K\HAA Konferensef 34:l0l-120. Stockholm 1995 The Archaeological and Ethnoarchaeological Interpretation of a Ceramological Enigma: Pottery Production in Sinai (Egypt) During the New Kingdom Period By Yuval Goren*, Eliezer D. Oren** and Rachel Feinstein** * lntcrdisciplinary Research Division Isracl Andquities , Jerusalem 910(N,Isracl Authoritt P.O. Box 586, **Archaeological Division, Ben-Gurion University ofthe Negev, Be€r-Sheva 84105, Israel Introduction Over the last decade, considerable efforts have been invested in the attempt to understand the mechanisms controlling the development of pottery production. The interest in this issue stems from the recognition of pottery as a key artefact for the archaeological interpretation. Fundamental studies, based predominantly on ethnographic observations of present-day potters in traditional societies, were published (amongst many others) by Shepard (1954), Nicklin (1971:.1979), Amold (1985) and Rice (1981; 1984; 1987). All these studies indicate that the location of specialised pottery manufacture is dictated mainly by the proximity to resources, their workability and qualify and the location of the markets. The first and the most crucial aspe€t to be examined in any ceramological study is the availability of resources. In order to produce pottery, the minimal n€cessary raw materials are clay, water and fuel for firing (Shepard 1954, p. 18). Amold divides the resources feedback mechanism into two parts: their quality for ceramic production and their availability, as measured in the distance to their source locations (tunold 1985, p. 20). As he points out (ibid): "Readily available ceramic resources, along with those of suitable quality for making pots, provide deviation amplifying feedback for the origin of the craft and its development into a full-time speciality. Conversely, resources of poor quality or those too far from the potter's household provide regulatory feedback for ceramic production and prevent the craft from developing". Therefore, the question of the availability of resources seems to be crucial in any ceramological study. Additionally, the potential of the markets in terms of demand and the potential quantity of consumers is also crucial for the development of ponery mass-production. In the present paper, we will investigate the issue of pottery production in 102 Y. Goren et al, /\ "r. lrf, t6, ,$\ w a >. 4i1:" -l*tlg..*a '--&5 PFT .b.-! .rq-l .#.4 ffi" }^ffi :ffi"' # Wr pilt V .4 arx^r gjivaY Fig. 1. Map of the Nonh Sinai survey. North Sinai during the New Kingdom Period (ca. 1550-1150 BC.). In this period, evidence for considerable cerarnic production was found in an area, which categorically lacks the components required for specialised pottery manufacture. In order to explain this phenomenon we will review the relevant environmental, historical, archaeological, and analytical data, These wiu be used in order to explain the occurrence of this craft in North Sinai during this specific period and to rcsolve its conflict with the ethnoarchaeological record. The research background The Sinai peninsula is a large desen area extending between the Israeli Negev on the east, and the Egyptian Nile valley to the west. This key location, and especially its northem coast, functioned as the most important overland route bridging Africa and Eurasia, particularly Egypt and the Irvant, from the PreDynastic period onward. Over the ages numerous convoys of traders and army expeditions have travelled along the northern Sinai coast on their way to conquer either Egypt or Canaan. Consequently, it has long attracted scholarly interest in its archaeology and history. Over a period of nearly 15 years (1967-1982) extensive systematic archaeological surveys werc carried out in North Sinai @ig. 1), sometimes accompanied by soundings and several large-scale excavations (Oren 1987; 1992). These yielded an enormous body of data which has supplied significant information about the settlement pattems of Sinaitic sites and the nature of the interrelations between Egypt and the Levant throughout the ages. In 1992, ateaty conceming the Sinai antiquities was signed between Israel and Egypt. This agreement obliged Israeli archaeologists to complete the study of the Sinai finds within a period of two years. The project included a comprehensive petrographic study, conducted by one of us (Y.G.), of the pottery assemblages from numerous sites of all periods. The results of this study re- The Archaeological and Ethnoarchaeological Interpretation 103 Fig. 2. Map of the New Kingdom sites and clusters in North Sinai. vealed evidence for local mass production of pottery in North Sinai only dur- ing the New Kingdom period, only emphasising the peculiarity of this phenomenon. Provenance studies of the Protodynastic and Eady Dynastic ceramic assemblages from this region (Porat l989a,b) have yielded no evidence for such a production. Our recent petrographic studies (norv in the process of preparation for publication) have proved that in other periods, most if not all of the pottery had been of exotic origins, usually from Egypt or the Levant. Therefore, a special consideration of the historical and technological background is deemed essential for the understanding of this unique episode. It may also shed some light on the nature of the NK Egyptian administrative systern which controlled the production centres in Sinai. The historical background: a brief overview One of the most dramatic chapters in the history of Egypt and Canaan began with the fall ofAvaris, the Hyksos Capital in the Delta, and the conquest of the Hyksos strongholds in southem Canaan. Subsequent military campaigns into Canaan by the kings of the early