During June–July 2005, the twenty-third season of excavations was conducted at Tel Dor (the third season of excavations of the renewed expedition to the site; License No. G-60/2005). The excavation, undertaken on behalf of the Institute of Archaeology of the Hebrew University of Jerusalem and the Zinman Institute of Archaeology of Haifa University, in cooperation with the Weizmann Institute of Science, the University of California at Berkeley and the University of Washington at Seattle and under the auspices of the Israel Exploration Society, was underwritten by the Berman Institute of Biblical Archaeology, the Zinman Institute of Archaeology, the Kimmel Center for Archaeological Science, the University of California at Berkeley, the University of Washington at Seattle, the Stella and Charles Guttman Foundation, the Roswell family of Baltimore and an anonymous donor. The excavation was directed by I. Sharon and A. Gilboa, with the assistance of E. Bloch-Smith (supervision Area D2), T. Goldman and A. Haeim (supervision of sub-areas in Area D2), B. Givon (recorder Area D2), Y. Shalev (supervision Area D5; former Lower Area D1), D. DiPietro and B. Haug (recorders Area D5), A. Stewart, S.C. Stroup and A. Estes (supervision Areas D1 and Upper Area D4), S. Rebecca Martin (supervision Upper Area D1), R. Karberg, M. Allen and L. Cannon (square supervision in Upper Area D1), E.J. Haug (recorder Upper Area D1), M. Wells (supervision Area D4), E. Dintino, N. Arrington and J. Pearson (square supervision Area D4), D. DeVore (recorder Area D4), I. Mitler (administration), H. Karesh (photography), R. Gross (artifact registration and pottery restoration), S. Matskevich (architectural drafting), A. Behar (management of a special project of high-resolution excavation of industrial waste-pits in Area D2), S. Weiner, E. Boaretto, R. Shachak-Gross and F. Berna (mineral analysis in the field), S. Silstein (XRF and metallurgy), R. Albert (phytolith analysis), G. Bar-Oz, N. Raban and I. Zohar (archaeozoology), R. Linn (conservation), V. Damov (artifact drafting) and I. Hirshberg (director of the Center for Nautical and Regional Archaeology, Dor). Participating in the excavation were students from the Universities of California at Berkeley, Washington at Seattle, Florence and Haifa, as well as volunteers.
Other activities undertaken during the season, which are connected to the excavation, were funded by the Israel Science Foundation, the Bikura program of the Israel Science Foundation, the White-Levy Program for Archaeological Publication, the Getty Foundation's Collaborative Research Program and Bob Roswell.
The excavation season focused on four sub-areas in Area D, which is in the southwestern part of the tell (Upper Area D1, D5—previously dubbed Lower Area D1–D2 and D4; Fig. 1). The general plan for the excavation at the site included the opening of new areas to connect Areas H and D, forming one large area. As part of a long-term conservation policy, it was decided to leave the deeply excavated parts in Area D2 as they are and to excavate the area around them until the Iron II or the end of Iron I layer is exposed. North of Areas D2 and D5 other units were excavated—Upper Area D1 and Area D4—with the goal of reaching the Hellenistic and Persian strata, while leaving the northern strip in Area D1 as representative of the layer from the Roman period, together with Area H. To evaluate the possibility of exposing remains from the Bronze Age, several squares, which have not been excavated since 1997, were reopened in the southern part of Area D1. During the season, this area was referred to as Lower Area D1; however, by the end of the season it was decided to continue the excavation in this area and designate it as Area D5.
