• 1 Besides the Tell al-Sour excavations, the author also conducted archaeological surveys between 197 (...)
• 2 The first season lasted six weeks (from 16.06.1984 to 31.07.1984), and initially the work was conc (...)

1The author conducted three archaeological excavation seasons at Tell al-Sour (fig. 1) between 1984 and 1986. The site (hereafter TS, coded 62) is located about 45 km east of Homs city in precipitation Zone III (fig. 2), which receives an average of 200 to 300 mm rainfall per annum. The tell is one of eighty tells and sites that have been documented by the author’s archaeological surveys in the Homs region.1 This paper contains new and unpublished data from the excavations at Tell al-Sour, in particular from Sounding 1 (Trench I), opened in 1984 in Area P.2

• 3 Two of the TS city gates, Gate 1 (Area P: fig. 1 and 3) and Gate 3 (Area I), have been excavated, (...)

5Table 1 shows TS phases I–VIII. These represent the stratigraphic-chronological sequences of the settlement determined by the author’s excavations between 1984 and 1986. The earliest phase is TS phase VIII (Late Neolithic/B2: approximately 6000–5500 bc), so far found only in Area P (fig. 3 and 4). This was succeeded in the same area (at least in Trench I) by TS phase V (EBA I and II: approximately 3200/3150–2950 bc). This paper will focus on Area P Stratum B2 only (fig. 4). Stratum B2 (TS Phase VIII) was succeeded chronologically by Strata A (TS Phase V). Area P/Strata A in Trench I (field coded A1–A3 and A4, including the mixed layer A4/B1: the city West Gate 1 floor ; study coded Strata C) will be the subject of a future publication.3

• 4 This estimated date was proposed based on the dating of the pottery from Strata B1 and B2.

6Sounding 1 (covering an area of 3 m²) was opened at the southern end of Trench I, where the West City Gate 1 floor was uncovered (fig. 3 and 4). The aim of opening this exploratory trench was to reach virgin soil, and bedrock was reached at an average depth of 35 cm. The gate floor, field coded A4 (also marked ‘Stratum B1’), comprised mixed deposits (see below and fig. 4, 6 and 7) resulting from the preparation for the laying of the floor, while Stratum B1 was the result of the partial destruction of the underlying Stratum B2 (at least of its upper part, fig. 4). A study of the pottery from both Strata B1 and B2 indicates that they belong to two different phases of settlement, separated by a hiatus of approximately 2 500 years.4 Floor A4/B1 is 10–20 cm thick, and is a hard layer consisting of small limestone stones mixed with the laid-down rubble and with a fine-to-coarse white-washed cemented surface (see below and fig. 6 and 7). The construction of the gate required the preparation and levelling of the topsoil. Underneath the floor, Stratum B1 (fig. 4) was revealed, comprising a mixture of dark brown to black soil, pebbles, assorted crude limestone stones of various sizes and pottery sherds. From this we concluded that we were dealing with more than one phase of deposit, and that, when the floor was made, it was partly mixed with the earlier, pre-existing deposits. This floor was overlaying Stratum B2 (fig. 4).

• 5 Study of the flourishing TS phase V (Areas P, U and I/Strata A) will be published under ‘Strata C: (...)
• 6 Van Loon, Meijer and Akkermans 1986/1987, p. 310.
• 7 Fig. 3, also cf. Van Loon, Meijer and Akkermans 1986/1987, p. 309.

7The excavation of Stratum B2 (Layer 2 in fig. 4) revealed rubble and deposits of debris from the collapsed super-structure of a wall as well as its intact foundation, field coded Foundation Wall 1 (fig. 3, 4, 7 and 8). A study of the pottery assemblages from both Strata A5 and Stratum B2 (fig. 5), indicated that they were of different cultures and time periods. Stratum B2 (fig. 4) is 20 cm deep on average, and sits directly over the bedrock (fig. 3, 7 and 8). The northern part of the wall runs in a straight line north-south, but the south-western part bends a little to the east. The plan of the wall resembles the early Halafian architectural plan found at Tell Sabi Abyad,6 where a rectangular structure was built in combination with a round structure.7 These structures (also called ‘tholos’) are related to characteristic Halafian architecture. This finding suggests that further stretches of wall may be uncovered in future excavations to the north, south and west.

