Middle Assyrian Occupation at Tell Fekheriye
Excavations carried out this season concentrated on an area of the Middle Assyrian occupation on the terrace of the western slope of the high mound, where trenches C-I to C-V are situated. The architectural remains start to appear directly underneath the modern surface but are superimposed by deposits of Late Bronze-Age – Iron-Age transitional layers from which several graves were dug into the Middle Assyrian levels. These will be discussed separately. Apart from that, due to the erosive effect on the slope and agricultural activities carried out over a considerable time, only few remains such as early Islamic or Byzantine wells, as well as fragments of floor levels and a tannur-installation in the eastern part of the excavation area are preserved (Fig. 1). On the western part of the terrace, close to the lower town of the Late-Antique city of Rhesaina, deep foundations made of huge limestone blocks cut into the slope and disrupt the Middle Assyrian occupational levels. Embedded between the early Iron-Age and Byzantine layers, mud-brick installations with several kilns have been excavated along the central part of the terrace between Trench C-III and Area D in the south. Preliminarily they have been dated to the Roman period based on the pottery contained within (2009: C-III: Roman period, Fig. 3). Despite this none of these activities have drastically destroyed the Middle Assyrian architecture, which is well preserved with walls standing up to a height of approximately 1.2m. These remains were excavated along the terrace at an elevation of about 354,90m a.s.l. and over a length of about 110m between Trench C-I in the north and Area D in the south. The western facade of these buildings can be tracked over the whole terrace and subdivided into at least two architectural units in Trenches C-I to C-IV and another in Trench C-V and Area D, with several building phases and phases of occupation in each unit. In the north House I covers Trenches C-I and C-II, further south a building labelled as House II extends over Trenches C-III and C-IV, whereas the remains of a building formerly known as House III now can be assigned to House II. The surface of the Middle Assyrian occupation was barely reached in Trench C-V as well as in the lowest step of the step trench in Area D, where the room of a building was excavated in 2006 and 2007 (Plan).
Parts of the northern building unit (House I) had already been excavated by the American expedition in 1940 (McEwan et al. 1958) and later re-investigated in 2001 by the Syrian-German mission (Pruß/Bagdo 2002).
Our research in this field season aimed at an in-depth investigation into the stratigraphy and architecture of the northern part (C-I) of the Middle Assyrian House I to better understand its construction, extension to the north and relation to a ‘production area’ exposed in the 2009 season in the centre of the trench and last but not least, to be able to assign the numerous seal impressions and fragments of clay tablets found in the previous seasons more accurately to the different phases of occupation. Excavations carried out in squares 6746 and 6846 were supervised by Francesco del Bravo, Carolin Jauß and Christine Kainert.
In Trench C-II the central part of House I with a courtyard (Room 2) and parts of the reception-room (Room 1) had been excavated earlier and were consequently removed completely in the 2009 season (Fig. 2). The southern facade, a room unit in the southeast and a second courtyard in the east have been the focus of archaeological investigations in this season. This area was supervised by Rosa Reising and Assad Alaw in squares 6645, 6745, 6644 and 6744, as well as Daniela Crasso in squares 6845 and 6844. Trench C-II has been enlarged to the south by another 2.5m in order to include the whole area of House I, thus cutting the area of Trench C-III.
The architecture of the Middle Assyrian House I is very distinct and has previously been described (McEwan et al. 1958, 4-6; Pruß/Bagdo 2002) as being grouped around a central courtyard, having two floor levels with associated finds and a reception-room to the south, as well as being terraced due to the slope of the ancient mound from east to west.
The northern part of the western room unit with Rooms 4, 5, 6 and Corridor 7, was still preserved at the beginning of the 2010 season (Fig. 3). Despite the fact that the floors of the rooms were destroyed to a certain extent by earlier excavation activities and the ensuing exposure to the elements, renewed research has presented a more detailed picture of the internal stratification of the building. The area to the north under the corridor and bathroom shows several floor levels. Is hard to tell which of them can be assigned to the Middle Assyrian House I, since the American excavators partly continued excavating under the foundation level of the walls disrupting the occupied floor levels (McEwan et al. 1958, 4-6, note 2; Pruß/Bagdo 2002, 322, note 23). Nevertheless at least one of these levels has to be assigned to an earlier building phase of House I. It acts as a multilayered foundation for the pavement of the bathroom (Fig 4). Also noteworthy is a sewer east of the room unit that runs parallel to Rooms 5 and 6 and connects the central courtyard (Room 2) with an unexcavated area north of Room 6. It is made of two rows of stones bordering the drainage canal that is about 30cm in depth and 15cm in width. The top is covered by a layer of larger capstones and shards from large storage vessels. A possible continuation north of a diagonal wall in the northernmost part of Trench C-I will be subject of investigations in the next season.