Area D5. This area was excavated in the past as part of Area D1 by a team from Canada, within the framework of excavations directed by E. Stern (ESI 14:61–71). A large courtyard (in excess of 5 × 10 m), flanked by rooms on the north and west (Fig. 2), extended throughout most of the excavation area. The walls of the rooms (width 0.9 m) were built of fieldstones. Three layers were excavated in the middle of the courtyard. Only a very small section of the earliest layer was exposed in the southern part of the area; it included a large patch of bright red fill, probably burnt mud bricks, in whose middle was a dark area, probably ash. Fragments of a large Cypriot-style pithos were discovered in situ, inside the red fill. Overlying the earliest layer was a thick layer of thin gray phytolith surfaces. The potsherds discovered in the upper layer dated to the transition period between Iron I and Iron II (Horizon Ir1/2 according to terminology of Dor excavations). The upper layer included a series of surfaces, usually composed of a kurkar foundation and covered with a thin layer of phytoliths. These surfaces were deposited unevenly so that a gap of up to 0.15 m between two surfaces was created on one side and on the other side they were compacted into a single surface. However, the general topography of all these surfaces was identical and they all sloped from south and west toward the center, where a trough-like depression was created. A sliding board-like installation with a U-shaped cross-section descended from the north toward this depression, bisecting the northern part of the area. The dozens of olive pits that were gathered from one of the surfaces may point to the function of this installation. The use of the inclined surfaces as installations is uncertain and it cannot be completely negated that the steep slopes of the surfaces were caused by some post-depositional processes. The ceramics gathered from these surfaces dated to Iron IIA (Horizon Ir2a), or slightly earlier than that (Horizon IR1/2). Several large pits from the Persian period had cut through the phytolith surfaces and the walls of the building in the northern part of the area.
Area D2. The key to understanding the stratigraphy in the area lay in two massive perpendicular ashlar walls (W10606, W04D2-065; Fig. 3) that were exposed in previous seasons. The two walls were built of large dressed kurkar stones (0.35 × 0.55 × 1.00 m), set in place as headers;
W10606 was slightly lower than W04D2-065. The juncture point where the two walls met had not yet been excavated and therefore, it is still not possible to determine if they were built together as part of a large ashlar building, or if W04D2-065 was later than W10606. When W10606 was excavated, it was noted that it covered the walls of a building whose latest floors dated to Iron IIA (Horizon Ir2a) and therefore, it was clear that W10606 also belonged to this period or postdated it. During this season, a series of thick crushed kurkar floors was exposed on both sides of W04D2-065; remains of a stone pavement were preserved on several of these floors. The connection between the floors and the wall is unclear due to later pits in this area (see below); however, contact between the floors and the wall was noted in several places. No artifacts were discovered on the floors but the ceramic finds from the fill between the floors mostly dated to Iron IIA (Horizon Ir2a), with only a few potsherds possibly dating to a slightly later period, i.e., the beginning of Iron IIB (Horizon Ir2a/b and Ir2b). Among the ceramic finds were several Cypriot potsherds that dated to the Cypro-Geometric III period (CGIII).
The ashlar walls and the adjacent floors were cut by several pits, whose internal stratigraphy and ceramic changes indicated that the pits were used over a long period of time. Three pits (L05D2-517, L05D2-544, L05D2-802) were exposed: Pit 05D2-517 severed Pit 05D2-544 and both were earlier than Pit 05D2-802, which was discovered replete with ceramics that included complete vessels, mostly Phoenician commercial jars from the seventh century BCE. The bowls from Pits 05D2-517 and 05D2-544 were carinated and had a flat rim and a wheel burnish or bowls with a red slip (fine Samaria bowls). The bowls discovered in Pit 05D2-802 were mostly curved bowls with a folded rim; some were in the Assyrian style. The jars from the two early pits were cylindrical and had a short ribbed neck and a rounded rim (Hazor jars), contrary to the jars from the later pit that had a square-rim and no neck. Assyrian influenced pottery types or types that are known from Assyrian destruction assemblages in the country were almost completely absent from the two early pits, indicating that they were filled-in prior to the Assyrian conquest of Dor, whereas Pit 05D2-802 postdated the destruction. This pit, which actually consisted of a series of pits and installations (diam. in excess of 10 m, depth 1.5 m), was excavated intermittently from 1988 onward. It was full with industrial debris and a large quantity of pottery vessels. A methodical and detailed excavation of the pit this season showed that about a third of the excavated area included high concentrations of metal and vitrified sediments, while the rest of the area in the same levels, was mainly composed of layers rich in phytoliths and concentrations of pottery, without any hint of metal. The waste in the pit pointed to high-temperature industries and included materials that underwent vitrification and calcification, as well as many hundreds of metal slag, which mostly had a high percentage of iron and a low percentage of copper. The large quantity of slag and its high iron content distinguished the finds in this pit from those found in the recycled copper region of the Early Iron Age in Area G at Tel Dor. The excavation findings are being studied in an attempt to learn about the kinds of industry represented in the debris of the pit and to gain further knowledge with regard to the formation and composition of the phytolith layers.
