk Muslim military patronage states”. This meant that “political power was concentrated in, and emanated from, the household (bayt)” and that “its political economy was marked by a relatively weak bureaucracy and ruling establishment, partial control of land revenue by horse warriors and religious leaders, indirect rule through religious and military magnates, and parcelized and derived sovereignties”. ¹⁰³ His chapter “Military Patronage States and the Political Economy of the Frontier, 1000–1250”, first published in 2005, subsequently engaged along similar, practical lines with these post-Saljuq military patronage states in Islamic West-Asia, and with their Saljuq predecessors. In fact, that chapter explained even more clearly Chamberlain’s explicitly non-legalistic understanding of Hodgson’s ‘military patronage state’, as he insisted on the fact “that we should not think of the Seljuks as a unitary state, but rather as a collection of powerful households kept in check by the head of the most powerful among them”. He furthermore also refined his own earlier minimalist understanding of the state as a so-called ‘abstraction’ of particular social relationships, claiming that “the ruling household’s adoption of monarchical and legal arguments for legitimate authority was one way of fending off the claims of the other households within the ruling family”. ¹⁰⁴ Chamberlain thus argued that rather than simply being marginalized or regularly broken by social practice’s key features of competition and struggle for power (as in a way Lapidus and Hodgson had suggested), the legalistic state was actually reproduced by those features. Like the aforementioned Charles Tilly, Chamberlain thus also believed that “war made the state and the state made war”. For Chamberlain at least, the state did so in highly minimalist ways, as an empowering set of ideas of legitimate political order and sovereignty that was part of the tools that were strategically deployed by powerful competing households. “The military patronage state”, Chamberlain concludes, “thus permitted the ideal of the universal cosmopolitan empire to survive within a political-economic context that tended towards fragmentation”. ¹⁰⁵

It remains difficult to gauge the actual impact of these more sociologically inspired approaches of Lapidus and Hodgson, and more recently of Chamberlain, on the different dynastic historiographies of post-Temür West-Asia. A recent survey article of Syro-Egyptian social history identified the enormous impact of Lapidus’ Muslim Cities in redirecting the field towards the study of informal social networks, urban elites and social equilibrium. At the same time, this survey acknowledged that there has been a lack of critical engagement with Lapidus’ systemic thinking. It furthermore questioned the value of Chamberlain’s contribution in this respect, claiming that “Chamberlain’s Damascus is ‘no longer a society’, but only a space for elite struggles, and therefore an unsatisfactory model of social action”. ¹⁰⁶ Even though the latter assessment is certainly too harsh, especially as far as the invigorating impact of Chamberlain on the study of Islamic knowledge practices and their practitioners is concerned, it is undoubtedly true that scholars today have tended to easily take for granted Lapidus’, and also Chamberlain’s, more general theoretical concerns, or even have forgotten them, whenever their more specific analytical tools were deployed in contemporary research. This observation of theoretical quietism regarding the state is certainly also valid for many more understandings of fifteenth-century West-Asian state formation, even those that do not engage with Lapidus or Chamberlain at all. Mostly unwittingly, uncritical legalistic readings of the state have arguably remained the norm, not just in modern Syro-Egyptian historiography, but—as Abou-El-Haj suggested—also in work on the Ottoman dynasty, and to some extent even in the arguably more theoretically informed work on Timurid and Turkmen dispensations. As will be demonstrated below, in the latter dynastic contexts, and also in the Ottoman case, some basic organizational principles related to ideas about princely courts, military households and dynastic dispensations, and cultural patronage have certainly been around for a long time, even long before Hodgson modelled them as ‘military patronage states’. ¹⁰⁷ However, just as Hodgson’s own engagement with those ideas continued to be rooted in that legalistic framework, so did most of the engagements of any dynastic specialist with the state in fifteenth-century Islamic West-Asia. In general, such engagements have therefore restricted their readings of the state to a theoretical quietism that favors merely descriptive accounts of diverse bureaucratic institutions, meritocratic practices and legal norms or of their roles in governing groups, communities and resources. They have prioritized detailed empirical reconstructions of the ranks and careers of dynasts, administrators and courtiers or of how centripetal phenomena expanded or contracted or were reproduced. In short, the dominant question has always remained how states successfully—or unsuccessfully—maximized their relations with, and their autonomy from, West-Asia’s societies, without genuinely contemplating the validity and implications of this type of question. ¹⁰⁸