XVF Dynasty, particularly Tutmose m (1504-1450 B.C.E.), paved the way for rhe establishmenr of the Egyptian empire of the New Kingdom, that held sway for almost three hundred years over the entire area between the fourth cataract of the Nile and the Euphrates River. Numerous contemporary documents, coupled with the rich archaeolog- 104 Y. Goren et al, ical record frorn Egypt and Canaan, provide a detailed picture of Egypt's foreign policy and of the close commercial and cultural contact between Egypt and its provinces to the east (weinstein l98l, pp.18). The first direct references to North Sinai appear in documents of the New Kingdom, clarifying its importance in the geopolitical system of the Egyptian empire. During the New Egyptian Kingdom the coastal strip of northem Sinai became the major land-bridge, over which the military and commercial traffic between Egypt and Asia flowed. The route along the Meditenanean coast of Sinai, known in the Egyptian sources as "The Ways of Horus", was the vital and efficient artery through which Egypt communicated with her Asian province. Judging from the Egyptian documents and the archaeological discoveries, it is evident that Egypt's interests in this inhospitable land were secured by a network of fons and supply and customs stations that were established along it between the eastem Delta and southwestem Canaan. Indeed, Thutmose Itr's fint campaign from the border fonress of Sile to Gaza, about 250 km away, in the record time of a mere nine to ten days, testifies to the effectiveness of Egypt's organisation in the "The Ways of Horus" (Wilson 1955, p. 235). The rebuilding of the military occupation in North Sinai, on a scale far surpassing what had existed previously, is probably to be assigned to Seti I who conducted punitive expeditions against the nomadic tribes in Sinai and was responsible for renewing Egypt's control in Canaan, in an area extending from Sile on the eastem frontier, to Upi in the Damascus area. The two most important documents on this subject are Papyrus Anastasi I from the reign of Ramesses II, and the reliefs on the north wall of the great Hypostyle Hall in the temple of Amun at Kamak from the time of Seti I (Gardiner 1920, pp. 99). The papyrus, a kind of "geographic guide" written in a satiric language, lists the stations in North Sinai as well as the major fortified cities in southem Canaan (Mlson 1955, pp.475). The Kamak relief's, on the other hand, record graphically the first campaign of Seti I into Sinai and beyond, and his victorious retum to Egypt (Fig. 3). The depiction in the Kamak relief's of the stations and forts along the "The Ways of Horus", most likely commemorated the renewal of Egypt's military and administrative presence in North Sinai at the beginning of the XIX& Dynasty. The subsequent reigns of Memeptah (1223-1213 B.C.E ) and of hrs successors Seti II and Twosret (l 199-1185 B.C.E.), mark a period of weak control over Sinai ard Canaan (Faulkner 1975, pp. 217; Kitchen 1982, pp. 215). Nevertheless, the latter rulers, in particular Seti II, are represented by a number of monuments in Egypt proper, near the turquoise and copper mines in Sinai and tbe Arabah and at sites in Canaan Eothenberg 1972, p. 163; Fig. 49:6; Faulkner 197 5, p.237). Egyptian administration in Canaan was restored briefly by Ramesses III, but had soon declined entirely in the intemational arena, and by the time of Ramesses V or M had retreated to its old border in the Delta The Archaeological and. Etlmoarchaeological Interpretation lO5 Fig. J. Stations and forts along the "Ways of Horus". Relief on the nonh wall of the great Hypostyle Hall in the temple of Amun at Karnak. In conclusion, the historical sources conceming the "TheWays of Horus" in North Sinai indicate an extremely well organised system of forts and road stations, established by the Pharaohs ofthe New Kingdom to secure the major artery of communication with the Asiatic provinces. The archaeologic al data From 1972 to 1982 the Ben Gurion University expedition, under the direction of one of us @.D.O), explored more than eighty New Kingdom settlement sites in the coastal and sand-dune area between the Suez Canal and Gaza (Fig. 1-2). The survey results enable us for the first time to delineate the course of the "The Ways of Horus" in accurate detail and to reconstruct the history of settlement and the degree of Egyptian activity on that land bridge between the Delta and southem Canaan. The largest concentration of New Kingdom sites is 106 Y. Gorm et aL reported in the southwestern comer of the survey area, in the triangle formed by Port Said, Rumani and Qantan. These sites are represented by industrial and domestic installations. Some of the larger sottlements yielded building remains in stone and brick, including sections of granite columns that probably belonged to some public structures (Oren 1987, 1992). The re-discovered ancient canal flowing tluough this region (Sneh 1977), made it part of the fertile and densely populated eastem Delta. New Kingdom sites were recorded east of the Delta plain and south of the Bardawil Lagoon, in an area roughly parallel to the modem road and railway line between Raphia and Qantara. The distribution map of settlement in the New Kingdom is characterised by clusters of sites in which the central fort or station is surrounded by many smaller camp sites for caravans and by many seasonal encampments for the local inhabitants who depended for their protection and supplies on the Egyptian authorities in the main settlement. At least ten different clusters like these have been recorded between the Suez Canal and Raphia. Numerous New Kingdom sites were also encountered between Raphia and Gaza. To