The distribution of power: hierarchy and its discontents

CARLA M. SINOPOLI, RODERICK J. MCINTOSH, IAN MORRIS, AND ALEX R. KNODEll

Over the last few decades, archaeological and historical research in many regions of the world has challenged long-standing ideas concerning the nature and organization of ancient states and cities.

The three chapters in this section - on proto- and Early Historic cities of South Asia, Greek city-states, and West African cities - all address early urban formations that for various reasons have been viewed as lying outside of the normative structures of “typical” ancient cities (for example, in Mesopotamia).

For the first millennium bce Greek city-states, discussed by Morris and Knodell, their exceptionalism has been seen to lie in their presumed position as the originators of Western democratic political thought: cities whose inhabitants consciously and creatively rejected hierarchical despotisms. However, as Morris and Knodell argue, numerous other factors, including geography, environment, and the larger geopolitical and economic contexts of the eastern Mediterranean region, may better explain the distinctive forms and long-lived success and expansion of Classical Greek cities. More­over, conceptions of exceptionalism concerning distributions of power in Greek cities are at least partly a product of Athenocentrism, as ancient Greece was home to a wide array of very different political systems. It seems as though the contrast between democratic and more centralized or palace-oriented societies is more applicable to the Classical period and preceding Bronze Age than to early Greek cities and their contemporaries (for example in Etruria or Phoenicia). These factors and the divergent disciplinary trajectories of classical and anthropological archaeology have made cross-cultural studies including cities rather rare.

In contrast, the West African cities discussed by McIntosh and the early South Asia Indus or Harappan cities discussed by Sinopoli, with their ambiguous or absent evidence of “expected” urban features, have often been viewed as both “other” and less than the ideal exemplars of ancient cities, such as Mesopotamian Ur with its monumental palace and ziggurats and elaborate royal burials. Indeed, lacking such features, both the Indus cities and West African cities such as Jenne-jeno have frequently been characterized as not quite cities and/or not quite states. Unlike the Greek cities, however, the African and South Asian communities who built and inhabited those urban places were presumed by earlier generations of scholars to have lacked the creativity to establish distinctive and original forms of urban life, leaving scholars without models to understand or explain their settlements.

In this chapter, we draw on our case studies and other recent research to consider alternate ways of being urban and to advocate for models of urbanism that recognize the existence of a broad range of organizational structures and institutions - both vertical and horizontal - through which power could be distributed in early cities. We begin by briefly reviewing how it is that the kinds of cities that we address came to be viewed as aberrant or somehow less than other more “typical” ancient cities. How is it that archaeological scholarship came to argue that (with the exception of the exceptional Greeks), ancient cities: (ι) must have been ruled by kings who reside in elaborate palaces and belong to a restricted hereditary elite; (2) must have had central and powerful religious institutions (that is, state religions); and (3) must have been governed by rigid administrative hierarchies? Certainly, many ancient cities had all these features; but many others did not, and it is this latter category we explore in this chapter, with the larger goal of understanding why and how different urban forms developed and were sustained.

Shaping many early expectations of ancient civilization is the long legacy of Orientalist understandings of a despotic, totalitarian East - an “other” dominated by powerful, tyrannical, kings and priests. Orientalist thought had its origins in Classical Greek political and historical writings on Persia, authored by such historians and philosophers as Herodotus and Aristotle, among others. Their writings, composed at a time when Persia posed a significant military threat to the Aegean region, articulated the anxieties their authors felt about the enemy empire and also helped to mobilize independent Greek states to accumulate and deploy military and economic resources that could resist their powerful neighbor.

Through a long and complex genealogy,^[348] Classical Greek views of tyran­nical eastern kingdoms migrated to Rome and medieval Europe and were transformed into eighteenth- and nineteenth-century European understand­ings of the essential natures of the polities and peoples of Asia, Africa, and the Americas.

A second and related strand that has contributed to the long-standing views concerning the nature of early cities lies in the history of archaeology as a colonial enterprise. Numerous scholars have written of the develop­ment of archaeology in colonial contexts and of the deployment of archaeo­logical knowledge in nineteenth- and twentieth-century colonial and postcolonial politics.^[354] Understandings of South Asian and African cities and early civilizations in particular were shaped within colonial discourses that constructed and reified particular interpretations of historical change and social forms in these regions, in ways that explicitly or implicitly legitimized colonialism by constructing arguments for changeless and, indeed stagnant pasts. This is not to discount the tremendous amount of valuable archaeological research conducted by colonial era scholars in these areas, or even to say that all early archaeologists adhered to such views.

