Excavating tailing piles at Kakal Spur (Kerem Ben Zimra) locality in the Nahal Dishon prehistoric flint extraction and reduction complex, northern Galilee, Israel

Highlights

• Report of an excavation in a prehistoric flint extraction and reduction tailing pile in the Dishon Complex, Galilee, Israel.

• Evidence of intense exploitation during the Lower and Middle Paleolithic, and limited Neolithic/Chalcolithic activities.

• Estimate of ~300,000 flint items weighing ~24 tons for the whole tailing pile tested

Abstract

Recent research has demonstrated that the Eocene Timrat Formation outcrops in northeastern Israel, which appear as an extensive land ‘strip’ west of and parallel to the Jordan Rift Valley, was a major source of flint in prehistoric times. This is supported by the identification of three large-scale extraction and reduction (E&R) complexes (Nahal Dishon, Mt. Achbara and Sede Ilan) that offer direct evidence of intense exploitation during the Lower and Middle Paleolithic, and limited Neolithic/Chalcolithic activities. We present the results of an excavation performed in two E&R tailing piles at Kakal Spur (Kerem Ben Zimra) in the Nahal Dishon Complex. Extrapolation of the amount and weight of knapped flint items from the excavated part of Pile 1 results in an estimate of some ~300,000 flint items weighing ~24 tons for the whole pile. Basalt and limestone wedges found in the excavation suggest their use in the extraction process. A description of the stratigraphy of a 10 m long trench cut into Pile 2 is presented as well. The new finds enable a better reconstruction of the ubiquitous and continuous flint E&R activities in the ‘Flint Depot’ of prehistoric northeastern Israel.

Introduction

The Eastern Upper Galilee and the northern Jordan Valley host many important prehistoric sites such as the Lower Paleolithic Gesher Benot Ya'aqov (Goren-Inbar et al., 2000; 2018), Ma'ayan Barukh/Hamara (Stekelis and Gilead, 1966; Ronen et al., 1980; Rosenberg et al., 2014); Baram (Ohel, 1991) and Yiron (Ohel, 1986) and North of Bridge Acheulian (NBA) (Sharon, 2010); and the Middle Paleolithic sites of Nahal Amud Cave (Suzuki and Takai, 1970; Hovers et al., 1995); Zuttiyeh Cave (Turville-Petre et al., 1927; Gisis and Bar-Yosef, 1973); and Nahal Mahanayeem Outlet (NMO) (Sharon and Oron, 2014; Sharon, 2018). This region is home to Late Epipaleolithic/Natufian sites, such as ‘Eynan (‘Ain Mallaha (Valla et al., 2007; Valla et al., 2017) and Jordan River Durijat (JRD) (Marder et al., 2015). The region was also extensively inhabited during the Neolithic and Chalcolithic periods at sites such as Beisamoun (Lechevallier, 1978; Rosenberg et al., 2006, Rosenberg et al., 2008, Rosenberg et al., 2010; Bocquentin et al., 2011, Bocquentin et al., 2014; Khalaily et al., 2009, Khalaily et al., 2015;) Hagoshrim (Getzov, 2008; Rosenberg et al., 2008, Rosenberg et al., 2010) and Tel Te’o (Eisenberg et al., 2001). All of these sites (see Fig. 1), and many more, have yielded rich flint assemblages; however, research efforts aimed at locating the exploited flint sources have been lacking or inconclusive, and only pointed to the geological outcrops available in the area without directly associating them with the archaeological sites (Delage, 1997, Delage, 2007a, Delage, 2007b). Recent research focusing on the Middle Paleolithic Amud Cave provided important geochemical data on Eocene flint from a ~ 6 km radius around the cave (Ekshtain et al., 2016).

Terms used in this paper vis-à-vis quarrying sites and features and lithics are briefly defined at the beginning of the Methodology section. Nahal Dishon Complex covers an area of 25 sq km, while Mt. Achbara and Sede Ilan Complexes cover smaller areas (3 and 1 sq. km respectively). Designated activity areas within the complexes, which usually hold dozens to a few hundred tailing piles, are termed E&R localities. To date, E&R complexes have been discovered in six locations in northern Israel: Sede Ilan in the Eastern Lower Galilee (Barkai et al., 2006; Barkai and Gopher, 2009); Giv'at Rabbi in the Lower Galilee (Ekshtain et al., 2012; Yaroshevich et al., 2017); the Nahal Dishon extensive complex in the Upper Eastern Galilee (Finkel et al., 2016 and references therein); Mt. Achbara Complex (Finkel et al., 2018) (see Fig. 1); Daliyat el-Carmel 3/Site 164 on Mt. Carmel (Barkai et al., 2006; Olami, 1984: 147; Rosenberg et al., 2009); and the Kela locality on the northern Golan Heights, located on an Eocene Bar Kokhba Formation ‘island’ within the basaltic area (Finkel et al., 2020). Although northern Israel contains other flint-bearing geologic formations, including the Cretaceous Deir Hanna, Sakhnin, Yagur and Yanuh Formations (on the suitability of their flint for knapping, see Delage, 2007b), five of the six E&R complexes mentioned above are located on the Eocene Timrat Formation, which Delage (2007b) specified as one of the best flint types for knapping in northern Israel. Following Delage (2007b), as well as Ekshtain et al. (2016), and considering the location of the E&R complexes mentioned, it seems that Eocene flint was preferred by local prehistoric knappers, especially when large items (see Methodology for a definition) were the target (Finkel, 2020).