The excavation this season also entailed the removal of several walls that were ascribed to later phases. It resolved the chronological issue of the architectural complex that included the buildings designated as the ‘Persian Palace’ in Area D1 and the so-called ‘Big Mama Building’ in Area D2. These structures were first dated to the Persian period, although it was suspected that their dates should be pushed forward to the Hellenistic period. It has been known for a long time that no clear connection was discovered between the walls of the ‘Persian Palace’ and the floors from the Persian period, exposed inside and around it, which gave it its name. The post-Persian period date was evidenced by potsherds of the Hellenistic period, several of which were inside one of the walls of the building and below it. However, since the potsherds were discovered in a wall that stood exposed for many years, the chronological issue remained open until this season, when the removal of the wall revealed a pit that was sealed beneath it. The pit contained finds from the Hellenistic period, including three intact vessels: a lamp and two potholders. These finds have conclusively settled the date of the wall that sealed them and provided a terminus post quem for the construction of the ‘Persian Palace’. It was definitely ascertained that this assemblage is dated to the Hellenistic period and probably to its later part.
Upper Area D1(West; Fig. 4). The aim of the excavation was to complete the exposure of the building from the Hellenistic period, dubbed the ‘Monument’, which was excavated intermittently from 1995 onward. This structure included two massive ashlar walls (W16020 and W16850) whose bases were molded and stuccoed. For this purpose, the excavation squares were opened north of the area that had been exposed in the previous seasons. Due to its close proximity to surface and later construction, only the bottom course of the building’s foundation was preserved. The northern continuation of W16850 was uncovered and the width (12 m) of the structure was ascertained. The building’s foundations were built of large ashlars, arranged widthwise. This method of construction and the use of architectural elements fashioned in the Greek tradition are unique to the construction of the Hellenistic period at Tel Dor, which is mostly Phoenician rather than Greek in character. The building’s foundations had cut through layers of fill from the Persian period. The bases of the walls were shaped and molded, indicating that the building was planned as a single free-standing structure, as was the custom in Hellenistic architecture. However, it was quickly absorbed into the agglomerative local tradition, when walls in the Phoenician ashlar-pier style had abutted it with complete disregard to the molded decoration and the open space between it and the complex to its west was converted into a series of rooms.
A small section of a paved Roman street and the beginning of a sewage channel below it were discovered north of the northern end of the ‘Monument’. This may be the continuation of a north–south street that bisected Areas F and H or a section of an east–west street that intersected it. The end of the street corresponded to the northern wall of the ‘Monument’ and therefore, it seems that this structure also continued to be used in the Roman period or another building was built on its foundations.
The large Roman-period building that severed the ‘Monument’ was also examined this year. The two long walls of the building, aligned north–south, were constructed from flat stone slabs that were set on a foundation of fieldstones and probably served as a stylobate for a row of columns. A course of stone slabs in the eastern of the two walls protruded eastward from the foundations that were built of roughly hewn stones; it is possible that the floor east of this wall was one step lower than the floor to its west. If this was indeed the case, it may be possible to reconstruct the building as a stoa that faced east. This stoa probably replaced an earlier stoa that was erected in the Monument Building of the Hellenistic period. A wide concrete floor that was built on a foundation of pebbles was exposed west of the building’s western wall. Two levels of similar superposed floors were exposed to the east of the western wall and it seems that the earlier floor of the building was superseded with a floor at an identical elevation as the one to the west of the wall. The artifacts discovered below these floors dated to the Roman period.
A courtyard floor was exposed east of the building and a large well-preserved and stone-built kiln was discovered below it (Fig. 5). The kiln’s opening faced north and its firebox was divided into two clay-lined basins, where the clay was baked by the heat of the fire. Another installation, also lined with stones, was exposed below the floor to the west of the building’s western wall, yet it displayed no signs of the use of fire.