• 8 Braidwood and Braidwood 1960.
• 9 Van Loon, Meijer and Akkermans 1986/1987, p. 310.
• 10 Nieuwenhuyse et al. 2018, fig. 12: 13.

12The B2 pottery vessel shapes are geographically widespread types within the same and adjacent regions, as well as across the wide spectrum of the northern Levant and in the Mesopotamian regions. Comparable forms include the Syrian-Cilician Amuq A assemblages;8 the Balikh region assemblages from Tell Sabi Abyad (no pottery figures have been published) and Tell Damishliyya (the coarsely made plant-tempered ceramics and the ‘husking tray’ sherds;9 bowls with ledge-handle from Shir10 (Hama region); and the assemblages from sites in the upper Zagros mountainous region, such as that from Gird Banahilk. The 1984 excavations at Tell as-Sur add one more site to the index of sites with Neolithic and Late Neolithic pottery, in the Homs region where there had previously been no known sites.

• 11 At the time of conducting the TS 1984 excavation season, there were no lab-means available to the (...)

15In order to propose a general chronology and relative chronology for TS stratum B2, the two study components (Foundation Wall 1 and the pottery analysis) will be compared with likely counterparts to be found in relevant archaeology sources. These can be used in the process of proposing the intended relative dating supported by discussion and strong arguments. Such dating will result from consideration of the above-mentioned two study components along with the negative approach (the exclusion of the wares that have not been found amongst the B2 assemblage). In any case, the relative dating proposed will, unfortunately, lack the support of lab-means data.11

• 12 Al-Nahar and Kafafi 2015, p. 58, and table 2 below.
• 13 Al-Nahar and Kafafi 2015, p. 61 and 68. 

18The excavations at Tell Abu Suwwan yielded pottery known in the relevant sources as ‘Yarmoukian pottery’ and the assemblages belong to the ‘Pre-Pottery Neolithic B’ phase, dated to between 8250 and 7500 bc.12 Tell Abu Suwwan ware is described as coarse, for example item 20 is made of a clay paste tempered with straw and basalt grit.13 Some vessels are large, and the vessel type is the bowl, their lip and body profile generally display the splaying wall section type. Some have red paint. This description seems almost identical in the ware attributes to TS B2 assemblage item 3 (fig. 5). Another similarity between the compared items is the large size of the vessels, but there is a lesser degree of similarity in the shape of the vessel lip. Yet, in spite of the similarities, the Yarmoukian ware differs from its TS counterpart (item 3) in being decorated with impressed of patterns (for example the herring-bone motif).

• 14 This ware is also called the ‘Syrian-Cilician DFBW’; its first appearance is in the Amuq A, which (...)
• 15 Braidwood and Braidwood 1960, p. 49.
• 16 Braidwood and Braidwood 1960.
• 17 ; Braidwood and Braidwood 1960.
• 18 Braidwood and Braidwood 1960, p. 51.
• 19 Braidwood and Braidwood 1960, p. 53.

19The Amuq Phase A pottery assemblages are mainly of Coarse Simple Ware and the Dark-faced Burnished Ware (hereafter DFBW).14 The latter ware was not found in the TS B2 assemblage, while the Coarse Simple Ware was found in the assemblage. TS item 1 (jar rim sherd with a diameter of 280 mm) can be compared to the rim shape of Amuq A DFBW fig. 22 : 19.15 TS item 3 (bowl with rim diameter of 240 mm) can also be compared with the Amuq A jar fig. 24 : 5, with a rim diameter of 260 mm, of the same Coarse Simple Ware.16 The ledge-handle shape of TS item 5 (bowl sherd) can be compared with that of the Amuq A ware made from red clay, fig. 22: 19 and 20 of the DFBW17 however, it displays a smaller ledge-handle. The TS bowl rim sherd no. 3 (with a rim diameter of 240 mm), which has light red paint-slip applied to its exterior surface, can be compared with the Amuq A DFBW jar fig. 24 : 5, with a the rim diameter of 260 mm.18 The Amuq A pottery (some shapes and tempering grits of the Coarse Simple Ware) continued during the Amuq B as well, which allows the comparison of the TS items with counterparts found in the Amuq B (the Coarse Simple Ware). Generally, the Amuq vessels differ from their TS counterparts in that they are larger in size.19

• 20 Akkermans 1988, p. 43.

