The unexpected discovery in 1958 of the now famous “Hasalu Gold Bowl” in a burned occupation level at that site led to extensive excavation of the early Iron Age settlement. This prehistoric cultural period at Hasanlu, located in northwestern Iran, begins in the second half of the 2nd millennium B.C. and ends around 800 B.C. Later remains, also of the Iron Age, come from the historically known Urartian and Achaemenid periods. The excavations were part of the Hasanlu Project which ran from 1957 through the summer of 1977; they were sponsored by The University Museum and the Metropolitan Museum of Art of New York, and the Archaeological Service of Iran.
When the Project was initiated, its general goal was to reconstruct the cultural history of the Solduz-Ushnu valley. This was to be achieved through excavation of a series of stratified occupation levels that spanned the prehistoric period in this region, beginning with the first Neolithic settlements around 6000 B.C. and ending with the conquest of Iran by Alexander the Great in the 4th century B.. Excavations were carried out at several sites in addition to Hasanlu in order to accomplish this broad objective; the most intense research focus was, however, on the Iron Age. The study was long-range and interdisciplinary, and formed a model for several later projects in other parts of Iran.
The description of cultural change was to be based directly on archaeological evidence, independent of linguistic arguments and speculation based on historic documents from adjacent regions. Subsequently, other scholars have used such arguments to identify ethnic groups within northwestern Iran, and to interpret the Iron Age remains at Hasanlu. In view of the continuing process of analysis of the Hasanlu materials, their conclusions must be regarded as provisional.
The Process of Discovery and Interpretation
In any archaeological project, the process of discovery and interpretation involves a complex interaction between excavation strategy and ongoing analysis of the recovered data. As excavation proceeds, accidents of preservation directly affect the interpretation of the deposits encountered, raising a series of technical questions often resolved only by further seasons of excavation.
At the same time, the study of excavated architecture and artifacts is fundamentally affected by the discovery of new remains at other sites, and by the consequent revision of broad historical or theoretical structures of interpretation. This interactive process is one of continuous learning, requiring periodic reassessment of the conclusions reached at any given stage of the investigation. Hypotheses formulated about the Iron Age in western Iran in the 1980s, for example, are now being reformulated as the result of the accumulation of new data (compare Young 1963 and 1985). The process explains why continuing field work is so valuable and why so much time is required to produce integrated final reports when the work is done.
The Problem of Preliminary Conclusions
A season of field work provides the excavator with a set of initial impressions and some seemingly obvious conclusions based on the information in hand. Following each field season, those people interested in the project want instant interpretation of the finds: how old they are, what they were used for, how were they made, who were the people responsible, and so forth. Usually the excavator produces some reasonable response to these demands in a preliminary report. If work continues, such responses are inevitably modified or even proved wrong.
At Hasanlu, for example, we encountered a major fortification wall 3 m thick, with massive stone foundations, bastions, and towers, surrounding the top of the High Mound (Figs. 4, 6). In 1958 we associated this wall with the burned Iron Age level of 800 B.C. (Hasanlu period IVB); it is now clear that this conclusion was erroneous, and that the wall belongs to a later Urartian occupation (period IIIB; see Fig. 5). How did we make this mistake? The answer illustrates the process of discovery and interpretation just described.
When the wall was discovered, no comparable architecture had been described in northwestern Iran. It was only after 1960 that securely dated Urartian structures began to be published as a result of surveys carried out by field teams from the German Archaeological Institute in Tehran. The results of these surveys led in part to a reassessment of the date of the Hasanlu wall, a re-evaluation already suggested by continuing excavation. The fortification wall was found during the excavation of Burned Building I-West (Figs. 4, 11). Excavation had progressed across the scorched and blackened floor of its central columned hall, and had ended neatly against the face of the stone foundation of the fortification wall. The latter lay at a right angle to the hall’s side walls, and appeared to form the back wall of the room. In the absence of evidence to the contrary we concluded that we had cleared a unified structure that belonged to the period of the great fire (Masanlu IVB).
This conclusion seemed to be confirmed by the recovery of burned debris lying against other parts of the fortification wall, inside the nearby West Gate and at the foot of Tower 4 (Fig. 6). Four years later (in 1962) Burned Building HI was excavated in the northwest quadrant of the High Mound and the pattern appeared to be repeated: the inner face of the fortification wall seemed to form the rear wall of a rectangular room, located at the back of the building.
It was not until 1972, when excavation was undertaken on the south side of the mound behind Burned Building II, that a well-preserved stratigraphic section revealed unequivocally that the fortification wall sat in a huge foundation trench dug into the burned ruins of Hasanlu LVB (Figs. 7, 11). The wall’s construction could now be seen to have taken place during the subsequent Urartian period of the 8th/7th century B.C. (Hasanlu period IIIB; Figs. 6, 8). This interpretation was based on the wall’s stratigraphic position (within a foundation trench cut from above Burned Building H), the pottery associated with its use, and its plan. The latter could be seen to duplicate features of Urartian fortresses of the same date newly documented elsewhere in northern Azerbaijan. The stratigraphic sequence was confirmed in 1972 and in 1974 while exploring the road system on the west slope (Fig. 11). The whole process of discovery and final interpretation had taken 6 field seasons spread over 12 years!
