Diversity in the past

Several recent works on historical international relations have expanded IR's historical and geographical horizons beyond traditional disciplinary ‘benchmark’ dates such as 1648 Europe.Footnote ¹⁵ Such studies revealed hitherto neglected alternative world orders,Footnote ¹⁶ demonstrating how practice and agency differs across time-space,Footnote ¹⁷ whilst also helping us challenge long-standing ontologies of world politics by better appreciating the significance of diversity and relations. Key themes in this literature include discussions of order and disorder in world politics, especially in relation to hierarchical political units. World order, as defined by Ayşe Zarakol, should be understood as ‘the (man-made [sic]) rules, understandings, and institutions that govern (and pattern) relations between the actors of world politics’.Footnote ¹⁸ Zarakol's volume on ‘Chinggisid’ forms of world order forces us to rethink traditional understandings of order that are centred on Westphalian-like sovereign states.Footnote ¹⁹ Zarakol demonstrates that order is possible in settings where other forms of sovereignty are dominant, including – as is the case with the Ottomans, Safavids, and Mughals – when sovereignty is conceived by populations linked to the Eurasian steppe. This is significant because many have assumed that an apparent lack of hierarchical socio-political units on the steppe indicates political disorder.Footnote ²⁰ IR, following realist thought, has commonly assumed that hierarchy is order whereas its absence is disorder.Footnote ²¹ This is, for Spruyt, a consequence of a preference for positivism that leads to interpretations of history being based on preconceptions of what should be found rather than on an understanding of the actual context, actors, and contingencies involved.Footnote ²² Heterarchical systems, meanwhile, ‘remain outside positivist purview’.Footnote ²³

Zarakol's focus is primarily on the ‘great powers’ (so to speak) of the post-Chinggisid empires. Andrew Phillips and Jason Sharman,Footnote ²⁴ meanwhile, demonstrate that international order is also possible when multiple diverse polity-forms are present. Their study of the Indian Ocean reveals that order can stem from diversity, thereby challenging assumptions that durability and order are only possible when the international is dominated by Weberian sovereign states. Elsewhere, Phillips notes a ‘general pattern’ of peripheral polities overcoming core polities to establish empires in processes involving local collaborators and diverse forms of political organization.Footnote ²⁵ Other recent studies similarly provide examples of times and places wherein the international involved a myriad of actor-forms playing important roles in historical settings.Footnote ²⁶ Importantly, these undermine assumptions that Western global predominance dates back to the fifteenth century, whilst simultaneously revealing the global significance of Asian polities until at least the mid-eighteenth century.Footnote ²⁷ Such broader perspectives on historical international relations reveal that ‘most international systems have been defined by durable diversity’.Footnote ²⁸ This is in contrast to realist, rationalist, sociological institutionalist, and constructivist assumptions that international relations lead to unit-type homogenization and conformity.Footnote ²⁹

This links to a second theme in the new historical IR literature: the integral role of ‘liminal’ and nomadic populations in world orders. Again, the steppe was the geographical source of many of these liminal polities but their world ordering incorporated and transformed polities along almost its entire perimeter. What are now Russia, Turkey, Iran, Pakistan, India, and China were all subject to steppe-derived political systems.Footnote ³⁰ Alan Kwan, for example, reveals non-Westphalian ‘international systems’ based on ‘adaptable hierarchy’ in pre-modern China and its neighbours.Footnote ³¹ Notably, many of Kwan's liminal actors are ‘nomadic’, thereby demonstrating the significance of polities that differ from the sedentary geodemographic norms of traditional thinking in the social sciences that has equated nomadism with ‘backwardness’ and ‘barbarism’ rather than the supposed ‘civilization’ of sedentary states.Footnote ³²

A third theme is the hybridization of political forms as a result of such contacts and conquests involving liminal polities and sedentary ‘cores’. This underlines the relational ontologies of much of this literature; an aspect also prevalent in other disciplines that have explored global inter-polity relations across time-space.Footnote ³³ The Mughals, Chinese Yuan and Qing dynasties, and the British Raj all demonstrated aspects of hybridization as a result of contact between different polities.Footnote ³⁴ The Bulgarian state ‘emerged as a result of a merger between tribute-takers from the steppe … and local tribute-giving tribes’.Footnote ³⁵ Imperial hierarchical systems depended on nurturing social ties with local collaborators.Footnote ³⁶ Relations between polities involve mutually constitutive influences, as Heather Rae demonstrates in the case of contemporaneous relations between Europe and the Aztecs in one hemisphere and with the Ottomans in another.Footnote ³⁷ Similarly, polities are often composites, involving multiple identity groups whose differences are sometimes managedFootnote ³⁸ and frequently at the root of significant political change.Footnote ³⁹ When imperial administrations sought to forcefully assimilate populations it often led to revolt, and even despite military victory over the rebels, empires such as the British in India reformed their means of governing diversity by becoming more inclusive of collaborators in their approach.Footnote ⁴⁰

These studies of historical international relations reveal that clearly demarcated territorial political units are historically contingent,Footnote ⁴¹ whilst their relational ontologies allows them to recognize the significance of actors beyond the Weberian state. They thereby expose the fallacy of IR theory's essentialization of a state system to ‘advance an ontological view of the international system as consisting of discrete and mechanically interacting elements’.Footnote ⁴² Nevertheless, their historical focus is primarily on post-Genghis Khan Eurasia which, although vital for their challenging of traditional Eurocentric narratives, remains insufficient to meet Phillips' and my calls to appreciate successive waves of hemispheric interaction and integration before 1500 or the deeper past.Footnote ⁴³

Seeking the ‘first’ international

Studies of the deeper past in IR remain scarce, although there have been notable attempts to identify a ‘first’ international. Exploring the earlier international(s) is challenging for a discipline that has been labelled ‘tempocentric’ for its fixation on the near-past.Footnote ⁴⁴ Nevertheless, claims around the ‘first’ international have established a dominant disciplinary narrative on the ancient world in IR that privileges the emergence of the state and relations between states. These works include realist claims that the ‘Amarna system’ of the mid-fourteenth century BCE was the ‘first international system’.Footnote ⁴⁵ As part of either broad trans-historical analyses of balance of powerFootnote ⁴⁶ or multidisciplinary analysis of early statecraft,Footnote ⁴⁷ such works helpfully nudge IR beyond the de-facto historical and geographical boundaries of Thucydidean Greece. However, realists offer little on relations beyond the inter-state. Significantly, once the state emerges the logics of relations are unchanging, as demonstrated by Waltz's claim that the ‘texture’ of international relations does not alter over time because ‘patterns recur, and events repeat themselves endlessly’.Footnote ⁴⁸

World Systems Theory (WST), a sub-field of several social sciences, also explores the deeper past. Following Immanuel Wallerstein,Footnote ⁴⁹ WST emphasizes causal links between the whole (world) and its constituent (local) parts over long timespans. Both local and global levels share the same causalities,Footnote ⁵⁰ contradicting those who insist on distinctive causalities for each level.Footnote ⁵¹ Studies are broad in both temporal and geographical scope, including suggestions of a 5000-year-old world systemFootnote ⁵² spanning ‘southern Central Asia, the Harappan civilization in the Indus valley, the Persian and Anatolian plateaus, Mesopotamia between them, and Egypt’.Footnote ⁵³ Sumer is a ‘core’ for broader ‘peripheries’, with Early Dynastic Mesopotamia (c.2900–2350 BCE) marking the ‘earliest emergence of a centre/periphery structure’.Footnote ⁵⁴ The approach has been employed by archaeologist Guillermo Algaze to explain how one Mesopotamian city, Uruk, rapidly expanded in size and cultural-economic influence from c.3500 BCE; the so-called ‘Uruk expansion’.Footnote ⁵⁵ However, the global level takes precedence: world-systemic relations trump individual and local agency. Scholarship therefore focuses on the division of labour at a world-systemic level rather than local contingencies.Footnote ⁵⁶ Peripheries are deprived of agency, with military, political, and economic domination by the core understood as absolute.Footnote ⁵⁷ Therefore, despite the promise of WST to think big in historical terms, it remains hamstrung by its neglect of micro-level causalities.