All the above-mentioned E&R complexes are characterized by tailing piles that were created as a result of long-term flint extraction and reduction, as part of the management of the extraction landscape (Gopher and Barkai, 2014). How were these complexes actually formed? Based on surveys of E&R localities in recent years (Finkel et al., 2016, Finkel et al., 2017, Finkel et al., 2018), and combined with the data from excavations conducted in the tailing piles at Mt. Pua and Sede Ilan (Barkai et al., 2002, Barkai et al., 2006; Barkai and Gopher, 2009; Gopher and Barkai, 2006, Gopher and Barkai, 2014, and see below), the following chain of operation is suggested:

• Locating specific desired flint extraction sources.

• Extracting flint from the limestone bedrock and karrens, sometimes employing limestone or basalt tools.

• Creating stone waste piles (backfill piles) from the large amounts of broken limestone – during, or immediately after, flint extraction.

• Aligning backfill piles on top of the exhausted parts of the extraction fronts (and between them), leaving unexploited flint fronts free for future use.

• Flint knapping – core shaping for later use, or for immediate use; blank production and/or tool making conducted on top of backfill/tailing piles (see discussion in Gopher and Barkai, 2011).

The thousands of tailing piles found at these sites vary in size from small (1–2 m in diameter and up to 1 m in height) to large (tens of meters in diameter and 3–5 m in height). The piles are concentrated within relatively restricted areas, which create a highly visible extraction landscape that stands out as a ‘place’ rich in one of humanity's very basic resources – a hallmark on the landscape. One of the most common features of these piles is the flint-knapping waste, with knapped flint artefacts found on the surface of the piles, and within them. Basalt wedges, which were probably used to enlarge natural joints in the limestone bedrock and karrens, are also found on and within the tailing piles, sometimes in considerable numbers (e.g., at Sede Ilan, Lower Galilee: Barkai et al., 2006) and sometimes in a limited number, as in the case of Achbara (Finkel et al., 2018). Of special interest are two caches of flint artefacts found at the base of an elongated tailing pile that have been reported from Mt. Pua (Barkai and Gopher, 2011, see below).

The extensive exploitation of the Timrat Eocene flint was mainly assigned to the Lower and Middle Paleolithic periods, based on: 1) the finds of the intensive Paleolithic E&R activities in the Dishon Complex (see Fig. 2) (Finkel et al., 2016, Finkel et al., 2017; Finkel and Gopher, 2018); 2) the geochemical similarities between the Dishon Complex flint and the flint found in Lower Paleolithic sites located both within the E&R complex and 20 km to the north of it (Finkel et al., 2019); and 3) the location of the Achbara E&R Complex (Finkel et al., 2018) within the Middle Paleolithic Amud Cave daily exploitation territory (Ekshtain et al., 2016), and the fact that the flint used by the cave's inhabitants is of the same Eocene origin. Based on thorough surveys of the eastern Galilee and the Hula Valley, and to the best of our knowledge, no later (post Middle Paleolithic) E&R complexes were found in the region. A small E&R locality in the Golan Heights within an Eocene ‘island’ was described recently—see Finkel et al., 2020). It was therefore suggested that the Eocene ‘strip’ was the main raw material source for the production of large, high-volume artefacts during the Lower and Middle Paleolithic of the Eastern Galilee and Hula Valley, and perhaps also for part of southern Lebanon (Finkel, 2020). These very same localities of Lower and Middle Paleolithic E&R were occasionally used in Neolithic and Chalcolithic times (see for example Finkel et al., 2017), but this is the exception rather than the rule.

Excavations in E&R complexes in northern Israel were rare and limited in scope (see below). This paper reports an excavation in one E&R tailing pile and a trench (cross-section) in another one in the Nahal Dishon Complex. A 70–120 cm deep stratigraphy is described and discussed, raising new questions vis-à-vis tailing piles stratigraphy. Preliminary results of a sorting of the flint assemblages recovered are presented as well.