20Tell Damishliyya is situated two kilometres from Tell Hammam et Turkman (the Balikh region). The excavation in this site delivered a Neolithic pottery assemblage that, on the grounds of its similarity to the TS B2 pottery (mainly regarding the ware and the rim section shape), can be compared with the TS item 1. Another possible comparison is between TS item 3 (fig. 5; bowl of the coarse ware open type) and the vessel fig. 8–9: plate 6 in relation to the rim shape and the exterior surface finishing treatment (Tell Damishliyya: Period II: the Neolithic Pottery).20

• 21 This pottery has been identified with ‘pottery from the pre-Halaf stages’, which comes from Area C (...)
• 22 Gómez-Bach et al. 2019.
• 23 Gómez-Bach et al. 2019.
• 24 Gómez-Bach et al. 2019.
• 25 Gómez-Bach et al. 2019, p. 14.

21Gird Banahilk pottery comes from a Halafian site that was excavated a second time. The new excavation yielded coarse pottery21 displaying strong similarities to TS sherds, particularly regarding the large size of the jar (no. 1), the bowl with splaying wall (no. 3) and the bowl with ledge-handle (no. 5). Besides the crude materials used in the production of jars (in particular), strong similarity to the Banahilk items was found in the vessel rim profile. Other examples: TS sherd no. 1 (jar rim sherd of the open type) displays a strong similarity to the Banahilk item: fig. 7: 3,22 and TS item 4 (jar rim sherd) also displays strong similarity to the Banahilk item: fig. 7: 13.23 TS bowl rim sherd item 3 also looks similar to the Banahilk items: fig. 7: 7 and 17.24 The pottery ware from Banahilk is plain coarse ware, which has been related to the Late Neolithic context. The radiometric data proposes dating this Banahilk ware to between 5559 and 5367 bc.25

• 26 Van Loon, Meijer and Akkermans 1986/1987, p. 310.
• 27 Van Loon, Meijer and Akkermans 1986/1987, p. 311.
• 28 Van Loon, Meijer and Akkermans 1986/1987, p. 310.
• 29 Van Loon, Meijer and Akkermans1986/1987, p. 310.

22The virtual absence of finds in the Tell Sabi Abyad tholoi and the presence of cereals around them26 suggests that could have been used as storage units,27 while the TS tholos of Area P/Sounding 1 (fig. 3, 7 and 8), in the 1984 season, yielded a few sherds of Coarse Simple Ware. The early Halaf remains from Tell Sabi Abyad were discovered sitting immediately above the Late Neolithic levels, and the interesting point here relates to the similarity between the architectural plan of TS Foundation wall 1 and the Tell Sabi Abyad tholos; both have plans that display duality composed of rectangular and round buildings. The Tell Sabi Abyad tholoi have been described as ‘showing the characteristic trait of the Halafian architectural tradition’,28 and these were used as kitchen units or for storage purposes; perhaps, the same use could also have applied to the TS structure. The study of the TS pottery assemblage (in particular sherds nos 1–4) showed that these were used as storage and kitchen vessels. Equally, the study and comparison of Foundation wall 1 (with its incurving part, fig. 3, 7 and 8) lead us to infer a dual structure (combined units) comprising one domestic unit. This interpretation needs confirmation through future excavations. The Tell Sabi Abyad pottery assemblage fits into early Halaf settlement material culture, which has been dated to the second half of the 6^th millennium bc.29 At TS, however, neither Halaf painted pottery nor Amuq B DFBW was found. Nevertheless, the close similarity between the TS architecture and that from Tell Sabi Abyad (mentioned above) allows us to infer that both were of the same or similar culture, and chronologically both could have been more or less contemporaneous. This tradition, which began at Tell Sabi Abyad during the Late Neolithic period, was accompanied by early Halaf low-fired coarse ware pottery, and these cultural elements continued during the second half of the 6^th millennium bc at Tell Sabi Abyad. However, whether this dating is applicable to the TS structure will be discussed below.