Turning specifically to archaeological research on early cities, Marxist archaeologist V. Gordon Childe certainly counts among the most influen­tial voices shaping approaches to ancient urbanism in the mid-twentieth century. In his influential writings on the “urban revolution,” also as viewed largely through the lens of Early Dynastic and later Mesopotamia, he was an inheritor of the two-millennia-long traditions of Orientalist thought. His writings (though often shorn of their nuance), and particu­larly his oft-quoted list of the ten criteria of urbanism, have influenced generations of subsequent work on early cities, both positively and nega­tively. And indeed, it was in part a rejection of what came to be critiqued as Childe's “trait list” approach that led archaeologists such as Henry Wright, Gregory Johnson, and others,^[356] to shift their archaeological focus from cities to states during the 1960s and 1970s, defining states as systems of hierarchies of differentiated decision-making and administration rather than by a set of traits.

Ultimately, the three chapters in this section, as well as others in this volume, seek not to reargue old debates but instead to recontextualize and broaden our perspectives on early cities. We do this by addressing the limits of hierarchy and the existence, and indeed long-lived success, of other ways of being urban - and the specific times, places, scales, and forms in which some of those other urban places emerged, persisted, or failed.

Other ways of being urban: distributed power in early cities

Many of the cities discussed in this section were places where power was distributed in ways that worked to challenge or undercut straightforward linear hierarchies of authority and administrative control.¹¹ This is not to say that hierarchy was not recognized or did not exist in the Greek city-states, the Indus world, or the African clustered cities that are the focus of the chapters in this section.

In first millennium bce Greece, for example, Morris and Knodell argue that the success of Classical period cities was fostered both by the ability of city dwellers to mobilize long-distance economic resources through exten­sive maritime market networks and by their citizens' philosophical commit­ment to male egalitarianism. The latter entailed the creation of mechanisms to assure that access to leadership appointments was circulated among the (relatively) large population of eligible male citizens. Governance of many Classical Greek cities was highly structured and institutionalized, with positions of leadership in both religious and political offices rotating among adult male citizens. Indeed, Morris and Knodell argue that such decentral­ized political systems and the high level of inter- and extra-cultural connect­ivity may help to explain the impressive versatility and success of the Classical Greek cities in addressing specific urban and regional problems, first independently, then as part of Athenian, Macedonian, and Roman imperial expansions.

While we know less about the formal political and social mechanisms through which African clustered cities were integrated and governed, McIntosh suggests that at Jenne-jeno, networks of horizontal relations among diverse, economically specialized, corporate groups created an urban system based on multiple and complex relations of reciprocity shaped within a widely understood “social contract.” The result was an intricate network of dyadic relationships that contributed to the creation of a geographically dispersed “urban cluster” comprised of multiple zones of specialized resi­dential and occupational areas. These clustered cities lacked a central core and distinctive administrative architecture, and looked and operated quite differently than more governmentally centralized urban forms. Even so, Jenne-jeno's expansive city wall reminds us that in such seemingly loosely integrated cities, considerable labor could be mobilized in a variety of ways without requiring recourse to central authorities. Such dyadic networks of inter-cluster links lent considerable fluidity and flexibility to clustered cities - allowing parts to break off or change while the whole endured.

II

12

Multiple corporate groups of merchants, artisans, and other urban resi­dents may also have formed a distributed structure of power relations for the Bronze Age cities of South Asia, though here too we know little about the specific mechanisms or administrative structures through which these cities were governed. With their dense residential architecture, sophisticated urban infrastructure, and massive platforms and enclosures, the Indus cities, especially the well-known site of Mohenjo Daro, more closely resemble traditional expectations for early cities than do the dispersed West African places McIntosh describes. The elaborate and dense architecture and routes of movement and sophisticated hydraulic infrastructure that characterized these cities would certainly have required significant coordination for both their construction and long-term maintenance. Nonetheless, numerous scholars^[359] have emphasized what Indus cities appear to have lacked: most often, unambiguous palaces and temples,^[360] elaborate royal or elite burials and evidence for kingship, and warfare. If these traits are essential characteristics of cities and states, the argument goes, then the Indus tradition cannot have had either. An alternative interpretation, however, might suggest that the particular forms that various Indus cities took - including the lack of evidence for elaborate material expressions of elite institutions and elite bodies - may be evidence for some very different political and urban dynamics. Specifically, current interpretations suggest that within Indus urban systems, power may have been variously distributed among competing and fluid social or economic groups rather than being highly centralized within a single ruling dynasty.^[361]

