The goal of the University of Chicago’s excavations at Zincirli in the first phase of the project from 2006 to 2013 was to obtain new data to complement existing information from previous excavations and ancient inscriptions, in order to answer questions about: (1) the chronology of the site’s occupation and its regional context and contacts in each period; (2) the origin of the site’s inhabitants and their ethnocultural interactions; and (3) the socioeconomic organization of the Iron Age lower town and the changes in this organization over time, from the ninth to the seventh century BCE, as the kingdom of Sam’al went from political independence to client status within the Neo-Assyrian Empire and then was ultimately annexed as a directly governed province of the empire.
Since 2014, the Chicago-Tübingen joint expedition has added the following topics of research: (4) the pace and process of the city’s refoundation and urban expansion in the Iron Age II, within the broader context of Syro-Hittite state formation; and (5) the earlier settlements at Zincirli in the Middle Bronze Age and the Early Bronze Age, with the goal of comparing the urban organization, resource use, and local environment of Sam’al in the Bronze Age vis-à-vis the Iron Age in light of both ethnopolitical changes in the Levant and long-term climate change and anthropogenic modification of the ecology of the region.
Chronology and Context
Although a basic chronology for Zincirli was established by the German expedition in the late nineteenth century, a more refined stratigraphic sequence was needed to determine the phases of occupation at the site and to isolate the architecture and artifactual assemblages associated with each phase. The first stage of our excavations from 2006 to 2013 focused on the final phases of the Iron Age occupation in the eighth and seventh centuries BCE, and also established a new outline of the stratigraphic and material-culture sequence from the Early Bronze Age to the early Hellenistic period (i.e., from ca. 3000 to 300 BCE).
We have now collected enough evidence to show that the pottery of the site, although broadly similar to the pottery of other sites in the region, is highly local in character, with many idiosyncrasies and very few imported wares. This is in keeping with Zincirli’s relatively isolated geographical position in a narrow valley hemmed in by mountain ranges. Occasional imported wares, such as Cypriot Black-on-Red, Cypriot Bichrome IV, Cypriot White Painted IV, Samaria Ware, Assyrian Palace Ware, and East Greek imports, provide a means to link the local sequence to ceramic sequences elsewhere.
Meanwhile, the CRANE project based at the University of Toronto and directed by Prof. Timothy Harrison has helped to fund radiocarbon analyses (conducted by CRANE project member Sturt Manning of Cornell University) to establish an absolute chronology for the architectural phases and local ceramic sequence of Zincirli. This work on absolute chronology is being continued by the Chicago-Tübingen Expedition with funding from the Deutsche Forschungsgemeinschaft.
Our chronological research, although fundamentally descriptive, is essential for understanding the site within its wider context, as well as for making temporal correlations from one excavation area to another within the site. Not only will it provide an essential framework for future archaeologists who excavate in the same valley, but it will also be essential for a regional survey project planned for the future, in order to date the periods of occupation at the scores of settlement sites in the vicinity. The Zincirli ceramic sequence will be used to understand the settlement history of the region, comparing the Zincirli material with surface artifacts collected from surveyed sites identified on the ground and via satellite imagery. A chronologically refined settlement history will be of great value for understanding the economic and political functioning of the kingdom of Sam’al in the Iron Age, and of earlier polities in the region, by indicating how many and what kind of settlements existed in a given period. With a sufficiently precise chronology, the wider settlement history and the architectural changes at Zincirli itself can be related to broader political and cultural changes of the Bronze and Iron Ages, such as the imperial conquests and population migrations documented in ancient texts.
The kingdom of Sam’al was located in a border region between the ancient Anatolian and Levantine cultural zones. Bounded on the north and west by the towering Taurus and Amanus mountain ranges, it was historically the most northwesterly region of habitation of Semitic-language speakers. In many periods, this region had cultural ties with inland Syrian population centers to the south and east. However, within the Semitic-speaking royal dynasty of Iron Age Sam’al were kings who bore non-Semitic Luwian names, reflecting the powerful political and cultural influence of Luwian-speakers of Anatolian extraction who had migrated southward into the region centuries earlier from across the Taurus Mountains under the aegis of the Hittite Empire and continued to dominate North Syria in the immediate post-Hittite period. The Semitic-speaking rulers Sam’al in the Iron Age II themselves adopted the Neo-Hittite style of architecture and iconography favored by earlier Luwian-speaking elites, indicating the continuing prestige of that cultural tradition.