Another installation for draining wastewater is the soakage pit at the northern end of the western room unit. It is located outside the narrow part of the northern wall of Room 6 opposite the toilet and allows for waste-water to be drained into a sewer paved with fragments of baked bricks that passes under the mud-bricks of the wall. The upper part of the pit was slightly conical, tapering towards the top and covered by several large capstones and shards (Fig 5). Such installations are known from many other sites such as Tell Asmar, Tell Beydar, Tell Hariri (Mari), Tell Bi’a (Tuttul), Yorgan Tepe (Nuzi), Tell al-Uhaymir (Kish) and Tell Telloh (Girsu) and have mostly been identified with toilets. Other interpretations are plausible yet little convincing, for instance the use of the installation for craft activities (see: Margueron, in: Syria 85 (2008), 175-222).
Research also continued on the eastern side of the courtyard in Room 8, whose southern half had already been excavated during the 2007 and 2009 season revealing numerous clay lumps with typical Middle Assyrian seal impressions and several fragments of clay tablets (Bonatz et al. 2008). Three consecutive door socket stones next to an opening in the eastern wall, as well as a repair in the wall and a succession of tramples indicate three phases of use in the room.
The known part of House I in the area of the former American sounding VI, re-excavated in the 2007 and 2009 seasons was the starting point for further investigations into the southern part of the building. During the removal of the refill in Room 1, accumulated from the second uncovering by Alexander Pruß in 2001, it was observed that the southern wall tilted northwards and that modern debris and intrusive material had gathered in the crack between it and the adjacent building to the south (House II). As already indicated in the last season, Room 1 is part of the oldest building-phase of the building, whereas the part north of it must have been built separately and added to the existing architecture (2009: C-III: The Middle Assyrian House I, Fig. 2).
Previously unexcavated terrain was unearthed close to the eastern section (squares 6845 and 6844). A rectangular room with two doorways opening to the south and north was found (Room 9). The opening to the north leads to the second pebbled area of the building known as Room 3, which has predominantly been interpreted as a second courtyard by the excavators (McEwan et al. 1958; Pruß/Bagdo 2002). The whole western wall of Room 9 has been exposed over a length of 4.5m and it can be observed that it abutted the eastern end of Room 1. Detailed stratigraphic studies of the architectural construction phases, rebuilding and decay were carried out in this area. They indicate that the room has a foundation made of two and a half rows of mud-bricks, two to three layers in height. On top of the foundation a mud-brick wall was constructed that continues into the eastern section. So far only the western part of the room has been excavated, thus its original dimensions are still unknown. Two floor levels have been observed, the older one consists of a layer of trample with scattered pebble concentrations covered by a thin layer of ash and white plaster. The interior wall in the foreground has been added in a later phase.During that period both doorways existed, the southern with a width of 1m and the northern with a width of 1.5m. In a later period the room was subdivided into an eastern and western part by a small wall with a doorway that has been blocked in a third phase. The door opening leading to the courtyard was also partly blocked in a later period, probably at the same time the partition wall was built. In that moment the broad opening in the northern wall had to be narrowed. This is attested to by two consecutive door socket stones along with the general layout of the threshold. Each new phase is indicated by another floor and after the abandonment of the building the room was filled in several stages visible in several occupational and accumulation levels (Fig 6).
The function of this room is still unclear, yet it is conceivable that at least during the last phase of occupation it acted as an access point to the second courtyard or the whole building.
Generally speaking, the rather preliminary observations made by the American expedition under McEwan and the first joint Syrian-German fieldwork were affirmed, amended and enriched by our own results.
The Middle Assyrian building designated as House I is built on the terraced slope of the ancient mound following an agglutinating construction technique. So far four room units have been excavated. After levelling the area for the different terraces, Room 1, the huge reception-room was constructed. This room unit, consisting of a single room, was abutted from the north by the aforementioned western room unit (Rooms 4 – 7). This unit has approximately the same size as the southern room unit. To the east the situation is more complex. A single room about 4x4m in size (Room 8) was built northeast of the central courtyard (Room 2) and allows passage from this courtyard to the area eastward where the second courtyard (Room 3) is on a slightly higher terrace. Already in 1940 a concentration of several clay tablets was found in this general area and our investigations revealed a further nine fragments that can largely be dated to the reign of Tukulti-Ninurta I (1233-1198 BC). Moreover numerous seal impressions on clay sealings for jars or other storage facilities and clay lumps associated with administrative sealing activities have been found. With these new finds it becomes clear that Room 8 has to be connected to administrative activities, possibly accounting and the preparation of clay for sealings. This room’s connection to the rest of the building is, thus far, only given via a small wall to its south that connects it with the reception room and at the same time separates both pebbled courtyards. The fourth room unit is the aforementioned part east of the reception room. As this unit has not yet been completely excavated it is hard to give any satisfactory interpretation (Fig 7).
