Back to Research Summary

2000 Survey in Wadi Ziqlab


The Wadi Ziqlab Project conducted geomorphological and archaeological survey of wadi terraces in Wadi Ziqlab from 28 May to 29 June 2000 under the co-direction of Lisa Maher and Ted Banning. Its goals were reconstructing late Pleistocene and early Holocene paleoenvironments and how the valley changed over time, and relating these changes to the distributions of sites, especially those of the Epipalaeolithic, Neolithic and Chalcolithic periods. The survey discovered 18 previously undocumented sites, including ones with Geometric Kebaran, Pre-Pottery Neolithic B (PPNB), and Late Neolithic material, and also found that the extent of Neolithic and Early Bronze occupation at Tell Rakan (WZ 120) was greater than previously believed. It documentecd the alluvial and colluvial histories of several places in the wadi that are key to reconstructing the prehistoric topographies associated with these sites.

The 2000 field season was a short one explicitly addressing the reconstruction of the valley's changing morphology, principally by examination of terraces, their stratification, where visible in section, and the artifacts in them and on them. Other geological and geomorphological information includes changes in water course, location of springs, tectonics, sediment transport and deposition. The survey team also collected bulk sediment samples, sediment thin sections, and samples for micro-artifacts. We know that the wadi has changed dramatically over time through the interplay of natural factors (climate, channel course, vegetation) and human use (deforestation, agriculture, irrigation, settlement). Ultimately, the paleoenvironmental reconstructions will be used to examine human land-use patterns and to predict the location of archaeological sites in less intensively surveyed areas in the region.

A second focus of the 2000 work was specifically to reconstruct how the Neolithic, Chalcolithic and Early Bronze Age inhabitants of Tell Rakan, where small excavations took place in 1999, were using the surrounding landscape and its different elements. Consequently, a large proportion of the fieldwork took place in the neighbourhood of Tell Rakan and Tell Abu Fokhkhar.

Methods

Survey was mainly by groundwalking on foot, with routes planned to examine a series of terraces previously identified in aerial photographs, rather than following arbitrary transects. At each terrace, a team of six spread out and intensively examined the modern surface, which was usually plowed. In addition, the team paid close attention to any sections exposed by road cuts, erosion gullies or the wadi channel, and to terrace margins where erosion may have exposed buried material. Initial inspection of these terraces took place during the first two weeks of the survey, and was followed by subsurface and column sampling at selected locations.

Subsurface sampling of selected terraces involved several different approaches, depending on the situation. One terrace on which prehistoric settlement was likely (WZ 140) was sampled by two small test trenches, 1 m x 1 m in area, excavated stratigraphically by trowel and hand pick. On another (WZ 148), where a road cut exposed the stratification, two stepped trenches, 1.5 m wide and intruding up to 40 cm into the section, were cut into the steep slope. Again this was done stratigraphically with trowels and handpicks. On terraces where the goal was to obtain a stratified column of bulk sediment samples and possible micro-artifacts, but no material culture was obvious, column samples were excavated from the sides of exposed sections, usually ones cut by the modern wadi channel. In a few cases we attempted to obtain bulk samples by augering, but the high stoniness of deposits made this quite difficult. At some sections exposed by wadi downcutting or road cuts, the section was cleaned by pick and trowel and then overlapping samples for thin-section analysis were taken in large blocks enclosed in plaster as it was found that Kubaniyya boxes were ineffective at removing intact sediments here.

Terraces on which we found reasonably substantial evidence of occupation received site numbers in series with those from the original 1981 survey, beginning with number WZ 133. As in previous seasons, where we tested terraces with no obvious cultural remains, we called them "localities" or "sections" and gave them higher numbers, in this case in the 400 series, to distinguish these primarily geological samples from the "archaeological sites."

Archaeological Results

Not only do archaeological remains help us to date terraces and other features that provide evidence for the paleoenvironmental reconstruction, some of them turned out to be important in their own right. These fill gaps in our knowledge of site distributions based mainly on the 1981 surface survey and improve our understanding of how people used the Ziqlab drainage basin in late prehistoric times. In addition, most terraces turned up some slight evidence of use in Roman, Byzantine, or Umayyad times, in the form usually of small body sherds, and, where it was more notable, we mention finds of later periods.

Paleolithic sites

A number of remnant terraces show evidence of Paleolithic, and sometimes specifically Middle Paleolithic, use. These will help us determine the location of the wadi floor at some point prior to about 50,000 years ago.

Levallois points and large flakes with similar white patina occur on terraces WZ 131, WZ 133 (both near ad-Dakhadil, west of Tell Abu Fokhkhar), and WZ 133 (north of Marhaba). Other lithics with a high probability of belonging to the Middle Paleolithic occur at WZ 134 and WZ 135, terraces on the south side of the wadi, opposite Tell Abu Fokhkhar. The terraces on both sides are at closely similar elevations (0 m ASL), providing strong indications that they represent the margins of the valley floor in this part of the wadi at the time the lithics were deposited or, alternatively, belong to the same episode of colluviation when the valley floor was considerably higher than today.

Epipaleolithic sites

The survey discovered three previously unknown Geometric Kebaran sites (ca. 14,000 to 13,000 BP), which, in combination with the previously excavated Geometric Kebaran material from Tabaqat al-Buma (WZ 200), brings the inventory of such sites in Wadi Ziqlab to four. Their assemblages contain many backed bladelets, including rectangles and trapezes, and bladelet cores, and some endscrapers.

