POTS AND POLITICS: CERAMICS FROM ASKUT AND EGYPTIAN COLONIALISM DURING THE MIDDLE THROUGH NEW KINGDOMS STUART TYSON SMITH The island fortress of Askut provides an unparalleled opportunity to docu- ment the presence of Egyptians in Nubia from the Middle through the New kingdoms, and thus gain an important insight into the changes in Egyptian colonialism in the region (fig. 3.1). The site was excavated from 1962-1964 by the late Alexander Badawy as a part of the UNESCO Aswan High Dam Salvage Campaign and under the sponsor- ship of UCLA (Badawy 1964, 1965, 1966). Due to a generous division with the Sudan Antiquities Service, virtually the entire collection from this project is curated in the Fowler Museum of Cultural History at UCLA. Unlike the majority of other sites in the area, there was no "winnowing" of "undesirable" or "uninformative" objects. Preser- vation was relatively good, and the standards of excavation were excellent for the time, better than virtually any of the major Egyptian sites in Nubia.' The result is one of the largest collections of well provenienced Egyptian and native Nubian domestic pottery outside of Egypt or the Sudan. I estimate that there are some thirty thousand Pharaonic sherds, with approximately twenty thousand from good stratified contexts. There are one thousand or more whole vessels, or vessels with complete profiles. Many of these are bowls and plates, of course, but other vessels and larger jars are also represented. This situation allows for a far better reconstruction of the total assem- blage and its stratigraphic associations than has previously been the case for Egyptian sites in Nubia. PART ONE: POTS This report is based on a preliminary sort and presence/absence analysis. Quantifica- tion is ongoing, and the results will be refined considerably in the final report. Natu- rally, some changes may be required, but the basic patterns described here should hold true. The illustrations are intended to give a good idea of the most common and char- acteristic types from each period, but establishing the full range of variability in vessel form, fabric, and decoration for such a large and varied collection was not possible Pots and Politics 39 Elpantine A FiM Cataract Lower Nubia . Loe Kubban Buhen 0 Anib Fams/ ?Second Catart, Wad.iAnaqi Kor Sai Gold Fields O Sai f?Duweisbat ' * GoldMines ' -Upper Nubia M irj nWCatarac Ka eKwugus Dongola \ **.ataact Fift Cat aot Rewch Ata 0 0 Askutt o Saras /po * GoldMines Shalfak 0 Heh 0 Uronarti Semna 13 Egyptian Fort 0 Km 0 Kerma and C-Group g Kumma Setlement Semna South L 0 5 10 Kilometers Figure 3.1 Nubia and the Second Cataract. 40 Egyptian Pottery within the scope of this article. Pot stands, one of the most common components of all the assemblages, are not considered in any detail. Many are of a very simple form (similar to fig. 3.3S) which defies any attempt at seriation, but others do show some promise for dating. They will appear in the final analysis of the ceramics and other artifacts, which will appear as a publication of the Fowler Museum of Cultural His- tory at UCLA.2 FABRICS The organization of the fabrics follows the Vlenna system (Nordstrom 1985, Nordstrom and Bourriau 1993), which, by and large, seems to work fairly well for the Askut material. The characterizations were carried out with a hand lens at lOX magnifica- tion. A series of thin sections concentrating on the marls is planned, which will serve to describe the types more precisely and identify any substantial deviations from the system. Foreign and native Nubian fabrics are not considered here. NILE SILTS The full range of Nile silt fabrics appear, including Nile B 1, B2, C, and D. To some extent, the categories of Nile B1-C represent a continuum based on the amount and size of chaff present. As in Egypt, they were by far the most common material used (e.g., Hope 1989, 4). Nile Silt B] is a very fine fabric with sand and small amounts of fine chaff (up to 2 mm; rarely, if ever, larger). It is most characteristic of the Middle Kingdom, particularly the hemispherical bowls (figs. 3.2A, 3.3A), but it also appears in a variety of small bowls, stands, and jars (figs. 3.2C-D,G; 3.3B-D,N,R). It is typically brown in color (7.5YR 5/4) and often without zoning. Nile Silt B2 is a medium fabric with sand and moderate amounts of small chaff, 2 mm to 5 mm, with occasional larger pieces. It appears rarely with hemi- spherical bowls, normally in a wide variety of small to medium-sized bowls, plates, stands, and jars (figs. 3.2B,I,M; 3.3G-I,M,O,P; 3.4H,J; 3.5A,C,D; 3.6A,C-I,K; 3.7A-D,G-J; 3.8J,K; pl. 3.4a,b,d). It is the typical fine silt in the New Kingdom. It usually varies in color from a reddish yellow (7.5YR 7/6 to SYR 7/8-6/8) to red (2.5YR 5/8), with weak red to reddish brown zoning (IOR 5/4 to 2.5YR 5/4). This latter variety, which is well-fired, seems to be more common in the New Kingdom. Poorly fired examples exhibit grey-black zoning. Nile Silt C is a coarse, chaff-tempered fabric with sand and copious amounts of large chaff, 5 mm and larger. It is used occasionally for medium, and even small, jars, plates, and stands, but most often for large to very large plates, stands, and jars, especially the water/beer jar, large coarse platters (pl. 3.1c), crude "bread dishes" from all periods, (not illustrated here, but see Bourriau 1991, 18-19, fig. 4.1; Hope 1989, 4, fig. lb), and cookpots (figs. 3.2H,K; 3.3J,K,S; 3.4B-D,E,G,K; 3.5H,J; 3.6J,L,N; 3.7D,F; 3.8C,F,O; pls. 3.1a-c, 3.3c, 3.4c). In the Middle Kingdom, its color is typically in the reddish yellow range of the Nile Silt B2, often with grey-black zoning. In the New Kingdom, better fifed examples have the reddish yellow to red coloring. Nile SiltD is a fine, hard fabric with significant amounts of crushed fine-coarse limestone and sand, but very little, if any, straw. It appears in large storage vessels, perhaps replacing those normally found in a Marl C during the late Middle Kingdom Pots and Politics 41 X/~ ~~~~~~~~~ ) ___- W- -IBLrnt Figure 3.2 Middle Kingdom cups and bowls from Askut. 42 Egyptian Pottery ' _ Dm~~~~~~~~~~~ F G cm Concretion. QFe Pots and Politics 43 (cf. fig. 3.41,K, both Marl C), as well as other vessels in the New Kingdom (fig. 3.1OE). It varies typically from red to reddish yellow (5YR 6/6 to 2.5YR 5/6), with grey to strong brown zoning (lOYR 5/1 to 7.5YR 5/6). It is very similar to Hope's (1989, 4-5) mixed silt and marl type Marl A4.2, and is the same as the Marl D-like fabric at Deir el-Ballas described by Bourriau (1990, 21). Nile Silt variants include a rare type with small amounts of large (usually 1 mm to 2 mm, up to 5 mm) crushed limestone added in what would otherwise be a Nile Silt B2 (figs. 3.5E, which is also rather sandy, and 3.6B) and C (fig. 3.4F). A sandy silt appears only with cookpots which closely resemble the Palestin- ian "holemouth" jar in both shape and technology, with a brushed on white slip which is often obscured by the soot (fig. 3.4L; cf. Cole 1984, 63f, fig. 18, pls. 24-25; the "upright rim" type also occurs more rarely at Askut, ibid., 65 f, fig. 17, pl. 26). The fabric, clearly a Nile silt, contains abundant quantities of rounded sand, mostly from 0.5mm to 1.0 mm. The amount of sand often appears to exceed fifty percent of the fabric. Such an abundance of rounded silicates would normally cause instability dur- ing firing, but under the right conditions, it can add to the durability of the vessel. Other cooking vessels, including imitations in the holemouth shape, were of Nile Silt C (figs. 3.2H, 3.3J), which makes much more sense, as the openness of the fabric would allow for expansion and contraction as the pot was repeatedly heated and cooled (Rice 1987, 96-97, 105; Rye 1981, 26-27, 34-35). Other variants no doubt also exist, but have not been identified. Of particular interest would be the possibility of distinguishing between locally produced and im- ported Nile silt wares. Pottery production is attested at both Mirgissa and Serra dur- ing the Middle Kingdom, and at Askut in the New Kingdom (Vercoutter 1970, figs. 23-24; Williams 1987).3 The fabric of the "Gilded" ware (fig. 3.3Q and pl. 3.ld; see below) might provide a control over local clays, since it only appears in Nubia. MARLS The marls are particularly significant because, unlike the silts, they must have been imported, the Marl A and B family from Upper Egypt, Marl C and D from Lower Egypt (Bourriau 1991, 129-30). Marls