In studying, or simply taking for granted, such specific issues and questions, many of these legalistic readings of state formation in fifteenth-century Syro-Egyptian, Ottoman, Timurid and Turkmen contexts actually quietly adopt an interpretation that considers the process of the endless development of the state in rather structured and linear terms. Interestingly this is at odds with the insights which, as explained above, now appear as mainstream in both the long history of the theoretical study of the premodern state and the study of state formation in late medieval Western Europe. In their maximalist, even absolutist, constitutionalism, these legalistic readings also often impose assumptions of what states do, or what they wish to do or should do, that go far beyond Tilly’s interlocking practices of exclusion, integration, reproduction and appropriation. This divergence may well be a function of the West-Asian particularity of these fifteenth-century contexts. At the same time, however, this may also be a result of the lasting impact of a particular generation of historical scholarship in Islamic West-Asian studies that once also made Europeanist readings of the state prioritize a dominant line of thinking that is identified as legalistic here. In any case, as seen above, when detailing the value of Chamberlain’s contribution, from a historical and historiographical perspective these interpretations and assumptions are not without their challenges. As Abou-El-Haj also lamented in the early 1990s about Ottoman studies, this is especially true when such approaches are adopted without much critical reflection. Nowadays any active stakeholder in fifteenth-century West-Asian scholarship obviously tends to readily acknowledge many of these more theoretical challenges and their rootedness in some resilient anachronism, or even in “‘the four sins’ of modernist (social) theory: reductionism […], functionalism […], essentialism […], and universalism […]”. ¹⁰⁹ Nevertheless, it is not easy to provide a constructive response, especially not when the exigencies of the legalistic framework continue to intervene. Making abstraction of these macro-historical challenges in the light of other, less arcane and more down-to-earth matters of scholarly interest appears therefore as an easy escape route for many engagements with fifteenth-century West-Asian history.

However, in each of the dynastic historiographies of fifteenth-century West-Asia there certainly also are exceptions to this general tendency of combining legalistic readings of the state with a kind of theoretical quietism. It is worth mentioning a few of these more explicit engagements with the notion of state formation, if only to illustrate how the trans-dynastic sensitivities found in Lapidus, Hodgson and Chamberlain did not operate in conceptual voids. This also serves to show how their more theoretically informed ideas did not go entirely unnoticed, or how these did have at least some counterparts in dynastically oriented scholarship.

3.2 Defining Early Ottoman ‘state formation’

For Ottoman fifteenth-century history, Hodgson’s aforementioned modelling was very much informed by foundational debates among (early) Ottomanists on charismatic origins in frontier warfare, fluid identities and loyalties, and how early Ottoman groups transformed into something more than simple warbands. ¹¹⁰ In fact Hodgson’s understanding still aptly summarizes the dominant paradigm of fifteenth-century Ottoman state formation. As explained before, this is the paradigm of the rise of a centralizing state around the Ottoman dynastic court, which integrated an ever-wider diversity of social elites, practices and ideas into its stabilizing military structures. In recent decades, one important factor has been interpreted as sparking this fifteenth-century dynamic of centralization and integration, namely the shifting of the balance between the centrifugal forces of marsher lords and old power elites and centripetal forces coalescing around the Ottoman household in favor of the latter, especially after the conquest of Constantinople by Mehmed ii. ¹¹¹ For Cemal Kafadar, this was part of a long process of Ottoman state formation, beginning in the early fourteenth century, and marked by the diversity of social forces and the fluid and mobile relationships that define any frontier, as well as by the local resilience “of sedentary administrative traditions in both the Perso-Islamic and the Byzantine modes”. ¹¹² On the basis of a critical reading of the ideological and narrative agendas of extant contemporary sources, Kafadar chose to understand this process as that of “a coalition of various forces, some of which were eventually driven to drop out of the enterprise or subdued or marginalized”. He concluded that this interaction of centripetal and centrifugal mobilities was determined by “the Ottoman success [to harness] that mobility to their own ends while shaping and taming it to conform to their stability-seeking, centralizing vision”. ¹¹³

