Kerma
This site is notable for two well preserved mudbrick ruins termed "Deffufa" in the local dialect. These were noted by earlier travelers, but it was not until the excavations of the Harvard University-Museum of Fine Arts, Boston Expedition to the Sudan under the direction of G.A.Reisner in 1913-16 that the nature of the site was revealed. Much more has been added to our understanding of Kerma and its development from the ongoing fieldwork of Charles Bonnet and the Expedition of the University of Geneva.
For much of the third and second millennia BC, the most important state in the Nile Valley aside from Egypt was the Kingdom of Kush, centered in the area of the Third Cataract, which seems to be the principal rival of the Egyptians for control of Nubia. In all likelihood, their state was centered at the site of Kerma, in the Dongola Reach. Unlike much of Nubia, in this region there is a broad floodplain that allowed rich agricultural production, thereby enabling the growth and maintenance of a very large population and accumulation of surplus wealth.
The town itself was situated on the east bank of the Nile. It is roughly circular in plan and built along an east-west axis. The central portion covers an area of about 0.5km in diameter and is surrounded by a dry moat and ramparts. Outliers of the settlement have been found to the southwest and may have consisted of cultic buildings and temple workshops. Other remains close to the river may have included port facilities. At the center of the town was a vast mudbrick temple known as the Lower or Western Deffufa, measuring over 50x25m and preserved to a height of almost 20m. It was surrounded by a series of workshops and storerooms. Nearby was a large circular structure over 15m in diameter, of post and thatch construction, that may have served as an audience hall. Perpendicular to the main axis of the town and to the Deffufa, a large palace was recently discovered that combined both Egyptian and local architectural features.
About 3km north-northeast of the center of the ancient town, on the desert edge, lies the associated cemeteries. Over 1km in length from north to south, the necropolis is estimated to contain over 30,000 separate burials. It was divided by Reisner into Cemetery N, the northernmost of the cemeteries, Cemetery M, the central portion of the necropolis, and the large tumulus burials of the so-called "Egyptian" or South Cemetery. Included in the last are the isolated "B cemetery" graves, which are intermediate between the M cemetery and the earliest tumulus burials of the South Cemetery.
In accordance with Nubian tradition, the deceased was not mummified but was placed in a contracted position, occasionally on a bed, under a large mound. At Kerma these mounds were decorated with geometric patterns in black and white pebbles and were sometimes adorned with painted skulls of gazelles, goats or cattle. Grave goods were placed with the deceased, including sacrificed animals and humans. Offerings could also be left outside the graves and eventually small chapels were built to contain them. These evolved into large cult chapels, including the monumental Upper or Eastern Deffufa.
The chronology of these burials has confounded scholars since Reisner’s excavations there. Reisner himself had reservations about the historical position he hypothesized for the culture, and admitted that he was at a loss to date the site by ceramics in the traditional way because the exact dating of Egyptian pottery of the 11th-18th Dynasties had not yet been worked out. Without pottery to serve as a dating tool, Reisner was led astray by imported Egyptian statues and inscribed monuments. This problem was further compounded by flawed anatomical analysis provided by D.E.Derry, who postulated a racial difference between the populations buried in the southern ("Egyptian") cemetery and those to the north. Based on these false clues, Reisner surmised that a garrison was installed at Kerma by Amenemhat I or II (12th Dynasty) to act as a trading post and to safeguard the string of fortresses to the north along the Second Cataract; later Hepdjefa, nomarch of Asyut, was made governor of the colony, "went native" and was buried there in Nubian style.
Reisner’s evolutionary scheme has been revised and the sequence he proposed, beginning with the great burial tumulus, K III and ending with the group K XVI-K XX, has been reversed, with K III being the last of the great tumuli. These are certainly the graves of the Kerma kings, datable to the end of the Second Intermediate Period in Egypt. Likewise, the M and N cemeteries are earlier, not later, and date to as early as the First Intermediate Period. Bonnet has also discovered a pre-Kerma phase that may date to the Old Kingdom and be related to the Nubian A-Group.
