The Decline and Disappearance of Chipped-Stone Tools: New Insights From Qubur el-Walaydah, a Late Bronze/Iron Age Site in Israel

ABSTRACT

Technological analyses of the lithic assemblage recovered from the excavations at Qubur el-Walaydah offer new insights into the nature of the specialized production system of sickle blades during the Late Bronze-Iron Age, and provide new data sets concerning the decline of chipped-stone technologies. Although the knapping technology used for the production of flint sickle segments was relatively simple, the spatial distribution of this system, the hafting procedures, and other technological considerations suggest that the specialists were no longer solely flint knappers, but had become sickle artisans. They produced the complete sickles that the farmers, who in turn lost their role in the manufacturing process, used. This specialized system can be characterized as a kind of vertical production integration. The organization of the production of flint sickles, moreover, may offer insights on the last stage of the stone-metal replacement process. If the intrinsic properties of iron and its availability must have played an important role, the flint-iron substitution might be facilitated by the fact that the production system was already characterized by a form of specialization where the users of these objects were no longer the producers. In this regard, the simultaneous fall of ad hoc flint tools represent another facet of the same phenomenon. If within the lithic production system the manufacture of sickles and ad hoc tools represent two distinct sub-systems, the large-scale adoption of iron implements suggests the emergence of a single specialized system producing different types of cutting tools, then used for different tasks. Thus, the end of chipped stone tools is related not only to the efficiency of iron implements, but also to a structural change in the relationship between tool consumers and producers.

KEYWORDS:

• Flint
• specialization
• Late Bronze Age
• Iron Age
• Southern Levant

Acknowledgement

We thank journal editor and enonymous reviewers for their constructive comments which helped in improving an earlier draft of this paper. We would like to thank Jacques Pelegrin for the photos of modern gun-flints.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Francesca Manclossi has defended her doctoral thesis “From Stone to Metal: Dynamics of technological changes in the lithic industries in the South Levant in 2016,” in a joint program between the University of Paris Ouest-Nanterre la Défense and the Ben-Gurion University of the Negev. She is currently a post-doctoral fellow at the Centre de Recherche Français in Jerusalem. Her researches focus on the shift from stone to metal, and integrating archaeology, sociology, anthropology with material analysis of stone and metal tools, her aims is to reconstruct historical scenarios concerning one of the key moments in technological evolution.

Steven A. Rosen holds the Canada Chair in Near Eastern Archaeology at Ben-Gurion University of the Negev. He received his doctorate from the University of Chicago (1983). Before joining Ben-Gurion University (1988), he led a survey team for the Negev Emergency Survey. Current research focuses on the archaeology of mobile pastoralists in the desert of the southern Levant and lithic industries from the Metal Ages. Among field projects, he directed excavations ranging from Middle Paleolithic through recent times. His most recent book is Revolution in the Desert: The Rise of Mobile Pastoral Societies in the Southern Levant (Routledge 2017).

Gunnar Lehmann is Professor for Ancient Near Eastern Archaeology at Ben-Gurion University of the Negev. He received his doctorate from the Freie Universität Berlin (1993). Current research focuses in ancient Egyptian imperialism during the Late Bronze Age in the southern Levant, Phoenician archaeology and the Assyrian empire in the Levant.

ORCID

Francesca Manclossi http://orcid.org/0000-0001-6132-7371

Notes

1. In our analysis, we did not consider 215 artifacts (ad hoc blanks, tools and waste) that derived from the topsoil.

2. For a general overview of the periods and their sub-phases see Fantalkin, Finkelstein, and Piasetzky (2015); for the chronological division between the archaeological phases see Finkelstein (2005, 2011); Mazar (2005, 2011).

3. In this regard, is important to distinguish between debitage/waste and retouched tools. If the ad hoc production waste dominates the debitage assemblage, the sickle segments dominate the tool assemblage.

4. In our technological analysis, we included also 16 large geometric sickle elements and five unretouched blanks which were collected in unclear stratigraphic contexts.