Excavations on the High Mound
Hasanlu is the largest site in the Gadar River valley, which runs from the Zagros mountains on the west, eastward to the marshy southern shore of Lake Urmia. The western half of the valley is called Ushnu, while the eastern half is called Solduz. The valley provides a direct route from the borders of Assyria on the west to highland Iran on the east.
The site of Hasanlu consists of a high central mound that rises 25 m above the plain and is about 200 m in diameter at the top (the “High Mound,” sometimes referred to as the “Citadel”). Around its base is a low, flanking mound that rises about 8 m (the Lower Mound, also referred to as the “Outer Town”). At its widest point the site extends approximately 600 m from one edge to the other (Figs. 1, 3). These existing limits are, however, artificial since vineyards have been cut into the sides of the Lower Mound, and part of the site is known to run under the present village of Hasanlu. between the various architectural units; this chronology is independent of pottery and artifacts. Radiocarbon determinations also provide evidence for the absolute chronology of the site (Fig. 5; see also Dyson and Muscarella 1989.
The Challenge of Interpretation
The preservation of so many artifacts in architectural contexts, and their abrupt burial as the result of the sudden collapse of the burning buildings, provides a unique opportunity for the study of a corpus of co-existing styles, artifact types, and methods of manufacture. The burned ruins of Hasanlu IVB provide the final resting place for these various objects; however, their paints of origin in time and space, and their function remain as problems of interpretation that must be studied category by category and even object by object. This is a complex task, but one which can be undertaken by considering the geographical location of Hasanlu and adopting a model of analysis developed for similar sets of conditions in adjacent regions of Asia.
The syncretistic nature of the architecture and artifact assemblages at Hasanlu may be understood by recognizing its location on trade routes leading from the kingdom of Urartu to the north, from Assyria and Syria to the west, and from Manna and Media to the southeast (Fig. 13). These trade routes (used also for military campaigns) facilitated the movement to Solduz of materials, objects, craftsmen, teachers, and officials from neighboring centers.
There is little reason to doubt the presence of foreigners at the site either as occasional visitors or as residents, a pattern already well documented for later periods at Pasargadae, Persepolis, Susa, and Babylon (Fig. 13). For example, ethnic groups that worked for the state and were present in Achaemenid cities included Cap Cappado-cians, Lydians, Carfans and Ionians from Asia Minor, Sogdians and Bactrians from Central Asia, Babylonians, and Egyptians. Given its location in a border area, it seems likely that Urartians, Assyrians, Hurrians, Mannaeans, Medes, and others could have been present as individual craftsmen in a center like Hasanlu. The political role that such foreigners may have played is not known, since we lack written sources at the site. What we do know is that objects from as far away as Elam and Assyria reached Ilasanlu during the Iron Age (Figs. 10, 12).
The model of cultural dynamics most likely to be useful for our purpose is one in which newcomers blend their own traditions with preexisting ones in an area. Groups in close proximity to more prestigious political and cultural entities have historically sought to increase their own prestige by adopting symbols, practices, and objects from these more powerful neighboring states (see Marcus, Pigott this volume). More specifically, political leaders in such new centers have consciously copied elements of architectural style in order to transfer visual symbols of power from established political centers to them selves, a practice amply documented by the actions of Assyrians, Elamites, and other rulers in ancient Mesopotamian and Iran (see Dyson, “Architecture,” this issue).
The work of Boris Marshack and V. I. Sarianidi at early historic sites in Bactria (Afghanistan and Central Asia) provides explicit examples of such a syncretistic process at work. Sarianidi’s recent study of tomb contents from Tillya Tepe provides a set of categories that can be used for an analysis of artifacts at Hasanlu (see Sarianidi 1990). The Tillya necropolis dates to the end of the Graeco-Bactrian kingdom, preceding the Kushan Empire (1st century B.C. – 1st century A.D.). The objects found in this cemetery fall into the following categories, defined on the basis of origin: (1) imported objects from Parthia (Iran), India, China, Siberia, and the Roman Empire; (2) booty given to the conquering Yueh-chi Chinese overlords, which includes local jewelry and heirlooms; (3) locally made objects in styles derived from the classical Greek traditions of the Craeco-Bactrians; (4) objects with syncretic images, created through the combination of symbols and stylistic elements drawn from neighboring high cultures and pre-existing local traditions; and (5) objects in a purely local style derived from persisting Bronze Age traditions.
This same kind of mixture, reflecting multiple cultural sources, appears among the objects at Hasanlu: some objects are imported, some are local imitations, some are from earlier local traditions, some are syncretistic, and some are heirlooms. The full understanding of their cultural and historical significance requires a consideration of their ultimate origin, as well as an examination of their role in the community in which they were found. The articles that follow in this issue of Expedition address the problem of interpretation from varying points of view. The wealth of hidden information that is being teased out of this material forms the current process of rediscovering Hasanlu.