One of the few Ottomanist scholars who have interpreted this more socially oriented understanding of Ottoman state formation through the lens of an explicitly sociological reading is Karen Barkey. Her work provides an ambitious endeavor to better understand ‘the longevity of empire’, the longue durée of Ottoman political continuities between the fourteenth and early twentieth centuries. For the Ottoman fifteenth century, Barkey explained that she actually built her novel interpretation mostly on the empirical work presented by Heath Lowry in his The Nature of the Early Ottoman State. In his analysis of Ottoman state building in the fourteenth and fifteenth centuries, Lowry was working in a dialogue with many Ottomanists, including Kafadar. Above all, Lowry stressed the hybrid nature of the ‘Ottomans’ themselves, and the fact that Byzantine and other Christian elites had become active stakeholders in Ottoman centripetal forces. ¹¹⁴ For Barkey, however, “Lowry’s framework lacks a sociological account of how a few men make a revolutionary change in their immediate social relations and transform them into relations of power and influence”. ¹¹⁵ Focusing on these relationships and their transformation, Barkey explained that “state transformation then is the resolution of organizational and boundary problems realized in the intermediate zone by state actors and social actors embedded in networks of negotiations”. ¹¹⁶ The state in this context is explicitly defined by Barkey with Weber’s “monopoly of the legitimate use of physical force over a given territory”, to which however she added “Bourdieu’s conceptualization of the state, where he talks of ‘symbolic violence over a definite territory and the totality of its corresponding population’”. ¹¹⁷ Barkey’s analysis is further informed by connecting “historical institutionalism with network analysis, because the mechanisms of institutional continuity, flexibility, and change are embedded in the meso-level network structures that link micro-level events and phenomena to macro-social and political outcomes”. ¹¹⁸ With the assistance of these analytical tools, Barkey explained how Ottoman political success is one of flexible and adaptive “brokerage across structural holes”, which offered a means for the successful vertical integration of diverse networks into Ottoman leadership and at the same time the horizontal segmentation of these networks so as to maintain Ottoman priority. As far as the Ottoman fifteenth century is concerned Barkey argued that this integration and segmentation took the shape of “‘institutional layering’, which involves attaching new elements onto an otherwise settled institutional frame”. ¹¹⁹ More particularly, as a consequence of “the layering of Byzantine and new Ottoman organizations and practices”, ¹²⁰ Barkey explained her vision on fifteenth-century Ottoman state formation as follows:

The empire that was built after 1453 became a robust, flexible, and adaptive political entity where a patrimonial center, a strong army, and a dependent and assimilated state elite interconnected with many diverse and multilingual populations ensconced in their ecological and territorial niches. The Ottoman imperial order was to be found in the three components of empire—legitimacy, control over elites and resources, and the maintenance of diversity—each forged through the relations between state forces and social forces, center and periphery, state and regional elites, and central officials and local populations. ¹²¹

