by Eckart Frahm
Figure 1.4 Sites, canals, and other features in the Assyrian imperial core.
The first canals constructed under royal impetus appeared in river valleys already in the Late Bronze Age, in association with Kar‐Tukulti‐Ninurta and Ashur in the heartland of the Tigris Valley (Bagg 2000), and possibly also along the lower Khabur River (Ergenzinger et al. 1988, Kühne 2012). New systems were created in the ninth to early seventh centuries alongside the new capitals and the deliberate deportation of captured populations into the cities and their hinterlands (Oded 1979). The canals that are best documented with regard to the textual and archaeological record are the construction projects attributed to Aššurnaṣirpal II (884–59 BCE) and, especially, to Sennacherib (704–681 BCE). The fields surrounding the new capital at Kalḫu (Nimrud), although the city itself was constructed on a terrace of the Tigris River, were irrigated with water from either the Upper Zab or its right bank tributary the Khazir River (Oates 1968; Davey 1985; Ur and Reade 2015). This 35 kilometer canal, which was called the patti ḫegalli, followed the right bank of the Lower Zab until the river approached its confluence with the Tigris River, at which point the canal turned north to the city. Most of the canal’s course was open, but at one place its engineers tunneled through a rocky outcrop to maintain the canal’s gradient (Davey 1985).
Sennacherib was the most prolific canal builder of the Neo‐Assyrian kings and claimed to have ordered the construction of a vast array of canals throughout the northern hinterland of his new capital at Nineveh, which remade a large percentage of the hydrology of northern Assyria (Bagg 2000; Oates 1968; Reade 1978; Ur 2005). The canals were dug in four increasingly ambitious phases (Reade 2000, 2002). The first was the Kisiri canal on the Khosr River, which involved the excavation of 13.4 kilometers of canals immediately upstream from Nineveh. The second phase watered the plain east of the city, but has not yet been located by archaeologists. The third phase, which focused on the northwest, and the fourth phase, which was aimed to the northeast, were massive undertakings. The third phase, called the “Northern System” included a chain of canals that tapped rivers and springs along the foothill fringes from Maltai near Dohuk to Tell Uskof. While not all of these canals were interconnected, the last two canals on this chain redirected some or all of the upper courses of the Wadi Bandwai and the Wadi al‐Milah and transferred them into the Khosr River, where they could flow to Nineveh (Ur 2005; Morandi Bonacossi 2012–13; Morandi Bonacossi and Iamoni 2015). The excavated length of these canals is just under 25 kilometers, but three of them involved the excavation of 100 meter‐wide earthworks through watersheds, which were up to 20 meters deep in some places.
The fourth phase system extended from a weir on the Gomel River at Khinis, where it is associated with Sennacherib’s inscriptions and reliefs (Bachmann 1927; Jacobsen and Lloyd 1935; Fales and del Fabbro 2012–13). This system flowed through 55 kilometers of excavated canals before adding its water into a tributary of the Khosr. Along its course, the water passed over a stone‐built aqueduct at Jerwan (Jacobsen and Lloyd 1935). The Jerwan aqueduct was constructed with an estimated half million cut stone blocks and included several short inscriptions naming Sennacherib as its mastermind as well as mentioning several elements of the surrounding landscape.
The imperial capitals were not the only beneficiaries of imperial canal construction. Sennacherib also commissioned a system that redirected water over 22 kilometers from the Wadi Bastura to the outskirts of Erbil (Safar 1947; Ur et al. 2013: 104–6). Unlike Sennacherib’s other canals, the Bastura canal was largely subterranean, and was accessible via vertical shafts at 42 meter intervals. In the eastern provinces, long canals watered the river terraces on both sides of the lower Khabur River, although these constructions have not yet been attributed to a particular Assyrian ruler (Ergenzinger et al. 1988).
Imposed water features served multiple purposes. Of particular importance was their basic economic function: to raise crop yields and to lower the risks that are inherent in rain‐fed farming. Economic factors have been downplayed in previous studies, which have emphasized the ideological role of the system (e.g., Bagg 2000; Oates 1968; Reade 1978), but evidence from remote sensing shows that offtakes from these canals were found throughout the system, and would have provided local irrigation water at substantial distances from Nineveh, for example in the regions of Girepan and Jerwan (Ur 2005: 341–2). A transportation function has also been proposed (Ur and Reade 2015). These irrigation canals would have been critical for sustaining cities that had far outstripped the demographic thresholds of the Bronze Age (e.g., Wilkinson 1994).