make up generally only a small percentage of the total ceramic assemblage. Marl A3 and A4 dominate the Marl A group and are particularly common during the Middle Kingdom. Marl A4 is similar to A3 in both inclusions and color, but is coarser, and it can be difficult to distinguish between the two. It, and/or a very coarse A3, appears from the Middle Kingdom through the New Kingdom, mostly in large bowls and small (very fme Marl A3 only) to medium and large storage jars, including amphorae (figs. 3.2L; 3.8A,B,M(?),N; pl. 3.2b). The A3-4 fabrics have a moderate content of rounded sand and abundant angular limestone, which appears either as a solid white inclusion or as a void, depending on the degree of firing. The color typically ranges from white (5Y 8/1) with reddish yellow zoning (5YR 7/6) to white to pale yellow (SY 8/2-3 to SY 7/3) with no discernible zoning. This difference is related to kiln placement, and should not be used to indicate separate fabric types (Nicholson and Patterson 1989, 80, fig. 8). Marl A2 has also been tentatively identi- fied in the collection, but thus far only in whole vessels which naturally could not be subjected to fresh breaks for description (fig. 3.7K). Marl B is similar to the Marl As, but can be easily distinguished by the large 44 Egyptian Pottery A EFine Silt cm .~~~~~~~~~~~~~~~~ ,: H) >X Figure 3. M Pots and Politics 45 r J~~I Aw I w C White OjRed cm Figure 3.5 Second Intermediate peniod pottery from Askut. quantity of fie-medium rounded sand, mostly at 0.5 mm or less, but occasionally up to about 1 mm. Color is very similar to the Marl A3-4. It ranges from pale olive (5Y 6/3) with pale yellow zoninga (2.5Y 7/4), to white (5Y 8/1) with pale to reddish yellow orlgtre 25 83t YR76o 2YR6)zoigItcurfomheSon Intereit perod whe it beistXia h alA n 34fbi.Iper MarluCri a . d enseohad faberic,diathie sandiandpo mediufrm Akto.orewhtn 46 Egyptian Pottery a. b. c. d. Plarte 3. Potteryr of theMidle- Kingdonm at Askuit: a) "berbttle,-" note cursoryr wett smohn;b)bedml,nt _h igrmrsa h o rmpesn h lyaon wode fom_)lreta,acasccaftmprdNl itCd idd aesad Pots and Politics 47 red inclusions up to 5 mm in length. It typically has a white surface (5Y 8/2) with a yellowish red background (2.5YR 5/6) showing through. The fracture is often dis- tinctly zoned, with a grey to black core. It is used for large to medium storage jars and "fish plates" during the Middle Kingdom (figs. 3.2N; 3.4A,I,L; the last are not illus- trated but do occur in small numbers in both Marl C and Nile Silt C).4 Marl D is another dense fabric with numerous white calcareous inclusions and sand. It is usually brown to reddish grey (7.5YR 5/2 to 5YR 5/2) with a pale yellow (5Y 8/3) coating, sometimes burnished, which often flakes away from the surface. It has appeared thus far only in amphorae of the New Kingdom (fig. 3.8G,L). THE MIDDLE KINGDOM The pottery from the Middle Kingdom was easier to characterize than that of the Second Intermediate period or New Kingdom. This is partly due to the presence of several large groups of sherds and whole vessels, often in excess of five hundred pieces per level, which made the process of visual sorting much easier. The standard- ized nature of the ceramic repertoire at this period may also be a factor (Bourriau 1981, 55). The Askut assemblages closely parallel the Lower Egyptian corpus seen in the typologies established at Harageh (Engelbach 1923), and, more recently, at Dashur (Arnold 1982). TECHNOLOGY, SURFACE TREATMENT, AND DECORATION Most of the vessels were thrown on the wheel, with the exception of certain types that were typically handmade at this period. The bodies of large vessels were usually handmade, often in several pieces with the rim finished on the wheel (figs. 3.2N; 3.41,K). Beer jars, on the other hand, were usually thrown (fig. 3.4C was handmade but is unusual in this regard). Small vessels, like figure 3.4K, were rarely handmade. Footed incense burners were made in two parts (figs. 3.2I,J; 3.4M). Bread molds were handmade on a conical form and had a fine layer of silt inside, presumably to ease removal of the bread (fig. 3.4E, pl.lb). Bases had been trimmed with a knife (figs. 3.2A,E; 3.3A,G,H; and traces under compaction of 4J) or reed brush (figs. 3.2K, 3.4C; inside stands 3.3K,N,R,S), or cut with string while turning on the wheel, but were never finished on the wheel. Sometimes a crude ring was pinched onto the base of small cups and bowls (figs. 3.2B, 3.3C). Large jars and bowls show string impres- sions from support while drying (fig. 3.2N; cf. the Second Intermediate period jar in pl. 3.3c). Most vessels have received at least a cursory wet smoothing, although this was often very rough, still allowing the wheel marks to be seen (pl. 3.la). Moderate to heavy polishing was found in a variety of forms (e.g., figs. 3.2A,H; 3.3M,P,R; 3.4F,G,J), but only on a small number of the vessels overall. Regular rilling patterns as on figure 3.4J show that this was occasionally done on the wheel. Most vessels were undecorated, but at least a handful of decorated vessels occurred in every large group. Red painted decoration was the most common and included a weak red to red (2.5YR 7/8-6/8 to lOR 5/8) wash on the interior and/or exterior of a variety of both closed and open forms (figs. 3.2-4). Plates and bowls occasionally had a red rim. This element was particularly common in the hemispheri- cal bowls. White decoration occurred occasionally on or below the rim and in crosses or similar patterns. Unfortunately, the white wash used is unstable, and so it is difficult to estimate the amounts originally present. Black painted lines occur very rarely, most com- 48 Egyptian Pottery monly as a border around a white band on the necks of beer/water jars (like fig. 3.4C). Incised straight and wavy lines, often in combination, were also a standard motif (figs. 3.3 and 3.4). They occur below the rims on the exterior of carinated cups and bowls, and also on jars along the neck or at the base of the neck. Incising was sometimes combined with rim pinching and applied ridges on large carinated bowls (fig. 3.3G,J). Very rarely, applied small cups were added to the rim of Marl A3 jars and carinated bowls, sometimes in combination with combing (not illustrated). Two pieces of clay were sometimes applied below the rims of small cups, sometimes in combination with the incised decoration on carinated cups (fig. 3.3E,F). They might either be related to the Hathor vase with applied nipples, which does occur rarely at Askut (pl. 3.2a), or perhaps served to tie off a chord securing a cloth or leather cover. They certainly do not represent real or vestigial handles. Applied crocodiles appear very rarely on spouts and the rims of bowls (pl. 3.2b), and may be related to the worship of Sobek near Armant (Vercoutter 1957), although they are a typical votive offering at various sites in Upper Egypt (Kemp 1989, 72).5 'Gilded' ware also appears as a regular, if rare, component of the Middle Kingdom assemblage at Askut. This type of decoration only occurs in Nubia in the Second Cataract forts and in pottery of the Kerma Classique (Gratien 1990, 234, no. 335). It would appear to be a wash or slip with a high content of golden mica. Since mica appears abundantly in Nile silts at Askut, it would presumably not have been difficult to extract it through levigation or some other means. This 'gilding' was applied not only to very fine, but also to fairly coarsely made, bowls, jars, stands, and lids (e.g., fig. 3.3Q, pl. 3.1d; ibid.). FORMS AND CHRONOLOGY The use of painting, incising, and applied decoration is typical of the period from Amenemhet III onwards (Bourriau 1981,55). This is consistent with the founding of Askut, probably during the reign of Senwosret II (S. T. Smith 1991b, 118). Close parallels can be found for distinctive forms and decoration from both Dashur Com- plex 6 of the late Twelfth to early Thirteenth Dynasty (cf. figs. 3.2-4; Arnold 1982, esp. figs. 6.11,21; 8.1,2,7), and Complex 7 of the advanced Thirteenth Dynasty (cf. ibid., esp. figs. 10.7,8,15; 11.1,3,4). The vessel indices of 81 hemispherical bowls range from 117 to 180, with a mean of 148.4, encompassing both Complex 6 at the high end and Complex 7 at the low end (fig. 3.9; ibid. 1982, 60, fig. 17). Two large groups from individual contexts correlate well with Arnold's minimum-maximum analysis (1988, 140-1), running from 142 to 180 (mean 161) in Room 12 (16 vessels), and from 117-145 (mean 135; a single outlier was 152) below Room Southeast 8 (7 vessels). A group of 7 vessels from Room 4 represents an intermediate phase with a range from 133-158 (mean 147). Several other groups show a similar distribution, indicating that Askut contains a mid Thirteenth Dynasty phase not represented at Dashur. This corpus may fill the gap between the end of Complex 6 in ca. 1760 B.C. or later, and the start of Complex 7 around 1700 B.C. (Arnold 1982,40). Future work will concentrate on correlating these hemispherical bowl groups and vessel types in the several secure Middle Kingdom contexts available at Askut. As quantification proceeds, it should be possible to define better the total ceramic assemblage occur- ring in these phases. Arnold places the end of Complex 7 at about 1650 B.C., and Tell el-Yahudiya Pots and Politics 49 ESoot Mincense m G - J -h _ e~~~~-- _ ~ - - LEi-zRed FoBurnt N cm t4o__~Fgue 3. e igo usanbolfrmAkt 50 Egyptian Pottery .V. . . c ~~~~~~~~~~~I z D i. ! a ' ~~~~~~~~~~~~~a l K gMY?? M554, . 