Recent excavations, notably at Sai, Mirgissa and Kerma itself, have markedly improved our knowledge of the Kerma culture. Based on results of the excavations at Sai, Brigitte Gratien has proposed a new chronology for the Kerma culture subdividing it into four main phases: KA (Archaic Kerma), dating to the First Intermediate Period and early Middle Kingdom; KM (Middle or Formative Pre-Classic Kerma), coeval with the Middle Kingdom; KC (Classic Kerma), the high point of the culture equivalent to the Second Intermediate Period; and KR (Recent Kerma or Post-Classic Kerma), belonging to the period of the Egyptian reconquest of Nubia in the early New Kingdom. During the earliest periods of development, there are great similarities between the Kerma culture and contemporary C-Group and "Pan-grave" Nubians. Through time, however, a growing differentiation in material culture can be seen between these groups.
During the Second Intermediate Period, the Kerma state extended its sphere of influence into southern Egypt and, according to Egyptian sources, made an alliance with the Hyksos in the north to divide and conquer Egypt. Eventually, the Kerma threat was removed after a series of campaigns by the pharaohs of the early 18th Dynasty. Bonnet has discovered remains beneath the modern town that appear to date to this later period. Napatan and Meroitic remains discovered by the Swiss expedition attest to the continued importance of the site to later Nubian cultures.
Kharga Oasis, Late period and Graeco-Roman sites
Kharga Oasis is located in the Western Desert (24°-26°N, 29°26′E) within the great Libyan depression. The oasis consists of a series of spring-fed areas about 100km in length. Ancient Egyptian terms used for this area were the hieroglyphic ‘ (oasis), rsy.t (southern oasis), and perhaps knmt (though the last probably refers to both Kharga and Dakhla Oases). In Graeco-Roman times it was referred to as "e prote Auasis" (Strabo), "Oasis Megale" (Ptolemy), and the Latin "Oasis Major." In Arabic terminology, Kharga also had the names "Biris" and "Bihit" (perhaps deriving from Coptic). In modern times it is known as Kharga or Wadi Gadeed (New Valley).
Numerous prehistoric peoples lived in and around Kharga Oasis. There is, however, little archaeological evidence surviving in Kharga until the Third Intermediate Period, although major archaeological sites from the Old Kingdom are found in Dakhla Oasis to the west. The important archaeological sites in Kharga all date from the Late period. In the 6th Dynasty the royal agent Harkhuf used the desert route through Kharga in order to travel south to Nubia. In the New Kingdom, the oasis was known to export a fine variety of wines to the Nile Valley. The area is also known as a place of exile (for example, by the Christian heretic Nestorius, or as discussed in the New Kingdom literary work, Tale of Woe).
The principal importance of Kharga Oasis throughout history remains its sweet water wells, which supplied the numerous desert caravan routes that intersected and went through the region. All of the major sites in Kharga are located at sources of water along desert routes to wadis leading up to the desert escarpment. The ancient city of Hibis at the northern end of the oasis clearly developed due to its water sources. Although Dakhla Oasis was the capital of the western oases during Old and Middle Kingdom times, in the Late period the sites around Kharga rose in importance as a result of the conscious attempt to better control the desert areas.
The Egyptian Antiquities Organization (EAO) has been conducting numerous excavations throughout the oasis, but most of these are unpublished. Except for Hibis Temple and Bagawat, all of the sites described here are not well known and are in need of major archaeological excavations.
ed-Deir
Located 26km northeast of Kharga City, the Roman period fortress of ed-Deir sits at the opening of the major wadi leading from the oasis up to the desert plateau. The fort served to control the desert caravan route to the Nile Valley which passed by the site. The impressive ruins feature 10-12m high mudbrick walls with round towers at the four corners. To the north of the site are several Roman period tombs and the unexcavated remains of a town site.
Hibis Temple
Located just to the north of Kharga City, Hibis Temple is the largest temple in the oasis and the only relatively intact structure in Egypt that survives from the Saite and Persian periods. Temple construction began in the Saite period (26th Dynasty), probably during the reign of Psamtik II, with additions by Darius I, Hakor, Nectanebo I and II, Ptolemy IV(?) and at least one Roman emperor. Several blocks found at the site indicate that an earlier temple dating to the New Kingdom also stood on the site.
Hibis Temple is dedicated to the syncretistic deity, "Amen of Hibis" and "Amen-Re of Karnak who dwells in Hibis." The architecture and religious inscriptions are closely related to temples of New Kingdom and Ptolemaic date from Thebes.
The site was excavated in 1909-11 by the Metropolitan Museum of Art. Epigraphic work, begun in 1985 by Eugene Cruz-Uribe, continues at the present. The EAO recently excavated the area between the two front gates, revealing an open courtyard surrounded by columns. The ancient town of Hibis, which surrounds the temple on the south, west and north, has never been excavated because it lies under modern cultivation.
