Tel Dor Excavation Project

Google Custom Search
Staff login:

2004 - End of Season Report - Area D1

In 2004, the UC Berkeley team excavated nine squares in Area D1 (AO/15-16; AP/15-16; AP/13 (W half); AQ/13 (E half); AQ/14-15; and AR/16). Most of these squares were located in the unexcavated area that separated Area D1 and Area D2. One of the goals of the 2004 season was to link the two areas so that stratigraphic correlations could be made between them, especially with regard to the Roman and Hellenistic levels. In particular, we want eventually to clarify the relations between the major Hellenistic buildings in the two areas - the big building in Area D1 and the "Big Mother Wall" in Area D2 - and provide a more complete plan of the Hellenistic levels on this side of the site.

The nine squares were organized into three large blocks, which we designated units; since this designation was found to be confusing, for the purposes of this report they will be termed sectors. Sector 1 encompassed squares AO-AP/15-16; sector 2 was AP-AQ/13-14; and sector 3 included AQ/15 and AR/16. Opening nine squares from the surface of the tell required an enormous physical effort. Owing to heavy disturbances in these squares, it was necessary to remove more than a meter of overburden in several of the squares before intact architecture was reached.

The major result of the excavation in 2004 was the discovery of a large Phase 1 complex that for the moment is interpreted as a bath house. This interpretation is based on the discovery of at least four thick plaster floors, roughly one meter square, that appear to be the remnants of basins and the discovery of an intact plaster basin. Unfortunately, most of the walls relating to these floors had been robbed, but several clearly delineated robber trenches established wall alignments. In Sector 1 a clear earlier subphase of Phase 1 was identified. This consisted of three large plaster floors laid parallel to each other and sloping down to the north, with the southern floor (F19422) highest and the northern floor (F04D1-085) lowest. The upper floor (F19422) was tiled with large tesserae and a portion of the floor makeup consisted of the top of a very large column drum in reuse or a column-like stone cut specifically to serve the function of the room. The middle floor (F04D1-131) was connected to the lowest floor (F04D1-085) by a small plaster channel that had been constructed through an earlier phase 2 wall (W04D1-052). Fresco remains were found on the Phase 2 walls in sector 1 (W04D1-050 and W04D1-052) that had been reused in the Phase 1 complex.

Sector 1:AO-AP/ 15-16

Excavators: Yiftah Shalev (IFY); Erin Dintino (EKD)

This sector contains three phases: the first of them contains more than one stage. The excavations reveal a big building that was built in phase II and was reused in phase I. The function of this building is unclear but it seems that at least in phase Ia it was bathhouse and industrial complex of some kind.

Phase I

W04D1-067, W04D1-068, W04D1-169, W04D1-070, F04D1-001, F04D1-071, F04D1-103, F04D1-130, L04D1-069, F04D1-085, L04D1-166, F04D1-021, F04D1-113, F04D1-131, F04D1-104, F04D1-106, W04D1-105, W04D1-050, W04D1-052, W04D1-143, W04D1-144

This phase is represented by two stages.

Five ashlar walls remain in this phase in reuse from phase II. W04D1-068 and W04D1-169 are two walls running east-west parallel to each other. W04D1-067 is a south-north wall abutting W04D1-068 from the north. W04D1-052 abuts W04D1-050.

Phase Ia: A drain, L04D1-069, was built between walls 04D1-068 and 04D1-169, using F04D1-085 (phase Ib floor) as its floor. A shell floor, one of a set of three floors, was found north of W04D1-068; the three floors (F04D1-071 #15.80/15.75, F04D1-103 #15.78/15.68 and F04D1-130 #15.74/16.64) represent the floor level of phase Ia. These floors were cut by the R.T. of W04D1-068 (L04D1-095) but probably reached this wall from the north. (The floors were sloping up toward the edge of the R.T). None of the floors presently reaches W04D1-067, but probably once did so. These floors might represent three stages in this phase but more likely are a single floor with two sub-floors/foundation. A small basin, L04D1-166, is located east of these floors; the relations between the basin and these floors is unknown, because a R.T. (L04D1-101) breaks the connection between them.