Upper Area D1 (East). This area was entirely plundered; numerous robber trenches crisscrossed it and not a single wall was preserved in place. Nevertheless, two construction phases from the Roman period were discerned in the area this season. Each phase may possibly comprise several secondary phases. The later phase included several sections of concrete floors and sections of wall foundations that consisted of a conglomerate of concrete and fieldstones. The early phase was characterized by horizontal, un-tamped levels and dry-built stone walls that incorporated column drums and capitals in secondary use. It seems that these architectural elements originated in the colonnaded building from the Hellenistic period (the ‘Monument’ or another building that has not yet been uncovered). At this point, it is unclear which of the many levels of the early phase are floors, floor beddings, or constructive fill.
Area D4. The excavation in this area, as part of Area D1, began in 2004 and a large building from the Roman period was exposed (Fig. 6). The walls of the building extended in the direction of Area D2; however, it constituted an independent unit both from an architectural and stratigraphic standpoint. Hence, it was decided to accord this area a separate number. The large size of the building (width in excess of 15 m) and the depth of its foundations seem to indicate that this was an important public building. Thick floors of hydraulic concrete, coarse white mosaic floors and several drainage channels and basins in the building indicate that liquids were extensively used in it. Two very different proposals were offered for the use of the building in 2004; one said it was a public building, probably a bathhouse, and the other suggested it was a large industrial building.
To evaluate the plan and use of the building, it was decided to excavate additional squares north and east of the dug area. Four other rooms of the building were uncovered. A terra-cotta water pipe (diam. 0.15 m; Fig. 7) was exposed in the northeastern room. It was lying in the robber trench of the room’s western wall and was set beneath the room’s concrete floor. A spout that faced upward was discerned in the middle of a section of the pipe. Close to it were the remains of an installation, probably a vat of some sort in the concrete floor of the room. The southern end of the pipe was robbed. At a later phase of the building or after it was no longer in use, two installations that had cut the concrete floor were exposed. A circular hollow (diam. 2.5 m) was preserved in the southern installation and had cut the floor down to its pebble foundation. A hole in the center of the depression was surrounded by a raised stone pavement. During a visit to the excavation, Y. Dray suggested that this installation was the base of a Pompeian mill (‘donkey mill’). The northern installation was a large tabun (diam. 0.9 m) dug into the ground and preserved to almost its entire height (c. 1.2 m). The tabun was ventilated by means of a terra-cotta pipe of the same type and diameter as the destroyed pipe from the first phase in the room. The exposure of a water pipe below the floor of the building reinforced the supposition that large quantities of water were used in it. The pipe was identical to the main water supply pipe of Tel Dor, discovered in Areas A and C, which led from the terminus of the aqueduct, located further south in Area B. The presumed water level at the end of the aqueduct was 12 m above sea level. This elevation did not facilitate the direct transfer of water to the top of the tell, whose elevation today is 14.5 m above sea level. It is believed that the pipe in Area E passed along the foot of the northwestern corner of the tell, where a bathhouse was discovered, and continued to Area F, where a section of the pipe was discovered opposite the main entrance to the temple complex. It is possible that a terminus of the circuitous route of the pipe was in the southern Area D. It is therefore feasible that the building was used as a public bathhouse or as a spring house from which the water was distributed to different pools and installations. Part of the building may have been adapted for use as a commercial bakery in the later phase.
Several pyrotechnical installations were discovered in the adjacent room to the west. One was a domed installation, lined with clay, whose shape resembled a tabun. This installation was smaller than the tabun in the eastern room and it seems to have been built on the floor rather than embedded into it. The second installation, built of a heap of small stones and hypocaust tiles, did not seem to have had a lining. The northwestern room was different than the two others. Two floor levels were exposed in it; the upper one consisted of shells (glycimeris) and the lower was a plaster and crushed kurkar floor. Overlying the floors and between them was a rich assemblage of pottery vessels and a few coins that dated the use of the building to the end of the first and second centuries CE.
Only a small part of the southeastern room, which had a thick concrete floor, was exposed.