The canals must, however, be seen as part of an ideological transformation of the landscape of Assyria in order to imprint upon it the power of the Assyrian kings and their divine legitimacy. No rural farmer, whether he was a native Assyrian or a forcibly transplanted Aramaean or Babylonian, could have failed to recognize the awesome power of a king who could redirect rivers and could recreate conquered landscapes in his own country (Ur 2005: 342; Wilkinson et al. 2005: 50). This ideological connection was made explicit with inscriptions and reliefs associated with, or inscribed upon, many of these canals’ features. Most famous is perhaps the series of reliefs associated with the weir at Khinis, including the so‐called “Bavian inscription,” wherein Sennacherib gives his most lengthy description of his irrigation constructions, along with information about other events in his reign. Most importantly, the associated monumental relief (Figure 1.5) shows Sennacherib standing before Assur and Mullissu, who bestow upon him the symbols of kingship. The iconographic message is that this canal was constructed by the divinely‐installed Assyrian ruler. Similar depictions of the Assyrian king and the gods occur in association with canals at Maltai, Bandwai, and especially Faida, where the reliefs are immediately adjacent to a sluice, inescapably visible to the farmer who draws water out of the canal and down onto his fields (Reade 1978).
Figure 1.5 Austen Henry Layard exploring Sennacherib’s monumental relief at Khinis, as depicted by Frederick Cooper.
The Assyrian landscape was crossed not only by water but also by human movement. It can be assumed that localized movement took a form similar to the radial patterning of the Early Bronze Age, as described above, although few such systems can be dated unambiguously to the Iron Age. Some linear features have been captured on satellite photographs around Nineveh and Ashur (Wilkinson et al. 2005: 32–7; Altaweel 2008), but most evidence comes from textual sources, which describe “royal roads” (variously transliterated as ḫarrān šarri or ḫūl šarri) between the Assyrian capitals and the major administrative towns of the provinces (Fales 1990: 98–9; Kessler 1997; Graf 1994: 171–2). Despite the use of the term “road” in English translations, there are few indications that these features were constructed or planned. Most probably, they were tracks that hosted royally maintained way stations (bīt mardēti) along them.
Many aspects of this imperial landscape were tied closely to the royal dynasty and appear to have disintegrated almost immediately upon its collapse in the late seventh century. The capitals were so thoroughly vacated that they had largely disappeared from memory only a few centuries later. Most of the small villages and hamlets were abandoned, and, thus, the extensive agricultural settlement pattern also dissolved. Emerging modeling results suggest that the dispersed rural settlement pattern may not have been viable from a long‐term ecological perspective (M. Altaweel, personal communication), and so, in the absence of Persian or Babylonian royal coercion, rural villagers may have had both social and economic cause to abandon their settlements. With the de‐urbanization of the capitals and the apparent abandonment of the countryside, the Median and Babylonian conquerors had little interest in maintaining the major irrigation works. They did, however, choose to maintain the network of “royal roads,” which was the basis for the Achaemenid system (Graf 1994).
Unlike the Early Bronze Age cultural landscape, which emerged without central planning, the Assyrian landscape was, to a grea
t extent, the intended product of imperial decision makers. Many elements were planned and imposed in a top‐down manner, reflecting underlying visions of the proper way for humans to inhabit their world (Wilkinson et al. 2005). The Assyrian landscape often incorporated elements that had long been in existence, most notably settlements such as Ashur and Erbil, which had ancient populations and long‐standing religious and political significance. New capital cities and the extensive rural settlements were imposed atop and surrounding existing landscape features, the growth of which cannot be explained by natural demographic growth or settlement fission. Rather, both cities and countryside appear to have been forcibly settled by captured and deported populations from elsewhere in the empire, a practice that is well documented in royal inscriptions and letters (Oded 1979: 366–9; Morandi Bonacossi 2000; Wilkinson et al. 2005). In the case of the lower Khabur and adjacent Wadi Ajij, for example, it is likely that the expansion of rural sites can be related to Adad‐nirari III’s deliberate colonization program described on the stele found at Tell al‐Rimah (Page 1968; Morandi Bonacossi 2000; Kühne 2010).