These examples all point toward the multiple ways that ancient soci­eties constructed and inhabited urban places, and they open the door to comparative archaeological and historical scholarship that seeks to under­stand the diverse routes and factors that led communities in various regions of the world to build large, complex, and economically, socially, and functionally differentiated residential centers - or cities. Each of the three chapters in this section discusses contexts in which power - eco­nomic, political, ideological, military - was variously distributed, such that no single kin or social group could effectively dominate the individ­ual cities they resided in or the polities of which these cities were a part. Tensions and competition no doubt existed among the diverse kin groups and communities that constituted Greek, West African, and Indus cities, as they engaged in commerce, agricultural and craft production, diplo­macy, sanctioned violence, and local, regional, and interregional inter­actions. In such urban places, a variety of social and ideological mechanisms would have been required to mitigate competition and limit the accumulation of wealth and resources that could have contributed to the increasing power of one or another corporate group, or the subjuga­tion of others. Viewed from within this framework, we might point to the lack of evidence for elite burials in the Indus cities and the limited (but far from insignificant) range of material expressions of wealth and status differences as being the result of political and religious ideologies that discouraged the materialization of the hierarchies that did exist - part of an active and self-conscious political strategy that worked to maintain a non-hierarchical political structure (or - perhaps, more accurately - a political structure that sought to ideologically render hierarchy less visible), rather than as evidence for the absence of political institutions.^[362]

It is also important to point out that large-scale cooperation and coordin­ated urban activities were not precluded by the simultaneous coexistence of multiple loci of distributed power. We noted above the city walls of Jenne- jeno. In some Greek cities, citizens collaborated to form a government that built impressive urban monuments, created legal and juridical structures, launched major maritime expeditions, and went to war. The builders of Indus cities coordinated labor to construct and for centuries maintain massive public works, and were able to develop and agree upon large- scale standardized systems of weights and measures (and economic values) that allowed for economic interactions within and between Indus settle­ments, over enormous distances and involving numerous categories of commodities.

In sum, these urban places were successful and their residents developed organizational structures and accommodations that allowed them to endure for many centuries. Indeed, their success and resilience may well have been fostered by the lack of rigidity in their organizational structures and rela­tions, which allowed individual households, neighborhoods, corporate units, and the city as a whole to respond to emerging situations and opportunities with greater ease than in highly bureaucratic administrative structures.

That said, it is not insignificant to note that early Roman emperors typically justified their efforts to dissolve republican institutions during times of crisis by arguing that democratic institutions such as the Roman senate were not sufficiently agile to respond rapidly enough to existential threats, and the slow, consensus-building, consultative democratic process endan­gered the Roman state and capital during times of political and economic crisis. In the millennia since, many similar claims have been made by many other powerful rulers seeking to centralize power and justify the dissolution of decision-making bodies. While such arguments clearly have a transparent aggrandizing agenda, they do raise important questions about the limits of structures of distributed power that merit serious consideration. Below, we turn to a consideration of the constraints and limitations of cities organized around distributed power, addressing both internal and external challenges to the durability and success of these urban forms.

Can distributed power endure?

We take as our premise that the social, political, economic, etc. organization of all early cities entailed some degree of distributed power. Even in the most hierarchical and dictatorial of political systems, rulers cannot (and seldom try to) control all aspects of life, ceding (voluntarily or not) some degree of autonomy to various corporate groups and institutions. And urban residents will inevitably find ways to resist or subvert administrative, eco­nomic, or religious hierarchies - avoiding taxes, exchanging surplus food for extra pottery vessels a neighbor may have, worshipping family or household gods rather than at state temples. The distribution of power in urban centers may best be conceived as ranging along a continuum from more centralized to more distributed rather than as an either/or construction. Different realms of urban life may be situated at quite different locations along such a continuum (that is, military defense of urban boundaries may be organized hierarchically even as markets are not), and any city's overall position likely varied over both long and short time scales and in response to a variety of internal and external factors.

We take as a second premise that not generating hierarchies takes work, and that the long-term durability of distributed power relations requires both ideological commitments and material benefits to the actors involved. The Classical Greek city-states, Indus cities, and West African clustered cities all thrived for many centuries; as we have noted, they were by any measure large and successful cities. Until they were not. Below we consider the range of factors and conditions that may influence the durability of urban systems of distributed power. We focus on questions of scale, resource distribution, and long- and short-term temporal oscillations in production, consumption, and interregional interactions, as they played within the frame of local political histories and political actors. We begin by addressing how urban life itself may create conditions that foster the intensification and formaliza­tion of hierarchy, in ways that challenge the effectiveness of non-hierarchical modes of organizing residents and activities.

