Mersa/Wadi Gawasis, Mission 2007-2008 - Kathryn A. Bard and Rodolfo Fattovich et alii

Stone assemblage - G. Lucarini

General remarks
The entire stone assemblage collected at Mersa/Wadi Gawasis during the 2007-08 field season includes 329 lithic artifacts and 8 large stone tools (a term including saddle querns, upper grinders, hammerstones, palettes, anvils and other large stones of undetermined function) coming from the excavation units WG 32, WG 33, WG 46, WG 47, WG 49, WG 50, WG 51, WG 52, WG 53, WG 54, WG 55, WG 56, WG 57 and WG Niche 15. Only one piece was collected on the surface area of the terrace's slope. Considering the very low number of implements brought to light in almost all the trenches (from 1 to 17 items), the statistical analysis carried out on the categories of artificts and on each class of tools was elaborated only on the data coming from WG 55 excavation unit. This 6 m x 6 m trench, located at the base of the southwestern slope of the coral terrace, at the entrance to Cave 7, yielded the highest number of artifacts (268).
The lithic assemblage of the site was divided into the major classes of cores, debitage and retouched tools. Each category was then classified using a Microsoft Access database, elaborated according to the methods of analysis of the Italian Archaeological Mission in the Farafra Oasis, for Neolithic implements, and by D. Holmes (1989), for the Predynastic lithic industries of Upper Egypt.
The 329 lithic artifacts found in the trenches excavated during the 2007-08 season include cores (10 items, representing 3.1 % of the entire lithic techno-complex), debitage elements (314: 95.4 %) and retouched tools (5: 1.5 %).


Main techno-typological traits of the lithic assemblage
Raw material

With 319 items representing the 94.4 % of the total amount of artifacts recorded during the 2007-08 field season, chert played the main role in manufacturing the lithic artifacts.
Chert occurs in small and medium size pebbles and nodules ca. 10-20 cm in diameter and showing a spherical or oval shape and a thick cortex. These pebbles are available, in great abundance, along the bed of Wadi Gawasis.
The chert shows a fine homogeneous structure which makes it a particularly good workable kind of raw material. According to the Munsell Soil Color Charts (1990 Edition Revised), the most exploited types of chert go from a brown/dark brown quality to a greyish brown one.
Almost all the artifacts are well preserved. Only a few items show a strong white or very pale brown patina, which was already noticed in almost all the artifacts coming from the excavation units located on the eastern slope of the terrace. Moreover, it has to be stressed that 131 artifacts (38.8 % of the total) show traces of burning. Even if among these, the highest number is represented by chips and chunks, the presence of cores showing traces of fire could indicate intentional heat treatment. The thermal alteration of siliceous materials improves the flaking qualities of chert and could have facilitated the detachments of longer flakes and blades in pressure flaking.
The other materials used in stone tool manufacturing and detected during this field season are quartz (8 items: 2.4%), quartzitic sandstone (4 items: 1.2 %), and obsidian, this last one used for producing only one not retouched flake (0.2 %).
The few quartz fragments and chunks found during the 2006-07 field season led us to think that quartz was not intentionally knapped. On the contrary, the quartz core found in WG 55 (SU2) in 2007-08 is clear evidence of an intentional, even if very restricted, use of this raw material. Moreover, petrographic analysis of the obsidian flake, not an indigenous stone, will provide very important information about the quarrying area of this material, located along the trade routes of the southern Red Sea.
The last raw material, green siltite, is present in a great quantity all over the area, occurring in medium- and large-size pebbles or in flat tablular form. It was used in order to manufacture large stone tools found in the site such as palettes or anvils.