this we may add the late Bronze Age settlement remains at Tell Abu Salima Tell Riddan and Deir el-Balah (Dothan 1972, 1987). The North Sinai Expedition explored most intensively the site clusters of Bir el-'Abd and the Haruba area. The cluster at Bir el-'Abd consisted of a central site, including impressive granaries and other public architecture (Oren 1987, pp. 78), surrounded in a radius of 3 to 4 km by a dense grouping of some thirty small encampments. The remains at these settlements were extremely poor, including only fragments of hearths, refuse pits, stone grinders and ponders, large quantities of pottery and a few metal objects. Indications for the presence of a local pottery production centre were detected, although as indirect evidence, at the central site. There the dominance of a ceramic fabric, unfamiliar elsewhere, indicates the existence of a pottery production centre which produced pottery of a pure Egyptian typology, using slightly different techniques from those applied by the Egyptian potten (see below). In the Haruba area, the explorations were focused on two particularly impressive sites: 4.-289, a military fort, andA-345, an administrative centre. Both were unusually well-preserved and stand as models of Egyptian civil and military architecture on the "The Ways of Horus". The fort of Haruba extends over some 2500 sq. m. (Oren 1987, Fig. 6). Its general plan comprises a massive enclosure wall, a gatehouse, a wide courtyard, and a complex of compartments, including mag^zines that were built against the outer wall. The whole structure must have risen at least 6 m. Both the size of the building and the building techniques employed, bear witness to mastery over engineering problems and to the excellent local administration that, besides providing the necessary tools and materials, supervised the workers. The Arc hae olo gical and. Ethnoarchaeolo gical I nterpretation lO7 -+-r '4*7/ Fig. 4- Ge\enl view of site A-345 (ceoter of the picnue), looking west. Notice the recent sand dunes covering the northem and southem parts ofthe site. Site A-345 is located right in the middle of active coastal sand dunes (Fig. 4). The original boundaries ofthe settlement are unknown, but judging from its building remains and pottery deposits, we may suspect that it once covered an area of more than 6 acres (Oren 1980, pp. 32). In the course of three seasons of excavations (1980-1982) the expedition explored and identified three building units: a complex of magazines in the centre of the site, a "casemate-walled" area in the north-west, and an industrial quarter on the east (Oren 1987, Fig. 8). The architectural and planning features are purely Egyptian: the ground plan, building and bonding techniques, and the standard brick size. One of the most important discoveries at site ,4.-345 was the industrial quarter, which included a potter's workshop. The excavation made it possible to follow in detail all the methods and procedures employed in this workshop, including the preparation of the clay, the shaping of the vessels, and their firing in the kilns. The northem part of the industrial quarter was occupied by a complex of installations that served in all probability for storing the clay. On the floors of these installations were registered cloaks of clean, light-brown clay, closely resembling the kind used by local brick manufacturers. Adjacent to the storage installations, a complete pottery kiln was uncovered (Fig. 5). It was preserved to a height of 1.50 m with an outer circumfer- lO8 Y. Goren et al. Fi& 5. The northem potter's kiln at site 4-345 in the course of the excavation. ence of 1.80 m. Brick steps on the west side of the kiln gave the potter access to the upper chamber where the pots were fired. The lower or fuel chamber, was about 1 m high. It was dug into the sandy soil beneath the surface, a feature probably intended to aid in maintaining the heat within the chamber. That fired brick floor, or grating, was pierced by a network of srnall holes, each about 0.10 m wide, designed to allow hot air into the firing chamber above. The entire kiln area was covered with refuse from the potter's workshop: wasters. fired bricks. charcoal and ash. One room to the west of the kiln contained large quantities of industrial waste, as well as many fragments of pottery stands with a tall, trumpet-shap€d foot, including some unfued specimens. Another room nearby contained a group of especially large flower pots. The literally tons of pottery recovered in the industrial area are important for indexing typology in ceramic assemblages at Haruba. Moreover, this complex forms an almost unique example of the full contents of a NK ceramic workshop, including the raw materials, the kiln, unfired vessels and ceramic wasters. So far the remains of kilns in Egypt date mainly to the Old and the Middle Kingdoms (Holthoer 1917, pp. 16) and the recently found or reconsidered workshops from el-'Amama are more fragmentary and incomplete (Amold and Bourriau 1993, pp. I l2). The Archaeological and Ethnoarchaeological Interpretation lO9 A second kiln, somewhat smaller, was unearthed in the southem wing of the site. It too is a round brick structure, preserved to the height of approximately 1 m with plastered walls about I m in diameter. The fuel-chamber is similar to that in the large kiln, though its opening widens, like a funnel, towards the interior of the structure. The floor of the firing chamber was not preserved, but it appears to have been supported by brick "brackets", judging from the brick remains jutting out from the inside of the wall. Remains of sections from the kiln wall indicate that the upper part was either conical or dome-shaped. Technological