The English School also identifies Sumer. Adam Watson saw it as the ‘original states system’ due to it being ‘a cluster of separate communities within the framework of a common culture, each with its own distinct personality and corporate life’.Footnote ⁵⁸ Cultures different to this ‘common culture’ are, by implication, not part of the system, suggesting consistency with the English School's reading of present-day global politics. Differences, both cultural and in polity-type, are written out. Sumer also represents the ‘first full international system’ for Barry Buzan and Richard Little.Footnote ⁵⁹ Underlying their narrative are three evolutions in social organization: sedentism; the emergence of social hierarchy; and, finally, differentiation.Footnote ⁶⁰ Sedentism is claimed to have led to population growth, a defining factor in a social evolutionary ideal-type model that progresses from egalitarian hunter-gatherer bands (HGBs), to ‘tribes’, to more hierarchical ‘chiefdoms’, and, finally, ‘states’ and/or ‘empires’. When growth led to a settlement exceeding its optimum size, the population divided as some moved to establish a new settlement.Footnote ⁶¹ Such a multiplication of societal units from a common original village-based community produces another ‘level’ of social organization above the singular village whilst remaining ‘inside’ the social collective. This generates a hierarchy of communities dominated by a ‘paramount village’ that controls subsistence resources, typically due to being located near a strategic resource.Footnote ⁶² Buzan and Little propose a ‘feedback loop’ from the intensification of agriculture, through the acquisition of wealth, to the emergence of hierarchy and the state.Footnote ⁶³

Crucially, it was only the first city-states that were ‘sufficiently hierarchical in their internal organisation to generate “inside” and “outside” political realms’,Footnote ⁶⁴ and it is with their emergence that we find the transition from the ‘pre-international’ to the ‘international’.Footnote ⁶⁵ City-states were ‘more specifically military-political’ than previous units and consequently more sufficiently durable to survive the rigours of an international system.Footnote ⁶⁶ Durability lies not in diversity but in uniformity: a single ideal-type of political unit. It is only with the emergence of such hierarchical political entities that we arrive at the ‘full range of nested actors’ engaged in political, economic, military, and socio-cultural relations that is necessary for a ‘full international system’ to emerge.Footnote ⁶⁷ While Buzan and Little recognize the ontological primacy of relations, it is specifically conflict between city-states that make Sumer ‘a fully-fledged international system’.Footnote ⁶⁸ They downplay also-mentioned conflicts involving ‘nomadic pastoralists’ or ‘mountain tribesmen’.Footnote ⁶⁹ Indeed, ‘fully-fledged international systems … began with the rise of city-states’Footnote ⁷⁰ and the type of relation (such as conflict or trade) is less important than the type of actor(s) engaged in those relations. Thus, we arrive at what can be called the international's ‘big bang’ moment of emergence. Without the relations between the city-states, there are no ‘international relations’ and no foundational moment from which theory-making might begin. Conversely, recognizing the agency of actors different to the state ideal-type – for example, when they engage in the exact same types of relations as the states – would imply that the ‘international’ existed before the state.

Recent Trotskyist scholarship extends this analysis but falls into the same trap. Justin Rosenberg argues that political multiplicity – that is, a plurality of distinct but interacting political entities – emerged from a process of uneven and combined development.Footnote ⁷¹ At the point of origin, Rosenberg's international is a quantitative multiplicity of a specific type of unit rather than a multiplicity that includes multiple actor forms. Even if Rosenberg uses the term ‘the political’ rather than ‘states’, it is state formation,Footnote ⁷² ‘the third stage of Buzan and Little's developmental sequence’, that marks the emergence of ‘the political’ for him.Footnote ⁷³ Liminal populations are absent. Moreover, he prioritizes a specifically sedentary, urban, and hierarchical understanding of the state. In keeping with Buzan and Little, agricultural surpluses of sedentary communities could be accumulated and exchanged, with such exchanges subsequently generating differential access to prestige goods within social units.Footnote ⁷⁴ These goods were essential factors in the ‘consolidation of internal hierarchies through which “the political” emerges’.Footnote ⁷⁵ For Rosenberg and Buzan and Little, the existence of early states stimulated reactive proto-state formation elsewhere,Footnote ⁷⁶ thereby proliferating a multiplicity of states and expanding the international. Relations between ‘multiple interacting societies’ are therefore the engine of multiplicity and, consequently, of the international itself.Footnote ⁷⁷ In this regard, international multiplicity ‘imparts its own dialectical mechanisms and dynamics to the structure of world history’.Footnote ⁷⁸

Rosenberg justifies his choice of Sumer as the earliest ‘society’ in his ‘multiplicity’ due to the transition to sedentary-agricultural lifestyles in the region.Footnote ⁷⁹ This transition ‘alters the … interactive logics of social reproduction and development’.Footnote ⁸⁰ The ‘spatial logics of security’ were consequently changed to favour nucleation over dispersal, and the international's emergence again relies on a typological shift to an ideal-type actor. Yet Rosenberg also recognizes that the international ‘crystallizes within a pre-existing social landscape of interactive multiplicity’.Footnote ⁸¹ Thus, there may have been an ‘interactive multiplicity’ before the international and there may not have been a ‘big bang’. This suggests a contradiction in Rosenberg's thesis. The international emerges because of the state and is characterized by the five ‘consequences’ of multiplicity (co-existence, difference, interaction, combination, and dialectical changeFootnote ⁸²). However, those same five consequences must exist regardless of the state if they were instrumental to its emergence. To frame this in relationalist terms, the international is contingent on the emergence of a substance – the hierarchical, sedentary city-state – rather than on relations (the five consequences). Agent-type homogeneity (in the form of the state-society) rather than relations between diverse forms is the defining factor on which the existence or not of an international depends.

This has ontological implications regardless of time and geography. The international is construed as a substance defined by fixed characteristics: it must contain states. To paraphrase Patrick Jackson and Daniel Nexon,Footnote ⁸³ the state is the key unchanging and unquestionable ‘constitutive property’ of the international as a substance. Therefore, IR's dominant narratives of the ‘first’ international are ‘substantialist’.Footnote ⁸⁴ Because the international is determined by the emergence and existence of states, the starting point for analysis becomes the state's emergence as the constitutive property of the international. Relations are secondary. This is reflected in arguments that proliferation comes from differentiation and dispersal following nucleation within a given culturally homogenous population,Footnote ⁸⁵ as well as suggestions of a singular originating culture.Footnote ⁸⁶ Such thinking rules out: centripetal growth from migration or combination; constitutive interaction with outsiders; and ongoing processes of division (what David Graeber and David Wengrow call ‘schismogenesis’Footnote ⁸⁷). The dominant narratives imply that units are bounded and impermeable rather than being subject to the ‘multiple overlapping and intersecting socio-spatial networks of power’ identified by Michael Mann.Footnote ⁸⁸ These narratives' conceptual model is therefore centrifugal: a single culture expanding outwards rather than one that is simultaneously subject to the generative influence of relations with diverse populations and open to the organizational hybridization identified in other studies of historical IR. In true substantialist fashion, the international is understood to exist before relations, rather than as emerging because of and through relations.