Contributing to this process was the increasingly sedentary nature of Aramaean pastoral nomadic groups (Wilkinson and Barbanes 2000). For example, Adad‐nirari II’s account of his campaign across the Upper Khabur plain to Nisibin and Guzana includes many hints at the nomadic past of their kings, mentioning almost no settlement elsewhere on the plain (Postgate 1974: 234–7). Given the abundant and extensive settlement pattern revealed by archaeological survey (described above), it is likely that these nomads were settled coercively on their former winter pasturelands and compelled to adopt an agricultural lifestyle by the Assyrian conquest.
It cannot be assumed that all kings shared the same vision for the landscape. No king ever left a comprehensive treatise on the subject (Radner 2000: 233), although non‐textual iconographic clues appear repeatedly throughout Neo‐Assyrian royal art (Winter 2003). Common characteristics emerge, however, over several centuries of Assyrian imperial control: the construction of walled cities of great spatial extent; population expansion not by natural demographic growth but by the physical emplacement of conquered groups; an evenly settled and agriculturally productive countryside, also populated via forced settlement of captured peoples; the labor‐intensive reworking of Assyria’s natural hydrology to sustain cities and their agricultural hinterlands; and the dispersal of the symbols of royal power and its divine legitimacy through a program of monumental relief carving.
The origins of the Assyrian landscape vision may have come, at least in part, from emulation. When Aššurnaṣirpal II commissioned the creation of a new capital at Nimrud, large walled cities had been in existence in northern Mesopotamia for over a millennium, but were subject to a certain demographic carrying capacity imposed by the environment and socio‐technical limitations (Wilkinson 1994). Instead, Assyrian kings probably looked to the south, as they had for centuries. Since the late fourth millennium BCE, the plains of southern Mesopotamia had been characterized by cities hundreds of hectares in scale, densely populated, and surrounded by rich agricultural lands whose productivity was enhanced with elaborate systems of irrigation. Emulative aspects of the Assyrian landscape vision could also be specific. For example, Sargon II and Sennacherib commissioned replicas of north Syrian and Babylonian landscapes in their respective capital cities, including the simulation of marsh conditions and the importation of botanical samples (Brinkman 1995: 28–9; Radner 2000: 239–40; Thomason 2001).
Conclusions and Future Prospects
The transition from the Early Bronze Age landscape to the Iron Age landscape in the land of Assyria represents a dramatic shift between two particularly clear signature landscapes, each on opposite ends of a continuum between emergent and imposed landscapes. Despite clear evidence of centralized political authority and socioeconomic inequality, the cities of the Early Bronze Age have very few unambiguous signs of planning. Likewise, the simultaneously intensive and extensive agricultural system undergirding them bears no direct evidence of royal or any other form of coercion in its formation. Rather, it appears that both were largely an emergent result of widely held rules and values concerning household subsistence, land tenure, patterns of movement and communication, and spatial patterning. Centralized authorities did not impose urbanization, trackway patterning, and agricultural intensification, although they may have benefited from these processes. The Early Bronze Age model developed to its greatest extent in the second half of the third millennium BCE, but had its origins in durable and nucleated agricultural patterns of settlement that extended back to the fifth millennium BCE (Wilkinson 2003: 105–9).
The Neo‐Assyrian landscape of the ninth to seventh centuries BCE developed in an almost identical physical environment in terms of soils, hydrology, and climate. Nonetheless, the Neo‐Assyrian model presented a dramatically different signature, and shows many indications of being deliberately imposed by centralized planners, likely on the basis of a singular vision of the proper form of the Assyrian landscape.