3.3 Defining Turkmen and Timurid ‘state making’

Timurid and Turkmen political history writing is marked by a similar Weberian modelling of co-constitutive relations between state actors and social actors. In fact, among the three dynastic research traditions surveyed here, Turkmen and Timurid historiography is arguably the field which has most explicitly engaged with this paradigm and with the conceptualization of state formation more generally. Hodgson’s qualification of Timurid state formation as a historical process of state-society polarization offers a case in point. His ‘military patronage state’ model describes a successful and culturally efflorescent nomad-urban symbiosis, albeit not a peaceful nor continuous one, operating within an immanent institutional framework of absolutist tendencies. This model is not just widespread in modern Timurid and Turkmen historiography but as suggested above, it is simultaneously rooted in that historiography’s own centennial lineage. It is moreover the object of empirically informed refinements and adjustments by a group of historians who like Hodgson all have some link with The University of Chicago and its intellectual legacies. The work and teaching of John Woods, professor of Middle Eastern history at The University of Chicago, has been particularly seminal in this respect. In his widely acclaimed The Aqquyunlu: Clan, Confederation, Empire, Woods engaged empirically with Hodgson’s historical model of the erratic empowerment of an apparatus of centralizing power in a volatile context of Turko-Mongol rule and nomad-urban symbiosis. ¹²² Importantly, Woods did this by situating his historical work in a carefully considered conceptualization of the “Aqquyunlu social and political structure”. ¹²³ Central to Woods’ understanding of the Aqquyunlu system is “the political and institutional separation and accommodation” which marked the relationship “between the nomadic military elite (‘Turks’) and the urban elite of eastern Anatolia and Iran (‘Tajiks’)”. ¹²⁴ Woods argued that this mix of centrifugal and centripetal tendencies manifested itself especially “in the structure of both the central and the provincial administration” in the format of a division of labor and power, and of “interlocking interests of these two groups”. ¹²⁵ Identifying this division and interlocking explicitly with the symbiotic “a’yan-amir system as defined by Marshall Hodgson”, Woods also echoed Hodgson by suggesting that this “proved to be a remarkably effective mechanism for maintaining order and preserving the fabric of society in the absence of a powerful central authority”. ¹²⁶ In fact, the latter problem of “absence” is equally central to Woods’ understanding of the Aqquyunlu system, as it also was for Hodgson more in general. Just like Hodgson, Woods explained this struggle for “central authority” as a political pattern that was informed by the elite’s nomadism, by Turko-Mongol ideas and ideals of corporative power, and by the latter’s creative clashes with the absolutist claims of local administrative traditions and individual military successes. ¹²⁷ The historical process which emerged from this pattern was one of regular dynastic turmoil and, as suggested, erratic empowerment of an apparatus of centralizing power. More precisely, Woods concluded, in general, historicizing terms, that

[i]t is precisely the interplay of these two tendencies among the Aqquyunlu—the centrifugalism or segmentation associated with clan corporateness together with nomadic traditions and the centralizing forces characteristic of charismatic, personal, or bureaucratic concepts of sovereignty—that constitutes one of the major dynamics in both Principality and Empire Periods. Though such systems have been declared explosive and ultimately unworkable, it was precisely this volatility—the continual elimination of entire houses from political contention and the consequent renewal of the sovereign mandate in successive dispensations—that helped the system continue to function among the Aqquyunlu for almost a century. Viewed in terms of the preceding discussion, Aqquyunlu dynastic history must therefore be resolved into a succession of three dispensations to the houses of Tūr ʿAli, Qara ʿUsman, and Uzun Ḥasan. ¹²⁸

These particular centrifugal and centripetal tendencies are explained here as not just being an integral part of a state-society dynamic, but simultaneously as being internal to Turko-Mongol dispensations such as that of the Aqquyunlu. These tendencies re-appear as central building blocks for the highly inspiring analyses that Beatrice Manz has pursued for Timurid political history. In her Rise and Rule of Tamerlane she focused almost entirely on this internal dynamic to explain Temür’s exceptional political and territorial successes, from his rise to power within the Turko-Mongol Chaghatay tribal formation to his death as the undisputed leader of “a new imperial dispensation”. For Manz, the secret to Temür’s success lay not just in his charismatic personality and his successful “rule through people rather than institutions”. ¹²⁹ Above all, his achievement lay in “the careful division and limitation of power”. ¹³⁰ Manz explained how Temür managed to de-center Chaghatay nomadic politics, and to defuse the devastating impact of—as Woods had explained—“the centrifugalism or segmentation associated with clan corporateness together with nomadic traditions”, by managing to create an entirely new politico-military elite around his person and around the “charismatic, personal, and bureaucratic concepts of his sovereignty”. This new elite was composed mainly of his “own family and his personal followers” with their armies, thus subordinating the existing political practices, institutions and elites in and beyond the Ulus Chaghatay under an entirely new, trans-regional political order. ¹³¹ This new political order was implemented, fed, and kept under Temür’s control, Manz explained, “by undertaking a war of conquest which kept him and his followers out of the Ulus Chaghatay almost constantly for the rest of his life”. ¹³²