The Mesopotamian trap: evolutionary illusions and methodological blindness

Why is there this privileging of the sedentary urban state? The answer lies in how genealogies of modern Western academic disciplines predisposed them towards the royal propaganda of Sumerian rulers millennia ago. Social scientists often lack the skill sets necessary to access material evidence from the past, especially the deeper past, making interdisciplinary scholarship essential. Therefore, social scientists rely on other disciplines for primary research on the deeper past, especially archaeology and anthropology. Interdisciplinary engagement, however, demands an awareness of other disciplines' epistemological and ontological frameworks. A failure to understand other disciplines' frameworks risks unwittingly importing their weaknesses and biases into the social sciences to consequently undergird their interpretations of the social world.

Two interrelated problems have shaped IR's understanding of both Sumer and the ‘first’ international. The first is the importation by IR of an essentialist-colonial ontological framework developed by nineteenth- and twentieth-century Western archaeologists and anthropologists. For instance, Buzan and Little's analysis rests on the typological HGB-tribe-chiefdom-state social evolutionary model. Despite them not citing influential anthropologist Elman Service, Service proposed an identical band-tribe-chiefdom-state model 25 years earlier.Footnote ⁸⁹ Service claimed that the chiefdom-to-state transitional moment was the ‘great divide’ in the ‘evolution of human culture … when primitive society became civilized society’.Footnote ⁹⁰ Having a state equated to becoming ‘civilized’. Service believed that:

HGB-to-state models are part of a long, if contentious, interdisciplinary tradition of social evolutionary thinking. Many highlight the colonial roots of Service's framework, alongside those of fellow ‘neo-evolutionists’ Leslie White and Julian Steward.Footnote ⁹² Social evolutionary approaches derive from nineteenth-century anthropological attempts to classify populations according to imagined forms of socio-political organization. These include Lewis Henry Morgan's seven ‘ethnical stages’Footnote ⁹³ that legitimized colonial administration over ‘inferior’ peoples.Footnote ⁹⁴ Typological categorization was central to this process. According to Morgan, ‘civilization’ was achieved by a society through the development of writing. ‘Civilization’ marked the seventh and highest of his technologically determined ‘ethnical stages’, below which are (in descending order): ‘upper’/‘middle’/‘lower’ ‘barbarism’, and ‘upper’/‘middle’/‘lower’ ‘savagery’.Footnote ⁹⁵

Among Service's typologies, ‘tribe’ is problematic ‘because of certainly pejorative, if not outright racist, implications’, in addition to a lack of archaeological evidence to justify its universal application.Footnote ⁹⁶ For example, the mobile populations of the Zagros mountains of West Asia and Eurasian steppe nomads are commonly grouped together as ‘tribes’, despite significant differences in their socio-political composition.Footnote ⁹⁷ Others claim that the term ‘tribe’ ‘was so fundamental to colonialist discourse and the devaluing of “the other” that it [cannot] be used as an academic designation without perpetuating this practice’.Footnote ⁹⁸

Ideal-types such as ‘chiefdom’ and ‘state’ are heuristic devices intended to scaffold disciplinary enquiry, facilitating theory-building by simplifying diverse forms of socio-political organization. Buzan and Little appreciate that their terminology ‘mask[s] an enormous amount of variation’ within each ideal-type,Footnote ⁹⁹ but is nevertheless justifiable in order to ‘tell the story of the pre-international in a coherent fashion’. Similarly, there is no ‘overly determined story of evolution leading from one to the other and eventually to states and international systems’: not all HGBs become ‘tribes’ just as not all ‘tribes’ become ‘chiefdoms’ or ‘chiefdoms’ states. Therefore, Buzan and Little fail to recognize that any ideal-type is ‘dependent on the different purposes at hand, that is, the different value perspectives providing the puzzles that have initiated the study’.Footnote ¹⁰⁰ They ‘accentuate those aspects of the empirical case of particular interest to the researcher’.Footnote ¹⁰¹ In their case, the ‘purpose at hand’ is to narrate the international's emergence as part of a narrative of state-centric ‘international systems’ that are comparable to what we have today. Such exclusion of variation across cases in favour of ideal-types embeds presentism and the discriminatory practices that were fuelled by these ontologies.

The ‘state’ ideal-type does a lot of interdisciplinary heavy-lifting. Disciplinary History, for example, often associates the state and its emergence with ‘civilization’, whereas ‘barbarism’, ‘savagery’, and ‘uncivilized’ are typical both before and outside the state.Footnote ¹⁰² Michael Mann's influential Sources of Social Power is rife with references to ‘noncivilized’ and/or ‘primitive’ populations.Footnote ¹⁰³ Civilization is defined as combining the three ‘social institutions’ of ‘ceremonial centre, writing, and the city’,Footnote ¹⁰⁴ and is a crucial prerequisite for Mann's concept of ‘social caging’. This echoes prejudice against ‘stateless’ polities in ‘first’ international literature. Watson called populations neighbouring Sumer ‘wilder immigrant peoples’.Footnote ¹⁰⁵ They apparently lacked ‘a highly developed civilization of their own’ and were consequently receivers of ‘the advanced culture of Sumer almost entire’, including the ‘Sumerian tradition of statecraft’.Footnote ¹⁰⁶ Buzan and Little do not entirely write-out ‘stateless’ polities: they include an overview of the influence of ‘nomadic tribes and their empires’,Footnote ¹⁰⁷ and they recognize that nomads are neglected in IR scholarship.Footnote ¹⁰⁸ Nomads ‘played a crucial part in international relations’ due to the Mongol Empire and the spread of Islam. The term is also recognized to signify diverse forms of ‘tribal’ lifestyles, including groups that were sometimes part of the same economic system as sedentary populations.Footnote ¹⁰⁹ Nevertheless, their chapter devoted to nomadic groups evokes Morgan's ‘ethnical stages’ by labelling such actors as ‘barbarians’ in its title.Footnote ¹¹⁰ Echoing Service, who wrote of the threat to ‘civilization’ from ‘nomadic raiding bands of predators’,Footnote ¹¹¹ and Watson's ‘wilder immigrant peoples’, Buzan and Little disparage nomads as threats to sedentary populations and as constantly ‘feuding’.Footnote ¹¹² Crucially, nomad-derived alternative forms of sovereignty, as highlighted by Zarakol, are unexplored.

Social evolutionary models also conflate unit-type with temporality. Time becomes framed according to typological classification rather than by chronological measurement.Footnote ¹¹³ Thus, polities contemporary in time can be defined as ‘backwards’ or ‘advanced’, ‘developed’ or ‘undeveloped/developing’, depending on whether they correspond to the contemporary understanding of a ‘state’. Thereby, modern stateless ‘hunter-gatherers’ are presented as analytical equivalents of ancient communities.Footnote ¹¹⁴ This is what anthropologist Johannes Fabian calls ‘typological time’.Footnote ¹¹⁵ Such thinking falsely equates ‘states’ in the present with ‘states’ in the past, emphasizing common features across time that permit classification but ignore historical contingencies. For instance, despite social scientific claims of Sumerian ‘states’, neither the Sumerians nor their Akkadian successors had words for ‘state’, instead using their nouns for ‘settlement’, regardless of size (Sumerian: ur; Akkadian: alum).Footnote ¹¹⁶ Palaeontologist and evolutionary biologist Steven Jay Gould blamed such thinking on the ‘iconography of an expectation’ that dominates evolutionary studies for such practices: developmental ‘tree’ diagrams that strip populations of their historical contexts.Footnote ¹¹⁷ Modernity is the ‘hidden grand narrative’ of such schema, regardless of their actual location in chronological time.Footnote ¹¹⁸ State-building is consequently associated with ‘progress’ and ‘modernity’.Footnote ¹¹⁹ Persisting with social evolution frameworks is to base the international's creation narratives on ahistorical ‘typological time’, perpetuating ignorance of polity diversity. Thus, the full complexity of relations and diversity of actors that constitute global politics over time remain obscured.