examinations of the NK pottery assemblages from North Sinai The full results of the analytical examinations of the NK ceramic assemblages from North Sinai are far beyond the scope of this paper. These will be published in tum in a forthcoming volume, together with the detailed descriptions of the petrographic groups and the results of the multi-element chemical analyses. The following data is intended for the sake of discussion on the general issue of ceramic production in North Sinai. At the present stage of the research the main analytical method that was applied was petrography. Petrography is most valid as a primary analytical method, since it enables the examination of large amounts of samples at low costs and provides valuable information in a relatively short time. It is also one of the only methods which indicates at the same time the estimated firing temperature, temper selection and sorting, the possible provenance and a mineralogical deflnition of the fabrics used. For these reasons it has been chosen as the basic and chiefly applied analytical method in this study. In addition, some preliminary chemical examinations were used for selected fabrics using ICPAES analyses. The petrographic data base gained during this research will be used, in time, as a basis for more advanced multlelement chemical examinations that will be directed towards specific problems. Therefore, the data pre- sented here is at its preliminary stage. Petrographically, Egyptian pottery can be readily identified by its typical pastes and discriminated from Canaanite wares (Amiran and Glass 1979; Goldberg et aL..1986: Porat 1989b; Bourriau 1990: Goren 1991). The results of the detailed technological research of Egyptian pottery assemblages, as well as the increasing data on Egyptian ceramic raw materials (cf. Tobia and Syre 1974; Allen et al..19821,1989), enable excellent correlation between Sinaitic and Egyptian materials. As for the Canaanite pottery a rapidly growing petrographic record related to pottery assemblages of Israel (e.g. Glass er aL . I 993 ; Goren 19911 1992, Goren and Gilead 1989; Porat 1987; 1989a,b) enables a reasonable identification of the origin of the Canaanite vessels. l1O Y. Goren et el. The petrographic study included about 400 thin-sections. These were exam- ined under a petrographic polarising microscope. During the petrographic analysis, the samples were divided into petrographic groups. A "petrographic group" encompasses vessels which share similar petrographic affinities in both clay and temper. This classification is determined according to the qualities of raw materials alone, regardless of variables such as typology, chronology and geographic location of the site. The application of this method for archaeological assemblages has been explained elsewhere, together with the key petrG. graphic groups for Canaanite pottery and their correlation with specific lithological environments (Gilead and Goren 1989; Glass et aJ.. 1993; Goren 1991, 1992, 1995; Porat 1987, l989a,b). In addition, two local petrographic groups were defined: Group "Bir el-'Abd marl": This group was named after the assembly of sites in which it dominates the ceramic assemblage. Technically, this petrographic group is very similar to the Egyptian fabric Marl C compact of the Vienna System (Bourriau and Nicholson 1992; Amold and Bourriau 1993, pp. 179; Pl. VI:e). Nevertheless, it is distinguished by two varied factors. The first one is the presence of large (>l mm) fenuginous shale particles appearing in this fabric in significant quantities. The second factor is the relatively high firing temperature of most vessels (usually 900-l0m "C) which caused typical reshaping and even melting of the shale particles, perhaps due to the fluxing properties of their iron contents. The definition between the Egyptian marly clay and the Bir el-'Abd marl was based most ofall on the presence ofthese shales. The proximity between the Egyptian marly clay and the Bir el-'Abd marl has been disclosed also by preliminary chemical analyses, demonstrating that these two groups hardly differ in their mean mineral contents and even in their trace elements. This points to a close relationship between the marly formations used in Egypt for pottery production and the raw material used by the Bir el-'Abd potten. This point is of primary significance, since the BEA area lacks any clay formation or soil suitable for ceramic production. This issue will be discussed below. Group " Haruba marl ": This group was named after its key site and production centre, namely site A-345 near el-Haruba. The group is characterised by the use of a very calcareous, silty (usually >lOVo) marl, rich in iron oxide grains. This marl is consistently accompanied by a coastal sand temper, readily identified by its high proportions of typical sand-sized heavy minerals, which maintained their rounded grain shapes even in the sand fraction. This indicates a coastal sediment which was not subjected to any severe chemical or physical weathering process which usually affects these components inland. The presence of several unfired vessels of this $oup in the potter's workshop of site A345, indicates that its origin is in this site. It is also the most common petrographic group in this site. The Archaeological and Ethnoarchaeological Interpretation 111 The identification of the origin of the Haruba marl is not an easy task. The archaeological data clearly indicates that it is local, yet in purely technical terms the raw materials should have been of a foreign origin since no suitable clay for ceramic production exists in the el Haruba vicinity. This point will be discussed below. Ceramic resources in North Sinai As we