Bagawat
Just north of Hibis Temple, surmounting a large hill, are the impressive ruins of a Coptic cemetery known as Bagawat. The site contains several hundred mudbrick tombs as well as the remains of a basilica. The tombs date from the late third/early fourth through tenth centuries AD, and are mainly single tombs consisting of a vaulted structure (cupola) over a sunken burial pit. Several family tombs are known. Most of the tombs are undecorated, but several have biblical scenes painted on the ceilings. The site was excavated mainly by Ahmed Fakhry for the EAO.
Nadura
Located about 2km southeast of Hibis is the Roman period temple of Nadura. The dating of the temple is based upon the presence of the cartouches of Antoninus Pius (AD 13861). The temple site and surrounding structures (including mudbrick houses and vaulted granaries) are mostly unexcavated. There is much debate concerning the deity of the temple. Suggestions include Amen-Re, Mut and the Dioscuri (the Graeco-Roman cult of the Twins, Castor and Pollux).
Qasr el-Ghuieta
The most archaeologically unknown area in Kharga Oasis is the hilltop site of Qasr el-Ghuieta. Located 17km south of Hibis, the town site is mostly unexcavated. A temple (ancient Perwesekh) has been cleared within the 8-10m mudbrick walls of the town. The earliest part of the temple probably dates to the 25th or 26th Dynasty, with later work done by Darius I and Ptolemies III, IV and X. The temple is dedicated to the triad of Amen of Perwesekh, Mut and Khonsu.
The excavated area in front of the temple has some stone buildings, probably administrative rooms, which appear to have suffered fire damage during the sacking of the city, probably in AD 450 by the Eastern Desert tribe known as the Blemmyes. The remainder of the town within the walls consists of numerous archaeological strata 5-8m deep. The EAO recently excavated part of the town southeast of the wall, and cleared a number of rooms and passages with recorded finds of pots, bronze artifacts and ostraca in Demotic and Greek.
Qasr Zaiyan
Twenty-seven km south of Hibis is the mudbrick enclosure of Zaiyan (ancient Tchonemuris). The sanctuary of the temple is the only stone portion of the edifice except for the lintels of the two gates and the floor. The only dated inscription is from the reign of Antoninus Pius (AD 140), but the lintel of the inner gate is Ptolemaic in style. The inscriptions in the sanctuary are at the latest Ptolemaic in date, but may be earlier. The temple is dedicated to Amen of Hibis, Mut and Khonsu.
The EAO recently completed clearing out most of the temple precinct, revealing a large granary and associated administrative rooms. At the lowest layers of mudbrick are remains of painted plaster covering the inner walls.
Qasr Dush
At the far southern end of the oasis, 104km south of Hibis and 17km southeast of Baris, is the site of Qasr Dush (ancient Kysis). The large temple here is axial in plan with an inner and outer sanctuary, columned hall and two pylons. Hieroglyphic inscriptions indicate the temple is dedicated to Osiris, although the Greek inscriptions suggest it is a temple of Serapis and Isis. Surrounding the temple are the ruins of a large town. The site has a number of inscriptions dating to the reigns of the Roman emperors Domitian, Hadrian and Antoninus Pius.
The site is currently under excavation by the French Institute of Archaeology, Cairo, which has built a small excavation house there. Recently they discovered a cache of gold and other items during the excavations.
Kharga Oasis, prehistoric sites
Kharga Oasis, situated about 200km west of the Nile Valley, is the easternmost of the five major oases of the Western Desert. It is an elongated depression, approximately 185km long and 20-80km wide, lying with its principal axis in a north-south direction. On the east and north it is bounded by steep escarpments capped with Eocene limestone, and on the west by an irregular escarpment which lacks this limestone cap. The oasis is open to the south and southwest, where the floor of the depression, generally some 300-400m below the top of the adjacent Libyan Plateau, rises gradually to meet the sandstone floor of the Sahara.
Kharga appears in several memoirs of seventeenth- eighteenth- and nineteenth-century travelers interested primarily in pharaonic and Graeco-Roman antiquities. Laying the foundations for prehistoric research were early geological investigations by A.K.von Zittel, a member of Gerhard Rohlfs’ expedition to the Western Desert in 1873-4, Captain H.G. Lyons, whose stratigraphic observations were presented to the Geological Society of London in 1883-4, and John Ball, director in 1898 of the first systematic survey of Kharga. Early references to the prehistoric stone tools of Kharga appear in the work of H.J.L.Beadnell, a colleague of John Ball, and in H.E.Winlock’s account of his 1908 round trip by camel between Kharga and Dakhla Oases.