The second of these interrelated problems is a pervasive, deep-rooted presentism across the social sciences. Presentism involves interpreting the past as if it operated according to the logics of the present, whilst simultaneously prioritizing aspects that resemble the present. The present is the ideal-type, with alternative historical possibilities and ways of doing being ignored.Footnote ¹²⁰ It includes viewing the past an ‘inverted form of path dependency’,Footnote ¹²¹ a Whig history focused only on signs of the emerging present, rather than remaining open to what might be found. Nineteenth- and twentieth-century dialogues between archaeology and the social sciences reinforced the presentist foundations of many disciplines. Ideas from canonical social scientific thinkers, themselves often informed by colonial archaeologists, influenced early archaeological fieldwork. This includes an archaeological preoccupation with remains that seemingly confirm aspects of the Weberian state.Footnote ¹²² In Mesopotamian archaeology, a ‘pervasive materialist sensibility’ grounded in Marxist and capitalist notions of wealth and complexity produced assumptions that the quantity of urban material remains meant that social complexity was exclusive to cities.Footnote ¹²³ Other consequential influences include Weber and Engels' assumptions around the absoluteness of the monarch's power in Sumerian city-states;Footnote ¹²⁴ Weber's stress on a monopolization of the use of force and its encouragement of an understanding of the state that is separate from its members;Footnote ¹²⁵ and Karl Wittfogel's infamous ‘hydraulic-bureaucratic state’ and ‘oriental despotism’.Footnote ¹²⁶ Presentist social science therefore created expectations of which relations and power structures should be found in the ancient world. These expectations consequently influenced archaeological data collection,Footnote ¹²⁷ with everything framed by the European colonial ontologies of the time.

Presentism is encouraged and archaeological fieldwork is facilitated by the material visibility of certain relics from the past, notably ancient cities and their textual paraphernalia. The architecture of cities, including their monumental walls and palaces, allow Buzan and Little to claim them as ‘states’ and consequently the founders of the original ‘international’. Many ancient cities resonate with the urban lifestyles of modern literate scholarly communities, their ruins often embedded in the university cities of today and, consequently, in the lived experiences of researchers. Sumerian cities are visible through stone or mud-brick ruins, supported by texts, inscriptions, and pictographic monuments. These superficially conform to modern Western notions of how a city and its supporting social, political, economic, and cultural infrastructure should appear. Contrast these with the sprawling contemporaneous urban ‘mega-sites’ of Ukraine with their alien lack of temples, palaces, and fortifications.Footnote ¹²⁸ These are notably absent from most Western public and academic imaginaries. However, the apparent familiarity of other ancient cities encourages presentist ‘fantasies about the past’ involving ‘retrospective desire[s] for and misapprehension of things as they never quite were’.Footnote ¹²⁹ This has been reinforced by the means of presenting such discoveries to Western populations: by ‘relocating’ them to Western museums and capitals, themselves designed in part as symbols of imperial mastery over global pasts and presents.Footnote ¹³⁰ Material remains and our conceptual frames of interpretation are intertwined, whilst simultaneously being subject to the political–cultural currents and global orders of the moment.

Ruins and textual fragments are, however, mere crumbs from their originating times and cultures. Interpreting them is not straightforward. Palaces, temples, ziggurats, and city walls are spectacular compared to other ruins. Such monumental architecture was designed to endure by leaders to communicate their authority across timeFootnote ¹³¹ or even to ‘halt time’.Footnote ¹³² Architecture ‘is one of the most powerful instruments of political propaganda for presenting the world view of the ruling powers’.Footnote ¹³³ It is deliberate political messaging by its creators. When we engage with it, however mundanely, we are ensnared in the power relations of its creation. But many buildings were multi-use, simultaneously providing a service (such as public administration or defence) and communicating political messaging from its sponsor(s). City walls, for example, are sometimes assumed by studies of the early international to be obvious signifiers of warfare.Footnote ¹³⁴ However, they might equally be symbolic assertions of authority over their own population,Footnote ¹³⁵ or mechanisms of fanning popular fears of an imagined external threat. Monuments more generally have complex temporal implications, designed to remind people of some event, person, or deity across time.Footnote ¹³⁶ For example, Gudea, a late-third millennium ruler from the Mesopotamian city of Lagash, went so far as to commission a statue of himself in diorite sourced in Oman so that no local successor had the ability to rework it.Footnote ¹³⁷ In choosing such durable materials for their monuments, Sumerian rulers were therefore aware of their diachronic reach.

Together with architectural remains, textual relics have also been dominant influences on our understandings of early political orders. Much of what is debated in the humanities and social sciences from ancient Greece and Rome derives from written texts. Writing is also the main source of data from Sumer, thanks to the durability of clay, the principal early writing medium.Footnote ¹³⁸ Clay was easily prepared and, when dried, hardens to such a degree that discarded tablets were used as construction material, incidentally preserving their text in the fabric of ruins and providing valuable data.Footnote ¹³⁹ But mobile pastoralist populations needed different forms of record-keeping because clay tablets, often weighing several kilograms, were impractical for life on the move.Footnote ¹⁴⁰ Alternative media included systems based on sticks, knotted string, or bullae.Footnote ¹⁴¹ Such information technologies are known to have persisted in nominally ‘literate’ regions for centuries after the invention of writing in those areas.Footnote ¹⁴² However, the more portable, organic, and less monumental material remains of communities without writing attract little archaeological attention, being discarded or even destroyed by archaeologists in search of co-located remains from more fashionable periods, peoples, or social classes,Footnote ¹⁴³ or simply rotting away over time.

Notwithstanding these material limitations, evidence of writing remains fundamental to many modern academic disciplines. For disciplinary History, the advent of writing marks the boundary between history and prehistory;Footnote ¹⁴⁴ a time beyond disciplinary-historical inquiry. Archaeology is problematically sub-divided into a ‘prehistoric archaeology’ of pre-literate peoples and ‘historic archaeology’ of literate populations.Footnote ¹⁴⁵ For the social sciences, writing's emergence is often inseparable from the state and, by extension, the international, making writing's emergence a foundational disciplinary event just as it is for History.Footnote ¹⁴⁶ The emergence of writing determines Watson's choice of Sumer as his ‘original states system’,Footnote ¹⁴⁷ whilst the written ‘data’ of the Amarna letters allow Cohen and Westbrook to claim it as the ‘first international system’.Footnote ¹⁴⁸ As Porter notes:

Thanks to presentist readings of the past, writing is therefore one of the ‘constitutive properties’ of the substantialist state. Sumerian clay tablets play to methodological biases in modern academia to become the literal temporal boundaries for disciplinary theory-building.