stated above, the first aspect to be examined in any ceramological study is the availability of resources, namely the clay, water, and fuel for firing. Tempor is not as important as clay, because many clays include naturally occurring silt and sand (Nile silt is a very good example of this). In addition, elementary temper typs, such as sand, grog, and chaff, are very widespread in any location. Howevel by definition, fuel and water may be scarce or even absent in any desert area, including North Sinai. At least in the two site clusters where evidence for pottery production was detected, that is to say Haruba and Bir el-'Abd, no suitable clay is exposed to a distance of at least 40 km from the sites. The nearest clay exposures to Haruba occur north of Jebel Libni, about 50 km to the south (Eyal et aI..1980). Alluvium with clay contents, which may be suitable for ceramic production, is also found closer at the delta of the Wadi el-'Arish, which is exposed usually after large flood events (Sneh 1987, p. 190). In the case of Bir el-'Abd the clay formations that outcrop south of the sites, in Jebel Mghara and its sunounding, are totally inelevant as a possible source for the clay. The district between them and Bir el-'Abd is impassable, due to the steep sand dunes that cover it and the lack of any water resources. The most rational direction for importing clay is westwards, towards the Nile Delta. But in this case, what benefit was there in imponing clay, instead of importing the pottery produced at the Delta sites? The presence of water is also limited, although the possibility of the use of brackish water cannot be excluded. In fact, the whitish coating which appears on many of the locally made vessels, may indicate the use of such water. The presence of salt in the clay body causes it to concentrate near the !€ssel's surface in the course of dehydration. In the firing process the chlorine reacts with the iron and evaporates, and the lack of iron minerals (the main pottery stainers) causes the formation of a pale surface (Franken 1978). This phenomenon probably did not disturb the Sinaitic potters, b€cause white slips were not uncommon in the genuine Egyptian ceramic repertoire (Bouniau 1981, p.72). The availability of fuel seems to be morc problematic. The amount of fuel required for ponery mass production is enormous, and usually scorcs to a ratio of 3:1 and more, as comparcd to the weight of clay (Rice 19871, pp. 174). In ll2 Y. Goren et aI. many communities fuel is the only expense acquired for pottery production (bi . Of course, in an arid zone such as North Sinai, the problem of fuel increases to become the cardinal obstacle for pottery manufacture. At present, the principal vegetation around Haruba and Bir el-'Abd is mostly non-arboreal and includes mainly Retama raetam and Artemisid monosperma (Znhai 1980, pp. 2l3:Yeizel 1984, p.226). T\e Retama raetam plant is considered by the Beduins an efficient fuel. However, its shabby character and scattered distribution would have made its long-term utilisation for pottery kilns an extremely laborious task, and could easily lead to a rapid over exploitation of the nearby vegetal environment. This, in tum, would have forced the potters to import their fuel from ever-increasing distances. Arboreal species are extremely rare in the ar€a and mainly include palm trces (Phoenix dactylifera). Apart from being inferior as fuel, due to their fibrous nature, date palms had and still have a significant economic value. It is therefore very unlikely that they would have been extensively used as fuel for ceramic production. Other trees are found in the channel of Wadi el-'Arish at an aerial distance of 15 km. from the site, including the tamarlsk (Tamarix nilotica) and the acacia (Acacia raddiana). This distance increases when traversed by foot due to the sand-duned character of this area. The study of the charcoal remains from Haruba (N. Liphshitz, North Sinai survey records) revealed that the kiln debris included pistachio (Pistacia atlantica) and tamarisk (Tamar* sp.). The pistachio is found at present only in the central Negev highlands, from an elevation of 800 m. and above. The minimal distance between the distribution area of this tree and the Haruba region is above 80 km.. The nearest tamarisk trees are located some 15 km. from the site. Their gathering from such a distance would have been an even more laborious and time-consuming task. As the petrographic data indicates, this did not prevent the potters from firing a large proportion of the pottery at high temperatures (above l00O'C). This activity undoubtedly required vast amounts of wood. Thus in any case, the fuel has to be regarded as imported, had it been collected in Sinai or elsewhere. In the case of Bir el-'Abd, the origins of the fuel resources were not determined since the actual location of the kilns was not detected. In any case, similar problems occurred also there due to the similarity in the environmental characteristics of this area with Haruba. Moreover, in this region no wadi channel including Acacia and Tamarix trees is found. It may thus be assumed that the Bir el-'Abd and Haruba potters have faced severe problems of collecting a suffrcient amount of material to feed their kilns. To sum up, in purely economic terms, the availability and quality of resources in North Sinai should negate any attempt of fully specialised pottery production as a profitable employment. The Archaeological and Ethnoarchaeological T rltct Innrpraation S{7 C{9 AEA A3a5 A3,13 A' 113 A-28t 137t12 I Locll tr CyFittcrA. tr Egyptlen Zc.n||nilc Gt Unknown 9Can Crut .1. Warc distribution in northem Sinai NK sites. Distribution of the petrographic groups In