5 Figure 3.7NeKie p y fm A t. Pots and Politics 51 ware from these levels at Askut is consistent with this date. The common form is Piriform lb-c, which should run between ca. 1710-1650 B.C. The example in figure 3.3L, the best preserved, is somewhat unusual, with the rectangular decorative zones of the lb style but with only two decorative bands, as with lc juglets. It might repre- sent either a variant or a very early lc. Two associated hemispherical bowls with vessel indices of 128 and 136 indicate a date in the advanced Thirteenth Dynasty, and pottery from nearby deposits included many parallels with Dashur Complex 7. Three more sherds of Piriform lb, and a Middle Bronze (MB) II red polished juglet neck (fig. 3.12B,C; pl. 3.2d) come from similar contexts. These correspond to Strata G-F at Tell el-Dab'a, equated by Bietak to Dashur Complex 7 (Bietak 1984, 480). Hemi- spherical bowl vessel indices from these strata commonly run from 120-40, very much the same as the later Askut groups. Another similarity is the dominance of the Thir- teenth Dynasty "kettle" mouthed beer jar (cf. fig. 3.4B,C; Bietak 1991, fig. 7) over the funnel-shaped neck, occurring at Askut in earlier strata but not illustrated here (e.g., Arnold 1982, fig. 7.11). William Dever has recently challenged Bietak's dating of this material, placing Stratum G in the late Twelfth Dynasty and F in the earliest Thirteenth Dynasty. He also notes that the dating of Complex 7 is somewhat uncer- tain, possibly falling as early as 1760 B.C. (Dever 1991, 74, 76, and n. 7). The pres- ence of a mid Thirteenth Dynasty group at Askut, however, implies a substantial gap between the end of Complex 6 and the start of Complex 7, consistent with Arnold's suggested starting date of ca. 1700 B.C. for the latter. The Yahudiya ware clearly occurs at Askut in contexts well past the beginning of the Thirteenth Dynasty, sup- porting Bietak's position. IMPORTED AND NATIVE POTTERY Several types of non-Egyptian pottery appear as a regular, if minor, component of the Middle Kingdom assemblage. At least some of the Tell el-Yahudiya ware appears to be in a Palestinian fabric. Sherds of Palestinian storage jars have also been identi- fied.6 Of more interest from the point of view of Egyptian imperialism is the rare, but consistent, presence of handmade native Nubian pottery (fig. 3.10). Most of the types have closest parallels in the domestic pottery of the contemporaneous Kerma Moyen (cf. Type 7, Gratien 1978, 175,243-4; also Gratien 1985a, 419ff., fig. 313; and Maystre 1980, pls. XLVII-XLVIII). These sherds are primarily from open forms, often used as cooking vessels. This distribution implies relations with a settled group, rather than long-distance trade,7 and may indicate that the frontier softened towards the end of the Middle Kingdom occupation of Nubia. Alternatively, these designs could have been part of a broader cultural tradition. Some of the same patterns have been found at a C-Group settlement at Aniba (Steindorff 1935, 202ff., pl. 92ff.), but few of the other typical C-Group types, and only one sherd of the elaborate Polished Incised wares characteristic of this culture, appear at Askut (fig. 3.7F). Distinctively Kerma Moyen sherds, however, do occur, including a nearly complete Kerma Moyen beaker from a secure early Thirteenth Dynasty context (Room 12, see above). The implica- tions of this material will be discussed below. SECOND INTERMEDIATE PERIOD This period, although definitely present at Askut, has proven somewhat elusive. This situation is partly due to the lack of heavy deposition within preserved structures, as was the case during specific times in the Middle and New Kingdoms, but may also be 52 Egyptian Pottery 71. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~F I Red cm 0 5 l0 15s H K~- X 0' X 0$ Figure 3.8 New Kingdom jars from Askut. a factor of the smaller amount of time available for deposition. The Middle Kingdom strata account for about two hundred years (ca. 1850-1650 B.C.) and the New King- dom over four hundred (ca. 1550-1100+ B.C.), while the Second Intermediate period was at most one hundred years (ca. 1650-1550 B.C.). Also, as Janine Bourriau (1991, 130-i31) has pointed out, in Upper Egypt the Second Intermediate period has a slow transition from the Middle Kingdom and to the early Dynasty 18 assemblages. At this point it is possible to document a few of the more diagnostic types. Once these as- Pots and Politics 53 1 0 9 8 7 6 U 4 3 2 1 0- I I I I I I I 1 117 123 129 135 141 147 153 159 165 171 177 120 126 132 138 144 150 156 162 168 174 180 Vessel Index Figure 3.9 Vessel indices of hemispherical bowls at Askut. semblages are quantified, it will be possible to define better the full range of types appearing at this period. TECHNOLOGY, SURFACE TREATMENT, AND DECORATION The Egyptian pottery continues to be primarily wheelmade, with the exceptions noted above. Wheel finishing replaces the rough knife and reed trimming on the bases of bowls and other vessel forms, and a wheel turned ring base becomes popular (fig. 3.5D). Bases cut with a wire or string while the wheel is turning do, however, con- tinue to appear. Large vessels continue to show the use of rope supports while drying (pl. 3.3c). A polished red surface on plates, bowls, carinated jars, and stands becomes more common in the Second Intermediate period. Burnishing was often done with a pebble or similar hard tool, usually by hand (pl. 3.3a). One of the most characteristic decorative motifs is the use of combed wavy lines along with applied ridges or "dummy pots" along the rim of carinated vessels, usually in a Marl B (fig. 3.5B). The use of a white slip, often polished, on Nile Silt B2 and D in imitation of the marl clays also becomes common (fig. 3.5D,J). FORMS AND CHRONOLOGY Several vessel types are characteristic of the early to late Second Intermediate period (cf. Brunton 1930; Bourriau 1990, and forthcoming). Small and carinated bowls with a distinctive profile and the decoration noted above gradually replace the old hemi- spherical bowl and the other forms characteristic of the Middle Kingdom (fig. 3.5A-E, with polished interior, are particularly common). Bietak reports the slow introduc- tion of some Second Intermediate period types, particularly bowls with a ring base, in otherwise Middle Kingdom groups with Tell el-Dab'a Stratum E1/2=b/1 (ca. 1640 B.C.; Bietak 1991, 41, fig. 10). Some strata at Askut show a similar pattern, and 54 Egyptian Pottery Figure 3.10 Native Nubian pottery from Middle Kingdom N{ contexts at Askut. A cm c~~~~~~1 J ,;) represent a transitional phase dating to the very early Second Intermediate period or very late Thirteenth Dynasty. Carinated jars, often with a low, sharp profile and a distinctive rim, become popular in the early to late Second Intermediate period, often in a Marl B fabric (fig. 3.5G-J; pl. 3.3a; cf. Bourriau 1981, 29, 35, figs. 1-4; idem, forthcoming). The forms and rim styles of some of the large storage jars change into a Marl B (fig. 3.5F) or Silt C (pl. 3.3c) and D (jars related to the Marl C type repre- sented by fig. 3.4M but with somewhat different rim and shape; cf. Bietak 1991, fig. 9.5). IMPORTED AND NATIVE POTTERY The amount of native Nubian pottery increases substantially in these levels, although still in the context of a primarily Egyptian assemblage. Incised bowls now have patterns more