Universidad Autónoma de Madrid Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East Madrid, April 3-8 2006 Edited by Joaquín Mª Córdoba, Miquel Molist, Mª Carmen Pérez, Isabel Rubio, Sergio Martínez (Editores) Madrid, 3 a 8 de abril de 2006 Actas del V Congreso Internacional de Arqueología del Oriente Próximo Antiguo VOL. II Centro Superior de Estudios sobre el Oriente Próximo y Egipto Madrid 2008 Pierced clay disks and Late Neolithic textile production Kevin T. Gibbs, Toronto Abstract Pierced clay disks, fashioned from potsherds, are often found on Late Neolithic sites in the Near East. Several suggestions have been offered to explain the occurrence of these objects but the most common suggestion is that they were used as spindle whorls for spinning yarn or thread for textile production. A collection of pierced disks from the Late Neolithic site of Tabaqat alBûma in Wadi Ziqlab, Jordan, was examined to determine the production sequence of these objects and to assess their suitability for use as spindle whorls. Replicas of a number of the disks were affixed to wooden spindles and used to spin several types of fibres, including flax and animal hair. The results of the analysis and experimental testing suggest that the pierced disks could have been expedient spindle whorls. Keywords: spindle whorls, Late Neolithic, Wadi Ziqlab, production sequence. Introduction Fired-clay disks with holes through their centres are often found on sites in the southern Levant that date to the Late Neolithic period (approximately 7500-6000 BP). An early suggestion was that these pierced disks were used as spindle whorls (e.g. Kaplan 1969; Wheeler 1982). To test this hypothesis a small assemblage of pierced disks from Tabaqat al-Bûma, a Late Neolithic site in Wadi Ziqlab, northern Jordan, was analyzed and replicated (figure 1). The Wadi Ziqlab Disks Wadi Ziqlab is one of several tributaries of the Jordan River that drain the western part of the Irbid Plateau in north-western Jordan. The Wadi Ziqlab Project, under the direction of Professor E.B. Banning of the University of Toronto, has discovered a number of Late Neolithic sites in the wadi, including Tabaqat alBûma (WZ200), which was likely a small farmstead (Banning and Siggers 1997). Excavations at the site recovered several pierced ceramic disks. Ten of these were available for study, as was one from another nearby Late Neolithic site (WZ310). Eight of the examined disks are complete or near complete. Of the 11 observed disks, 10 are round to oval in shape while the remaining one is more angular. The mass of the disks (excluding the incomplete examples) ranges from 7.3g to 21.2g (mean=13.4g; figure 2) while the thickness ranges from 6mm to 17mm (mean=10.0mm). The maximum diameter (taken across the perforation) ranges from 30 to 54mm (mean=39.5mm). The colour of the disks is typical of other sherds from the site, ranging from pink to reddish yellow to very pale brown. None of the disks are decorated, although decorated pottery of any sort is rare from Late Neolithic sites in Wadi Ziqlab. 90 Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East The production sequence Macroscopic observation of the disks, combined with subsequent replication experiments, shed some light on their production sequence. First, the Tabaqat al-Bûma disks were fashioned from potsherds. While the disks’ fabrics are varied, they all fit within the range of fabrics of ceramic vessels at the site. Moreover, some of the disks show a slight concavity, resulting from the contour of the pot from which they were recycled. Second, although no evidence of the shaping technique was observable on the edge of the disks, the rough shape of a disk could easily be attained through chopping and scraping with flint tools, like ones commonly found on Late Neolithic sites (figure 3a). Third, the common hourglass-shaped perforation was the result of drilling from both sides. This could be achieved with a simple flint drill, or even a small unretouched flake held between the thumb and finger (figure 3b). One disk had obvious striations