Previous excavations were conducted in tailing piles at Mt. Pua (Dishon Complex – Fig. 2, Barkai et al., 2002, Barkai et al., 2006) and Sede Ilan (Barkai et al., 2006). The excavation at Mt. Pua included a 2 × 2 m square (G24) at the top of a 30 × 12 m size pile, numbered PW3, and a full excavation of a small pile 5.3 × 4.2 m in size, numbered PW100. The excavation of PW3 – G24 yielded 1146 knapped flint items, including retouched flakes and blades, scrapers, bifacial roughouts and a handaxe. It reached the top of the limestone extraction bed (‘bedrock’) at a depth of ~90 cm. The stratigraphy of this small-scale excavation consisted of a stony complex of limestone waste blocks with almost no soil/sediments, and no clear division of layers was observed. The average number of flint items per 1 m³ was 286.5. The two caches mentioned above were found on bedrock and consisted of 13 flint artefacts each, including rejected bifacial roughouts and Levallois cores (Barkai et al., 2006; Barkai and Gopher, 2011). A full excavation of the ~20 sq. m tailing pile of PW100 yielded 8825 items (~441 per 1 m³) and reached a depth of ~100 cm. The stratigraphy included the stony pile and bedrock below, with very little sediment. At Sede Ilan, a 2 × 2 m square test excavation was conducted in the margins of tailing Pile SE3 (measuring 15 m × 8.6 m and as high as 1.3 m). It was excavated to a depth of 45 cm and yielded 480 flint items (Barkai et al., 2006) assigned to Lower and Middle Paleolithic times. It is important to note that none of the three excavated tailing piles in the two E&R complexes mentioned above presented a clear stratigraphy. These excavations provide a background to the work reported here, which attempts to better understand the formation processes of a tailing pile, the stratigraphy and the E&R process.

Nahal Dishon (the ‘Dishon Stream’) is located on the western mountain flank of the Eastern Upper Galilee. The 96 sq. km drainage basin runs 32 km east from Mt. Meron to the Hula Valley. The area is an erosive surface divided into mountainous plateaus at altitudes of 650–750 masl and peaks of up to 830 masl – the Baram, Yiron, Dalton and Alma plateaus (Fig. 2). The surface is dissected by the Dishon Stream and its tributaries that run between the plateaus at elevations of 500–400 masl in deeply incised V-shaped gorges with occasional cliffs. Geologically, the study area consists of Eocene limestone and chalk. The Lower Eocene Timrat Formation is 400 m thick and is characterized by limestone and chalk karrens which contain large numbers of flint nodules (Delage, 2007b; Levitte and Sneh, 2013). The Dishon area has undergone erosion, and by the Late Pleistocene reached a stage that differs only slightly from that of the current topography (Yair, 1962: 195). For the purposes of our research (following Brosh and Ohel, 1981: 25; Ohel, 1991: 161), we assume that at the end of the Lower Paleolithic, and during the Middle Paleolithic, the topography of the plateaus was similar to that observed today, although the depth of the incision of the Dishon Valley and its tributaries must have been shallower.

The Mt. Pua E&R locality was discovered somewhat before the turn of the current millennium (Fig. 2; Barkai et al., 2002; Barkai and Gopher, 2011), and recently another eight E&R localities were identified within the Nahal Dishon E&R Complex, all within the flint-bearing Timrat Formation's outcrops (Finkel et al., 2016; Localities 1–8). We selected Locality 1 – the Kakal Spur (Finkel et al., 2016: 227 and Fig. 5, Fig. 6, Fig. 7, Fig. 8) for excavation, due to the abundance of flint items found on the tailing piles of this locality, and due the fact that the preliminary survey located an area with smaller than the usual knapping debitage (see below, Area C). To make sure that the piles selected for excavation were not damaged/disturbed by modern agricultural (or other industrial) activities, we compared recent aerial photographs (2016) with a British series from 1946 (Fig. 3). The excavated piles were found undamaged.

Except for the river valleys/gullies, the landscape experienced only minor erosion and (topographic) changes during the second half of the Pleistocene. In other words, the rocky terrain of the mountainous region of the Dishon has (apart from near-surface formation and/or erosion of soils) for the most part remained unchanged. In addition, post-depositional processes have been minimal due to the plateau nature of the higher (or moderately sloping) parts of the topography, which characterize many of the E&R complexes (Mt. Pua included). The E&R complexes and the thousands of tailing piles thus constitute an almost pristine landscape and the modern disturbances are minor since these rocky areas were not farmed until the second half of the 20th century. The major reason for this is that the rugged karstic landscape combined with grassland vegetation of the area, known by botanists as the ‘Arid Galilee’ (rather than with the common oak forests of the Galilee as a whole), due to the local soil's limited water carrying capacity (Rabinovitch-Vin, 1986). These areas were thus used throughout history mainly for pastoral agriculture and not for plant cultivation, which would have heavily damaged the E&R complexes. Farming in modern times was only possible after deep rooting/bulldozing of the rocky terrain and in many cases caused fatal damage to the E&R complexes as can be seen from early and mid-20th century aerial photographs compared to the present-day situation.