The last floor in this sector from stage 1a is an oval cement floor/installation, F04D1-001. This floor contains several layers of cement and is clearly intended to hold liquids. A channel is built into its northern part. The floor reached W04D1-070, a poorly built rubble wall, from the west; another floor of the same type (F04D1-021, AO/16) reached it from the east. F04D1-021 appears to have been either a rebuilding of an earlier phase I floor, F04D1-113, or F04D1-113 was subflooring for F04D1-021. This floor is related to the basin, the drain, the shell floor, and some hypocaust tiles that were found in fills in this sector and in situ in AQ/12. All of these elements hint that during this stage this part of the building functioned as bathhouse.

Phase Ib: The main features in this phase are F19422, F04D1-131, and F04D1-085. F19422 is a big, sloping mosaic floor at the southern part of the sector. This floor, comprising 2x2 cm mosaic cubes embedded in thick cement, probably served the same purpose as F04D1-131, and F04D1-085, to channel the flow of liquids. A column drum was found sunk into its eastern edge flush with the surface of the floor and was probably a part of the installation. Floors F04D1-131 and F04D1-085 were constructed in the same fashion as F19422, but without any evidence of mosaic. The surface of F04D1-131 appears eroded at its lowest and easternmost end. There is an abraded L-shaped depression leading from the west down to the northeast corner of the floor, which is possibly the footprint of a missing drainage pipe or other feature of the installation. The depression leads to a portion of the floor that seems to be a channel for draining liquids onto F04D1-085. The three of them were connected with a channel system. It is unclear if F19422 was still in use in phase Ia or if there was another floor above it, now eroded away. F04D1-085 reached W04D1-068 and probably also W04D1-169, forming a long narrow basin. F04D1-131 is connected to F04D1-85 by means of a small channel that was built through W04D1-052 during this phase of reuse. Also, F04D1-131 reached W04D1-144 and W04D1-143, which also appear to have been constructed during this phase of use. W04D1-143 was built up against W04D1-050.

East of W04D1-050, F04D1-104 and F04D1-106 reached W04D1-050, and W04D1-105 appears to have been built on top of the floors. W04D1-105 abutted W04D1-050. These features are provisionally assigned to Phase I, but could be Phase II. They lie outside the Phase I/Phase II building and correlations to features inside the building do not exist.

The main loci in this stage are the sealed loci below the floors and in the drain. These loci are L04D1-103 (floor make up of F04D1-071) , L04D1-130 (floor makeup of F04D1-103), L04D1-137 (floor makeup of F04D1-130) , L04D1-140 (fill sealed below L04D1-137) and L04D1-154 (fill in the drain). A few coins were found sealed under the floors and a large number of lamps of Dor types 22, 26 and 28. These finds date phase I to the 2nd century C.E. The floor makeup and fill below F04D1-021 produced one small, sealed locus, L04D1-112.

Phase II

W04D1-067 ,W04D1-068 ,W04D1-169, F04D1-178, F04D1-184. W04D1-050, W04D1-052.

This phase is represented by five ashlar walls and a floor. W04D1-067 and W04D1-169 run east-west parallel to each other, forming a corridor. W04D1-067 abuts W04D1-068 from the north. W04D1-068 probably related in some way to W04D1-050, although the connection between the two walls was destroyed by robber trench L04D1-49. W04D1-052 abutted W04D1-050. Both of these walls appear to have been covered with fresco during the earliest phase of use, i.e. prior to the building being converted into some kind of industrial/bathing installation. The only floor in this sector from this phase is floor F04D1-178. This floor reached both W04D1-067 and W04D1-068. In the northern part of this floor, a small patch of cement floor was found. This floor, F04D1-184, is probably a resurfacing of F04D1-178.

The main locus that represents this phase is L04D1-178. This locus, a fill above floor, contains a huge amount of pottery sitting above the floor. The lamps from this locus date the end of this stage to the 1st century C.E.