Following this remarkable analysis of how war-making and state-making went hand in hand in very particular ways in Temür’s personal politics, Manz studied the reign of Temür’s son and successor Shah Rukh from a similar perspective in her Power, Politics and Religion in Timurid Iran. In this study, however, Manz expanded her focus to again more explicitly include the state-society relationship that also interested Woods, and Hodgson before him. At the same time Manz engaged critically with some of the assumptions that had informed this bipolar model. Manz mainly argued that the relative absence of war-making during a significant part of Shah Rukh’s reign paradoxically weakened central authority, as this allowed for more autonomous political actors and centrifugal tendencies to reemerge with more power. In this context, however, she explained that any analysis of Timurid politics and state formation should acknowledge the fact that “the towns from which the Timurids ruled their dominions were like an archipelago within a sea of semi-independent regions, over which control was a matter of luck, alliance and an occasional punitive expedition”. ¹³³ In this diverse and multipolar, rather than uniformly bipolar, socio-political context, she claimed that “the division between military and civil responsibility and between city and state may have been less marked than [Hodgson’s] aʿyān-amīr dichotomy suggests”. ¹³⁴ More specifically, power was not the exclusive domain of any one social group or ruler, but was “an individual achievement”. Manz concluded that power “was what one person could make out of a variety of affiliations which sometimes went across the boundaries between military and civil affairs and between the religious and the governmental spheres”. ¹³⁵

At the same time, Manz explored the impact of this empowering mix of centrifugal and centripetal tendencies which co-existed within the relational constructs surrounding individual actors, also framed here as a “multiplicity of conflicting allegiances”. She argued that these prevented clear-cut social boundaries from establishing themselves and “thus promoted the cohesion of society as a whole”. ¹³⁶ Most importantly for the present context, she explained how this fluid social nature of power “helped to attach society as a whole to the ruling elite”. ¹³⁷ For Manz, this in fact almost means that the state-society dichotomy is resolved in favor of an interpretive integration of the state, and of “the practice of politics” more generally, in society. Manz therefore reconceptualized “the central government”—the “Timurid state”, here again identified with the chancery and financial administration and also, in conflicting ways, with military leaderships, dynastic members and their entourages and courts. ¹³⁸ She sees that state as “a source of money, employment, status, and military power, which might be converted into political capital within one’s own region or profession” whereas the usefulness and uses on many levels of central government, and even stakeholdership in government “helped to legitimize it and to further its influence”. ¹³⁹ In fact, Manz prioritized these far more fluid and co-constitutive conceptions of center and periphery as an alternative to the traditional state-society paradigm. They proved useful tools to construct more complex and constructive understandings of issues which, as shown above, Woods had also identified in a similar (and contemporary Turkmen) context as “the absence of powerful central authority”. To this end, Manz again ‘de-centered’ power, this time not from the clan to the individual chief, as in her study of Temür, but, as explained above, from the state or central government and the ruler to a multiplicity of individual social actors. At the same time, Manz acknowledged the implications of this conceptual shift. Proposing a far more minimalist understanding of the Timurid state, she considered it as the messy locus of multiple conflicting interests, resources and center-periphery relationships. In line with her archipelago metaphor that already suggests very vividly what this means for her understanding of the Timurid center, Manz advocated a kind of spatial approach to appreciations of Timurid authority and regional control, explaining that “Timurid government represented a spectrum from relatively direct rule over central regions, under princely governors, to a hopeful fiction of suzerainty over neighboring confederations”. ¹⁴⁰