characteristic of the Pan Grave and possibly C-Group (fig. 3.1 1A-I; cf. Brunton 1930, pls. LXXXII-LXXXIV; Sadr 1987, figs. 4, 5; Gratien 1985b, figs. 11-3). The Kerma Moyen style is replaced by Kerma Classique of the finest quality (fig. 3.11J-L; cf. Gratien 1985b, 204ff., figs. 61-62), equal to that found at Kerma.8 Beakers and various sizes of storage jars occur in the fine black topped fabric, as well as large globular jars with roulette impressed rims (fig. 3.11iL). An unusual type of bag shaped jar with a groove in the base may be of Kerma origin, although an exact parallel has not been found (pl. 3.3b). The more generic Nubian mat and chord im- pressed wares also occur, along with applied clay to roughen the bottom of cook pots (figs. 3.1lM-O; cf. Bourriau 1990, 16-8, fig. 4.1). Pots and Politics 55 NEW KINGDOM Since the New Kingdom occupation of Askut covers a very wide time period, it is not possible within the scope of this paper to show the full range of variability within the different assemblages. Therefore, this section will concentrate on some of the more representative and diagnostic types. Unfired Nile silt sherds of both well thrown and crude mat impressed bowls provide clear evidence for local manufacture of pottery in the New Kingdom and perhaps the Second Intermediate period. Although no obvious kiln site has been found (see note 5), open or small pit firing can reach more than adequate temperatures for the Nile silt fabrics and provide a great deal of control in the hands of a skilled potter (Rye 1981, 96-9, 102-3). TECHNOLOGY, SURFACE TREATMENT, AND DECORATION The Egyptian pottery is still predominantly wheelmade. The use of chord supports during drying is common in large bowls and jars (figs. 3.61-J,L,N; 3.7G; 3.8C). Bases are wheel finished, although string cutting continues to be used for simple bowls (e.g., pl. 3.4b; note rilling from wheel finishing running to the base). Wheelmade ring bases become more popular on both bowls and small to medium-sized jars. A type of jar or bowl with the base extending below the ring is introduced (fig. 3.7L). The bases of amphorae were sometimes made in a mold (cf. fig. 3.81; Hope 1989, 93, pl. Sc). The trend towards heavier finishing seen in the Second Intermediate period con- tinues in the New Kingdom. Polishing, often on a red coating and including "pattern burnishing" on the wheel, is common on bowls, stands, and jars (figs. 3.6H; 3.7A-G,L; pl. 3.4d). The incised decoration favored during the Middle Kingdom and Second Intermediate period falls off, and the use and variety of painted designs expands, especially on carinated bowls and jars (fig. 3.7; pl. 3.4d). The most popular are crossed line and line and dot motifs, sometimes in bichrome. A polished white coating is also commonly, although not invariably, used as a ground for these painted motifs on Nile Silt B2 bowls and jars (fig. 3.7B-D,H,I). A simple white coating is common on in- cense burners (fig. 3.6B; there were possible traces of white on 6F as well). FORMS AND CHRONOLOGY Vessel form and decoration reflect strong occupation throughout Dynasty 18 (figs. 3.6-8; cf. Holthoer 1977; Bourriau forthcoming; Hope 1989). The trends during the Second Intermediate period continue, with carinated bowls and jars being particu- larly common. The plain bowl with polished and/or painted interior completely re- places the hemispherical bowl of the Middle Kingdom, occurring with high frequen- cies in New Kingdom contexts (figs. 3.5A and 3.6A; the latter with red "splash" deco- ration is rare). Some vessel forms continue from earlier periods, notably the round shouldered, pointed base jar of the Middle Kingdom (fig. 3.4G, but not F). The "de- canter" has shrunk and now has a wheel finished base (cf. figs. 3.2E, 3.6G). Incense burners are taller and have a ledge-shaped rim (cf. figs. 3.21,J; 3.31; 3.6F). "Zir" type jars with a folded rim have become elongated (cf. figs. 3.4M; 3.8N). Simple lined motifs on carinated bowls and jars are typical of the early Eighteenth Dynasty through the reign of Thutmose III (fig. 3.7; Bourriau 1981, 72, 135). Bichrome pottery and the strap handled form shown here (fig. 3.7K) are characteristic of the mid Eighteenth dynasty, between the reigns of Thutmose Ill and IV, but drop off during the reign of Amenhotep Ill (ibid., 77-79; idem 1982,80; Hope 1987, 109; idem. 1989,7). Pilgrim flasks (fig. 3.8E) may occur as early as Thutmose Ill, but do not really become com- 56 Egyptian Pottery a. il Plate 3.2 a) hand holding nipple from a "Hathor" vase; b) crocodile on small bowl with incurving rim; c) Mycenaean pilgrim flask; d) Palestinian MBII red polished juglet. b.- :::- : ::F c. d. Pots and Politics 57 a. Plate 3.3 Second Intermedi- ate period pottery from Askut: a) red polished, low waisted carinated jar; b) unusual bag- shaped vessel (Kerman); c) large storage jar and stand, note the impressions from rope supports during drying. b. c. mon until Amenhotep II, continuing in popularity through the end of the New King- dom (Bourriau 1981, 75-6; idem 1982, 83). Askut continued to be occupied during the Ramesside period. Restricted carinated bowls are characteristic of the later New Kigdm (eg. Hoe18,fg_3-) n he-hnldapoarue sadan ag po ssmlrt yewihocr rmterino aessI c.fg .L ibd. 94 fi.32,wiealresoaejri eysmlrt neapefo on dation deposit of Ramesses IV (cf. fig. 3.8M; Holscher 1939, p1. 56).~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 58 Egyptian Potery Black/Red. C -i Figure 3.11 Native Nubian pottery from Second Intermediate period and New Kingdom contexts at Askut. Pots and Politics 59 IMPORTED AND NATIVE POTTERY A Cypriot Base Ring Ware IAa(i) juglet of the early Eighteenth Dynasty (fig. 3.12A; Merrillees 1968, 147ff., pl. III) and sherds of a Mycenean stirrup jar and pilgrim flask, probably Late Helladic I1A2 (pl. 3.2c, ca. 1350-1300 B.C.; Mountjoy 1986, 79ff.), were found. Palestinian amphorae also occur, including one shoulder from a Late Bronze (LB) II Canaanite jar (ca. 1300-1200 B.C.; cf. Amiran 1970, 142, pl. 43; and, in Egypt, Hope 1989, 95, figs. 4-5). Native Nubian pottery continues to appear throughout the Eighteenth Dynasty, with the same resemblance to Pan Grave and C-Group pottery as in the Second Intermediate period (fig. 3.11D-E,G,M-O). Kerma Classique disappears, apparently replaced by the cruder Kerma Recent, which has been tentatively identified in early to mid Eighteenth Dynasty contexts. PART Two: POLITICS Adams (1979) has cautioned that changes in a ceramic repertoire alone cannot be used as evidence of cultural or political change, since pottery production can be influenced by a small number of individuals and often follows its own dynamic. By taking into account the processes and patterns of deposition, however, pottery can provide key information about the chronological and cultural associations of a site where reasonable control over provenance is established (Adams 1986/87). When this information is combined with the historical record in a region like Nubia, it can document key political changes at different periods. At Askut, a rigorous analysis of the pottery allows us to document archaeologically for the first time the history of the Egyptians living in Nubia from the end of the Twelfth Dynasty through the end of the New Kingdom. Based on an analysis of the stelae and cemeteries of Buhen, H. S. Smith (1976, 67ff.) has proposed that the garrisons changed from rotating military units to permanent settlers sometime in the late Twelfth Dynasty. Modifications in the plans of several of the rooms in the "barracks" area of the Main Fort at Askut, dated by associated ceramics to the Middle Kingdom, confirm this conclusion (fig. 3.13, pl. 3.5a). Rooms 5-6 show the typical tripartite arrangement of the Twelfth Dynasty "barracks" unit, but there are several examples of remodeling. Upon the abandon- ment of Rooms 11 and 12 in the early Thirteenth Dynasty, a door was knocked through to Room 13 from Room 28. Sometime before the mid Thirteenth Dynasty, two doors were added from Room 2 to the Room 1 complex, and Rooms 2, 3, 4, and 10 were completely remodeled from two "barracks" units into one house with a completely different floor plan. These changes reflect patterns of private ownership attested at other periods in Egyptian history (H. S. Smith 1972). Evidence of an ancestor cult similar to that found at Deir el-Medineh in the Ramesside period (Bruyere 1939, 85, 151ff.) also