The stylistic and iconographic affiliation of the seal impressions found in House I are clearly Middle Assyrian and the motive of one example (TF 7746) can be reconstructed due to the high amount of over 40 fragments on which it is displayed. The motive consists of a contest scene depicting an anthropomorphic winged-lion and a winged-bull, as well as a smaller winged-bull or calf crouched underneath both rearing protagonists (Fig 8) (Bonatz et al. 2002, 111-112, fig. 11; Bonatz, in print; Bartl/Bonatz, in print). An identical impression was imprinted on a clay tablet found at Tell Sheikh Hamad (Dur Katlimmu) making an identification of the seal owner possible. The Assyrian vizier, later grand-vizier, Aššur-iddin has been identified as the owner by Eva Cancik-Kirschbaum (Cancik-Kirschbaum 1996: 22-23, note 76). He is said to have had his administrative seats at Dur Katlimmu (Tell Sheikh Hamad) and Aššukanni (Tell Fekheriye) from where he governed the province Hanigalbat under the reigns of Shalmaneser I (1264-1234 BC) and Tukulti-Ninurta I (1233-1198 BC). This identification, and the resulted dating of the architectural complex of House I, fits well into the glyptic repertoire we have uncovered so far. In the layers underneath House I, a deposit with even more clay sealings, or clay tablets with cylinder-seal impressions has been excavated. One of these seals has now been identified, through the comparison of a sealed letter from Tell Khuera (Ḫarbe) (D. I. Janisch-Jakob in Jakob 2009: 185, seal motive 3), as belonging to Sîn-mudammeq, a high official and vizier in the Assyrian administration. The seal impression found on two clay sealings and on two clay tablet envelopes shows an ostrich hunt (TF 6293). Since Sîn-mudammeq temporarily had his official residence in Aššukanni, this adds to the idea of both officials playing an important role in the administration of the western province at about the same time and that a part of these activities took place at Tell Fekheriye especially during the occupation period of House I, as well as in earlier times as is witnessed by the numerous glyptic and epigraphic evidences from the layers underneath (these objects will be discussed separately).
House I was founded on a preceding monumental mud-brick building, following the terraces of this building as they ascend from west to east according to the slope of the mound. Especially parts of the western room unit, which is erected on the lowest terrace of the slope and the southeastern room unit, which is situated on a higher level, (Fig 9) are close to the levelled walls of this older building, however, in other areas, for instance underneath the central and eastern courtyard and under Room 8 the levels are separated by a layer of accumulations, about 50-70cm in depth, compensating for the slope of the ancient mound. After removing the remaining walls of House I it became clear, that the whole area in trenches C-I and C-II was covered with a more or less homogeneous layer of debris varying in consistency, resulting from discarding activities of some kind before House I was constructed.
Several areas with different accumulations, finds and functions can be distinguished and will be discussed here in brief.
The area between the western room unit with the sewer canal and Room 8 in the east is bordered by a diagonally running wall in the north, which was already part of the previous seasons investigations. It is characterized by several small, single-rowed mud-brick walls and ovens (tannurs). They form part of a production area that is bordered by the aforementioned diagonal wall and comprises two standard sized tannurs and one slightly bigger oven. They have been rebuilt several times, encircled by mud-bricks standing on edge and filled with ashy deposits attesting to a long duration of use. The eastern part of this production area has partly been covered by all three building levels of Room 8 and cut by several Byzantine wells and Late Bronze-Age mud-brick cist graves (Fig 10). In 2010, thick deposits mixed with ashy lenses and traces of firing activities were excavated down to the bases of the ovens and walls, and to a level, which superimposes the monumental mud-brick building. From a stratigraphic point of view, it seems probable that the area can chronologically be placed in the intermediate phase between the abandonment of the monumental building and the construction of the Middle Assyrian House I as this layer has been covered by the courtyard of House I, which contained large amounts of broken, but complete vessels of the Middle Assyrian pottery assemblage (Middle Jezireh II A-B). In 2007 a similar octagonal oven with mud-brick enclosure was excavated in Trench C-II, further to the southeast, and assigned to a level directly under the architecture of the Middle Assyrian building, as was a tannur north of the reception-room (Room 1) of House I. Whether the latter was part of the same production area or part of an earlier phase of House I as suggested in Bonatz et al. 2002 (Bonatz et al. 2002, 112-113, fig. 13) cannot be said with absolute certainty, as the floor-levels were heavily disturbed by previous excavations and winter rain.