Defining an ancient polity as ‘literate’ is, however, elite-centric, denying the illiteracy of the majority along with the ethnic, gendered, and class power relations this presumes. Surviving texts from West Asia and the Mediterranean are almost exclusively written from a sedentary perspective, providing the modern reader with only a partial representation of the time.Footnote ¹⁵⁰ Accountancy records reflect the biases of urban elite bureaucrats, foregrounding their ways of recording and consequently limiting our understanding of less bureaucratically organized parts of the economy.Footnote ¹⁵¹ Much surviving textual evidence ‘is largely the self-representation of the elite aimed at asserting their authority’.Footnote ¹⁵² The material that has survived is often from later Babylonian versions, rather than the original Sumerian or Akkadian versions. There is therefore a double-layer of selectivity involved in their preservation. First, the texts were originally intended for literate elite Sumerian and Babylonian audiences. Second, they survived in their Old Babylonian iterations as school texts, meaning that they tell us ‘more about what teachers wanted their students to learn at the time than what was actually composed and performed [in Sumer]’,Footnote ¹⁵³ supporting the preservation of an ideal of kingship rather than necessarily reflecting actual practices of at the time of the Sumerians.

This is also true of various image-based texts from Sumer. Stone stelaeFootnote ¹⁵⁴ blend the textual with the monumental, with military victories over mountain peoples being notable among surviving examples. These include the stele of Naram-Sin – today found in the Louvre, Paris – depicting an Akkadian military victory over the Lullubi, a mountain people.Footnote ¹⁵⁵ Sumerian texts, both written and in stele form, construct a Manichean cosmology of civilized urbanism in competition with the dangerous, but sometimes tempting, wilds beyond.Footnote ¹⁵⁶ They portrayed settled regions, including walled cities and surrounding agricultural and pastoral lands, as the embodiment of order, whilst the steppe (Sumerian: edin) and mountain regions (Sumerian: kur, from which ‘Kurd’ and ‘Kurdistan’ derive) were ‘foreign, chaotic, and dangerous’, the realms of beasts, spirits, demons, and nomads. These are Sumer's liminal populations and the personification of an order/disorder dichotomy between urban/sedentary and rural/nomadic lifestyles. But we know, thanks indirectly to their persistent demonizations in Sumerian and Akkadian texts, that these liminal mountain peoples were resilient and durable, persisting despite the military defeats claimed in the Sumerian and Akkadian sources.

Subsequent centuries' literary traditions have added further layers of prejudice by presenting a dichotomous impression of relations between sedentary and nomadic populations.Footnote ¹⁵⁷ This includes works that are themselves centuries old, such as Ibn Khaldun's denigration of nomads in the Muqaddimah.Footnote ¹⁵⁸ These traditions are built on the foundations laid by the Sumerians and Babylonians and are frequently the products of literate urbanites (like Ibn Khaldun) writing for city-dwelling elites during times of general illiteracy. As Claudia Glatz and Jesse Casana conclude, ‘our understanding of ancient Mesopotamia as the proverbial “cradle of civilizations” derives predominantly from lowland-centric, text-informed self-representations’ designed to legitimate an ‘elite political and imperial ideology’.Footnote ¹⁵⁹ These elites needed to overcome the demographic instability of the ‘city-states’ and the potential transience of their populations by convincing their populations to remain and others to immigrate.Footnote ¹⁶⁰ These are, however, the data that have traditionally informed our narratives of the ‘Sumerian’ international, elite biases and all.

This is the ‘Mesopotamian trap’: social scientists, primed by colonial evolutionary ontologies and presentist inclinations, have rushed straight for the familiarity of the architectural and textual propaganda of Sumerian urban elites to imagine the state and the international based on literacy and urbanism. Colonial ‘civilizational’ social evolutionary epistemologies and social scientific preferences for textual data have driven scholars to the biases and self-aggrandisements of Sumerian texts. From inside this Mesopotamian trap, the emergence of ‘civilization’ – the basis of Western academic disciplines like History, IR, and Archaeology – has come to be seen as urban, sedentary, hierarchical, bureaucratic, and literate, with the ‘international’ being construed as a space of sedentary urbanized states. This is at the expense of liminal populations that neither built monumental stone architecture nor employed writing as an information technology, but nevertheless proved to be resilient neighbours that were integral to inter-polity relations in West Asia. Diversity and heterarchy are written out.

Dynamic multiplicity

Escaping the ‘Mesopotamian trap’ requires a shift to an approach of dynamic multiplicity that foregrounds the generative impetus of relations whilst removing remnants of colonial social evolutionary tendencies. Enquiry should not proceed from a substantialist search for ‘firsts’ linked to the sedentary agrarian state and social evolutionary typological hierarchies but rather from a recognition that the social world is dynamic. A dynamic ontology assumes the social world to be comprised of never-ending and always-there ‘unfolding relations’, in keeping with Mustafa Emirbayer's relationalism.Footnote ¹⁶¹ It builds on Norbert Elias' argumentFootnote ¹⁶² that the notion of a static person, let alone a society, is a myth: the interdependent nature of human existence means that they are always engaged in relations with others.Footnote ¹⁶³ Elias' concept of ‘figuration’ is useful as it challenges approaches that treat different levels of the social as ‘independently existing objects’.Footnote ¹⁶⁴ He criticized notions of the individual as a ‘closed personality’, preferring instead an

Due to these interdependencies, people exist ‘only as pluralities, only as figurations’.Footnote ¹⁶⁶ Approaches that treat individuals and society as separate, however, have ‘prevented us from thinking of people as individuals at the same time as thinking of them as societies’.Footnote ¹⁶⁷ Yet Elias said little about relations between societies; his focus was the ‘individual’ and ‘society’. In contrast, dynamic multiplicity applies his logic to the global, extends ‘openness’ to all substances so that we better appreciate the fuller extent of the multiplicity involved in relations.

We must also think of ‘social formations’ as processes rather than fixed, finalized entities.Footnote ¹⁶⁸ Groups are never permanently ‘bounded’ or ‘caged’;Footnote ¹⁶⁹ they are always somewhat permeable with neither they nor their inhabitants ever fully isolated from ‘external’ relations with their constitutive influences. Therefore, we should understand the greater caging capacities of sedentary polities in Mann's work in relative rather than absolute terms.Footnote ¹⁷⁰ Examples from different continents demonstrate that early ‘states’ always faced the prospect of populations leaving them, with these populations able to resort to alternative subsistence methods.Footnote ¹⁷¹ Even if modern territorial states have more control over population movement, cultural influences, and external relations, control is never so absolute that a polity is entirely ‘caged’. Thus, relations and the changes stemming from them persist despite caging.

From a dynamic multiplicity perspective, we never actually ‘arrive’, only continue from the foundations laid by previous relations. A diachronic perspective is therefore necessary to appreciate a polity's temporal dimensions.Footnote ¹⁷² This includes understanding relations as ‘unfolding, ongoing processes rather than as static ties among inert substances’.Footnote ¹⁷³ A polity's characteristics are therefore always evolving and change is human nature.Footnote ¹⁷⁴ Time constantly moves and relations are both never-ending and always constitutive. Therefore, everything is unstable to some degree, and nothing is permanently fixed. Processes such as state-formation that consolidate power relations in institutions help regulate and organize behaviour long enough for new identities to form and new relations to develop. Yet even these apparently stable institutions are also always evolving, subject to reform and revision, inclusion and exclusion. With time, the tiniest variations result in enormous differences, meaning that even the most complex socio-political structures undergo processes of integration and disintegration, growth and decay.Footnote ¹⁷⁵