indicates settlers. A handful of fragments from offering platters or "soul houses" occurs in Middle Kingdom contexts at Askut (fig. 3.14). These modest funerary monuments were found by Petrie (1907, 14) in great numbers in the cemetery at Rifeh. They also appear, along with funerary stelae and statuary, at Kahun, Buhen, and other settlement sites (e.g., Petrie 1891, 9, 13, pls. IV, XII). Their presence in a domestic context is usually explained as the result of looting from nearby cemeteries or temples for reuse as children's playthings or architectural components (Emery, H. S. Smith, and Millard 1979,98, 151). This is rather unlikely at Askut, however, since to loot the cemetery requires a boat trip to the 60 Egyptian Pottery B Figure 3.12 A. Cypriot A Base Ring Ware juglet; B. c m C MBII red polished juglet; cm C C. Tell el-Yahudiya ware juglet opposite bank of the river, and there is no temple or chapel dating to the Middle Kingdom. Room 12, filled with pottery of the late Twelfth to early Thirteenth Dy- nasty, has a niche with a cornice above a mastaba (pl. 3.5c). This group certainly represents a household shrine, the earliest example known, and the only one dating to the Middle Kingdom.9 A fragment of a Middle Kingdom stela was found in Room 11, which was apparently abandoned at the same time. Although some of the stelae, statuary, and offering platters from Buhen and other settlements of the period were no doubt really looted (as was definitely the case at Kahun) or placed as ex votos in local temples (H. S. Smith 1976, 66-77), it is likely that many, in fact, originally derived from household shrines like those at Askut. The association of official sealings with Middle Kingdom ceramic deposits demonstrates that control from Egypt was maintained until at least the advanced Thir- teenth Dynasty (especially in the Southeast Sector; S. T. Smith 1990,211-14). Goods were also flowing from both Upper and Lower Egypt, as shown by the presence of both Marl A and C storage jars. A considerable change takes place about the end of the Thirteenth Dynasty, with several structures abandoned and new ones built with very different plans. As noted above, the terminal Middle Kingdom assemblages, and thus this event, can be dated to ca. 1700-1650 B.C. by the presence of Tell el-Yahudiya ware (fig. 3.12c)and types of Dashur Complex 7. This correlates with an increase in the percentage of Asiatic pottery in Strata G-F at Tell el-Dab'a, which Bietak (1991, 38) interprets as representing an influx of settlers. Strains resulting from this immi- gration may have affected the central authority, draining resources away from Nubia and towards the Delta and forcing the Egyptian garrison-settlements to become even more dependent upon local resources for their maintenance. By Stratum E/2 at Tell el-Dab'a (ca. 1640 B.C.; ibid., 51-52), Hyksos kings ruled openly in the Delta, and the Egyptian settlements in Nubia were presumably left to fend for themselves as the central government fell into disarray. What happened to the expatriate garrisons who had regarded Nubia as their home for six or more generations? The stratigraphy at Askut indicates that they stayed on to serve the Ruler of Kush. Although we tend to think of archaeological stratigraphy as a kind of giant layer cake, with one stratum succeeding another in relatively orderly fashion, deposi- tion usually occurs in a complex pattern of peripheral disposal and abandonment, leading potentially to deposits of very different dates within and outside of buildings Pots and Politics 61 J~~~~~~~~~~~~~ Midl Kindo t ~~El Es /- a X u s . -. ...... ' R 5 5 tM ' f " ' ld~~~~~~~~~ecndIntermediate Period/ BftJOc _ NewEKingdom X.~. Scale Figure 3.13 Distribution of ceramics at floor of upper level of the lowest layer at Askut. (Dixon 1972; Hoffman 1974; Schiffer 1987; Kemp 1989, 301). Such layers often appear to reflect severe disturbance, and might be dismissed by excavators as unreliable. In Nubia, either this misinterpretation or genuinely poor preservation has hampered a reconstruction of events during the Second Intermediate period. H. S. Smith (Emery, H. S. Smith and Millard 1979, vii, 44, 93) concluded that at Buhen the general lack of stratification did not allow for reliable dating of deposits, since New Kingdom sherds often appeared at greater depths than those of the Middle Kingdom. Following the 62 Egyptian Pottery layer cake model, the different assemblages at Askut would also appear to be incon- sistent, and Alexander Badawy (n.d.) concluded in almost exactly the same terms that disturbance had rendered contextual analysis useless. A careful consideration of pro- cesses of deposition, however, reveals strong indications of abandonment in some areas and maintenance of floor levels in others, explaining the otherwise puzzling lack of consistency in the association of depth with date. Thus Emery and H. S. Smith may have been overly pessimistic in their assessment of the stratigraphy at Buhen (cf. Bourriau 1991, 131). Abandoned structures within a settlement present an ideal location for trash disposal, being both convenient and out of sight. Once started, deposition would continue until the rooms were filled (Schiffer 1987,63-64; Dixon 1972). Figure 3.13 shows the distribution of ceramics from different periods at floor level.10 The group of "barracks" style rooms discussed above shows a classic pattern of de facto aban- donment refuse, characterized by intact and reconstructible pots, with secondary refuse as fill, represented by large numbers of 'orphaned' sherds which do not mend with one another (e.g., pl. 3.6a,b). The gradual nature of the abandonment (see above) and large numbers of sherds from this area, often over one thousand to a room, indicate continuous occupation at the site throughout this process (Schiffer 1987, 58ff., 298ff.). We need not look far to see who was producing this trash. The floors in the house comprising Rooms 14-24 were maintained at roughly Middle Kingdom levels until the late Eighteenth Dynasty, more than a meter below the extant fill in the northern group." The original "barracks" complex that no doubt once stood here was remod- eled into a typical Egyptian "mansion," with rooms grouped around a courtyard (cf. Peet and Wooley 1923, pl. I; Frankfort and Pendlebury 1933, pl. III). Floors in the nearby "Commandant's Quarters" (Rooms 34-45) were also maintained, as was the sturdy structure itself. Middle Kingdom deposits in Rooms 38 and 39 represent floor raisings, attested by the presence of lintels and ramps leading up to the new level. Rooms 41 and 42 were either abandoned or had similar floor raisings. The Middle Kingdom buildings in the Southeast Sector were leveled off and new structures built above. A pattern of peripheral disposal appears around these later houses, with Second Intermediate period and Middle Kingdom pottery around and New Kingdom assemblages within them (fig. 3.15; pls. 3.5b, 3.6c). A rdi-R' scarab, characteristic of the first half of the Hyksos period and shortly before (Bietak 1991, 51),12 from one of the peripheral deposits confirms the date. The Middle King- dom material may indicate that these structures were in use by the advanced Thir- teenth Dynasty, although more work needs to be done on the quantification of the ceramics and correlation of the architectural plans with the field photographs to rule out the possibility that they are associated with the earlier structures.'3 The house comprising Rooms 29-36 shows a direct connection to the Thirteenth Dynasty (S. T. Smith 1992/93). The final building incorporates the wall stubs and tile floors of an earlier structure, itself built above the remnants of the first, Middle Kingdom struc- ture (Rooms SE 32a-b; pls. 3.7 & 8). A large Marl C storage jar set into the tile floor of Room SE 32b indicates a date in the advanced Thirteenth Dynasty (pl. 3.8b, fig. 3.4M; cf. Complex 7, Arnold 1982, abb. 11.3-4; Tell el-Dab'a "zir" type 4, Strata G-E, Bietak 1991, 36ff., fig. 9). Another pot was set into the floor in Room SE 32a. Although the rim was not preserved, several Middle Kingdom carinated cups placed within suggest a similar date (figs. 3.2D, 3.3E; cf. Dashur Complex 7, Arnold 1982, abb. 10.8, 11.2). These pots served as a drain for a household altar constructed on 40 Pots and Politics 63 Figure 3.14 Fragment of Middle Kingdom offering tray with a haunch of beef and granary (?). cm. of fill (pis. 3.7a, 3.8a). A Second Intermediate period style funerary stela was still in situ in the altar's niche, and two Second Intermediate period vessels were found nearby in a group of pots representing de facto abandonment deposition in the mid Eighteenth Dynasty (p1. 