However, it has to be noted that these installations, and the related architecture, are connected to the diagonally running wall in the north, whose foundations reach into deeper levels than those of House I. It seems not to have lost its function in later periods witnessed in its use as the northern perimeter of House 1. Despite these observations, the exact stratigraphic relation to House I was not completely clarified, as both the eastern end and connection to Room 8 have been destroyed by later graves and wells, and its western end is hidden in the section. Repeated repairs of this wall are visible and it is highly probable that earlier installations in the production area were covered by the courtyard during the building of House I (Fig 2).
Excavations in the deposits northeast of this diagonal wall continued and another wall in the eastern part of square 6846 was exposed, running parallel to the aforementioned one. A contemporaneousness of those two walls can be postulated but not yet proved, as the base of the accumulations and deposits between both has not yet been reached. It will therefore be subject of investigations in the next season (Fig 11).
Interesting finds were made in an area, which comprises the northeastern part of square 6845 and southeastern part of square 6846, approximately starting underneath Room 8 of House I. It covers the area to the west up to the production area and continues eastward into the section of the trench. Parts of this area had already been excavated in 2009. The predominantly ashy layers are the result of discarding activities and contained a high concentration of clay lumps with or without seal impressions, a few small clay tablets and fragments of clay envelopes amongst other finds. About 50 sealed objects and several fragments of sealed clay tablet envelops discovered this year, bear diverse impressions of mostly mature Middle Assyrian style seals that date to the mid 13th century BC (Fig 12). Amongst them are the aforementioned impressions of the seal of Sîn-mudammeq (TF 6293). Other impressions include captions in cuneiform letters and depictions of a hero-griffin contest scene or a hero-centaur contest scene.
Parts of this fine-layered accumulation covered a deposit in which a total of 48 clay tablets or fragments of clay tablets and three clay tablet envelopes have been recovered. Since the deposit, which continues into the eastern section, has not yet been completely excavated the total amount of texts might be larger and can only be accessed in the future.
For a more detailed description of the tablets see Highlight: New texts.
Towards the west, where in 2009 an open space alongside the facade of House I and under the western part of the reception room was excavated, the amount of pot-shards, bones and pebbles contained in the fill between the monumental mud-brick building and House I rose significantly during this year’s work. Within these accumulations numerous cylinder-seal impressions on sealings have been found along with three fragments of small clay tablets and a piece of an envelope. One almost complete clay tablet was associated with the skeleton of a beheaded piglet (Fig 13). This might, however, be a coincidence. The objects found so far support the proposed dating of the layers to the 13th century. It has to be considered, however, that the layers also contained intrusive materials such as a seal impression clearly of Mittani origin (Fig 14). Only a detailed iconographic investigation and statistic analysis of the huge amount of seal impressions can give further insight into the consistency of the accumulations and their process of deposition. Yet at the end of this process, the whole area was levelled for the construction of the Middle Assyrian houses on the terrace using low steps to compensate the slope of the mound.
The architectural remains uncovered in the northeastern corner of square 6846 consist of a room, which had several phases of use each associated with a replastered tannur and a stone bowl inserted in the pavement (Fig 15). These last elements point to some production activity connected with fire and liquids. The structure was connected with the younger of two distinctly separate phases of a wall running diagonally southeast–northwest through most of the northern part of the trench. The ensemble was probably part of an economic area that extends further to the northeast and can be dated by means of stratigraphy to the post-House I period, or to the end of the Late Bronze-Age – early Iron-Age.
In 2009 and 2010 similar installations have been excavated further to the south in Trench C-II (squares 6844 and 6845). Directly above Room 9 in House I two tannur-installations and several small walls encompassing them have been excavated. A second tannur is visible in the section directly above the walls of the Middle Assyrian building.
Therefore it seems that, before and after the occupation of House I, this area may have been used for food preparation and other economic purposes, or as burial ground as will be shown later.
West of the main facade of House I, an open space or alley was excavated during the 2009 and 2010 campaigns. It seems that this area was used for the same purpose during the preceding Mittani-period, namely as open space where huge amounts of broken pottery, process waste, and bones were discarded. Mixed with ashy lenses and scattered pebbles this area forms a sequence of occupational levels covering the Mittani and Middle Assyrian periods. The levels contemporaneous to the occupational phases of House I show a decreasing number of shards and process waste but numerous bronze fragments scattered in the consecutive layers (see 2009: The Middle Assyrian House I).
For further reading continue to Sounding C – Middle Assyrian House II & III
> Work in Areas C-I - C-IV
> House I – aims and supervisors
> House I – architecture, building-ph...
> Underneath House I – Production are...
> Late Bronze-Age – Iron-Age levels
> Open space in the west