Agriculture and the urban hinterland
In exploring the role of agriculture in early urbanism, we need a working definition of that new mode of settlement, the city, that emerged out of the previous landscape of (often undifferentiated) villages, hamlets, and more or less mobile camps of seasonal passage.
Most definitions coalesce around a common set of themes: a relatively large population size and density, a population engaged in diverse tasks providing services for residents and non-residents, and, in many cases, the existence of spatial features such as an urban core.[345] [346] However, given our emphasis on agriculture, we require a robust conception of ‘hinterland', one that allows us to link the urban core to its agricultural foundation.For some scholars, particularly those who emphasize the many similarities of cities throughout human history, the concept of hinterland becomes intertwined with industrial and post-industrial mega-plexes of the modern era, in which there is a significant spatial dissipation of what would have been considered ‘urban' activities in the pre-industrial era.15 These approaches, while useful in drawing connections within the shared urban experience of ancient and modern peoples, are functionally challenging for archaeologists actively reconstructing ancient landscapes who wish to determine the relationship between an urban core and its surrounding territory. It is for this reason that we engage the useful heuristic of city/hinterland, utilizing this dichotomy to expose the pairing as an intertwined whole. These linkages are central to the dynamic interplay of agriculture (specifically, decisions about production and disposal of surplus) and early urbanism (understood as an interplay of several dimensions of demographics and cities as arenas for the negotiation of authority in emergent, highly unstable, complex societies).
Given this centrality of hinterland in our agriculturally focused analysis, we take as our definition one drawn from Trigger and modified by us, characterizing the city as a large and heterogeneous settlement supported by, and providing a variety of services and manufactures to, a wider hinter- land.[347] Implicit in this definition is the two-way exchange in which commodities (agricultural products, raw materials), objects (e.g. manufactured goods such as ceramic or metal objects), and even practices (construction styles, religious cults, etc.) flow between the centre and the periphery. While this exchange will occur in both directions (and among hinterland communities) for each of these classes, in general there is an expectation that commodities will be traded to cities while goods and practices will spread from cities. Likewise, there is an assumption that there will be some relationship between the ‘distance' (whether characterized by length, travel time, social connections, or otherwise) of a community from the focal city and the intensity of exchange. These presumptions create challenges for archaeological visibility, as the movement of goods is frequently easier to track than the movement of consumables. This can in some cases lead to circular attempts to define a city-hinterland space when the size of the urban ‘catchment' is estimated by the number of residents who need to be sustained and the urban demographics are presumed to be a function of the rural sphere of influence.
Trigger provides us with not only a useful definition, but also an important distinction in characterizing cities' relationships with their hinterlands.[348] He argues that most pre-industrial urban societies can be divided into citystates and territorial states, with each having a distinct set of characteristics. City-states are small political entities that typically consist of just the city and a small surrounding territory.
As they are frequently more defensive in orientation (being often located in regions populated by multiple competing polities), and since their agricultural catchment is usually directly adjacent to the urban centre, these cities frequently include a high proportion of resident farmers, and it is not unusual for them to incorporate substantial fields within the city walls. For example, early Mesopotamian cities were known for their orchards, and some estimates suggest (perhapsoptimistically) that as much as two thirds of ‘urban' space was in fact farmland or gardens.[349] In contrast, territorial states encompassed numerous cities within comparatively secure territories. As a result, their populations were less heavily urbanized, with more farmers living in rural settlements and an elite that was frequently more isolated from the general population. Cities in each of these categories create different on-the-ground expectations for interaction, cohesion, and urban-hinterland boundaries.
Trigger's distinction focuses on the macro scale, but at the level of the individual farmer, the urban-hinterland distinction becomes more complex. Early urbanites not only had, in many cases, substantial garden plots in or near their residences (even within territorial states), but may have left the city on a daily or even seasonal basis to cultivate near or distant fields. Within kin groups, members of the same nuclear family could be distributed between residing in an urban household, cultivating a seasonally occupied farmstead, and tending to a mobile herd of livestock. Likewise, a rural household could send members to urban markets, to apprentice with urban craftsmen, or to fulfil military and/or labour obligations administered through an urban centre. This process becomes even more multifaceted when ‘family' is expanded to include people with whom urban residents maintained longterm relations of reciprocity, usufruct, or fictive kinship.