3.7b; fig. 3.5D, I; with the Eighteenth Dynasty pots in fig. 3.7B,K,L). This suggests a continuity of cult activity and occupation from the late Thirteenth to mid Eighteenth Dynasty, with the same house rebuilt and floor levels maintained at more or less the same level. A similar pattern appears at Deir el-Medineh, where houses were rebuilt on the same foundations over a period of four hundred years with no appreciable rise in house floors (Dixon 1972). This evidence has a number of implications for the reconstruction of the his- tory of Lower Nubia. Walter Emery, the excavator of Buhen, interpreted a massive fiery destruction layer there as evidence for a violent overthrow by the forces of the Ruler of Kush (Emery, H. S. Smith, and Millard 1979,3,92; H. S. Smith 1976, 80ff.).14 The civilization at Kerma by this time had indeed reached a considerable sophistica- tion, and might well have threatened the fort system.15 But at Askut there is no evi- dence of a violent overthrow, and every indication that the fort was continually occu- pied. In fact, none of the Second Cataract forts, including Mirgissa, show any solid evidence of a siege. This is odd, for we would expect this powerful chain of forts, designed especially to stop an invading force from the south in its tracks, to have born the brunt of the first Kerma assault. Vercoutter (1976,275,303) has remarked on this point, and his analysis of the cemeteries reveals strong indications of continuity of burial, although he apparently still argues for a short hiatus in occupation based on the evidence from Buhen. Other forts, notably Semna, show similar evidence of continu- ity (in the cemeteries and especially the transitional layer "b" at Semna on Plan V; Dunham and Janssen 1960). We have known for some time that Egyptians were serving the Ruler of Kush shortly after the Kermans gained control of Lower Nubia. Some were just mercenar- ies, but H. S. Smith (1976, 80ff.) traced the family of Ka and Sepedhor at Buhen, both of whom boast of serving the Ruler of Kush on their funerary stelae, back to the Thirteenth Dynasty. In order to reconcile their continuing presence with a violent overthrow of the fort, he suggested that they might have turned traitor, explaining the apparent ease with which the Kermans breached such powerful defensive works. They did replace another family in the office of Commandant, but this need not indicate that Sepedhor's father betrayed his Egyptian compatriots to the Kermans. It could 64 Egyptian Pottery a. b. c. d. Plate 3.4 New Kingdom pottery from Askut: a) incense burner, note the characteristic pattern of burning; b) bowl, note the rilling patterns from finishing on the wheel; c) red polished, carinated bowl with inurving rim; d) red "pattern burnished" jar of the late Eighteenth Dynasty. rather simply reflect a shift in politics at a critical transition. 1 _ !Naiv Nuinpteya su hw htteEyta xaraes _eel Pactezdte late Middl Kingdompteyfom Bokth the Egyptinse aundr nthe Khermn chadaeveris-i Pots and Politics 65 a.C I ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~11Ns110 |111-1 Plate 3.5 a) Askut, view across Main fort to Southeastern Sector; b) Askut, view of upper layer of Southeastern Sector; c) niche and mastaba in Room 12. c _ . ... . .. .. I~~~~~~~~~. . ....1 ... 1 _.. .... .................... re=.~~~~~~~~~~~~~~~~~~~~~~~~~ol . ...... . 1'* i w C v~~~~~~~~~~~~~~~. . ..-..:..i'-,. ,.S,:''........... '' ::. . -,: 66 Egyptian Pottery a. thing to gain from cooperation after the collapse of the Egyptian central administration at the end of the Thirteenth Dynasty. The expatriates were literate and had close con- tacts within Egypt, particularly the south. They were the ideal intermediaries with over a hundred years of experience in the Nubian trade. It is not surprising that Kushite rulers such as Nedjeh placed them in key positions. As for the Egyptians, they could see that the Kermans could impose their rule by force if necessary. The Ruler of Kush also still controlled the all-important sources of, or routes to, the luxury goods from the south. It is a measure of the trust the Kerma leaders placed in them that only light garrisons were established in Lower Nubia. Just as under the kings of the late Middle Kingdom, the brunt of both defense and trade was still undertaken by the expatriates. This in itself is more consistent with a system taken over intact than with one rebuilt after a hiatus. If the Kermans were not responsible, who did sack Buhen? The Egyptian army of Kamose, with its expertise in siege warfare, honed in the ongoing Hyksos wars, would have been well equipped to reduce the fortifications around Buhen per- haps the seat of Egyptian expatriate/Kerman rule. Unlike an attack coming from Kerma, the widely spaced fortresses of Lower Nubia could be easily bypassed by an invading Egyptian army. The Second Cataract, however, would block all progress south, and Buhen, the strongest fortress yet encountered, would have provided a natural rallying point for the dispersed Kerman forces. The Egyptian expatriates to the south, and Pots and Politics 67 b. Plate 3.6 a) Room 2, Middle Kingdom pottery in situ; b) Room 7, bread molds and other Middle Kingdom pottery in situ; c) Room Southeast 14, New Kingdom pottery in situ. C. ~ ~ ~ _1 !|1 68 Egyptian Pottery s . ~~~~~~~~~~~~~~~~~~~~~~~~~~~~. o~~~~t m mm .... .*. K Z s ...... ; .. ..~~~~~~~~~~....... < ) ~~~~~~~~~~~~~.. . -.A... . . ,~~~~~~~~~~~~~~~~~~~~~~~~~. o...... ,,,..... ..... -. ^ _ w ~~~~~~~~~~~~~~~~~~~~~~. . . . . . . . .. . . . . . . . . . . . ......... .. .XS.. , t ~~~~~~~~~~~~~~~~~~~~~~....... , .... ........ g ~~~~~~~~~~~~~~~~~~~~~~..... _..... ,Z....... .. . . . t.2.2.... UoU l - / . . o./....._ ~o 0 ~~~~~~~~~~~~1~~~~~..... ii"'~~~~~~~~~~~~~~~~~~~~~~~~... ... Pots and Politics 69 probably throughout all of Lower Nubia, would quickly see the inevitable and swear obedience to Pharaoh, even as they had done to the Ruler of Kush less than a hundred years before. High officials under the Kushites would, of course, be deposed, and Egyptians put in their place. Thus we hear no more of the family of Ka and Sepedhor. Some individuals, particularly at Buhen, may have been taken captive to Egypt or been slain in the fighting, but the majority of expatriates and their friends (and possi- bly relations) among the native rulers would have been key supporters and advisors of the new regime. Although these remarks must remain for the moment somewhat hypothetical, a reassessment of the evidence at Buhen does indicate that the sack could have taken place under Kamose, a possibility which both Emery and H. S. Smith admit (Emery, H. S. Smith, and Millard 1979, 3; Smith 1976, 81). Part of the problem with Emery's interpretation lies in his assumption that any light modifications, such as floor raisings, bins, narrow dividing walls, and vaulting, often but not always associated with Kerma pottery, represented the crude work of Kerma "squatters" at a time when the buildings were mere burnt-out shells. As Janine Bourriau has pointed out, one can hardly imag- ine Sepedhor building a new temple and trying to maintain an elite lifestyle in such a context (Bourriau 1991, 132). Her solution was to place Sepedhor after the end of Middle Kingdom control, but before the establishment of Kerma garrisons, marked by Buhen's violent destruction. This need not, however, be the case. Light structures and floor raisings up to a meter are commonplace at Askut in all periods, and Kerma pottery frequently oc- curs, but always in association with a predominantly Egyptian cultural assemblage. Modifications at Buhen reflect a similar pattern to those at Askut. The floors and walls of the "Commandant's House" were maintained into the New Kingdom. Fill interpreted at Buhen as debris from the weathering of abandoned structures is of a similar depth (40 cm to 90 cm) and description (sherds, brick detritus, and sand) as floor raisings in standing structures at Askut. It is also unlikely that the Kermans would go to the trouble of digging out the ruins to their original floors just to build light structures. Kerma pottery often appears at floor level and thus in association