Hodgson’s historical model of the erratic empowerment of an apparatus of centralizing power in a volatile context of Turko-Mongol rule and nomad-urban symbiosis arguably continues to linger at the background of Manz’ critical interpretations too. That model is also present in very conscious, yet also very different, ways in Maria Subtelny’s Timurids in Transition. Turko-Persian Politics and Acculturation in Medieval Iran. Focusing on the reign of the last Timurid ruler, Sultan-Husayn Bayqara, Subtelny took an explicitly Weberian approach to further conceptualize Hodgson’s notion of the Timurid military patronage state ¹⁴¹ and, more generally, “the process of transition and state formation under the Timurids”, which she pitched explicitly as “the process of transition from a nomadic empire based on a booty economy to a state on the sedentary Persian model”. ¹⁴² Weberian concepts of personal charisma and its “routinization”, of the traditional patrimonial household state, and of the rational bureaucratic state are Subtelny’s basic tools to analyze this so-called process of transition. ¹⁴³ The tension between centrifugal and centripetal tendencies again appears as the driving force behind her understanding of this process. However, in contrast to Manz’ multipolar and minimalist approach and in an interesting parallel to Woods’ understanding of “the political and institutional separation and accommodation” between ‘Turks’ and ‘Tajiks’, Subtelny interpreted this tension not just in maximalist and almost teleological ways, but also in the strictly dichotomous terms of a nomadic-urban polarity, which at the same time she framed within that resilient framework of legalistic models of the state as follows.

Once set in motion, the process of transition created a dialectic relationship between two basic and opposing tendencies—the one centrifugal, represented by the Turko-Mongolian military elite who wanted to preserve the decentralized patrimonial system and values embodied in the törä [Timurid custom]; the other centripetal, represented by the proponents of the Persian bureaucratic tradition who sought to establish a bureaucratic state on the Perso-Islamic model in which the Sharīʿa represented the chief ideological basis for centralization. ¹⁴⁴

Subtelny argued that this dialectic relationship faltered as “the Timurids would ultimately be unsuccessful in effecting […] the routinization of charisma”, and the Timurid transition from patrimonial to bureaucratic state never entirely materialized. ¹⁴⁵ Echoing Woods’ suggestion of “the absence of a powerful central authority” in the face of dynastic volatility, she suggested that the Timurid state, as well as its regional successor states, therefore remained stuck in an intermediate stage between so-called Turko-Mongol traditionalism and Perso-Islamic rationalism, conceptualized here as “a patrimonial-bureaucratic regime at best”. ¹⁴⁶

3.4 Defining the Syro-Egyptian Sultanate’s ‘state’

Unlike in these fifteenth-century Timurid and Turkmen contexts, and to some extent also in Ottoman contexts, most modern historiography of the Syro-Egyptian Sultanate has never really engaged with any of the more sociologically inspired models that have so far been surveyed here. Whereas, as explained above, the writings of Lapidus and Chamberlain have had an invigorating impact in redirecting the fields of Syro-Egyptian social and cultural history, in general this has not happened at all in the field of Syro-Egyptian political history. There, Lapidus’ legalistic statist model of a continuous central bureaucratic apparatus that organized protection and exploitation for Syro-Egypt’s urban centers and military elites has remained the quiet interpretive norm. This tends to be in direct continuity with conceptualizions of the state from before Lapidus, without really engaging with Lapidus’ particular framing of this bureaucratic state as marginalized by urban society due to a so-called “privatization of power”. ¹⁴⁷ Moreover, Hodgson’s model of the military patronage state has arguably not been directly engaged with at all in this fifteenth-century context. This is somewhat understandable, given the origins of this model in Hodgson’s readings into Mongol and post-Mongol history and his rather peripheral and therefore also superficial, interest in the Sultanate. What is remarkable, however, is the fact that Hodgson’s impression of a very strong state-society polarity—as manifested in his somewhat simplifying claim that “the great Mamlûk amîrs formed an exclusive oligarchy” and “the civilians of Cairo [were] despised and permanently excluded from power”— ¹⁴⁸ remains an unqualified assumption in many, if not most, scholarly engagements with the Sultanate’s state, its institutions, practices, norms and actors, and their eventful fifteenth-century histories. ¹⁴⁹