The problems of distinguishing the city spatially from its hinterland are likewise challenging on the ground. While some cities, such as Teotihuacan, may in general conform to the ideal of a centre surrounded by farmland, others, such as the classical Mayan centres, have farmland integrated throughout.[350] Recent studies have even found agricultural terracing between ceremonial pyramids, and field systems were organized not around the central urban core, but rather around households that are distributed in an increasingly diffuse arrangement as you move from the urban centre.[351]
Even more intriguing are the cities of the ancient Yoruba kingdoms of Nigeria.[352] These settlements consist of complex arrangements of earthen walled enclosures, spread in some cases over tens of kilometres, such that it is often difficult to distinguish an urban core from a hinterland. This landscape is reflective of the extensive agricultural practices of these societies, where many of the inhabitants of political centres practised agriculture both in fields adjacent to settlements and in others further away. Given the extensive nature of agricultural practice, there are complex histories of villages, towns, and cities moving or being moved by political centres, further complicating the process of identifying the ‘urban core'.
New technologies are increasingly aiding archaeologists in documenting the spatial networks of these urban centres. While classic, low-tech methods like pedestrian survey are still among the most thorough methods of locating archaeological sites within an urban catchment, this form of research is inherently limited in scale. In contrast, remote sensing technologies, including aerial photos, satellite images, and LiDAR, allow archaeologists to rapidly identify various human constructions, including habitation sites and agricultural infrastructure, over very large areas.22 For example, Jason Ur has used satellite images to document the numerous radial paths extending from third-millennium bce cities in the upper Khabur basin (Syria),23 demonstrating the intensity of movement from urban centres to farm and pastureland, and utilized the particular properties of soil on human-occupied sites (anthrosols) to remotely identify habitation sites in a systematic and comprehensive manner over the entire region.24 In the densely forested region of Mesoamerica, archaeologists are increasingly using LiDAR (a technology that analyses reflected light from a laser) to literally see through the trees and rapidly create detailed topographic maps of the ground surface.25 These maps have revealed significant investment in agricultural terracing and allowed scholars to more accurately analyse the distribution and density of the small residential groups around which most agricultural fields were located.
These expanded archaeological data sets on urban-hinterland relationships have both increased our ability to challenge the standard narrative and illustrated its persistence. The Yahwist conception of the impious
Nigeria (ca. 2000 bc-ad 1900): archaeological perspectives', Journal of World Prehistory, 19 (2005), 152-4.
22 D.C. Comer and MJ. Harrower, Mapping Archaeological Landscapes from Space (New York: Springer, 2013).
23 J. Ur, ‘CORONA satellite photography and ancient road networks: a northern Mesopotamian case study', Antiquity, 77 (2003), 102-15.
24 B.H. Menze and J. Ur, ‘Mapping patterns of long-term settlement in northern Mesopotamia at a large scale', Proceedings of the National Academy of Sciences, 109 (2012), 778-87.
25 A.F. Chase et al., ‘Geospatial revolution and remote sensing LiDAR in Mesoamerican archaeology', Proceedings of the National Academy of Sciences, 109 (2012), 12916-21. urban centre in opposition to a rural countryside does not allow for the extensive flow of people and ideas between city and hinterland, nor for the direct involvement of urban residents in agricultural or, particularly, pastoral production. However, we are still left with the despot in control of a large agricultural surplus, an idea which was elaborated in Weber's conceptualization of pre-industrial cities as ‘consumer' cities.[353] For Weber, the power of urban elites derived from extraction of wealth from the hinterland, and the overall urban economy was oriented towards the interests of a small group of wealthy consumers. In a sense then, our theoretical conception of a city-hinterland dichotomy aligns with his fundamental conceit distinguishing the urban elite from the general populace, and under a consumer city paradigm, the interaction with and investment in the hinterland documented archaeologically could simply demonstrate close control by an extractive power. Therefore, we must examine not only the integration of agricultural practice into the urban landscape, but also the integration of agricultural decision-making.