on the interior of the hole, likely the result of using a sharp flint implement on a fired sherd. Fourth, small chips around the holes of two disks could be the result of drilling with a back-and-forth motion (alternating clockwise-counter clockwise) without attaining a complete rotation with each alternation (figure 4). This can result in small, remnant points jutting into the hole in one or two places which, when finally removed with a complete rotation, take a piece of the surface as well. The reason for the incomplete rotation likely reflects the limited range of movement of the human wrist (Gwinnett and Gorelick 1981). Fifth, concentric striations on the surface of one disk could have resulted from the driller’s thumb nail coming into contact with the disk while holding a drill between the finger and thumb (figure 5). This is further evidence that the disks could be pierced with a small unhafted tool. Sixth, the shaping and drilling of a pierced disk could be completed in less than two minutes with simple, unhafted flake tools. Are pierced disks spindle whorls? The suggestion that Late Neolithic pierced disks are spindle whorls is a plausible one as whorls made from sherds have been observed in ethnographic contexts (e.g. Crowfoot 1931). However, other possible functions have been proposed (Verhoeven 1999:238-239). These include their use as gaming pieces (Erikh-Rose and Garfinkel 2002), accounting tokens (Schmandt-Besserat 1992), markers of identity (Tsuneki 1998), pendants or ornaments (Liu 1978), jar stoppers (Mallowan and Rose 1935; Garfinkel 1992), or as flywheels for pump-drills or children’s spinning-tops. Future research should aim to address these alternate hypotheses. Spinning in the Late Neolithic Hand spinning involves drafting or drawing out fibres from a mass and twisting them into a stable yarn or thread. The use of a spindle increases the speed of spinning and provides a convenient place to wind the thread to keep it from untwisting. Drop spindle spinning requires the spindle to rotate freely on the end Pierced clay disks and Late Neolithic textile production 91 of the thread being spun. The whorl’s role in the process is to act as a flywheel, stabilizing the spindle and maintaining its rotation. Unfortunately, direct evidence for spinning technology, in the form of preserved fibres, has not been recovered from Late Neolithic sites in the southern Levant. However, rare textile specimens from the earlier Pre-Pottery Neolithic B and the following Chalcolithic reveal what Late Neolithic textiles might have looked like (Schick 1988, 1998). Flax was the most probable source of fibre, although human, goat, or dog hair could also have been used (Barber 1991). Implements used in the spinning process, including whorls, may provide crucial evidence for spinning in the Late Neolithic. Published accounts of spindle whorls observed in ethnographic contexts suggest certain characteristics are desirable in a whorl (Liu 1978). As utilitarian objects, whorls are often undecorated. They are frequently circular although other symmetrical shapes also occur. Whorl size varies with the fibre being spun, with flax benefiting from a heavy whorl of 140 grams or more (Barber 1991). Hole size is generally 7mm or more to accommodate a spindle, and the hole’s shape is usually tapered or straight to fit snugly on the spindle. The disks from Tabaqat al-Bûma fit some of these criteria, but not others. While they are undecorated and roughly circular, some are not particularly symmetrical, too light to optimally spin flax, and most have hourglass shaped holes that would seem not to fit well on a spindle. Experimental Use To further assess the disks’ suitability for use as spindle whorls, a number of them were replicated and mounted on wood spindles and used to spin a variety of fibres, including flax, mohair, wool, and dog hair. After a very short time even a novice spinner was able to produce lengths of yarn of each kind of fibre. Even the most asymmetrical «whorls» could be used. The low mass of the whorls was not a critical impediment