Cores
Cores are represented by 10 items, all coming from WG 55, representing the 3.7% of the lithic assemblage of this unit. Almost all the items come from SU2 and SU3. The single platform type is the most frequent (4 items), followed by the 90° type (3 cores) and multiple platform cores (2 items). Finally, an initially struck element (rough precore) was also found in SU3 of the unit. All the cores are of small size and irregular shape. The platform surfaces do not show evident traces of preparation. The negatives of removals on flaking surfaces indicate that 8 items were mainly exploited for the production of flakes, while the remaining 2 were for blade manufacturing. A high or medium exploitation of cores seems to have been more frequent.


Debitage
As in the units excavated during the 2006-07 field season (see Fattovich and Bard 2007: 35-37), the debitage category (314 total artifacts) shows the highest percentage (95.4 %) of the entire lithic assemblage collected, with an almost equal ratio (95.5 %) specific to WG 55, where the debitage elements are 256 of the total 268 artifacts collected.
The debitage category is greatly made up of flaking discard products, named as chips (flakes smaller than 15 mm) and chunks (unclassifiable but intentionally detached elements). 190 total debris products were brought to light reaching a percentage of 60.5 % of the debitage category. WG 55, with its 160 chips and chunks, shows a similar ratio (62.5 %).
Flakes are the second class of debitage represented. 101 total specimens have been recorded, with a general frequency of 32.2 %. In WG 55, where 78 flakes have been found, the specific ratio comes down to 30.4 %.
Flakes are mainly from single platform cores (49 total specimens), with a general frequency of 48.5 % of the total flakes detected. 37 single platform flakes have been found only in WG 55 (47,5 %). Primary flakes are the second well represented class. 23 total artifacts have been found, 21 of which are in WG 55. The general ratio is 22.8 % of the total of the flakes detected and in WG 55 is 26.9 %. Multiple platform flakes are also quite well represented (19 total elements: 18.8 %). 90° platform flakes (4 specimens: 3.9 %) and opposed platform flakes (3: 3 %) are much less frequent. It was not possible to determine the type of the remaining 3 flakes.
The blade index is quite low, considering both the general situation (18 blades representing 5.7 % of the debitage category) and in WG 55 (14: 5.5 %). Almost all specimens are from single platform cores (10 total pieces: 55.6 % of the blades). 5 primary blades (27.8 %) and 3 opposed platform blades (16.6 %) follow. Among the blades, the presence of a very stardardized, elongated, single or opposed platform type found in WG 32 and WG 54 has to be stressed. These blades, already found during previous seasons in excavation units WG 16 and WG 27 (Fattovich and Bard 2007: 36), seem to have been produced using a pressure flaking technique. Similar blades were also found at Old Kingdom sites on Elephantine Island (Hikade 2002) and Dakhla Oasis (Kobusiewicz 2007). In both contexts these artifacts have been considered as semi-finished or complete products for lamellar sickle inserts.
Core rejuvenation practice is not well represented. During the field season only 5 core trimming elements have been found (1.6 % of the debitage category). They are all core sides and 4 of them come from WG 55.


Retouched tools
The 5 retouched tools brought to light represents only 1.5 % of the total lithic assemblage collected at Mersa/Wadi Gawasis in 2007-08. They include 2 denticulates, 1 notch, 1 perforator and 1 composite tool sidescraper-notch.
The two denticulates come from WG 55. The first one, coming from WG 55, SU3, is manufactured on a large primary flake showing an obverse, dexter, invasive, semi-abrupt retouch.
The second one, found in WG 55, SU8, is a bit smaller in size and is manufactured on a naturally fractured pebble. The tool shows four detachment points caused by strokes all blown on the same face. Such strokes are the cause of the denticulate pattern on just one end of the tool. The detachments are slightly invasive and semi abrupt.
The notch and the perforator come from WG 33, the first from the general surface of the excavation unit, while the second one is from SU4. The notch is manufactured from a 90° platform, tertiary flake, fractured on the proximal end. It shows an obverse, sinister, single blow, slightly invasive and semi-abrupt notch. The perforator is manufactured on an opposed platform, tertiary blade, fractured on the proximal end. The distal end of the blade, which is pointed, has two obverse, dexter, invasive, semi abrupt, lamellar parallel detachments, which could be interpreted as the result of a first stage of preparation of a perforator.
The last multifunctional, sidescraper-notch-tool comes from the SU1-SU2 interface of WG 57. The tool is manufactured on a multiple platform, tertiary flake. It shows an obverse, continuous all over distal end, non invasive, semi abrupt retouch. On the dexter side, an obverse, single blow notch is present as well.