Phase 3?

The only feature assignable to this phase is a small wall, 1.5 m long. This wall, W04D1-180, is a trapezoid ashlar stone wall cut by W04D1-068 from the north and by W04D1-169 from the south. This wall was probably running north-south.

It is unknown yet if this is really a phase III wall; it could be a phase IIb wall. The phasing of this wall and its relation still need to be clarified by further excavation.

Sector 2: AP/AQ-13/14

Excavators: S. Rebecca Martin (SRM); Rebecca Karberg (RAK)

This sector, comprising squares AP/AQ-13/14, was opened in order to correlate the stratigraphy between areas D1 and D2 and to expose more Hellenistic strata in area D1. AP/AQ-13 was also opened to define the limits of a large pit excavated in 2000 immediately to the west and to recover more of the Hellenistic mosaic found there (see Stewart and Martin 2003).

In the 2004 season, mostly Phase 1 features were uncovered. The main structure excavated was a Roman bathhouse. The corner of its western and southern walls was found in AQ-13, although these exterior walls are robbed out elsewhere in D1. The southern wall, W17591, was previously uncovered in 1996? in AO-13, but its function was not understood until the 2004 season. Several partial floors were revealed, although only two reached Phase 1 walls (F04D1-029, F04D1-088). One was preserved only to subfloor level (F04D1-029) but did produce a sealed context from which several pieces of fresco, molded plaster, and Hellenistic mosaic were found. A thick plaster floor (F04D1-088) reached one interior bathhouse wall (W04D1-125; Phase 2 and reused in Phase 1). On the floor was piled rubble of water pipes and hypocaust tiles, at least one of which was in situ; impressions of the hypocaust tiles could still be seen on the surface of the floor.

Outside the Roman bath house was another series of Phase 1-2 floors and walls. Some of the Phase 1 features were clearly associated with the bathhouse complex, such as a small room built off the southwestern corner (W26245, W04D1-110, W04D1-150). Another small, Phase 2 room (W04D1-127, W04D1-165, F04D1-129) was uncovered at the end of the season; it may have been cut either during the construction of the bathhouse or by the robbing out of the western wall (L04D1-042). The majority of the area was highly disturbed by later robbing activity that cut through both Phase 1 and Phase 2 features (RT L04D1-128, L04D1-042, L04D1-038).

The architectural dump in which the Hellenistic mosaic was found was also cut by a later robbing trench for the southern wall of the bathhouse complex. The dump continued to produce many architectural fragments (ashlars of various sizes, column drums, thresholds, voussoirs, and rafter-supports); only the larger ashlars can be attributed confidently to the bath house complex, whereas the other fragments could have come from any number of other structures on the tell. The dump, which also included numerous large fragments of the Hellenistic mosaic partially excavated in 2000, was used as constructional fill for both the bathhouse and the exterior rooms south and west of the complex. Most of the recovered mosaic was white limestone with haphazard tessellation; presumably these areas of the mosaic came from the areas of the floor that would have been obscured by the klinai, or dining couches. These fragments of haphazard tessellation ranged in size from a few tesserae to large fragments measuring over 0.15 m in length. Fewer than one dozen fragments of colored mosaic were recovered from the dump. One of these was found sealed by the aforementioned Phase 1 subfloor. Other mosaic fragments were also clearly part of the constructional fill of the southwestern room; they were found above and below the three walls that made this small room and within a hard-packed shell and kurkar construction surface for the easternmost wall of the room (W04D1-110). It seems likely that more mosaic fragments will be found within the walls of this room, should these be removed.

The combination of four years of erosion and cleaning in the western half of AQ-13 (excavated in 2000) revealed some key relations between areas D1 and D2. The southwestern corner of the "Big Mother Wall" (initially excavated in D2) was recognized in 2000; in 2004, it became clear that the easternmost wall of the "Persian Palace" (W26230) covered the foundation of the westernmost "Big Mother Wall" (W26000); one ashlar of W26230 was beveled in order to overlap the foundation course of "Big Mother." Several field stones from the Persian Palace's pseudo-atelaio construction also overlapped "Big Mother's" foundation, which indicates that the "Persian Palace" is a (slightly) later construction (Phase 4a to "Big Mother's" Phase 4b); but it is likely that these two monumental structures were essentially contemporary and together dominated the southern part of the tell. Both buildings are cut by the corner of the Roman bath complex (W17591), as are the Phase 2 walls (W16550a-b) built on top of them.