This is lost on substantialist ideal-type narratives that tend towards synchronic comparison wherein the substance is immune from the vagaries of time and relations. Dynamic multiplicity, however, is diachronic, taking the emergence, change, and entropy of political orders for granted. Every object is permanently contingent, emergent, entropic, and the result of processes. Even where matters appear stable – such as, for example, when a ‘state’ is said to exist – they can only ever be relatively stable. Objects are constantly changing due to the complex constitutive nature of the ongoing transversal, overlapping, and multiplicitous relations of their constituent individuals and groups.Footnote ¹⁷⁶ This is already recognized in disciplines beyond IR, not least among global historians.Footnote ¹⁷⁷

This emergentist ontology deepens the relational aspects of the existing literature on the ‘first’ international by moving away from substantialist, social evolutionary, big bang-type creation myths that prioritize ideal-types. Such myths are, as Michel Foucault might say, sovereignty-centred understandings of power, flawed for their failure to appreciate the ‘strictly relational character of power relationships’.Footnote ¹⁷⁸ Foucault encourages us to appreciate a different form of multiplicity, the ‘multiplicity of points of resistance’ that ‘are present everywhere in the power network’.Footnote ¹⁷⁹ Rather than speak in terms of a multiplicity of specific actor-forms, we should instead adopt an agnostic approach to actorness that allows for the potential of a ‘multiplicity of points of resistance’ that may be present anywhere in the social world. This produces an openness to forms of agency that may be context-specific and variable over time.Footnote ¹⁸⁰

By rejecting sovereignty-centric understandings of power, dynamic multiplicity recognizes that power relations are both hierarchical and heterarchical. Dynamic multiplicity is thus consistent with heterarchical interpretations of world politicsFootnote ¹⁸¹ that are, in part, derived from archaeological theory.Footnote ¹⁸² This archaeological work also stems from critical responses to social evolutionary frameworks which assumed that any ‘complex’ society, including ‘chiefdoms’ and ‘states’, was hierarchical.Footnote ¹⁸³ This assumption, Crumley argues, ‘has provided the intellectual and moral rationale for scientific racism, colonialism, and other forms of domination, in that “complex” societies (e.g., nation-states) were considered more advanced than “simple” (e.g., pastoral) societies’.Footnote ¹⁸⁴ Archaeological fieldwork, informed by new methods and decentred ontologies, have demonstrated that ‘interaction is organized not just by core states but by the actions of all participants in the network’.Footnote ¹⁸⁵ Heterarchical thinking involves exploring the contexts in which substances emerged, and understanding the factors that maintain or weaken substances over time.Footnote ¹⁸⁶

Appreciating context, in turn, allows us to understand how polities reacted and adapted to different pressures, including ecological, geographical, political, or social. As Elias noted, actors and their social relations ‘can be considered separately, but not as being separate’ from everything and everyone else around them.Footnote ¹⁸⁷ Thus, dynamic multiplicity allows a broader understanding of a polity's relations combined with those of other polities, capturing a fuller range of relations essential to actors' emergence, persistence, and entropy. The dynamic processual character of relations therefore avoids being relegated to a secondary concern, as Jackson and Nexon put it,Footnote ¹⁸⁸ whilst also including diverse actor types. This allows for the possibility that the state is only one of several possible outcomes to socio-political processes, and not always the most desirable. Power hierarchies, such as those of the ‘city-states’, are embedded in multi-directional relational networks, organized around unstable nodes rather than centrifugal cores of superior cultures and their ‘peripheries’. Analysts are therefore able to appreciate the fuller range of relations in which actors emerge and exist.

Heterarchical approaches also question the tendency to understand the social world according to ‘levels’. Philip Cerny argues that thinking according to ‘levels’ is an ‘oversimplification’ that prevents us from grasping the complexities of twenty-first century global politics.Footnote ¹⁸⁹ A heterarchical approach captures a fuller range of actors involved in global politics, including mini- and meso-hierarchies and hybridized actor-forms that disrupt traditional understandings of public and private.Footnote ¹⁹⁰ It chimes with Elias' call for us to avoid the temptation to view different levels as separate from each other, with the associated analytical and methodological limitations that brings. However, whereas Cerny views this as a recent, twenty-first century development linked to globalization and neoliberalism that supersedes the state-centred international relations of the twentieth century,Footnote ¹⁹¹ dynamic multiplicity recognizes this to be the case across time. The shift from twentieth-century world order(s) to the present is thus better understood as a continuation of never-ending relational processes of consolidation and differentiation.

Indeed, Cerny's suggestion of an erosion of state dominance in late- and post-twentieth global politics demonstrates both the emergent and entropic aspects of dynamic multiplicity, including the varying roles played in those processes by non-state actors. Studies suggest that, even at the apparent zenith of state-centric international order, the supposedly most powerful actors were dependent on non-state, public–private hybrid actor-types that Cerny associates with the present, post-Cold War world. Examples include Eric Grynaviski's exploration of the ‘middlemen’ in US foreign policyFootnote ¹⁹² and my highlighting of the historical dependence of European empires on auxiliary, non-European military forces.Footnote ¹⁹³ Recognizing such auxiliaries' agency blurs public/private distinctions whilst highlighting liminal populations and spaces' roles in world order. Such actors include hybrid forms of Western/non-Western agency similar to those identified in historical IR scholarship, further disrupting simplistic dichotomous readings of global politics. Dynamic multiplicity recognizes a multi-scalar web of relations between diverse agents (state and non-state) that is incessantly generative, regardless of hegemonizing efforts by a particular actor-type or specific actor. These relations persist irrespective of efforts by those who seek, in Cerny and Prichard's words,Footnote ¹⁹⁴ to radically and misleadingly oversimplify global politics into neat, bounded levels of analysis.

Dynamic multiplicity ultimately forces us to rethink the idea of ‘firsts’. In temporal terms, ‘first’ not only indicates the beginning of something but also the end of something(s) that came before. Yet relations are never-ending and always-there, therefore a relational approach cannot include a search for firsts. Such a search is inevitably a substantialist exercise, initially requiring the definition of the constitutive properties of the substance in question, and consequently trying to locate those properties in time-space. Finding the ‘first’ international, therefore, depends on identifying the achievement of a defined set of characteristics consistent with the definition of a particular ideal-type. Those not conforming to the ideal-type are excluded or essentialized under problematic alternative typologies, such as ‘tribe’ or ‘barbarians’. Various parts of the social world are oversimplified, treated as unconnected, and allocated to different levels and, consequently, other disciplines. Aspects of the relational processes (e.g. liminal populations) crucial to actors central to the analysis (e.g. city-states) are thereby neglected, cast aside for different disciplines and contrary to both Elias and Cerny's warnings.

Dynamic multiplicity avoids this mistake. Broadening the scope of enquiry beyond a single level – such as the international – to be multi-scalar also undermines the premise of a ‘big bang’ beginning to the international. Elias was explicit in this regard, stating that ‘there is no point zero in the historicity of human development’.Footnote ¹⁹⁵ Similarly, Foucault contended that all existence is predicated on ‘countless lost events.’Footnote ¹⁹⁶ We should instead think about ‘moments of emergence’Footnote ¹⁹⁷ that ‘may appear as a culmination, but they are merely the current episodes in a series of subjugations’.Footnote ¹⁹⁸ Thus, emergence is processual, dependent on multi-scalar relations. The ‘international’ should therefore be understood not as a substance with strict constitutive properties but as a relational setting: ‘a relational matrix’ with ‘no governing entity according to which the whole setting can be categorised; it can only be characterised by deciphering its spatial and network patterns and temporal processes’.Footnote ¹⁹⁹

Appreciating the multi-scalar and diachronic dimensions of the social world, however, demand much of any single discipline, especially methodologically. Another step towards escaping the Mesopotamian trap is therefore a commitment to critical interdisciplinarity. Interdisciplinarity was important for Elias, who noted that approaches that treat individuals and society as separate ‘may block possible channels of communication’ within and between disciplines.Footnote ²⁰⁰ The multi-scalar and diachronic dimensions of dynamic multiplicity demand persistent interdisciplinary communication alongside a willingness to abandon shibboleths following new findings. This includes an openness to new forms of data about the past beyond architectural and textual remains. Such remains are always only partial accounts of reality. In the Sumerian case, they were often created by a select group of people in service to urban leaders. This does not mean their complete rejection as data but rather a recognition both of their partiality and the possible partiality of the archaeologists who interpret them. A continued interdisciplinary dialogue involving archaeologists and social scientists would help expose the peculiarities of such data, enable critical constructive dialogues across disciplines, and thereby keep individual disciplines abreast of developments in others.