with and not above the burnt layers, and in at least one case was sealed under a col- lapsed burnt vault (e.g., House E, and in Block J; Emery, H. S. Smith, and Millard 1979, 61, 72ff.).16 There is also no solid evidence, in the form of associated pottery or datable artifacts, that the defensive system was damaged and allowed to decline in the Second Intermediate period. Indeed, Sepedhor's boast of building a new temple of Horus, Lord of Buhen, hardly indicates neglect. Some sort of reconstruction is indi- cated not long after the New Kingdom reconquest by an inscription of Year 3 of Kamose (H. S. Smith 1976, 206). This need not, however, indicate that the defenses and buildings had been completely rebuilt. A thorough restoration might well have waited until later, allowing enough time for the debris to build up. Indeed, Smith implies that the final restoration of the fort was only completed by the reign Thutmose I under Commandant, later Viceroy, Turi (ibid., 208). It is not possible within the scope of this paper to present a complete reanaly- sis of Buhen, but, all in all, it is at least plausible that the light modifications were made while the structures at Buhen were still standing and in use, before the violent destruction of the fort by burning.17 The amount and some concentrations of Kerma Classique pottery do indicate that there were probably Kermans living inside the walls, but most likely in the context of a thriving community of Egyptian expatriates. 70 Egyptian Pottery a. _~~~~b Plt . )Ro otes 2,ata ihsea ria loee in bakron b)Scnlnemdaepro n NwKndmpteyi iunab Pots and Politics 71 Plate 3.8 a) View of altar in Room Southeast 32a, showing stratigraphic section with mud floors, fill, tile floor; b) view of the tile floor in Room 23b with jar inset, taken from Room Southeast 32a. Note the step up from 32c into 47 in the background, which had been partly filled with rubbish in the Second Intermediate period. b. 72 Egyptian Pottery Emery himself noted that the amount of Nubian pottery only indicated a small occupa- tion by the Kermans (Emery, H. S. Smith, and Millard 1979, 3). This reconstruction also solves the problem of the Buhen horse, which was found at the base of the Middle Kingdom defense walls, under burnt debris related to the fort's sack, and thus dated to the late Thirteenth Dynasty.'8 Horses are otherwise unattested in the Middle King- dom, and chariot warfare is nowhere mentioned or depicted. 19 Yet the Buhen horse had been broken to a bit, implying that it was part of a chariot team (Clutton-Brock 1979, 192). Even if horses were used that early, which seems unlikely, what would it be doing on the Nubian frontier? A horse would, however, be expected to appear in an army of the late Seventeenth Dynasty, as chariot warfare became the norm. It was old, and might equally well have died of natural causes or in combat and been placed or buried at the base of the wall before most of the debris had accumulated. Tthe continuing presence of Egyptians in Nubia helps explain the rapid accul- turation of the Nubian elite (S. T. Smith 1991a, 90ff.). Save-S6derbergh (1949, 57- 58) has long argued that Egyptians serving the Ruler of Kush were a key factor, de- veloping a familiarity and taste for Egyptian goods and customs within both the Kermans and especially the C-Group. At Askut, for the first time, we have seen that these were no mere mercenaries, come from Egypt in temporary service, or even a small clique of Egyptian "collaborators," but full-time residents who had called Nubia home for at least eight generations. They survived both the fall of the Middle King- dom and the reconquest of Nubia at the beginning of the New Kingdom, prospering all the while. Their presence profoundly altered the existing social, economic, and political structure of Lower Nubia, opening new opportunities to the conquering Egyp- tians. Still culturally Egyptian, but with profound contacts with C-Group and Kermans, they would have provided the needed infrastructure to make acculturation colonial- ism more appealing than the simple occupation of the Middle Kingdom. The settlement flourished during the Eighteenth Dynasty. The house com- prising Rooms 51-56 was apparently built early on, before trash had accumulated in front of the main gateway (fig. 3.15). Rooms 50-51 were added later on trash depos- its of about half a meter. Sealings indicate that the Chapel was added by at least the reign of Thutmose Il, and ceramics indicate that it was in use until the late Eighteenth Dynasty. By the end of the Eighteenth Dynasty the Main Fort had apparently been completely abandoned, but the houses of Rooms SE 5-21 and SE 29-36 were filled with trash and rebuilt using the tops of the old walls as foundations. They continued to be occupied into the Twentieth Dynasty. Construction on a grand scale was still undertaken in the Ramesside period, as a pot used for drainage set in the floor of the poorly preserved house(s) of Rooms 58-73 indicates (fig. 3.8L). The occupation at Askut during the Twentieth Dynasty touches on another question, the supposed "de- population" of Nubia in the Nineteenth Dynasty (Adams 1977, 241-43). Although there is a decline in the number of graves and settlements starting in the Eighteenth Dynasty, it would appear that this is more likely the result of the acculturation process rather than environmental or political stress causing emigration. With an increasing number of natives impoverished and concentrated in a few important towns, the num- ber of settlements and graves with substantial funerary trappings, allowing their tombs to be dated, would diminish. In contrast to the archaeology, the historical record reveals ample evidence of agricultural and administrative activity throughout Nubia at this period (Kemp 1978, 39ff.; Morkot 1987). It is also rather unlikely that Askut would continue to thrive while Lower Nubia became a wasteland. Pots and Politics 73 Relatively intact domestic architecture and associated settlement debris are rare from Nubia, and indeed in Egypt itself. A preliminary analysis reveals a thriving community stretching from the Thirteenth Dynasty through the late Second Interme- diate period and on into the New Kingdom. It also provides us with a large group of stratified Egyptian and native Nubian ceramics spanning the period of ca. 1850 B.C. to 1100 B.C., unbiased by the practice of discarding "non-diagnostics" and "dupli- cates" so common in early excavations. The ongoing study of this small community can reveal much about the development of Egyptian pottery during this period, as well as answer broader questions about the history and development of Egyptian co- lonialism in the region. Acknowledgments: I greatly appreciate the help of Dorothea Arnold and, most re- cently, Janine Bourriau in the characterization of the Askut fabrics. The identifica- tions are based principally on a discussion between myself and Ms. Bourriau, who had the opportunity to examine the sherds in June 1992. I remain responsible for the descriptions and final identifications. I am grateful as well to Dorothea Arnold, Manfred Bietak, Jack Holladay, Pamela Rose, and Janine Bourriau, who patiently looked through drawings and pottery from Askut and answered many questions. The recovery of the Askut field notes and photographs from Badawy's papers at the American Research Center in Egypt (ARCE) in Cairo was made possible in part by the Southern Califor- nia Chapter of ARCE. Many thanks also are due for the support and help given by volunteers from this group and the Friends of Archaeology at UCLA, notably Noel Sweitzer, Ed Johnson, Jack Lissack, Art Muir, and Maissa and Roger Sanders. Initial drafts of several ceramic drawings were made by Jill Ball, Jack Lissack, Robert Masterson, Cynthia Fowler, and Alexander Badawy and/or Martha Wilcox. Final drawings were made by myself, Jill Ball, and Jack Lissack. The field and pottery photographs were taken by the late Alexander Badawy. I am also grateful for con- tinuing support from the Fowler Museum of Cultural History at UCLA. NOTES Mirgissa was apparently excavated to a high standard, but unfortunately only the cemeteries have appeared in a final publication. 2 Forthcoming. It has not, however, proven possible to publish Badawy's Askut manuscript, which will remain on file at the Museum. 