Click on small images to see larger version.

During June and July of 1995, the Wadi Ziqlab Project carried out excavation of a Chalcolithic site on the western slopes of Tubna, southwest of Irbid, Jordan. Lying on some agricultural terraces that overlook a deep tributary of Wadi Ziqlab, about halfway between the Jordan Valley rift and the Ajlun plateau, the site was discovered through sub-surface testing by a 1 x 2 m trench in 1993. ¹ Its location is reminiscent of Hanbury-Tenison's observation that in the Jerash region "Chalcolithic sites are high up towards the top of south or west facing slopes, with access to water at some distance, and covering at least two hectares." ² The nearest water source would have been the stream some 100 m below the site, or a spring that locals tell us once existed about 1 km to the south. Excavations were part of the ongoing Wadi Ziqlab Project, a regional investigation of the late prehistory of al-Kura district, Jordan, and were initially intended to test for the presence of Late Neolithic occupation on the site.

The site has been disturbed by recent agricultural activities, including pitting, the removal of stone and the planting of olive trees, but still preserves Chalcolithic deposits more than 30 cm thick over an area of at least 1 ha. In the southern part of the site, excavations over an area of 90 m² revealed several stone-lined pits, many of which had been paved over with small flagstones, many more unlined pits, and much of a long building, with stone foundations more than 1 m in thickness, that had been rebuilt several times. The stone-lined pits were filled with very ashy sediment containing high densities of sherds. Similar stone-lined pits that occurred at other Chalcolithic sites, such as Shiqmim in southern Israel, ³ have turned out to be cist-graves, but the ones at Tubna appear to have had quite a different function. The foundations may be part of a large broad-room domestic structure typical of Chalcolithic sites in the southern Levant. In a more northly part of the site, excavations over an area of about 30 m² revealed the corner of a similar Chalcolithic building into which two stone-lined pit features had been inserted after its abandonment. One of these had been covered with small flagstones, like some of the pits farther south, and contained many fragments of Chalcolithic pottery, most of which appear to come from two large storage vessels, each with a row of lunate spatula impressions below their out-flaring rims. Smaller excavations elsewhere on the site were used to test for the extent of Chalcolithic settlement, which seems to cover at least 1 ha.

The deepest deposits at the site, only reached over a small area towards the southern end of the site, contained small numbers of sherds that appear to be similar to Late Neolithic pottery from Tabaqat al-Buma (site WZ 200), about 3 km away, including widely splayed strap handles. The earliest settlement here appears to have had shelters built into natural hollows in the bedrock.

Later Chalcolithic pottery includes large storage vessels with rope moulding over parts of their bodies and medium or large jars with a row of applied scallop-like bands or spatula-impressed lunates below their out-flaring rims. Other common forms are V-shaped bowls, hole-mouth jars with slightly or heavily out-flaring rims, and small cups. There are broad strap handles as well as the small pierced lugs typical of Chalcolithic sites in the region, and bases are typically flat or disk-bases. There is one example of a bowl with a "pie-crust" rim. Fabrics are usually very coarse, with large inclusions of chert, limestone or grog. The most common sherds have red, yellow or orange paste with angular chert inclusions and no slip. Less common fabrics are yellow or light brown with a brown or buff grog or clay temper, or grey and black with angular chert inclusions. Neither cornets nor string-cut bases are apparent in the assemblage. Much of the pottery has close parallels in form, but not in fabric, with pottery from the sites of Abu Hamid ⁴ and Tell Fendi ⁵ in the Jordan Valley and Pella XXV ⁶ in the foothills.

Among the approaches we are using to investigate the pottery is the chemical analysis of residues that the pots absorbed during use. During excavation we collected pottery samples along with samples of associated soil specifically for these analyses, with great care to avoid contamination, and have kept them refrigerated continuously since excavation to slow down any chemical alteration of the residues. We will now subject it to analysis of absorbed amino acids by a combination of gas chromatography and mass spectrometry. ⁷ It is our hope that these analyses will not only help us to identify the functions of various vessel types at the site, but will contribute to our understanding of the origins of dairying, beer-making and wine-making in the region

Although utilized flakes dominate the lithic assemblage at the site, there is also a large repertoire of bifacial adzes, axes, picks and chisels and smaller numbers of sickle blades and cortical scrapers. The last are made on broad cortical flakes removed from flat nodules rather than on the tabular flint commonly used for "fan scrapers" in the Chalcolithic of the Levant.