5. The presence of a single core of this type and the relatively low frequency of sickle blanks and blades (c. 10 per cent) resulting from this simple reduction strategy are not enough to support the hypothesis of different knapping schemes corresponding to various craftspeople. As suggested by the type of raw material (flint of group 3), the differences in knapping procedure are probably linked to the configuration of the blocks and/or to the stage in the reduction sequence.

6. In this analysis, we excluded about 30 per cent of the pieces which are partially fractured and do not permit definition of their original shape.

7. According to the direction of the truncations, it is possible to recognize some variants in the lateralization of the edges. Thus, the parallelogram elements can be // or \\, and the trapezoidal segments can be / \, \ /, |/, |\, \|, or /| (see Rosen, 2004, pp. 2204–2205).

8. Sickle segments are identified by the presence of gloss (Curwen, 1930). Although it is generally agreed that the harvesting of cereals results in the formation of gloss (e.g. Anderson, 1980), and that its formation does not require especially intense use (Unger-Hamilton, 1985), other possible functions linked to other type of plants (e.g. reeds, bulrushes, grass, etc.) have been suggested. In the case of the large geometric elements, their use in composite sickles is well proved by the discovery of complete items (e.g. Petrie, 1932, pl. XXIII: 50), by contemporaneous representation in Egyptian paintings (see Coqueugniot, 1991, p. 196), and by some use-wear analyses (Rosen et al., 2014; Coqueugniot, 1991, pp. 199–200).

9. Although the two materials are similar, the main difference between gypsum (plaster of Paris) and lime plaster (calcite plaster) is related to the technology used to produce the “powder” that, mixed with water, creates a plastic and easily manipulated substance which, once dry, becomes compact and solid. This powder is obtained by heatin limestone which according to the temperature and its composition (e.g. sulfides, calcite, carbonates, silicates, etc.), undergoes different chemical/physical reactions. If low temperatures are enough for the production of gypsum, calcite plaster requires a treatment at more than 800°C (Endlicher & Tillmann, 1997).

10. Coqueugniot (1991, pp. 159–163, 195–198), Mozel (1983), and Rosen (1986, p. 260) describe in detail the hafting arrangement of sickles made of large geometrics in the Levant during the 2^nd–1^st millennia bce. This hafting system is the last of a long development that started during the Natufian and continued during the Neolithic, Chalcolithic and Early Bronze Age (for a more detailed analysis of different sickle hafting arrangements see Fisher, 2008, pp. 350–352; Groman-Yaroslavski, Weiss, & Nadel, 2016; Kadowaki, 2005; Vardi, 2011; Winter, 2006; and beyond the Levant see e.g. Astruc, Ben Tkaya, & Torchy, 2012; Borrell & Molist, 2007; Gurova, 2016; Holdway & Douglass, 2005; Maeda, Lucas, Silva, Tanno, & Fuller, 2016; Mortensen, 1970; Nishiaki, 2003).

11. In the Southern Levant, during the 2^nd–1^st millennia bce, the plaster (either gypsum or calcite plaster) seem to be the only adhesive used in the manufacture of composite sickles. This adhesive replaced bitumen, commonly found until the end of the third millennium bce (e.g. Marder, Braun, & Milevski, 1995; Milevski, Marder, & Goring-Morris, 2002). However, in other regions such as the Northern Levant, the bitumen continues to be used for the hafting of large geometric sickle elements (Adam et al., 2013).

12. In this regard, for example, we can observe that 5–6 per cent of the waste category is composed of pieces that, according to the type of raw materials, could belong to the manufacture and/or repair of the large geometric sickle elements.

13. Although it is possible to recognize some analogies with the large geometric cores, the ad hoc cores differ from those ascribed to the production of sickle elements in the type of raw material, the size and morphology of the detached blanks (e.g. ad hoc blanks are usually smaller and more irregular than those transformed in large geometrics), the organization of the reduction sequence, and the use of metal hammers.

14. The organization of the knapping and the level of execution of this group of cores show some analogies with the cores used for the production of large geometric sickle elements. Although they could be the last stage of a more intensive exploitation (that is, a “reconversion” of large geometric cores for a more expedient production of flakes then used as ad hoc tools), the size and morphology of the removals (too small and irregular to be used as sickle elements) seems to indicate a production with a different aim. Another argument suggesting a different production is the lack of technical features indicating the use of metal hammers.