The Neo-Hittite cultural tradition in North Syria was transmitted within the rump kingdoms that emerged after the collapse of the Hittite Empire in the early twelfth century BCE, which were ruled by Luwian-speaking elites. In particular, Carchemish on the Euphrates River, 100 kilometers east of Zincirli, which was formerly the seat of the Hittite viceroy (and prince of the imperial dynasty) who ruled Syria on behalf of the Hittite king, became the capital of a powerful local kingdom with a Luwian-speaking dynasty. The presence of Luwian hieroglyphic inscriptions at various sites in the Zincirli region suggests that it was initially part of the Luwian-dominated Carchemish kingdom, or at least was controlled by a local Luwian-speaking elite, until a Semitic-speaking dynasty took over in the late tenth or early ninth century BCE and chose the abandoned Bronze Age mound of Zincirli as the capital of their kingdom.
The newly expanded and fortified city was called Y’DY (possibly its Luwian name, pronounced Yādiya) and was also referred to interchangeably by the Semitic name Sam’al (ŚM’L, meaning “north”). It is possible that one of these two names denoted the city proper while the other denoted the kingdom of which it was the capital. In any case, the new Semitic-speaking rulers adopted Neo-Hittite iconography and decorative styles similar to those of Luwian-ruled Carchemish, as shown by the basalt orthostat reliefs lining the gates of their city, indicating the continuing prestige of that cultural tradition, which was widely imitated even by non-Luwian elites.
Further complicating the ethnolinguistic situation, it is worth noting that the Semitic place-name Sam’al was in use already a thousand years earlier, as shown by an Old Assyrian text from the Middle Bronze Age, which indicates that there was a Semitic-speaking population in the area long before the Hittite Empire and subsequent Luwian domination, and that the old Semitic name was never forgotten. This is confirmed by the strongly Amorite character of the artifacts and architecture found at the Middle Bronze Age royal capital excavated at the site of Tilmen Höyük, located just 8 kilometers south of Zincirli (see Marchetti 2006). The rulers of Tilmen had close cultural and political links to regions to the south in inland Syria, which was dominated at that time by the kingdom of Yamhad, whose capital was at Aleppo. Similar Middle Bronze Age material has been found at Zincirli itself, which was apparently a fortified center in the Amorite kingdom ruled from Tilmen. The place-name Sam’al (written ś-m-i-r-w) appears also in the Late Bronze Age in an Egyptian list of North Syrian toponyms carved on the wall of the temple of Amun at Karnak to celebrate the exploits of Thutmose III, a pharaoh who campaigned repeatedly in the region, conquering the kingdoms along the Orontes River and its tributaries and marching as far as the Euphrates River in his eighth campaign around 1450 BCE (see Astour 1963, p. 233).
For the Bronze Age Amorites of the Levant, Zincirli was in the “north” (ŚM’L), just as much as it was for the Iron Age Arameans who supposedly invaded in the tenth or early ninth century BCE. But the only reason to think that the Iron Age rulers of Sam’al were invading Arameans, as opposed to indigenous Semitic-speakers who had been resident there for a millennium or more, is the linguistic classification of the Sam’alian dialect (attested in local Iron Age inscriptions) as a branch of Aramaic. This is a problematic assumption, because there is some question as to whether Sam’alian is actually Aramaic (Huehnergard 1995; Pat-El and Wilson-Wright, forthcoming). It does not possess a number of morphological innovations shared by other Aramaic dialects, so it could instead be an otherwise unattested branch of Northwest Semitic spoken in the Zincirli region since the Amorite period. In that case, Gabbār, the founder of the Iron Age kingdom of Sam’al, may well have been, not a roving Aramean warlord, but a local resident of Amorite heritage who threw off the Luwian yoke and restored his Semitic-speaking compatriots to a position of power. There is certainly no archaeological hallmark of the Arameans as an invasive ethnic group that can be detected at Zincirli. If Gabbār and his followers were not Aramean, it would change the historical picture considerably. To the extent that Sam’al became Aramaized, as is indicated by the use of Official Aramaic in King Barrākib’s last known inscription, this would have occurred much later, in the late eighth century, under the aegis of the Neo-Assyrian Empire and its Aramaic-speaking administrators who hailed from northern Mesopotamia.