even when spinning flax. Likewise, the hourglass-shaped holes did not create a critical problem as the spun fibre could be wrapped around the whorl in the manner commonly used with low-whorl drop spindles, keeping the whorl in place. Conclusions This preliminary research suggests that Late Neolithic pierced disks could be quickly formed from a potsherd using simple, unhafted flint tools. Replication and experimentation demonstrates that, even though the pierced disks from Tabaqat al-Bûma may not be optimally suited for use as spindle whorls, the possibility cannot be ruled out. In fact, as expedient tools, the disks suit hand spinning, which can be an intermittent task, picked up and carried out during episodes of free time between other jobs (Barber 1991). Acknowledgements I would like to thank E.B. Banning, Heather Miller, Lisa Maher, and the Toronto Guild of Weavers and Spinners for their assistance in making this research possible. 92 Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East Bibliography BANNING, E. B. AND J. SIGGERS 1997 Technological Strategies at a Late Neolithic Farmstead in Wadi Ziqlab, Jordan. In The Prehistory of Jordan II: Perspectives from 1997, edited by H. G. K. Gebel, Z. Kafafi and G. O. Rollefson. Ex Oriente, Berlin. BARBER, E. J. W. 1991 Prehistoric Textiles: The Development of Cloth in the Neolithic and Bronze Ages. Princeton University Press, Princeton. CROWFOOT, G. M. 1931 Methods of Hand Spinning in Egypt and the Sudan. Bankfield Museum Notes Series 2, Number 12, Halifax. EIRIKH-ROSE, A. AND Y. GARFINKEL 2002 The Pottery. In Sha’ar Hagolan, Volume 1: Neolithic Art in Context, edited by Y. Garfinkel and M. A. Miller, pp. 86-138. Oxbow, Oxford. GARFINKEL, Y. 1992 The Pottery Assemblages of the Sha’ar Hagolan and Rabah Stages of Munhata (Israel). Cahiers de Recherche Français de Jerusalem 6. Association Paléorient, Paris. GWINNETT, A. J. AND L. GORELICK 1981 Beadmaking in Iran in the Early Bronze Age. Expedition 24(1):10-23. KAPLAN, J. 1969 ‘Ein el Jarba: Chalcolithic Remains in the Plain of Esdraelon. Bulletin of the American Schools of Oriental Research 194:2-39. MALLOWAN, M. E. L. AND J. C. ROSE 1935 Excavations at Tell Arpachiyah. Iraq 2:1-178. LIU, R. K. 1978 Spindle Whorls: Pt.I. Some Comments and Speculations. The Bead Journal 3:87-103. SCHICK, T. 1988 Cordage, Basketry and Fabrics. In Nahal Hemar Cave, edited by O. BarYosef and D. Alon, pp. 31-43. ‘Atiqot. vol. 18, Jerusalem. SCHICK, T. 1998 The Cave of the Warrior: A Fourth Millennium Burial in the Judean Desert. IAA Reports 5. Israel Antiquities Authority, Jerusalem. SCHMANDT-BESSERAT, D. 1992 Before Writing, volume I: From Counting to Cuneiform. University of Texas Press, Austin. TSUNEKI, A. 1998 Other Objects. In Excavation at Tell Umm Qseir in Middle Khabur Valley, North Syria: Report of the 1996 Season, edited by A. Tsuneki and Y. Miyake. Pierced clay disks and Late Neolithic textile production 93 Al-Shark. vol. 1. Department of Archaeology, Institute of History and Anthropology, University of Tsukuba, Tsukuba. VERHOEVEN, M. 1999 An Archaeological Ethnography of a Neolithic Community: Space, Place and Social Relations in the Burnt Village at Tell Sabi Abyah, Syria. Nederlands Historisch-Archaeologisch Institut, Istanbul. WHEELER, M. 1982 Appendix F: Loomweights and Spindle Whorls. In Excavations at Jericho, volume four: The Pottery Type Series and Other Finds, edited by K. M. Kenyon and T. A. Holland. British School of Archaeology in Jerusalem, London. 94 Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East Fig. 1: Late Neolithic pierced disks from Wadi Ziqlab. Fig. 2: Scatterplot of mass and diameter of the seven most complete disks. Pierced clay disks and Late Neolithic textile production Fig. 3: Using simple flint tools to shape the edge of a replicated disk (a) and drill a hole (b). Fig. 4: Chips around the hole of a Late Neolithic disk (a) and a replicated disk (b). 95 96 Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East Fig. 5: Striations on the surface of a Late Neolithic disk (a) and a replicated disk (b).