Two phases were excavated in AP/AQ-13/14. The phases fall into two categories: the Phase 1 bath house and features within the complex, and the Phase 1-2 structures outside the bath house complex.

Phases 1-2: Within the bath house complex

W26073 (N-S) and W17591 (E-W) form the southwestern corner of the structure. W26073 is completely robbed out by L04D1-042 in AQ-14/15. At least five meters of W17591 are also robbed out by L04D1-089; the other corner of the building lies in the western baulk of Area D2's AO-13 (it will be necessary to excavate AP-13 east in order to determine the extent of the robbing activity). At the end of the season, robber trench L04D1-089 came down on a small area of wall that is, for now, considered part of W17591. F04D1-029 was only preserved to its subfloor of small fieldstones; patches of cement on the floor and its construction against W26073 suggest that its cement surface once reached the cement-and-ashlar wall. F04D1-029 seals L04D1-126, L04D1-158, and L04D1-174. Pottery recovered from L04D1-126 suggests that the floor was constructed in the 2nd century C.E. A thick plaster hypocaust floor (F04D1-088) reached two walls within the bathhouse complex: the possible continuation of W17591 and a N-S ashlar wall, W04D1-125. W04D1-125 is completely robbed out by L04D1-038 in AQ-14 and is cut on the south by robber trench L04D1-089, but two courses of the wall reemerge in AQ-15. Only in AQ-15 is it clear that W04D1-125 was a Phase 2 wall (because of its association with a Phase 2 pavement (FO4D1-149) reused in Phase 1.

Also within the bath house complex was a series of walls and floors that could not be directly associated with the bath itself because of the cutting action of robber trenches. The main numbers associated with the N-S robber trenches are L04D1-042 and L04D1-038. W04D1-015 is considered Phase 0 because it partially covered a concrete Phase 1 floor, F04D1-014. F04D1-014 seals L04D1-086; L04D1-086 came down on two Phase 2 features, W04D1-096 and F04D1-086. F04D1-086 reaches W04D1-096 and covers and seals L04D1-098. It is cut by two robber trenches: L04D1-060 to the north and L04D1-042 to the west.

Phase 1-2: Outside the bath house complex

One Phase 1 room was built off the southwestern corner of the bathhouse. Only three walls are preserved: W04D1-110, W26245, W04D1-150. A fieldstone wall (foundation?), W26245, abuts fieldstone and ashlar wall, W04D1-150; W04D1-110 abuts both W26245 and the bath house corner wall, W17591. W04D1-150 is cut by the large E-W robber trench, L04D1-089; it is possible that this wall also originally abutted W17591. In AQ-12 a field stone wall/foundation was excavated in 2000. It was removed during that season, but it is likely that it was part of the same construction as W26245 (although exact relations will be impossible to determine post mortem). In AP-12 (also excavated in 2000), W26245 covered a Phase 2 ashlar wall with a threshold, W26246.

A small patch of Phase 1 pavement was recovered to the west of the bathhouse: L04D1-019. The fill below the level of the pavement (L04D1-164) covered a Phase 2 floor, F04D1-164. This floor reaches a Phase 2 field stone wall/feature W04D1-185. These Phase 2 features are cut on the east by robber trench L04D1-042 and on the north by robber trench L04D1-128.

Also west of the bathhouse wall W26073 was a Phase 1 floor, F04D1-044. This floor covered Phase 2 wall, W04D1-127. F04D1-044 sealed L04D1-129, which came down on the remainder of the small Phase 2 room: W04D1-165 and F04D1-129. F04D1-129 probably reached both W04D1-127 and W04D1-165, perhaps only at the second course (as yet unexcavated). These Phase 2 walls are cut by the same robber trenches, L04D1-042 and L04D1-128.