At the very least we can say that, over the course of the Late Bronze Age II and the Iron Age I and II, there was some form of coexistence and ethnocultural interaction among what were originally Luwian-speaking and Semitic-speaking populations, with Luwian elites politically dominant during and after the Hittite Empire, in the Late Bronze Age II (after 1340 BCE) and in the Iron Age I, and a Northwest Semitic-speaking elite dominant both before the Luwian regime, in the Middle Bronze Age, and after it, in the Iron Age II (after 900 BCE). Scholars have long suspected that the rivalry between these two groups is reflected in the enigmatic reference to the downtrodden muškabīm (literally, “those who lie down”) and a presumably better-off group called ba‘rīrīm in the Phoenician inscription of Kulamuwa, king of Sam’al in the late ninth century BCE. Kulamuwa claims to have aided the muškabīm (the down-at-heel Luwians?) and to have fostered mutual respect between the two groups. Moreover, despite a long period of coexistence and perhaps even intermarriage, as has been suggested to explain the appearance of both Semitic and Luwian names within the Sam’alian royal dynasty (e.g., Kulamuwa himself, whose father Ḥayyā had a Northwest Semitic name but whose own name is Luwian—perhaps he was the son of a queen who belonged to an old Luwian noble family), the original ethnic identities were not forgotten and could surface in various forms. The assertion of a Luwian Neo-Hittite identity may have been politically advantageous for Kulamuwa in the late ninth century BCE, in the aftermath of the devastating Assyrian invasion under Shalmaneser III. This phenomenon is widely attested in other historical periods, including the modern Middle East.
A striking example of ethnocultural interaction at Zincirli is the Katumuwa Stele, on which was carved a 13-line alphabetic inscription written in Sam’alian, the local Northwest Semitic dialect. Katumuwa was a Sam’alian royal official with a Luwian name who served a king with a Luwian name, Panamuwa II, whose dynastic predecessor Barṣūr and successor Barrākib both had Semitic names. Katumuwa’s inscription refers to both Semitic and Anatolian deities (including the goddess Kubaba, known in classical times as Cybele), and he was apparently not buried in a communal tomb to which the offerings were brought, according to the traditional West Semitic custom, but was most likely cremated, a practice attested in culturally similar cities of the period such as Carchemish on the Euphrates, Hamath on the Orontes, and Guzana (Tell Halaf) in Upper Mesopotamia. Cremation was an Indo-European practice generally regarded as abhorrent in the Semitic world (for example, in ancient Israel and its neighbors). In both his name and his mortuary customs, if not his language, Katumuwa had plainly not forgotten his ancestral traditions, even though he lived some two hundred years after the Luwian elite had lost control of the region. It seems that Luwian names and Luwian gods were still favored by some in Sam’al, even at this late date, along with distinctive Anatolian mortuary practices. Perhaps these names and practices were adopted even by people of non-Anatolian extraction for political or social reasons, just as people today will adopt foreign names and identities to enhance their social standing. Or, we must reckon with the ongoing vitality of Luwian cultural traditions in the midst of an equally vital and dynamic Semitic-speaking cultural milieu. The discovery in our recent excavations of non-monumental epigraphic material in both Luwian (a lead document fragment, a Hittite-period stamp seal, and a short epigraph on ivory) and Northwest Semitic (a jar inscription, a scaraboid stamp seal, and an inscribed cosmetic container) suggest that both languages were in use in Iron Age Sam’al, even though Northwest Semitic languages (Phoenician, Sam’alian, and Offical Aramaic) were employed in monumental inscriptions.