light of the archaeological background, the technological study was oriented towards the following problems: 1. The proponions of locally made and imported wares: For the sake of discussion, the two main assemblages examined here (e.g, the Haruba sites and the Bir el-'Abd cluster) were chosen for comparison, together with some smaller assemblages ftom the T-site cluster, sites S-57 and C-69 (Chart 1). The latter were included as representatives of the S and C locations in the survey map (Fig. 2). The two primary site groups, the Haruba sites (A-3451343), and Bir el-'Abd (BEA), represent locations of potlery production, one to the west ofthe survey area @ir el-'Abd) and one to the centre-east (Haruba). Chart 1 demonstrates that the proportions of the three important wares, namely the Egyptian, Canaanite and local, change gradually in space. As for the imported wares, the Egyptian pottery plays a predominant role in most of the assemblages and its relative quantity increases westward. In the T:sites, located on the eastem edge of the Nile Delta, it becomes the principal component of the assemblage. As the quantity of the Egyptian wares in the larger sites decreases from west to east, the proportions of the locally made wares increase reslrctively. This is best shown when the T-site cluster, the BEA sites and sites A-3431345 are compared. It must be stressed that this model relies on the assumption that no pottery was produced of the typical Nile silt or genuine Egyptian marly clay outside the Nile Valley. This point should be emphasised since the T:sites are in fact located in a province where Nile Silt constitutes the local sediment. 114 Y. Goren et al, I Loc.l O Eevpt EC.m.n Ecfprbr. E unlnonl Chart 2. Ptoye\ance of the Haruba ceramic assemblases. 2. Imported vs, locally-made vessel types: A comparison between the two pottery production centres, namely BEA and A-345 in Haruba, reveals that changes occur even when the provenance of specific vessel types is being examined. For the sake of comparison, bowls and storage jars were chosen. The dichotomy of bowls versus jars most probably signifies on a general level, table ware versus storage ware respectively. The provenance of jars, when imported, may indicate the location of administrative centres which supplied liquids and other commodities. Chart 3 demonstrates the orovenance of these Eloc|l OEgypf El6m.n C,hart 3. Provenance ofjars versus bowls in the main NK sites. The Archaeolo gical and Etlnoarc haeoln gical Interpretation 11 5 Fig. 6. A Canaanite jar (left) and an Egyptian Drop-Shaped Vessel from site A.-345. two vessel types in the BEA sites, the Haruba siteA-345 and the slightly later site 4-289. A comparison between the BEA and ,4-345 bowls and jars reveals that in both sites there is a conspicuous difference between the provenance of these two vessel types. While most of the jars (in BEA) and all of them (in A-345) are imported, a high proportion of the bowls is local. Moreover, in both sites a significant component of the jars (abo'tt l2%o-14%o) is Canaanite (Fig. 6), whereas none of the bowls is attributed to this provenance. This clearly indicates that pottery production in both sites was restricted to household functions. If supply was imported to these sites in storagejars, it had been imported mostly from Egypt and less from Canaan. Nevertheless, "table ware" pottery was partly produced at the site, to meet the daily demands for these vessels due to their higher breakage rates (Amold 1985, pp. 152). 3. The influence range of each local workshop: It has been noted that the distribution of locally made pottery is significantly reduced in the smaller sites, where pottery was not produced. This phenomenon is most pronounced when the adjacent assemblages of A-343/345 and A-1371142 are compared (Chart l). This implies an intemal use of the pottery produced in the larger sites by their population only. This phenomenon may indicate a more sedentary use of thes€ sites as comDared to the smaller ones. 116 Y. Goren et aI. 4. Changes through time: These changes arc most visible when the Haruba sites A-3431345 are compared to the adjacent site A-289, since the latter is dated to a slightly later phase within the NK sequence (Dynasties XIX-XX). The proximity of these sites negates any possibility of changes due to geographical factors. In Charts I and 2, the assemblages of sites A-343l5 and A-289 are compared. The quantity of the locally made pottery decreases dramatically in A289 in favour ofthe Canaanite wares. This definitely indicates a wealening of the dircct Egyptian influence and an increasing dominance of the Canaanite centres. This issue is, of course, reflected also in the typology of these assemblages. Moreover, it clearly relates the trend of local pottery production in northem Sinai to the Egyptian, rather than to Canaanite interests. This inclination is well rcflected in Chaft 3, which indicates a significant penetration of Canaanite table ware to northem Sinai, replacing almost entirely the locally made component. Moreover, the proportion of Canaanite jars, as compared to Egyptian ones, alters to a clear dominance of the former If the provenance of the storage jars indeed reflects the location of the supply sources, this phenomenon may indicate an inven in the geo-political influence on the region and the weakening of Egyptian authority. Discussion Information obtained from this study, sheds some light on the question of the essence of Egyptian authority in North Sinai during the New Kingdom period, providing a picture with important historical and archaeological implications, going far beyond the rnere question of the nature of ceramic production. A preliminary conclusion