This should not simply involve archaeologists informing the social sciences with nothing passed back in return. The interdisciplinary is its own space of mutually constitutive scholarly relations with long-lasting ontological legacies and further shibboleths that need slaying. This space is one of disciplinary interdependencies as disciplines turn to each other to account for their own methodological limitations. Archaeologists schooled in nineteenth-century social scientific epistemologies contended that a lack of material relics from mobile pastoralists was evidence of a lack of complexity, economic poverty, and an inability to engage in the wealth-accumulation that is sometimes deemed necessary for political hierarchies and state-formation.Footnote ²⁰¹ Archaeology's traditional prioritization of the monumental urban landscapes of ‘great civilizations’ produced theories and fieldwork methods designed to recover their ‘high cultures’ at the expense of less fashionable populations.Footnote ²⁰² Only recently have some archaeologists moved to reform a discipline that has, for example, been ‘fundamentally incapable of investigating … nomadic cultures’,Footnote ²⁰³ as well as other groups associated with gender, class, and racial categories. Consequently, archaeology has often been dependent on other disciplines such as ancient history and ethnography for knowledge of ancient pastoral populations,Footnote ²⁰⁴ even if those disciplines themselves have been limited in their own approaches to the same population groups. It is hardly surprising that there is a cross-disciplinary ‘historical absence’ of populations that may not have built stone cities or produced textual remainsFootnote ²⁰⁵ if each discipline tells the other that such peoples are insignificant.

In sum, dynamic multiplicity can be understood to rest on seven pillars. First, building on Rosenberg and others, global politics involves both a quantitative and qualitative multiplicity of relational actors. Second, this multiplicity is dynamic in that relations are never-ending and always unfolding. The constitutive influence of relations is therefore also never-ending. Third, any object, including social formations, is a relational process that cannot be fully ‘bounded’ or isolated from external influences. Substances are consequently only ever relatively stable as they are constantly evolving, facing inevitable entropy. Fourth, social action has diachronic effects, and therefore analysis also needs to be sensitive to temporal dimensions. Polities across time-space may share particular features but each iteration must be historicized and their constitutive properties opened to negotiation. Fifth, power relations can be hierarchical and heterarchical. Sixth, the social world is multi-scalar, and no social ‘level’ can be isolated from others. Therefore, investigations of global politics also need to be open to agents across a range of ‘levels’, with agency not limited to a particular type of actor, such as ‘states’. Seventh, there is a methodological imperative for interdisciplinary dialogue. It is only through such dialogues that any single discipline can overcome its parochial expertise in particular ‘levels’ of the social to comprehend the full extent of the dynamic multiplicity that shapes the human social and political world.

Visualizing dynamic multiplicity and moving beyond the ‘first’ international

This final section demonstrates how dynamic multiplicity reconfigures our understanding of the Sumerian ‘first’ international by revealing diverse forms of agency and strategies of resilience. Furthermore, the longevity of alternative forms of polity undermines claims that only states are sufficiently durable to survive the rigours of the international. Two cases are highlighted to demonstrate alternative political forms in the same time-space as Sumerian city-states.

The first of these cases is superficially a city-state, but one that has recently been rethought. Notably, textual evidence – the preserve of ‘civilized’ sedentary states, according to traditional perspectives – has informed this rethinking. Here, texts reveal complex power relations and practices of government that disrupt notions of liminal populations and the city-state ideal-type. The texts in question are a cache found at the city of Mari on the Euphrates, in present-day southeastern Syria. Dating mainly from the first half of the eighteenth-century BCE, the ‘Mari letters’ involve correspondence between individuals located across the region and are ‘therefore “Mari” letters only insofar as they were found at that city’.Footnote ²⁰⁶ They are significant for both their number – the Amarna documents represent only around 10% of the total found at Mari – and the details found within them.Footnote ²⁰⁷ They reveal Mari to have been an important regional political centre for a polity with significant rural dimensions.Footnote ²⁰⁸

Architectural remains suggest that for much of its history it lacked a sizable residential component. It is thus believed to have been a sedentary political-administrative hub for a larger rural and at least partially nomadic population. This suggests a diversity in the functional logics of Mesopotamian cities consistent with modern cities, from political centres such as Rabat and Washington DC to economic hubs like Casablanca and New York. Notably, political organization in Mari was a blend of the collective and the hierarchical, involving overlapping and shifting affiliations to towns, ‘tribes’, and realms (Akkadian: mātum). So involved were the population in the political life of an ostensible ‘monarchy’ that Mari has been called ‘democracy's ancient ancestor’.Footnote ²⁰⁹ Moreover, the lines between dichotomies that social science often takes for granted – including urban/rural, sedentary/nomadic, state/tribe – are so ambiguous in the ‘Mari letters’ that trying to determine who was one or the other ‘is an exercise in frustration, and … misses the point’.Footnote ²¹⁰ People had multiple and shifting affiliations that could each acquire prominence according to context, although kinship ties were more important signifiers of belonging than mode of life or place of residence.Footnote ²¹¹ This form of politics that transcends modern assumptions of urban/rural separability in Mesopotamia, whilst reinforcing suggestions that the practices of rule in the cities were in fact imported from rural communities.Footnote ²¹² Mari therefore undermines notions that city-states mark a transition from kinship modes of affiliation. It is also consistent with findings that across historical Europe, Africa, and Asia, sovereignty was frequently not a zero-sum territorial matter in non-Westphalian world orders.Footnote ²¹³ Furthermore, Mari demonstrates continuities of heterarchical political organization despite an urban revolution that has traditionally been associated with the emergence of the hierarchical ‘state’ ideal-type.

Relations with non-urban, non-Sumerian populations were essential to Sumerian cities, none of which were economically self-sufficient.Footnote ²¹⁴ These cities never existed in isolation, regardless of what their rulers sought to convey. A city was simultaneously a focal point for a surrounding region whilst also dependent on its hinterland and beyond to survive. It was a point of collection, storage, and redistribution of resources and a source of services.Footnote ²¹⁵ Whilst urban political economies demonstrated hierarchical organizational patterns, such hierarchies were frequently localized rather than generalized, with horizontal connections between hierarchies crucial for production, and the overall political economy less homogenized that often suggested.Footnote ²¹⁶ This enabled a diversity of practice, ideas, and materials vital to food production and, by extension, polity resilience.Footnote ²¹⁷ Cities were nodes in relational networks, combining local practices and materials with those from other polities, rather than constituting an exclusive network of like-units.