The assemblage also includes animal bone, fragments of basalt grinding stones and chalice-like vessels, and two broken haematite mace heads. No burials were encountered.

The analyses we are now carrying out on other materials from the site include microrefuse analyses of sediments from gridded surfaces within the large structure in the southern excavation areas. We plan to use the distribution of microscopic fragments of bone, shell, pottery, lithics, charcoal, basalt and other refuse across housefloors and outside houses to investigate small-scale spatial variation that might help us to infer activity areas and site-formation processes. Our methods involve painstaking counting of subsamples to arrive at mean densities and standard errors but in other respects are similar to the methods that Rosen has used on some Chalcolithic and Early Bronze Age sites in Israel. ⁸ Our work to date, involving dozens of undergraduate volunteers, suggests that the larger artifacts are rarely found where they were used, but instead represent trash dumped into houses after they were abandoned. During occupation of these houses, regular sweeping kept floors relatively clear of macroscopic artifacts, but tiny flakes from manufacture of stone tools, microscopic chips of bone, pottery and charcoal from cooking activities, snail shell and insect parts related to microclimatic variations within houses were too small to be swept up completely. When complete, the microrefuse distributions from Tubna will be compared with those already analyzed from two Late Neolithic house floors at nearby Tabaqat al-Buma ⁹ and with distributions of phosphate and pH levels as determined by analysis of sediment samples we collected in the field.

DEPARTMENT OF ANTHROPOLOGY
UNIVERSITY OF TORONTO
TORONTO, ONTARIO M5S 3G3

Fig. 1. Map of Jordan showing the locations of Tubna (WZ 121) and Tell Fendi (WZ 126) in the Wadi Ziqlab drainage of northern Jordan (drawn by E. Banning) Pl. 1. View of two Chalcolithic stone-lined pits in Area P22 at Tubna in Wadi Ziqlab, Jordan (photo by P. Racher) Pl. 2. Vertical view of a portion of the east wall of a large Chalcolithic structure in the southern portion of the site (photo by E. Grossmith)

² J.W. Hanbury-Tenison, "Jarash region survey 1984," Annual of the Department of Antiquities of Jordan 31 (1987) 129-157.

³ T.E. Levy, Shiqmim I: Studies Concerning Chalcolithic Societies in the Northern Negev Desert, Israel (1982-1984). BAR International Series 356 (Oxford 1987).

⁴ G. Dollfus, Z. Kafafi, "Recent researches at Abu Hamid," Annual of the Department of Antiquities of Jordan 37 (1993) 241-255.

⁵ M. Blackham, K. Fisher, D. Lasby, "Excavations at Tell Fendi, a late Chalcolithic site in the Jordan Valley, Jordan," this volume; J. Kareem, "Tell Fendi: Jisr Sheikh Hussein Project, 1986," Annual of the Department of Antiquities of Jordan 33 (1989) 97-109.

⁶ A.W. McNicoll, P.C. Edwards, J. Hanbury-Tenison, J.B. Hennessy, T.F. Potts, R.H. Smith, A. Walmsley, P. Watson, Pella in Jordan, 2 (Sydney 1992) pl. 14.

⁷ Alicia Beck will be carrying out this work as part of her doctoral research in the Department of Near and Middle Eastern Civilizations, University of Toronto.

⁸ A. Rosen, "Ancient town and city sites: A view from the microscope," American Antiquity 54 (1989) 564-578.

⁹ P. Racher, E.B. Banning, "Sampling theory and microrefuse analysis: Neolithic house floors in Wadi Ziqlab, Jordan," paper presented at the 62nd Annual Meeting of the Society for American Archaeology, 6 April 1997, Nashville.