These two signature landscapes, one largely emergent and one largely imposed, both proved to be fragile and ultimately unsustainable. In the case of Early Bronze Age urbanism, the largest settlements grew to scales beyond what the environment and the subsistence economy of the time could sustain in the long term, despite economic adaptations toward expanding and intensifying production; only a few of the largest settlements survived more than a half millennium in an urbanized state. The Assyrian Empire developed technological and social means to overcome earlier urban demographic limits (irrigation, water transport, the efficient spatial distribution of agricultural labor, and a quasi‐monetary economy), but the Assyrian cultural landscape dissolved nonetheless, coincidentally (it would seem) with the political collapse of the empire. In this case, the shared values and motivations of the sort that had enabled the emergence of the Early Bronze Age landscape were missing. The citizens of Neo‐Assyrian cities and their hinterlands did not necessarily share the landscape vision that had been imposed on them, many of them having been brought against their will from their homelands in Babylonia, Judah, and elsewhere, or forcibly settled on their former pasturelands (Yoffee 1988). The Assyrian landscape was unsustainable not because of environmental limitations but because the imperial authorities and the bulk of the population did not share common identities, values, and ideas about what the land of Assyria should look like.
In archaeology, conclusions are rarely final, but rather are (or should be) the best that can be drawn from the incomplete dataset at hand. The conclusions regarding the evolution of the Assyrian landscape presented in this paper represent generalizations based upon a particularly uneven archaeological record. In particular, the divergent histories of scholarship within the Republic of Iraq on the one hand, and in the Syrian and Turkish Republics on the other, mean that many conclusions drawn on extensive data from the latter two nations must be extrapolated to far less vigorous data from the first. The degree to which urban dwellers at Early Bronze Age sites currently in Iraq (especially at Tell Khoshi, Tell Taya, and Tell Baqrta) modified the hinterlands of their cities is a subject for future research. The same can be said about the nature of Neo‐Assyrian rural settlement and land use in the imperial core along the Tigris River, which is currently modeled from urban and irrigation data from excavation and remote sensing in Iraq and from rural settlement data from Syria and Turkey (although see now Ur et al. 2013; Ur and Osborne 2016; Morandi Bonacossi 2012–13).
These conclusions should be taken as points of departure for further testing, rather than as established facts. At the time of writing, there is at least some reason for some optimism, as a new generation of Iraqi scholars of ancient landscapes is emerging (e.g., Al‐Hamdani 2008) and foreign research is resuming in the Kurdistan Region (Mühl 2010; Ur et al. 2013). Remote sensing analyses employing CORONA and more recent satellite imagery have revealed a vast array of sites and landscape features in northern Iraq (e.g., Altawe
el 2008) that are only just recently receiving systematic study in the field using the new methods that have been developed in Syria and Turkey over the last two decades. If sociopolitical stability, governmental priorities for cultural heritage, and archaeological research agendas can coincide, northern Iraq may yet see its golden age of landscape archaeology.
Acknowledgments
I am indebted to Mark Altaweel, Hartmut Kühne, John MacGinnis, Susan Pollock, and Tony Wilkinson for references and critical comments on an early draft of this manuscript.
References
Adams, R.M. 1981. Heartland of Cities, Chicago: University of Chicago Press.
Adams, R.M. 2005. “Intensified Large‐Scale Irrigation as an Aspect of Imperial Policy: Strategies of Statecraft on the Late Sasanian Mesopotamian Plain,” in: J. Marcus and C. Stanish (eds.), Agricultural Strategies, Los Angeles: Cotsen Institute of Archaeology, 17–37.
Akkermans, P.M.M.G. and Schwartz, G. 2003. The Archaeology of Syria: From Complex Hunter‐Gatherers to Early Urban Societies (ca. 16,000–300 BC), Cambridge: Cambridge University Press.
Al‐Hamdani, A. 2008. “Protecting and Recording our Archaeological Heritage in Southern Iraq,” Near Eastern Archaeology 71, 221–30.
Al‐Quntar, S., Khalidi, L., and Ur, J.A. 2011. “Proto‐Urbanism in the late 5th Millennium BC: Survey and Excavations at Khirbat al‐Fakhar/Hamoukar, Northeast Syria,” Paléorient 37, 151–75.
Altaweel, M. 2005. “The Use of ASTER Satellite Imagery in Archaeological Contexts,” Archaeological Prospection 12, 151–66.