We know of Mari's complexities through the traditional data-form of texts. Meanwhile, new research methods are further expanding our awareness of liminal, non-urban populations in and around city-states.Footnote ²¹⁸ The second example presented here, the so-called ‘Elamites’ of southwestern and south-central Iran, is a case in point, speaking directly to that other corpus of historical IR work on populations with links to the Eurasian steppe. Geographically, the historical region commonly known as ‘Elam’ links West Asia with the South Asian subcontinent and Central Asia,Footnote ²¹⁹ including areas of later Iran that have been the subject of other studies of historical IR exploring non-Westphalian world orders.

Referring to a singular ‘Elamite’ polity is problematic, especially during the third millennium when multiple diverse groups unlikely to have shared a common language have been identified as inhabiting the space associated with them.Footnote ²²⁰ ‘Elam’ is likely a Sumerian or Akkadian word incorporating the meaning ‘high’ in geographical terms,Footnote ²²¹ serving as a catch-all term for the populations of this region neighbouring Sumer. The spatial dimensions of ‘Elam’ probably varied over time, incorporating lowland, highland, urban, and pastoralist elements. Nevertheless, their presence in Sumerian texts under the singular term ‘Elamites’ from the mid-third millenniumFootnote ²²² demonstrates their inseparability from the West Asian relational setting,Footnote ²²³ with surviving accounts suggesting alternating Sumerian invasions of Elam and of Elamite control over parts of Sumer during the height of the ‘city-state’ era. A coalition of these polities consolidated in the second half of the third millennium, leading to some form of Elamite ‘confederation’,Footnote ²²⁴ and they often served as antagonist neighbours in Sumerian texts. They were therefore dialectical contributors to the ‘international’ of the Sumerian cities.

Anthropologists and archaeologists often stereotype ‘mountain dwellers’ as socio-politically less complex,Footnote ²²⁵ largely due to a tendency to equate the lack of recognizable ‘modern’ features in their regions with a lack of complexity.Footnote ²²⁶ Sedentary agriculture, crucial to theories of state formation that stress resource accumulation, is often portrayed as a superior form of production, continually developing through technological innovation, in contrast to supposedly unchanging subsistence-level pastoralism.Footnote ²²⁷ Such essentializations stem from unfounded assumptions that urban life must have been perceived by those in the past as superior to pastoral life.Footnote ²²⁸ Elamites, however, developed writing around the same time as the Sumerians,Footnote ²²⁹ perhaps of their own accord rather than by inheriting it from the Sumerians.Footnote ²³⁰ The languages come from different language familiesFootnote ²³¹ and proto-Elamite script differs from Sumerian.Footnote ²³²

The Elamites developed resilient socio-political structures that mitigated the myriad geographical-ecological conditions of their landscape, being simultaneously ‘mountain peoples’ and city-builders, nomads and sedentists. Their ‘inverted’ model of nomad-centred rather than urban-centred Sumerian political organization involved farming communities being ‘enclosed within the much larger sphere of the nomadic society and ruled by a hierarchy that was drawn from various highland [peoples]’.Footnote ²³³ Cities including Awan, Susa, and Anshan supported a political system centred on a nomadic elite, demonstrating that pastoral-nomadic populations need not always be subservient to urban hegemony. This nomadic governing elite offers a precursor to the travelling Ottoman court that informs the ‘steppe tradition’ identified elsewhere in IR literature.Footnote ²³⁴ External relations were important, with their geographic location allowing control over lucrative Eurasian trade routes, of which the Mesopotamian cities were also part.Footnote ²³⁵ Outlasting ‘Sumer’ by frequently re-surfacing until ‘well after’ the Islamic conquest of Iran in the seventh century CE,Footnote ²³⁶ their persistence undermines claims that only the Sumerian city-state model was sufficiently durable to survive the rigours of the international.Footnote ²³⁷ Here also is durable diversity within a polity, consistent with the diversity identified in imperial systems by Philips and Sharman albeit not in a polity commonly typologized as an ‘empire’.

Non-hierarchical polity structures are evident across West Asia, including in supposedly hierarchical city-states. Examples of egalitarian and deliberative forms of government predated, coincided with, and persisted in Mesopotamia well into the first millennium CE.Footnote ²³⁸ Seth Richardson argues that if it is ever appropriate to use the term ‘state’ for this period, states were at best ‘low power’ and did not enjoy anything near the degrees of competency or capability traditionally attributed to them in social scientific literature.Footnote ²³⁹ Politics involved decision-making at different levels, ‘negotiated through complex webs of potential authority relationships’ inside, outside, through, between, and beyond the cities.Footnote ²⁴⁰ Economic rather than political centralization was likely to have been more comprehensive and widespread.Footnote ²⁴¹ The legal supremacy of city ‘kings’ was often tenuous, with little legal textual evidence from the ‘Old Babylonian’ period (c.2025–1595 BCE) that convincingly refers to the king's law as statute.Footnote ²⁴² Only ‘about twenty’ letters from one sample of 2800 legal texts made such claims,Footnote ²⁴³ whilst only one of the 279 ‘Laws of Hammurabi’, inscribed in stone around 1750 BCE, ‘reserves any specific powers to the king’.Footnote ²⁴⁴ The absence of such laws is notable as Hammurabi himself commonly boasted about his powers in other contexts.Footnote ²⁴⁵ Law, it seems, ‘was complicated and tempered by location and the involvement of other authorities’.Footnote ²⁴⁶ This chimes with the Mari documents' evidence that political power was negotiated between diverse interest groups operating along transversal lines rather than purely according to hierarchical, bounded models.

This is underscored by evidence from periods of reduced urban habitation. The Akkadian period (c.2350–2100 BCE) in Sumer witnessed a decline in the number of large urban sites that was not accompanied by simultaneous economic decline or cultural discontinuity.Footnote ²⁴⁷ Similarly, political fragmentation and conflict between Mesopotamian cities in the early second millennium BCE documented in textual sources did not affect high levels of economic interactivity.Footnote ²⁴⁸ This suggests a persistence of inter- and intra-polity relations – some personal, some large-scale – regardless of ‘state’-level turmoil and changes to the personalities noted in textual sources. Such persistence requires: resilience and complexity beyond the city-states; populations' abilities to change settlement practices when circumstances required; and the use of migration as a means of resilience. Decisions to become or remain nomadic rather than sedentary need to be understood as being possible strategic responses by populations to political and/or economic changes.Footnote ²⁴⁹

Thus, we find contemporary polities sharing some features with the Sumerians but also sufficiently different for them to be disparaged or ignored in social scientific narratives. A simplified, levels of analysis approach misses their complexities and thereby blinds the analyst to indications that diversity rather than ideal-type homogenization was the basis of West Asian world order at this time. Both Mari's and Elam's forms of sovereignty offer precursors to later non-Westphalian forms of sovereignty identified elsewhere in historical IR literature. Their exclusion from analyses that locate the ‘first’ international in Sumer is consistent with the wider malaise in IR of imagining non-Westphalian polities and orders as being without international politics and outside international order.Footnote ²⁵⁰ The Elamites, in particular, are emblematic of humanity's socio-political versatility and adaptability, along with the resilience that comes with versatility. Their recurring antagonistic role in Mesopotamian texts suggests that, just like ‘it is impossible to think Europe without including the steppe’,Footnote ²⁵¹ it is impossible to think Sumer without including wider West Asia. By revealing continuities of heterarchical and kin-based socio-political organization, along with diverse but resilient forms of sovereignty among liminal populations that were integral to the West Asian relational setting, dynamic multiplicity forces us to rethink both the idea of a ‘first’ international and of the international as a bounded level of the social world. This is all lost if one adopts simplistic ‘levels of analysis’ approaches.