• 30 Fig. 12: 13; Nieuwenhuyse, Daskiewicz and Schneider, 2018, p. 15-33.

23The site of Shir is situated north-west of Hama in the Orontes valley. The excavations produced Neolithic pottery that was dated to between 7000 and 6400 bc. The sherds came from the upper levels of the site, where test trenches were opened. TS jar rim sherd no. 4 (fig. 5) is comparable with the Shir item,30 a sherd of Coarse Unburnished Ware from the upper levels was identified with the ‘pottery from the pre-Halaf stages’.

• 31 Kenyon 1979, p. 46.

24According to Kenyon31 the development from the Pre-Pottery Neolithic to Neolithic A culture took place in Palestine following a hiatus of nearly one thousand years (6000‒5000 bc). Towards the beginning of the 5^th millennium bc this development led to the production of primitive ‘coarse pottery ware’ and the building of permanent rectangular dwelling structures combined with a round structure. They were erected on stone foundations with a superstructure of mud-bricks. The associated pottery (for example from Jericho) was found in pits. It consists of coarse bowls made of clay tempered with straw; these bowls also display the splaying wall profile, were low fired, have a red slip and crude surface smoothing. On the basis of this cultural development, it is inferred that Neolithic B culture in Jericho belonged to a historical period that could have fallen during or around the TS B2 post period. The TS B2 pottery, in terms of the ware and shape, can still be compared with the Palestinian counterparts. The Jericho pottery has a different chronology, and both the TS B2 and the Jericho assemblage are marked by the lack of Halaf painted wares.

• 32 Van Loon, Meijer and Akkermans 1986/1987, p. 309 and 311.

25Comparison of the TS stratum B2 architecture (Foundation Wall 1) with the Tell Sabi Abyad architecture showed strong similarities, as did the comparison between the B2 pottery items 1‒5 (fig. 5) and the Tell Sabi Abyad 2 pottery assemblage (bowls of the pre-Halaf or early Halaf).32 Comparisons with the Amuq A pottery (the Coarse Simple Ware and the DFBW) also showed strong similarities. The similarities between the pottery from TS B2 and that from Tell Damshliyyar and Gird Banahilk only concern the vessels’ shapes and materials. The strong similarities found between the TS architecture and pottery and their counterparts from Tell Sabi Abyad, Amuq A and B, Tell Damishliyya and Gird Banahilk, provide a solid argument to suggest possible contemporaneity. This statement is based on the findings of the study, which show that the vessel forms and large size of the mainly of coarse ware TS pottery found both inside and outside the building (Foundation Wall 1) were similar to the comparanda, whereas the exterior surface treatment was less so. However, the transition seen at Tell Sabi Abyad from the pre-Halaf 2 stage to the Halaf painted ceramics stage took place at TS during a time span that has so far not been confirmed due to an interruption in the occupation (in Area P, the very long hiatus mentioned in table 1). Therefore, it cannot be inferred why no Halaf painted ceramics and the DFBW pottery types have been found in B2.

• 33 The author’s PhD thesis (Mousli 2021), defended on August 12, 2021, has not been published to date.

30Foundation Wall 1 (mainly its plan and construction materials),33 which was found to be very similar to the Tell Sabi Abyad tholos, and the pottery described as ‘splaying wall’ ware and ‘ledge-handle’ ware, historically appeared between 6000‒5000 bc. No incised or impressed pottery, and no husked-tray pottery were found in TS B2. However, the TS assemblage did not include Halaf painted pottery or flints. On the basis of what was found in B2, the architecture together with the pottery are likely to remain undefined for the time being. Due to the small size of the sounding, a more accurate dating than a proposed dating of between 6000–5500 bc is not possible at this point in time. This period was followed by a long hiatus of approximately 2 500 years, which was followed by the site climax (the construction of Gates 1 and 3) that ended the evolutionary development (at least in Areas P and I).