of this study relies on data which demonstrate that Egyptian ceramics were produced in North Sinai to a considerable extent following Egyptian styles and typologies. This is evident from the potter's workshop at site A-345, and from the indirect, but clear evidence for pottery production around Bir el-'Abd. The technological examination of the Haruba and Bir el'Abd wares confirmed their technological distinctness from any Egyp tian wares, yet pointed to their similarity in choice of raw material to the commonly used Egyptian marly clays. These contradictions suggest that, pottery in fact was produced in North Sinai in a very rigid, conservative manner, following the basic Egyptian conventions of clay and temper selection as closely as the circumstances would allow. The question has to be raised as to which efforts had been invested in the project of pottery mass production in Nonh Sinai? Or rather, in purely economic terms, was it more profitable compared with transportation of pottery from the Nile Delta or from Canaan? As previously pointed out, in terms of resources, market demand and competition with the main production centres The Archaeological and Ethnoarchaeological Interpretation ll7 in Egypt, the development of pottery production in North Sinai must have been a totally unprofitable business. Therefore, technological data concerning cerarnic production in the North Sinaitic New Kingdom sites are intriguing and difficult to interpret. In light of these data, it has to be concluded that the mechanism of pottery production in Nonh Sinai was motivated by benefits other than purely economic ones. The historical and archaeological record demonstrates that the establishment of the NK sites of North Sinai was organised by the Egyptian authority as part of the colonisation of Canaan and the greater Levant during this era. Consequently, the organisation of ceramic production centres with highly skilled craftsmen, should have been part of these efforts which indicate purely political motives. The presence of significant pottery production sites in North Sinai, within administrative centres which also included fortresses, storehouses, and granaries, indicates that specialised ceramic production was conceived of as a public craft. Therefore, production centres were controlled by the communal bureaucracy and functioned under the supervision of the authorities. In this sense, the establishment of an administrative system along the overland route required, in the eyes of the Egyptian authorities, the reproduction of centres analogous to those in Egypt. These were replicated, with all their components, within the hostile environment of North Sinai. Such an activity most likely required an enonnous investment in terms of means and personnel, far beyond the direct economic value of the crafts. This may explain the seemingly illogical investment in the task of pottery production in North Sinai in this period, unknown in any other epoch in the history of this region. Whether the intention of the powers that be was to replicate the daily life of Egypt at these sites, or to create employment for their inhabitants or both, it is clear that this industry was based on extemal resources and support. It should be remarked that in s€veral cases communities that live in environments of poor agricultural potential are pushed into pottery production as an inferior, yet supporting occupation (Amold 1985, pp. 176). However, this was not the case here, since natural resources would not allow for specialised pottery production without a well established organisation practically subsidising it. The rnain conclusion of this study goes beyond the immediate question of pottery production in North Sinai, to the validity of ethnological data for the explanation of archaeological phenomena. Ethno-archaeological observations, needfully based on observations of present day pottery producing communities, tend to regard only social and economic factors and often ignore considerations that may have existed in past realities. The production and use of earthenware for domestic daily use, as well as for public purposes (such as cult), is practised nowadays mostly by rural communities. These societies, commonly regarded as "traditional", have on the one hand a marginal role in modem industrial economies. On the other hand, they are constantly influenced to a 118 Y. Goren et al. great extent by modem industrial societies, their potters having to compete with inexpensive containers of tin and plastic. Therefore, their output do not fully reflect the reality of ancisnt pottery production by the urban societies of Egypt and Canaan. In consequence, most ethno-archaeological studies have only a limited value for explaining mechanisms of ancient pottery production since they tend to overlook lifestyles and mental templates of ancient societies. In conclusion, ethno-archaeological data should be treated with caution when used for the explanation of motives and mechanisms of ceramic production, or indeed any other human b€haviour in a different time or sDace. References Allen, R.O., Rogers, M.S., Mitchelt, R.S. and Hoffman, M.A. 1982. A geochemical approach to the understanding of ceramic technology in Predytastic Egypt. Archteometry 24 199-212. Allen, R.O., Haruoush, H. and Hoffman, M,A, 1989. Archaeological implications of differences in the composition of Nile sediments.lt Archaeological Chemistry lY. Allen, R.O. (ed.). 33-56. Washington D.C. Amiran, R. and Glass, J. 1979. An archaeological-petrographical study of 15 W-Ware pots in the Ashmolean Mus€um. Tel-Aviv 6: 54-59. Amold, D. 1985 Ceramic Theory and CulturaL Process. Cambridge. Amold, D. and Bourriau, J. (eds.). 