The first systematic investigation into the prehistory of Kharga was conducted from 1930 to 1933 by Gertrude Caton Thompson and E.W.Gardner. Caton Thompson not only defined the archaeological sequence of stone tool industries but, with the aid of Gardner placed these cultural remains into a broader geological and paleoenvironmental context. Unfortunately, relative dating was made complex by the absence of site stratigraphy, and Caton Thompson’s sequence, therefore, is based on the comparison of characteristic artifacts from known assemblages.
Prehistoric research in Kharga Oasis was resumed in 1976 with the Combined Prehistoric Expedition (CPE) led by Fred Wendorf and Romuald Schild. In 1983 Fekri Hassan and Diane Holmes contributed additional data on the Khargan Neolithic. A short survey undertaken by the Western Desert Expedition (WDE) in 1982-3 contributed an important assessment of prehistoric settlement patterns in Kharga. Since 1992, members of the Dakhleh Oasis Project (DOP) have been engaged in reassessing Caton Thompson’s classic sequence, and in obtaining Uranium-series (U-series) dates on associated wadi tufas from Kharga.
The earliest cultural material found in the oasis is described by Caton Thompson as "typically evolved Acheulian" (Lower Paleolithic). Although none of the sites with Upper Acheulian tools can be dated directly, preliminary U-series dates on associated tufas suggest an age of over 400,000 years BP (before present). Well-made handaxes (lanceolate and pear-shaped) are the dominant tool-type. Other tools, such as choppers and flakes, are present, but in low frequencies. Levallois elements are rare, indicating that this technique (where a core is intentionally prepared in order to produce a "specialized" flake of predetermined shape) does not yet play an important technological role. It appears likely that Caton Thompson’s "Acheulio-Levalloisian" material, found only at Refuf Pass, is a natural admixture of Upper Acheulian and "Lower Levalloisian" artifacts.
The presence of wadi tufas overlying Acheulian deposits, and floral evidence including remains of plants such as fig, suggest that the climate of the oasis during the late Acheulian was considerably wetter than today. Favored site locations appear to be near wadi courses in the vicinity of the Libyan Plateau, or adjacent to spring vents on the depression floor.
Although the dating is inadequate, it is generally accepted that the Acheulian was replaced by the specialized flake industries of the Middle Paleolithic by at least 220,000 years ago. A period of hyperaridity which followed the late Acheulian wet phase was succeeded by multi-staged, Middle Paleolithic wet periods, interrupted by drier intervals. Caton Thompson believed that there were at least two stages of renewed spring activity at Kharga.
In Caton Thompson’s (relative) sequence, there are five Middle Paleolithic taxonomic units at Kharga: "Lower Levalloisian" (earliest), "Upper Levalloisian," "Levalloiso-Khargan," Khargan and Aterian (latest). More recent evidence, however, suggests that the Khargan postdates the Aterian.
Only two of the ten "Levalloisian" sites reported by Caton Thompson are found in spring mound deposits on the floor of the depression. The remaining eight are situated in four different scarp areas. Thirty-five additional components, mostly from the Libyan Plateau, were reported by the WDE.
The Aterian is considered a typologically late, Levallois-based industry, characterized by very thin, carefully shaped cores (triangular or discoidal), which produce correspondingly thin flakes, many with finely faceted butts. Stemmed (pedunculated) implements, primarily points, and finely flaked, leaf-shaped pieces (bifacial foliates) are characteristic tools.
According to the WDE, Middle Paleolithic settlement is oriented to the exploitation of a wide range of habitats. Of the forty-two Middle Paleolithic sites identified, however, 72 percent are associated with deposits of ephemeral lakes (called "playas"), 26 percent occur along wadi courses, and 2 percent are near spring vents.
In view of the fact that none of Caton Thompson’s "Levalloiso-Khargan" assemblages is in a secure context, the integrity of this cultural unit is considered doubtful. The validity of the Khargan itself has been questioned by some, primarily because of the random, ad hoc appearance of much of the retouch, which could be the product of natural agencies. Also, the industry as a whole is so morphologically variable that it defies rigid typological classification. However, since neither ambiguous retouch nor highly variable morphology invalidates the distinct technological traits present, others believe the Khargan to be a valid archaeological designation. Among tools which permit formal classification, scrapers (particularly end and nosed varieties) and borers are common.