Key loci:

Within bath house Outside bath house

Sector 3: AR/16 and AQ/15

Excavators: Jessica Nager (JNN); Melissa Thornton (MAT)

Sector 3 was highly disturbed, and it was only after several weeks of removing the disturbed overburden that intact architectural features were reached. Because these features are difficult to correlate with the rest of the Phase 1 complex, they are presented in their stages of construction.

AQ/15 Inside Complex

W04D1-125 abuts W04D1-175. The relation of F093 to W125 and W175 is not clear, thanks to the massive robbing action and pit digging that took place in this unit. However, it is likely that F04D1-093 reached these walls. F04D1-093 is a resurfacing of the pavement F04D1-149. However, F04D1-149 was not entirely covered by F04D1-093 - it seems part of F04D1-093 was destroyed in the North. Pavement F04D1-149 reaches W04D1-125 and W04D1-175, but is cut on the East and West, and partially in the North also. F04D1-160 is a fragmentary kurkar floor that reaches F04D1-149 and has a circle cut in the middle of it. The purpose of this floor is unclear. F04D1-162 is a crushed kurkar floor that seems to continue under the pavement F04D1-149. F04D1-122 is most likely a continuation of F04D1-162, but a large piece was cut out of its upper part by the extensive N-S robbing action. Thanks to all the disturbances in this unit, the relations are not clear. W04D1-175 and W04D1-125 were assigned to Phase 2, which means they were reused in Phase 1; three floors (F04D1-093 [stage i]; F04D1-149 [stage ii]; and F04D1-162 [stage iii] go with these walls. However, the walls were both extensively robbed and the floor relations are not certain.

The floors encountered in this unit have been provisionally assigned to the following stages:

AQ/15 Outside Complex

The relations of W04D1-094 and W04D1-136 are not clear. Only the following of the above features have been removed: F04D1-036, F04D1-074, F04D1-093.

As for loci with good contexts, the sealed loci are:

All the other fills are disturbed or possibly disturbed, including one that produced a modern floor tile.


AR/16 appears to lie completely outside the Phase 1 complex, so it cannot be correlated to constructional stages there. Also, the unit was highly disturbed.

The latest feature in the unit is the wall fragment W04D1-076, a cement and fieldstone wall similar to other late walls on the tell. Its only relation to any other feature in the unit was that it appeared to cut F04D1-170.

The floor sequence in AR/16 appears to be:

F04D1-056 floats over part of F04D1-079 and part of F04D1-079 covers F04D1-139. F04D1-079 is probably a continuation of F26049 in AR/15. L04D1-135 appears to be a late trench that cuts through this floor. However, it might be a robber trench for a wall that went with both F04D1-079 and F26049 (F04D1-079 is probably but not necessarily contemporary with F26049). F26049 covers F04D1-159. The fact that we didn't find an extension of F04D1-159 on the other side of L04D1-135 keeps this question open.

F04D1-177 is probably a resurfacing of F04D1-170. The relation between F04D1-177 and F04D1-139 is unclear. The elevation for F04D1-139 is about 15.30m, and F04D1-177 is at 15.22m. Also, F04D1-139 cuts F04D1-170. So for now, the most that can be said about F04D1-177 is that it appears to be contemporary with F04D1-139. Both F04D1-177 and F04D1-170 may be earlier than F04D1-139. Moreover, F04D1-139, F04D1-177 and W04D1-076 are all cut by the N-S robber trench L04D1-133, the robber trench for the N-S wall found in AR/15 (W26127). So the relations are unclear.

Only W04D1-076 is safely assigned to Phase 1. The floors are Phase ii or earlier. The tentative constructional stages in AR/16 are as follows:

Important loci:

Written by Allen Estes, with contributions from IFY, EKD, SRM, RAK, JNN, MAT, and Andrew Stewart
Edited by Andrew Stewart