Urban life can offer many opportunities to city residents, and ancient cities were often magnets that drew dispersed rural populations and families and individuals to them in search of a better life. The presence of producers of craft and luxury goods, and the centrality of cities as centers of accumu­lation and commerce provided their residents with access to a broader array of goods than available in towns and villages. Cities provided (at least some of) their residents with opportunities for social and economic mobility and were sites for the creation of new kinds of social and kin affiliations. And they were centers for the construction and performance of elite (and non­elite) cultural values, and sites of religious worship. Early cities were likely also places of crime and violence, of scarcities and poverty, and of social alienation. As we have noted earlier, even in relatively non-hierarchical cities, it should not be forgotten that urban life was no doubt experienced very differently by citizens as compared to slaves, by women as compared to men, by land-owners as compared to the landless, and by the well-off as compared to the impoverished.

In considering the trajectories of early cities, we might also consider whether cities, as a consequence of being cities, intensify and generate inequalities and hierarchy. In the 1990s, at a conference on Asian cities attended by Sinopoli, Rita Wright - a specialist in Mesopotamian and Indus archaeology - astutely observed that once the inhabitants of the densely packed houses in Indus cities had invented and adopted indoor toilets, which fed to sewer drains that ran through the cities' streets, they also created the necessity for people to clean those sewers so that the urban infrastructure could continue to function. Sewer cleaners, trash collectors, tanners - one can easily identify numerous kinds of undesirable occupations that would be required to maintain dense urban places (though perhaps less so in the more dispersed clustered settlements of West Africa). Urban life, in other words, may generate its own social hierarchies, which then require new organiza­tional and ideological responses. These hierarchies may be obscured in ideologies of democracy that exclude slaves, resident aliens, and women from participation, allowing for an urban structure that benefits a sizeable portion (but far from all) of their population. Or they may be regulated in non-hierarchical ways and/or by non-centralized self-regulating organiza­tions; here we think of guilds or religiously sanctioned castes such as developed in Early Historic South Asia and which proved tremendously durable and effective in regulating both behavior and local and interregional political and economic interactions. The corporate kin and occupational groups that occupied the various groups of West African clustered cities may have operated in similar ways. Or the “winners” in ancient cities may regulate their emergent underclasses through the creation of new, more centralized and hierarchical administrative structures that undercut earlier distributed modes of organization. And, of course, cities may fail - and residents may, for a wide variety of reasons, “vote with their feet” and walk away from particular urban places or, in rare cases, like the Indus, from urban life entirely.

We have suggested above that cities characterized by distributed power may have greater resilience to some stresses than more hierarchically organized cities, though we have also noted that in other circumstances, they may be more vulnerable.^[363] Comparative research on the long- and short-term histories and responses of early cities of diverse organizational forms to various kinds of stresses may allow us to better understand both individual urban trajectories and larger patterns of similarity and difference. Here, we only note a few of the larger themes around which such research might focus.

Most important among these are questions of scale and temporality: including both the frequency and intensity of temporal oscillations affecting a range of domains, from subsistence production to the social, economic, and political relations within and among any city's diverse communities. Internal challenges to the success of networks of distributed power may come from rapid demographic changes - population growth, reduction, or redistribution - that render unstable existing organizational modes. It is certainly the case that what constitutes “rapid” change may have been quite different in different historical contexts and environmental settings; and it is also likely the case that different urban organizational structures were likely better (and worse) able to respond to such change. These are questions for comparative empirical research.

Numerous external factors may also affect the durability of systems of distributed power. All early cities existed in larger networks of other urban (and non-urban) formations; and all these existed in larger worlds populated by numerous other cultures and societies with which they interacted. In their roles as centers of commerce, early cities were certainly affected by larger regional and “global” networks through which material resources moved, such as the maritime exchange relations that allowed the Classical Greek cities discussed by Morris and Knodell to grow, thrive, and feed their subjects on crops imported from throughout the Mediterranean. And as we mentioned earlier, the Persian Empire initially gave common cause to fifth­century Greek cities and in the fourth century provided a reason and conduit for the conquests of Alexander; this, as Morris and Knodell report, “drastically changed the politics of power in the Greek world.”

Large-scale threats and crises do of course pose challenges to all early cities - those organized hierarchically and those characterized by non- hierarchical structures of distributed power. And it is beyond our scope to address this topic. Here, we merely end where we began this chapter, arguing that the comparative study of early cities must take into account ancient cities that look different and were organized differently than our standard (and problematic) inherited model: cities that may have consisted of dispersed settlement mounds without a central core; cities that lacked kings, courts, palaces, and state religions; cities whose residents created systems and structures for leadership, administration, and international relations that were dispersed among diverse interlocking groups, rather than under singular linear systems of rule. Moreover, we must recognize that cities frequently grouped together as “Greek,” “South Asian,” or “African,” for example, varied markedly in their scale and organization through time. We hope to show that this diversity (both within and between ancient societies) is precisely what makes the comparative study of ancient cities interesting and relevant. Scholarship on all ancient cities will benefit as a result.