15. In this regard, it is interesting to observe that in the Southern Levant large geometric sickle elements are generally manufactured on the same type of raw materials, independent of the site in which they are found (e.g. Hammond, 1977). This uniformity could indicate that the sickle artisans, from different sites, went to the same geological outcrops (this kind of flint is quite frequent in the central-north part of the Southern Levant; Sneh et al., 1998), or that there were flint knappers who provided blanks to the sickle artisans. It is also possible to imagine that both systems existed and, according to the geographical location of the site, the production of sickles was differently organized (for an ethnographic example of a multi-central network with different but overlapping production systems see Karimali, 2005).

16. The hypothesis that the farmers were the producers of their own sickles, although the flint inserts were produced by specialized knappers, has been suggested for the Chalcolithic and the Early Bronze Age (Rosen, 1997, pp. 104, 107).

17. This system profoundly differs from that of the preceding the Early Bronze Age, where a few specialized knappers produced standardized Canaanean blades and supplied them to the farmers/consumers who shortened and placed them into the hafts, also retouching them and reversing the segments in order to prolong their use (e.g. Manclossi et al., 2016, and references within).

18. In this regard, it is interesting that neither the proportion of raw materials nor the techno-typological characters (e.g. the organization of the removals, the size and shape of flint elements, etc.) change from the Late Bronze to the Iron Age. The use of metal hammers is attested during both the LB and the IA I, and cores (although very few) have been found in layers dated to both periods.

19. Written texts indicate the presence of workers attached to the Egyptian estates. However, none of them mention flint knappers or sickle artisans, but rather they mention potters (Lesko, 1995) and other forced laborers from the surrounding city-states (Moran, 1992).

20. The production of large geometric sickle elements started at the beginning of the second millennium bce, during the Middle Bronze Age (Coqueugniot, 1991, p. 158; Rosen, 1997, p. 60), and replaced the specialized production of Canaanean blades.

21. In this regard, the extension of the excavated areas in Field 1 is of 600 m², in Field 1A it is 175 m², and in Field 2 it is 240 m². The occupation time varies considerably according to the periods: 70 years for the LB III, 190 for the IA I, 140 for the IA IIA (40 years for the earliest phase, and 100 for the latest), 75 years for the IA IIB, 85 years for the IA IIB-C, and 40 years for the IA IIC (compare to Rosen, 1996a, 1997, p. 152; Rosen et al., 2012, p. 506).

22. Although we cannot rule out the possibility that flint sickles continued to be used during the IA IIB and IA IIB-C, the very low number of large geometrics seems to reflect intrusion from earlier layers. This pattern holds for other sites as well (e.g. Rosen, 1996a).

23. In this paper, we do not deal with the complex and long process of flint-metal replacement which, in the Southern Levant, took place over more than three millennia (Rosen, 1996a, 1997, pp. 159–164; see also Manclossi, 2016).

24. The possibility that sickle artisans were also farmers is supported by various ethnographical and historical examples where the knapping (and tool manufacture) was a part-time specialization integrated with other activities (e.g. Darras, 2009; Karimali, 2005). However, all the other farmers (who represented the majority of the population) did not participate any more in the production of sickles but acquired finished tools, made by specialized artisans.

25. Most of the metal objects dated to the Bronze Age derive from tombs (e.g. Philip, 1988), and only a small percentage of items come from villages and cities.

26. In this regard, it is interesting to observe that at Qubur el-Walaydah the number of ad hoc tools seems to increase from the LB to the IA I, and then decrease, although during the IA IIA the quantity of chipped-stone tools is greater than during the late second millennium bce. Although we cannot explain the meaning of these differences (they could represent variations in the function of the site), these data clearly show that the decline of ad hoc tools was not gradual through time, but ceased quite rapidly after the IA IIA. The low incidence of ad hoc tools during the subsequent IA IIB, IIB-C, and IIC might correspond to some smaller productions of chipped-stone tools which survive the general disappearance of the domestic production of flint implements, or, more likely, they could be intrusions from older layers. The fact that their percentage seems relatively stable seems to support this second hypothesis.