The archaeological question is whether these enduring identities deployed by intermingled ethnic groups, and the cultural influences exerted by these groups on one another, can be detected in their material remains. We are approaching this question through the careful analysis of spatial and temporal patterns of architecture, artistic styles and iconography, cuisine (detected via pottery and botanical and faunal remains), mortuary customs, and other social practices revealed in extensive exposures of the Iron Age lower town. The goal is not to identify specific households or individuals in reductive terms as permanent members of this or that monolithic, crudely reified ethnic group, but to study processes of group-identity formation and identity maintenance from the point of view of the nonverbal social practices and habits that in every society accomplish the socialization of individuals into communities.
The Iron Age inhabitants and their ethnocultural interactions at Zincirli can be studied on a scale and at a level of detail that is unusual in Near Eastern archaeology because of the large quantity of Iron Age urban architecture readily accessible just under the modern surface. A modern village has grown up over the western part of the upper mound and lower town, but at least 20 hectares are available for excavation. In most places of the lower town, there are only about 1.5 meters of accumulated debris representing two or three architectural phases. A given area can be excavated down to virgin soil in two or three two-month field seasons, depending on the complexity of the stratigraphy. This is important because a quantitative increase in the scale of excavation will produce a qualitative leap in our understanding of urban life and culture in this formative period of Mediterranean history. An ancient settlement’s social and economic organization cannot be understood from a limited sample of individual houses. We have now excavated a total of 7,000 square meters in and around the site, providing a sample of architecture and artifacts large enough to permit meaningful conclusions about the social and economic organization of the city.
There are very few Iron Age sites in the Mediterranean region at which large-scale exposures of urban districts have been achieved. The Zincirli excavations, profiting from easy access to well-preserved Iron Age strata of both the Assyrian and pre-Assyrian periods, provide a large quantity of new evidence concerning the organization and use of urban space. This evidence will allow us to answer questions about population density, subsistence practices and food storage, craft production and economic specialization, livestock stabling, household size and composition, and neighborhood relations (as shown by multi-house architectural arrangements in relation to shared courtyard spaces and other shared facilities). At Zincirli, geophysical surveying methods have been shown to produce an unusually clear picture of buried structures in the lower town, at least for the latest architectural phases representing the last century of habitation in the lower town. This subsurface survey data augments the data obtained from excavated areas, which in turn can be used to interpret more accurately the geophysical maps of unexcavated areas.
A key question has to do with the existence (or not) of kin-based or quasi-kin modes of social organization. In other words, were there urban “clans” or patron-client household groupings governed by politically powerful patriarchs, forming economically autonomous and mainly agrarian sub-communities within the larger city, on the model of traditional Islamic cities or medieval Mediterranean cities in Italy and elsewhere? This model of urban farming clans (fictive or not) that had moved from their rural villages to land allotments within the city walls but had retained their traditional mode of life and kin-based social organization, is in marked contrast to the widely adopted model of urban organization that posits economically specialized urban households interacting and competing as individual units and integrated with one another, not by means of kinship, but by some form of market economy, or else by a top-down command economy focused on royal military and labor requirements and the large-scale distribution of royal rations. A careful study, not just of individual houses, but of groups of neighboring houses, can determine which model best fits the evidence.
Zincirli also provides the opportunity to examine changes in socioeconomic organization over time. The Iron Age lower town was in use for at least 200 years and witnessed three political stages, from independence under the rule of a local king, to client status within the Neo-Assyrian Empire, to the removal of the local dynasty and direct imperial rule by a provincial governor, and possibly also some measure of deportation and population replacement, although this is not textually documented for Sam’al, as it is for other kingdoms of the era. Did these major political shifts, from independence to provincialization, leave a visible mark on ordinary urban districts, reflecting the reorganization of urban elites and economic production in line with the demands (or incentives) created by the empire? Was there an upsurge in interregional trade as a result of the “Pax Assyriaca,” causing a restructuring of the Sam’alian economy, with ripple effects at the household level? Or were daily life and the use of space in the lower town largely unaffected, even though the royal citadel (and the members of the royal court) undoubtedly experienced drastic changes?
The site of Zincirli provides an ideal laboratory for investigating all of these issues: for examining ethnicity in an urban population of diverse origins affected by and adapting to cross-cutting cultural influences; for examining the material correlates of identity-forming and identity-maintaining social practices; and for examining Iron Age urban subsistence, and household and neighborhood organization, both before and during incorporation into the vast Neo-Assyrian Empire. Large-scale horizontal exposures of coherent architectural phases provides valuable new data to address these questions.