1993. Az introduction to ancient Egtptian pottery. Mainz am Rhein. Bourriau, J. 1981. Utwn el-Ga'ab, Pottery from the Nile Valley Before the Arab Conquest. Canbidge. Bourriau, J. 1990. Canaanite jars from New Kingdom deposits at Memphis, Kom Rabi'a. EretzJsrael2l (Ruth Amiran volume): 18*-26*. Bourriau, J. and Nicholson, P.T. 1992. Marl clay pottery fabricks of the New Kingdom from Memphis, Saqqara and Amama. ?,he Journal of Egyptian Archaeology 7E 2991. Dothan, T . 19'12. Excavations at the Cemetery of Deir el-Balah (QeAemMonographs of the Institute of Archaeology, the Hebrew Universitv of Jerusalem l0). Jerusalem Dothan, T. 1987. The impact of Egypt on Canaan during the l Sth and l9th Dynadties in the light of the excavations at Deir el-Balah. ln Egypt, Israel, Sinci. Rainey, A.F. (ed.).121-135. Tel-Aviv. Eyal, M., Barto\ Y, Shimron, A.E. and Bentor, YK. 198O. Geological Map of Sinai. Jerusalem. Faulkner, R.O., 1947. The Wars of Setosl.loumal of Egyptian Archaeology 33 3440 Faulkner R O., 1975. In Canbridge Ancient Hirtory (3rd ed.). Edwards, I.E.S. et al. (eds.) Vol. II, Pt. 2. Cambridge. Franken, H.J. l9?8. Pottery from a Middle Bronze tomb near Tell Nadid on the Euphates. In Arclueology in the lzvant. Moorey, R. and Parr P (eds.). London. Gardiner, .4.1920. Thc Ancient Mi l itary Road Between Egypt andPalcstine. Joumql of Egyptian Archaeology 6: 99-l l6 Gilead, L and Goren, Y 1989. Petrographic analyses of fourth millennium b.c. pottery and stone vessels from the Northem Negev,I{ael. Bulletin of the American Schools of Orie ntal Re se arch n, 5-l 4. The Archaeolo gical and Etlnoarchaeo logical Glass, J., Goren, Y., Bunimovitz, S. and Finkelstein, I. Interpretation I 19 1993. Pebographic analyses of Middle Bronze III, Late Bronze and Iron Age I cemmic assemblages.Il Shiloh, The Archaeology of a Biblical Site Finkelstein, L, Bunimovitz, S. and Lederman, Z. (eds.). 271-286. Tel Aviv. Goldberg, P., Gould, B., Killebrew, A. and Yellin, J. 1986. Comparison of neutron activation and thin-section analyses on Late Bronze Age c.eramics from Deir elBalah. ln Proceedings of the 24th Intemational Archaeometry Symposium Olin, J.S. and Blackman, M.J. (eds.).341-351. Washington D.C. Goren, Y. 1991. T/re BeginningE of Pottery Production in Israel, Technology and Typology of Proto-Historic Ceramic Assemblages (6th4th Millennia B.C.E.). Unpulr' lished Ph.D. dissertation. Jerusalem: The Hebrew University (Hebrew). Goren, Y. 1992. Petrographic study of lhe ponery assemblage from Munhata. In Llte Pottery Assemblage of the Sha'ar Hagolan and Rabah Stages of Munhata (lsrael). Garfinkcl, Y (ed.). 329-360. Paris. Goren, Y 1995. Shrines, ceramics and metal objects in Chalcolithic Israel: The view through the p€rographic microscope. Arcraeometry 37 (2). ln press. Holthoer, R. 1977. New Kingdom Pharaonic Sites: The Pottery-Uppsala. Kitchen, K.A. 1982. Pharaoh Triumphant: The life ond Timer o/nanesres /1. Mississauga Nicklin, 1971. Stability and innovation in pott€ry manufacturc, World Archaeology l3-98. Nicklin, K. I 979. The location of pottery manufacture . Man 14:, 436458 . 3: Oren, E.D. 1980. Egyptian New Kirgdom sites in Nonhem Sinai. Qadmoniot 13l 2633 (Hebrew). Oren, E.D. 1981. The architecture of Egyptian 'govemors residencies' in Israel under the New Kingdom. EretzJsrqel 15: 183-199 (Hebrew). Oren, E.D. 1987. The "Way of Horus" in Nonh Sinai .ln Egypt, Israel, Sinai. Raircy A.F (ed.). 69-120. Tel-Aviv. Oren, E.D. 1992. Not1h Sinai. Inr ?ie New Encyclopedia of Archoeological Excavations in Eretz-lsrael (Vol. 3). Stem, E. (ed.). I106-l I 16. Jerusalem. (Hebrew). Porat, N. 1987. Local industry of Egyptian pottery in Southem Palestine during the Early Bronze I period. Bulletin of the Egypmlogical SemiturE: 109-129. Porat, N. 1989a. Petrography of ponery from southern Israel and Sinai.ln L'urbanisation de la Pqlestine a l'age du Bronze ancien, bilan et perspectives des rechercheE actuelles I.Miroschedji, P de (ed.). Oxford: BAR Intemational Series 527(i). 169188. Porat, N. 1989b. Composition of pottery--application to the study d the interrelations between Can4an and Egypt during the 3rd millennium 8.C. (Unpublished Ph.D. diss€rtation). Jerusalem: the Hebrew University. Porat, N., Yellin, J., Heller-Kallai, L. and Halicz, L. 1991. Correlation between petrography, NAA, and ICP analyses: application to Early Brcrlze Egyptian pottery from Canaur. Geoarchaeology 6: 133-149 . P 1981. Evolution of specialized pottery production: a trial model Current A n t hro p o Io gy 22: 21 9 -24O. Rice, P 1984. Change and conservatism in pottery-producing systems. In The Many Dimensions of Pottery- van der l€euw, S.E. and Pritchard, A.C. (eds.). 231-293. Amsterdam. Rice,P. 1987. Pottery Analysis. A Sourceboo,k. Chicago. Rothenberg, B. 19'12. Timna: Valley of the Biblical Copper Mircs.lnndon. Shepard, A.O. 1954. Ceramics for the Archaeologist.Washingaorl Rice, l?l Y. Gorcn a aL Snch" A, 1987. Quatcnrary drainagp syst€ms in northem Sinai. In Stnai Vol. I Gvinzman, G., Shmueli, A,, Gradus, Y., Ben-Arieh, L and Har-El, M. (cds.) 185-192, Tel- Aviv. ad Syle, BV.. 194. An andyticd comparison of vuiors Egyptian soils, clay+ shales atrd somc anci€Dt pottrry by norfoo actirratioo oralysis. Ia Rcccnt Advones ia kicrtc ard Tecrvology d Mderiab. Bishan A.(ed"). 9-128. Nsw Tobia, S.K. Yo*. lcnrcl,Y. 1984. Ecolog g tlu Uegetation of tltc Ind of Isrzcl.Tel-Litt. (Ilebrcw). W€instcin, J.M. 1981, The Egyptian empirc in Palestine-A rcasse{rsrr'r,aL Bu etin of tlE Anpricot Schools d Ofiefial Rcscarch AL l8-2l. Wilron, 1955 ESDtian historical tcrb. InArci'r Near EasremTes Relati//lt to tlu OwTerwrcm,P'irfrd[d, J.B. (cd.), 173-182. pdncch. 'J,. 7 ili,tt 1980.lhgenl Lodsrycs of Isrwl,Tcl-Aviv. (ltuett).