Nevertheless, there are a handful of recent empirical insights that are worth referring to here briefly as promising avenues for further, more sociologically informed, reflection. This overview comes with the caveat that, conceptually, most of these studies equally continued, and continue, to situate themselves rather vaguely within the legalistic tradition’s structural models of absolutist bureaucratic power. These recent insights mostly involve the remarkable centralizing empowerment of the sultan’s household in the fifteenth century which scholars have seen as a function of its growing socio-economic weight, impact and autonomy and of the changing nature of its agents and their relationships. This approach is present in a survey chapter by Jean-Claude Garcin and also especially in detailed studies by Julien Loiseau and Francisco Apellániz. ¹⁵⁰ In Reconstruire la maison du sultan Loiseau has an interesting (albeit mainly empirical) way of describing the rise to prominence in the first half of this century of royal household officials and families of administrators (“l’avènement des grands commis civils”). Here he demonstrated how these officials appeared as increasingly autonomous royal household agents and, eventually, bureaucratic leaders, empowered by their success in expanding royal income and investments, ultimately becoming active stakeholders in “a new Mamluk order (‘un nouvel ordre mamelouk’)”. ¹⁵¹ Apellániz engaged with a similar model of fifteenth-century central socio-economic transformation. In his monograph Pouvoir et finance en Méditerranée pré-moderne he focused in particular on the “commercialization” of the Sultanate’s political economy, with another set of new royal agents and partners (including the Venetians) tapping into new sources of royal income and simultaneously opening up new avenues of political power along parallel old (military) and new (commercial) structures. However, in analyzing this kind of ‘layering of organizations and practices’ (as the aforementioned Ottomanist Barkey termed this phenomenon of parallel structures in the Ottoman context), Apellániz argued that these generated opposite, decentralizing tendencies towards the end of the century. ¹⁵²

Here we see at least an allusion to the challenging analytical horizons of the fifteenth-century Syro-Egyptian process of state formation, from the manifestation of deep socio-economic transformations and a widening political participation to the formation of a new political order and its construction out of variously interacting centripetal and centrifugal tendencies.

The most explicit theorization of these developments is arguably to be found in a joint publication by Van Steenbergen, Wing and D’hulster. “The Mamlukization of the Mamluk Sultanate? State Formation and the History of Fifteenth Century Egypt and Syria” presents an argument for modelling these different analytical horizons as “the growth of the state, both as a non-dynastic idea of hegemonic political order and as a coercive bureaucratic apparatus that was produced by, and that was set up to reproduce that order”. ¹⁵³ This conceptualization was inspired by Ibn Khaldun’s notion of the sedentarization of nomad power and by Pierre Bourdieu’s model of transformation “from the king’s house to the reason of state”. It sees the changing loci of central power in the fifteenth century as the manifestation of a radical social transformation. This again involved the above-mentioned idea of ‘layering’, but now in a social rather than institutional way. The range of empowering relationships was expanded during this period from personal service being given to individuals or households to specialized service being offered to the sultan’s court in Cairo, and to its agents and local offshoots, and their diverse interests. Further, “The Mamlukization of the Mamluk Sultanate?” suggested that this social layering was “part of a broader development in which the messy paths to power and influence, to a share in the central state’s political system, became ever more meritocratic and bureaucratized”. The paper also claimed that this involved, almost by default, the performative imagination of that central state in increasingly meritocratic, non-dynastic, and ‘Mamluk’ terms. ¹⁵⁴