The restricted distribution of Khargan material, both at Kharga and in the Western Desert as a whole, appears to reflect the increasingly arid conditions which followed the Aterian wet phase. At Kharga, all known sites are in the scarp deposits of the Bulaq Pass.
Caton Thompson claims that her term "Epi-Levalloisian" denotes all those regional, Levallois-related industries which are believed to fall "anachronistically" within the time-span of 30,000 to 10,000 years (represented in the circum-Mediterranean by the blade industries of the Upper Paleolithic). This term, however, seems to be a theoretical construct of Caton Thompson’s to bridge the gap between these apparently late, Levallois-derived industries and succeeding microlithic assemblages.
To date, there is no known lithic sequence in any of the oases, and only one from the Nile Valley, which documents a transition from the (Levallois-based) Middle Paleolithic to the blade-producing technologies which characterize the Upper Paleolithic in the lower Nile Valley. In fact, with only two exceptions, there is an almost total gap in the Egyptian archaeological record between 40,000 and 20,000 BP. Climatic conditions improved around 10,000 BP, and with the onset of the Holocene wet period there is renewed evidence of increased human habitation in the Western Desert. At Kharga, there is renewed artesian spring discharge and playa formation.
Archaeological assemblages of the early Holocene display considerable regional diversity. They are all characterized by blade and bladelet technologies, and by an emphasis on microlithic tool production. Grinding stones and ostrich eggshell beads are commonly found in these early Holocene sites. Caton Thompson classified such material as "Bedouin Microlithic," found at eight localities in Kharga (two on the depression floor and six at silt pans on top of the eastern escarpment). An additional twenty-one "Terminal Paleolithic" (Epi-paleo-lithic) components have been reported by the WDE.
Following the "Bedouin Microlithic," Caton Thompson defined the "Peasant Neolithic," which she regarded as the equivalent of the Predynastic culture in Upper Egypt, a conjecture wholly supported by radiocarbon dates obtained by the CPE. The "Peasant Neolithic" at Kharga, therefore, marks the latest period of prehistory.
Caton Thompson found Neolithic material associated with spring mounds and also at the chert-mining quarries on the eastern escarpment of the Libyan Plateau. Holmes has reported a single radiocarbon date of around 7,220 BP for the Umm ed-Dabadib area which has yielded fourteen Neolithic surface scatters. The lithic industry is reported to be very similar to that recovered by Caton Thompson.
Six Neolithic sites have been reported at Kharga by the CPE. Sites E-76-7 and E-76-7a at truncated spring vents in the airport area are the most notable. Site E-76-7 is the earliest, with one radiocarbon date of around 5,450 BP. This site consisted of a large concentration of lithic artifacts, a few bones and ostrich eggshell. Only one (upper) grinding stone was in evidence. Several potsherds were found nearby, as well as a hearth. Both Sites E-76-7 and E-76-7a have yielded fragments of a large bovid which has been identified tentatively as domesticated. The Neolithic industry reported by the CPE is characterized by a technology limited to unprepared cores, and by a stress on working tabular chert into bifacial tools such as foliates and large oval hoes. The major elements among the flaked tools are denticulates and perforators. Rare sherds of undecorated, coarse-tempered pottery are present.
The WDE has reported only nine Neolithic sites, the majority of which suggest ephemeral, task-specific occupations. They appear to be very unlike those reported by Caton Thompson or the CPE.
Data from playas elsewhere in the Western Desert suggest a complex pattern of alternating wet and dry intervals throughout the Holocene. The combined floral and faunal evidence, however, indicates that these wet phases were drier than earlier ones, and that the Western Desert was a semi-arid grassland. This paleoenvironmental situation is reflected in the Epi-palaeolithic and Neolithic settlement patterns in Kharga Oasis.
According to the WDE findings, most sites of these periods are associated with playas; a few are associated with springs on the depression floor. In contrast to earlier settlement evidence, none of these sites is associated with wadis.
The scarcity of in situ cultural material as opposed to deflated surface concentrations, the absence of radiometric techniques for much of the early time range involved in the prehistoric record, and the lack of datable material from sites which are amenable to radiocarbon analyses all continue to plague prehistoric research in Kharga Oasis. That we are able to reconstruct any part of it, even in a fragmentary way, is a tribute to the pioneering efforts of Caton Thompson and Gardner, and to the skill and patience of those who have followed in their tracks.