Urban Expansion and State Formation
A major objective of recent research at Zincirli by the Chicago-Tübingen Expedition is to investigate the relationship between urbanization and state formation at Iron Age Sam’al by means of excavations that will improve the chronological and spatial resolution of our reconstruction of the city’s expansion and renovation. The Iron Age Syro-Hittite cities of the Northern Levant illustrate the phenomenon of urban foundation (or re-foundation) that tends to occur in periods of political transformation and is attested in many historical periods. An aggressive policy of urbanization carried out by the ambitious rulers of new, small states has been recognized as a key element of secondary-state formation in this period. However, the details of this process, including how and by whom the cities were populated, are still poorly understood.
Previous excavations at Zincirli established this city as an archetypal example of the coincidence of Syro-Hittite urbanization and state formation. A goal of the current excavations is to produce a more nuanced and holistic picture of the re-founding and expansion of a walled city at Zincirli in the Iron Age II. This has been a major emphasis of our research in its second stage, since 2014, when the University of Tübingen partnered with the University of Chicago to continue the work at Zincirli. A comparison of stratigraphy, material culture, and radiocarbon dates from four excavation areas in widely separated locations is allowing us to refine the chronology of urban expansion and development, while shedding light on the socioeconomic organization of the first Iron Age settlers through contextual analysis of their architecture, installations, pottery, and artifactual and faunal remains. This research will contribute a new perspective on the role of urbanization in the formation of the Iron Age Syro-Hittite kingdoms and will contribute to broader theories concerning ancient practices of sovereignty as they affected everyday life.
Zincirli in the Early and Middle Bronze Ages
Since 2015, the ongoing excavations of the Chicago-Tübingen Expedition have, somewhat unexpectedly, found excellent points of access to the earlier periods of occupation at Zincirli buried under the Iron Age city. Our first forays into these strata are already transforming our view of Bronze Age Zincirli and are providing a better basis for studying the material culture of these periods.
The Middle Bronze Age town, although apparently restricted to only a few hectares on the upper mound, turns out to be much more monumental than expected. We are finding pottery and seals with far-flung connections ranging from the Middle and Upper Euphrates River to central Anatolia. The revelation that Hilani I was constructed in the Middle Bronze Age rather than the Iron Age raises new questions about the origins of the influential hilani architectural form and its transmission to the Iron Age kingdom of Sam’al. Also in question are the political and economic relationship of Middle Bronze Age Sam’al with the palatial center at nearby Tilmen Höyük (probably ancient Zalwar), as well as Sam’al’s role in the Old Assyrian caravan-trading network.
Furthermore, the recent discovery of an extensive settlement of the Early Bronze Age IV under the southern part of the Iron Age lower town compels us to regard Zincirli as a more significant settlement in this period, too, than scholars had previously thought. The Early Bronze Age lower town is accessible at a depth of only about 1.2 meters below the modern surface, immediately under the earliest Iron Age II phase. This presents us with the opportunity of doing extensive excavations in what were presumably residential districts of an Early Bronze Age city, which is otherwise known in this region almost exclusively from graves (e.g., at Tilmen Höyük and Gedikli Höyük). We have a valuable opportunity in future field seasons at Zincirli to learn a great deal about urban organization, social life, and economic activity in the third millennium BCE. We plan to pursue this line of research while keeping in mind that changes in settlement pattern and socioeconomic organization in this period may have resulted from the expansion of the powerful kingdom of Ebla in Syria, or from a major climate change around 2200 BCE that resulted in long-term aridification and drought, for which there is considerable scientific evidence.
Our ongoing excavations and analysis of excavated materials, in combination with geophysical remote-sensing methods (e.g., ERT and GPR) that can reach beneath the Iron Age remains in the lower town, will help us to answer these new questions. Eventually, these efforts will provide a basis for the cross-temporal comparison of urban organization, regional interconnections, and the character and utilization of the local environment, including the prized timber resources of the Amanus Mountains, in the third, second, and first millennia BCE.