In many ways, this theorization of fifteenth-century Syro-Egyptian social transformation (and its ‘Mamlukization’, as a form of ‘meritocratization’) builds on various scholars’ re-interpretations of thirteenth- and fourteenth-century Syro-Egyptian state formation through the lens of elite households, dynastic practices and Hodgson’s model of the military patronage state. ¹⁵⁵ Van Steenbergen also elaborated on the issue in two more generalizing theoretical essays which complement each other in their respective diachronic and spatial approaches to the theoretical problem of Syro-Egyptian state formation under the Cairo sultans between the thirteenth and early sixteenth centuries. “Appearances of Dawla and Political Order in late medieval Syro-Egypt” makes an explicit attempt to connect with the cultural turn in state studies, and with Chamberlain’s prioritization of social practice. This essay also pays particular attention to Mitchell’s above-mentioned Foucauldian understanding of the state as a structural appearance and “as the powerful, metaphysical effect of practices that make [actual] structures appear to exist”. ¹⁵⁶ Following Chamberlain, “Appearances of Dawla” proposes a useful way out of the longstanding state-society predicament and the legalistic structuralism that it implied by advocating for an analytical turn towards social practice. It is suggested that priority should be given to exploring practices of reproduction, integration and segmentation among Syro-Egyptian power elites, and that, as Mitchell suggested, this should be coupled with an interpretation of representations of statist institutions, practices and discourses as both structural and structuring effects of what these elites did. This is then suggested to imply that we should accept as an entry point for any historical analysis and interpretation the messy micro-historical diversity of elitist actors and agencies in which these practical arrangements of reproduction, integration and segmentation and their effects continued to manifest themselves. On this basis, “Appearances of Dawla” claimed that scholars ought to approach the long history of the Sultanate as a discontinuous one, in which a preponderance of state-making over the messiness of war-making happened only “in contexts of successful, relatively long and charismatic authority”. In the long fifteenth century, these contexts are suggested here to have emerged especially during the reigns of the sultans Barsbay and Jaqmaq in the 1430s and 40s, of Qaytbay in the 1470s and 80s, and of Qansawh in the 1500s. ¹⁵⁷ “An alternative model for understanding late medieval Syro-Egyptian political organisation that emerges from all this”, this essay concludes more generally,

is then one of sultanic political order—the state—as process, in constant flux as the structural effect and structuring embodiment of constantly changing practices of social reproduction, elite integration and political distinction, in contexts that range between multipolar and unipolar social organisation at and around Cairo’s court and its military elites. ¹⁵⁸

Furthermore, in the essay “Revisiting the Mamlūk Empire. Political Action, Relationships of Power, Entangled Networks, and the Sultanate of Cairo in late medieval Syro-Egypt”, a transregional and at the same time socially constructed center-periphery dimension is added to the diachronic model of Syro-Egyptian state formation that was formulated in “Appearances of Dawla”. Taking inspiration from, amongst others, Barkey’s aforementioned rethinking of the question of Ottoman longevity, “Revisiting the Mamlūk Empire” suggests thinking of Syro-Egyptian power groups and their patrimonial leaders as different bundles “of diverse but entangled networks, that appear to operate—or present themselves as operating—on a trans-local canvas of connectivity”. However, these networks were “always constructed in the micro-history of people and their negotiation of particular cultural, socio-economic and political relationships”. As a result, these relationships did not just connect central and peripheral elites (as Barkey suggested for the Ottoman case). They also actually constituted different groups, or entangled networks, as such central and peripheral elites, in continuously disputed and reconfigured ways. Furthermore, these particular sets of centering relationships always appeared as “permeable and as crisscrossed by many other, equally fluid, relational realities that seemed to be pulsating from other local or trans-local centers of social, economic, cultural or even political action”. ¹⁵⁹