Sites WZ 136 and WZ 138 are adjacent remnant terraces separated by a small gully at the east end of the modern Ziqlab reservoir, in an area now known as al-Qas'a. The exposed sections of each terrace clearly shows deposits of alluvial gravels and tufa overlain by recent colluvial material. Here the assemblages are eroding out of the side of the terrace from the water-lain sediments, as opposed to the virtually empty capping colluvial sediment. The assemblage of WZ 136 contains clearly Epipaleolithic material, including numerous bladelet cores, unretouched and backed blades and bladelets, scrapers, rectangles and trapezes (Bar-Yosef 1970). WZ 138 contains Epipaleolithic material in much smaller quantities, and possibly some Neolithic material, which originally brought the site to our attention.

We conducted small test excavations at site WZ 148, which has a high density of Geometric Kebaran lithics associated, in one level, with well preserved faunal bone.

So far, we have still not discovered any sites with lunates or other artifacts we would attribute to the Natufian or Pre-Pottery Neolithic A (PPNA), apart from a single Helwan point previously known from Tabaqat al-Buma (WZ 200).

Pre-Pottery Neolithic

Although some artifacts found on various terraces, and especially some of the long, slender blades, represent possible occupation or use during Pre-Pottery Neolithic B (PPNB), most of our new evidence for PPNB consists of the discovery that the Neolithic settlement at Tell Rakan (WZ 120) was larger than we previously suspected, continuing onto neighbouring slopes and terraces. We found no clear evidence of PPNA occupation during the survey.

Survey of terrace WZ 143, on the other side of the tributary of Wadi Antar and a little southwest of Tell Rakan, showed fairly abundant lithics and ground stone that most likely belongs to the PPNB. These include an unfinished adze. It is possible that this represents fairly intensive use of this terrace for agriculture in PPNB times, but the fairly high artifact densities in some parts of the terrace suggest that at least some Neolithic houses may have been located here. Today the terrace has a small olive grove on it.

The terrace a short distance west of the concrete fish ponds at Tell Rakan was impossible to survey in 1999 because of the jungle-like cover of reeds and the many snakes it hosted. Since then, however, this area has been cleared, unfortunately bulldozed, plowed and planted with vegetables, giving us the opportunity to sample it by walking parallel transects about 5 m apart. As it turned out, this new field contained really abundant PPNB material, including a large limestone saddle quern, many blades, and a small biface. Although we had expected this terrace to have served as an agricultural field for the PPNB occupants of Tell Rakan, it now seems very likely that PPNB houses occurred here. Unfortunately the bulldozing (to depths of 1.5 m, according to the landowner) has presumably removed most traces of architecture and all the stratigraphic context for these artifacts. Although this is presumably an extension of Tell Rakan, for convenience we gave this terrace its own site number, WZ 144.

Late Neolithic

Lithic assemblages of the Late Neolithic tend not to be particularly distinctive, consisting mainly of flakes and various by-products of "expedient" core reduction. This makes them more difficult to identify and distinguish from Chalcolithic and Bronze Age assemblages. However, we can tentatively attribute a number of the new surface scatters to the Late Neolithic on the basis of the high proportions of "amorphous" multi-directional cores and the presence of some types of sickle elements. The sickle elements found to date seem more consistent with an age quite late in the Late Neolithic, showing relatively fine denticulations more consistent with the Wadi Rabah facies, than with the Yarmoukian, the lithics of which usually show quite coarse denticulations. No pottery that we could attribute to the Neolithic with reasonable certainty was found on the surface, but that is fairly typical of the Late Neolithic; Neolithic sherds tend to be too fragile to survive exposure on the modern surface. However, test excavations at site WZ 140 did yield a rather large number of Late Neolithic sherds.

Chalcolithic

The 2000 survey did not discover any substantial Chalcolithic remains. Although some lithics and a number of body sherds from sites WZ 140 and WZ 141 probably belong to this period, none are diagnostic. The fabrics of some of the body sherds, however, appear similar to Chalcolithic ones found at Tell Rakan (WZ 120) only 1 km to the east, and closer laboratory examination may allow us to attribute them to the same types.

Early Bronze Age

Definite evidence for the Early Bronze Age comes from a fairly high terrace, WZ 150, above and east of WZ 143 and across a small wadi from Tell Rakan (WZ 120). Elsewhere, occasional small body sherds with gritty fabric probably belong to the Early Bronze Age but, as with possible Chalcolithic sherds, require more detailed laboratory examination and comparison of fabrics with sherds from Tell Rakan.

WZ 150 received only cursory examination during the survey, and the two diagnostic Early Bronze sherds came from the slope immediately below it and presumably derive from it. Very few lithics were seen on the terrace itself while the survey team was walking to nearby terrace WZ 143. It seems likely, however, that this terrace, like site WZ 130 on the spur above Tell Rakan I, was an extension of that settlement during the Early Bronze Age or at least was an agricultural field used by its inhabitants.

Iron Age

Several Iron Age sherds were found on the surface at terrace WZ 140, and possible ones elsewhere, but the main collection of Iron Age material during the survey occurred at a robbed cemetery, WZ 149.

A brief summary of the 2000 survey will appear in Annual of the Department of Antiquities of Jordan.

A gallery of photographs from the fieldwork is under construction.

Web access to the Wadi Ziqlab Database is also currently under construction.

 

Main Page  Publications  Ted Banning  UofT Anthropology


©2000 E. B. Banning, all rights reserved