Large stone tools
The 8 large stone tools discovered this field season include: 1 palette coming from WG 46; 1 hammerstone from WG 50; 2 upper grinders and 1 large stone anvil from WG 51; 1 palette, 1 small lower grinding stone and 1 not determined tool, coming from WG 52. All these excavation units, where the large stone tools were found, face the southern slope of the fossil coral terrace. The palettes, lower grinding stone, anvil and the undetermined tool are manufactured in green siltite. Natural pebbles and tabular elements of this material, often characterized by a quite regular shape, were available in great abundance on the top of the terrace and could be easily knapped and used to manufacture large tools.
The lower grinding element, circular in shape and small in size, could have been used for pigments.
The large stone anvil, parallelepipedal in shape, shows all the six faces naturally smoothed and the edges roughly chipped to regularize its shape. Its four main faces are marked by speckles resulting from beating by another stone.
The hammerstone, manufactured in chert, is characterized by a sub-spherical shape and medium size. Beside larger detachments present on most of its periphery, it also shows smaller ones probably caused by a first stage of another artifact's polishing action.
The two upper grinders, both manufactured in sandstone, are of medium size and could have been used for cereal processing. The presence of such an activity is also attested by the saddle querns found in previous seasons (Fattovich and Bard 2007: 35) and by the abundant Triticum dicoccum and Hordeum vulgare remains found in the site (see Fattovich and Bard 2007: 39).


Conclusions
Analysis of the 2007-08 lithic assemblage, almost entirely coming from WG 55, showed a very high percentage of debitage elements in which waste products resulting from knapping activities are the most represented class. This datum, together with the finding of ten cores (the only ones brought to light during this season), often highly exploited, and the presence of several, even if not numerous, primary flakes and blades, indicate that the WG 55 area could have been devoted to the manufacturing of stone artifacts. The finding of four core trimming sides, evidence of a prolonged exploitation practice of the cores, seems to support this hypothesis, even if a complete lack of hammerstones has to be stressed as well. As a matter of fact, the only hammersone found during this season comes from WG 50, located on the adjacent slope of the terrace.
The retouched tool kit, very poor in quantity, also does not show a particular standardization of the products and, on the whole, the very scanty retouched elements are not characterized by clear typological traits. Confirming the data collected during the previous field seasons, pieces to be used for scraping (denticulates, notches, sidescrapers) seem to be predominant, representing the majority of the retouched tools.
Considering the features of cores and blanks, as well as the almost complete absence of piece markers of well standardized stages in the chaîne operatoire, the Mersa/Wadi Gawasis chipped stone tradition shows quite primitive and opportunistic technological traits. On the other hand, the rare more standardized cores, debitage elements, as well as some of the very well manufactured and bifacially retouched tools found during previous field seasons, are a clear mark that more evolved technologies were also well known by the groups settled in the area.
The high presence in excavation unit WG 55 of a large quantity of wood debris, mainly located in SU2 and SU3, where the greatest part of the stone implements also come from, can support the hypothesis that these more primitive implements could quickly and roughly be knapped in the area and used, at once, on the wood's timbers to be scraped and cleaned.
The rare well manufactured items, such as the long blades and bifacial tools, could have been devoted to other and more specific activities, such as the exploitation of plants, which were carried out in the "harbor area."
Analysis at the metallographic microscope, which will be conducted on casts of the artifacts produced on site with a hydrophilic vinypolisiloxane impression resin, will provide more detailed information about the exact function of these implements.

Please find the statistics in the pdf attached file