3 Badawy was somewhat overenthusiastic in attributing any oven-like structure with associated sherds and burning as a kiln (e.g., 1964, 51). In fact, most of these are simply ovens, and nothing resembling a pottery kiln occurs at Askut at any period. 4 In contrast to examples from Deir el-Ballas and Memphis, the variety with convex interior does show consistent wear patterns, suggesting that these vessels were abraded during use, perhaps in grind- ing or mashing some sort of soft material. 5Simple figurines of crocodiles also occur at Askut, and Sobek appears in names and offering for- mulae on a stela, statue, and in graffiti (Badawy 1966, 25). 6 J. S Holladay, Jr. and Janine Bourriau, personal communications, 1992. 74 Egyptian Pottery 7 For a similar interpretation of such a pattem, see Bourriau (1991, 131). 8 Personal observation by the author at the Boston Museum of Fine Arts. 9 Household shrines are well known at Amama and Deir el-Medineh, and the earliest example be- fore Askut was during the reign of Amenhotep III (Badawy, 1968, 65-8, 94). 10 Rooms which were left blank were either disturbed or have not yet been assigned a secure date. The magazine-granary complex (Rooms E1-17) was subject to massive disturbance in later, perhaps Meroitic, times. Through careful analysis it may prove possible to assign dates to more of the rooms. New numbers were not given to the earliest buildings in the Southeast Sector. Instead, room numbers were projected downward from the plan of the upper level (see fig. 3.15). 11 Some Second Intermediate period groups may represent a combination of peripheral trash dis- posal, raising of house floors and/or fill used for the new construction. Ongoing quantification is focus- ing on providing a more detailed history of these structures. 12 Dever, along with Ward, also questions the Hyksos period date of the rdi-R' scarabs, arguing that they occur as early as the late Twelfth Dynasty (Dever 1991, 75). This, however, goes against the evidence from Lower Nubia, where this kind of scarab is clearly associated with the period of Kerma control (=Fifteenth Dynasty; Vercoutter 1976, 281ff.). 13 Some variation in the level of deposits can be expected since Badawy measured depth from the tops of the walls rather than a single datum point. Mixed Middle Kingdom and Second Intermediate period deposits might date either to the latest Thirteenth Dynasty or the very early Second Intermediate period (see above). 14 Adams (1977, 190) argues, however, that the evidence need not indicate a violent sack, but simply deliberate or accidental buming during the occupation of the site or upon its abandonment. 15 Emery, like many Egyptologists of his day, tended to view the Kermans as rude barbarians, a sentiment the ancient Egyptians would no doubt have heartily endorsed! It is clear from the work of Charles Bonnet at the settlement of Kerma, however, that they should be regarded as a highly complex state society, with a long tradition of urbanism (Bonnet 1990; O'Connor, 1991). It seems unlikely that a permanent occupying force would be any more content merely to "squat" than the Egyptians themselves. 16 Emery argued that this indicated two burning episodes, one by the Kermans destroying the original building, another by Kamose destroying the "squatters" light structures. This reconstruction, however, seems a bit forced, and one wonders how he could distinguish two episodes of burning when there was no debris from the first. 17 Given the inherent problems in the publication, outlined by H. S. Smith (Emery, H. S. Smith, and Millard 1979, vii, 44, 93), it may never be possible to make a positive statement. I hope to examine the original field notebooks and photographs, which may provide more detailed information to assess my hypothesis. 18 Note that a radiocarbon date from the burnt debris above the horse only provides a date for the wood burnt, probably from a timber parapet or other works on the defence walls. The corrected date of 2070 ? 160 B.C. indicates a period roughly contemporary with the construction of the defenses in the early Twelfth Dynasty. The true date of the horse thus rests on the date assigned to the buming from other evidence (Burleigh 1979, 196). 19 They first appear at Tell el-Dab'a at the start of the Hyksos period with Stratum E/2-b/1, ca. 1640/ 1620 B.C. Pots and Politics 75 REFERENCES ADAMS, W. Y. 1977 Nubia: Corridor to Africa. London: Allen Lane. 1979 "On the Argument from Ceramics to History: A Challenge Based on Evidence from Medieval Nubia." Current Anthropology 20: 27-44. 1986/87 "Times, Types, and Sites: The Interrelationship of Ceramic Chronology and Typology." Bulletin of the Egyptological Seminar 8: 7-46. AMIRAN, R. 1970 Ancient Pottery of the Holy Land: From its Beginnings in the Neolithic Period to the End of the Iron Age. New Brunswick, N.J.: Rutgers University Press. ARNOLD, Do. 1982 "Keramikbearbeitung in Dahschur 1976-198 1" Mitteilungen des Deutschen Archaologischen Instituts Abteilung Kairo 38: 25-65. 1988 "Pottery." Chap. 12 in The Pyramid of Senwosret I by D. ARNOLD. The South Cemeteries of Lisht 1. The Metropolitan Museum of Art Egyptian Expedition Publication 22. New York: Metropolitan Museum of Art, 106-46. BADAWY, A. n.d. Askut: An Egyptian Island Fortress of the Middle Kingdom in Upper Nubia. Museum of Cultural History, University of California at Los Angeles. 1964 "Preliminary Report on the Excavations by the University of California at Askut (First Season, October 1962-January 1963)." Kush 12: 47-53. 1965 "Askut: A Middle Kingdom Fortress in Nubia." Archaeology 18: 124-31. 1966 "Archaeological Problems Relating to the Egyptian Fortress at Askut." Journal of the American Research Center in Egypt 5: 23-27. 1968 A History of Egyptian Architecture. The Empire (the New Kingdom) from the Eighteenth Dynasty to the End of the Twentieth Dynasty 1580-1085 B.C Berkeley and Los Angeles: University of California Press. BIETAK, M. 1984 "Problems of Middle Bronze Age Chronology: New Evidence from Egypt." American Journal of Archaeology 88: 471-85. 1991 "Egypt and Canaan During the Middle Bronze Age." Bulletin of the American Schools of Oriental Research 281: 27-72. BONNET, C., ed. 1990 Kerma, Royaume de Nubie: l'antiquite africaine au temps des pharaons: exposition organisee au Muse'e d'Art et d'Histoire, Geneve, 14 juin-25 novembre 1990. Geneva: Mission archeologique de l'Universite de Geneve au Soudan: Musee d'Art et d'Histoire. BOURRIAU, J. 1981 Umm El Ga'ab: Pottery from the Nile Valley Before the Arab Conquest. Catalogue of the Exhibition organized by the Fitzwilliam Museum, Cambridge, 6 October to 11 December 1981. Cambridge: Cambridge University Press and the Fitzwilliam Museum. 1990 "The Pottery." Chap. 15 in Deir el-Ballas: Preliminary Report on the Deir el-Ballas Expedition, 1980-1986 by P. LACOVARA. American Research Center in Egypt Reports 12. Winona Lake, IN: Eisenbrauns, 15-22. 1991 "Relations between Egypt and Kerma During the Middle and New Kingdoms." In Egypt and Africa: Nubia from Prehistory to Islam, edited by W. V. DAVIES. London: British Museum Press in association with the Egypt Exploration Society, 129-44. forthc. "Second Intermediate Period to Thutmose I." In Introduction to Ancient Egyptian Pottery. 76 Egyptian Pottery BRUNTON, G. 1930 Qau and Badari III. British School of Archaeology in Egypt Publication 50. London: British School of Archaeology in Egypt and Bernard Quaritch. BRUYERE, B. 1939 Rapport sur les fouiles de Deir el-Medineh (1934-35). Troisieme partie: Le village, les decharges publiques, la station de repos du col de la valhEe des rois. Fouilles de L'Institut Francais du Caire sous la direction de M. Pierre Jouguet 16. Cairo: Institut Francais d'Archeologie Orientale. BURLEIGH, R. 1979 "The Buhen Horse - C. Attempted Radiocarbon Dating of the Buhen Horse." In The Fortress of Buhen: The Archaeological Report by W. B. EMERY, H. S. SMITH, and A. MILLARD. Excavations at Buhen 1. Egypt Exploration Society Excavation Memoir 49. London: Egypt Exploration Society, 195. CLUTI7ON-BROCK, J. 1979 "The Buhen Horse - B. The Description." In The Fortress of Buhen: The Archaeological Report by W. B. EMERY, H. S. SMITH, and A. MILLARD. Excavations at Buhen 1. Egypt Exploration Society Excavation Memoir 49. London: Egypt Exploration Society, 191-95. COLE, D. P. 1984 Shechem I: The Middle Bronze IIB Pottery. American Schools of Oriental Research Excavation Reports. American Schools of Oriental Research. Distributed by Eisenbraun's, Winona Lake, IN. DEVER, W. G. 1991 "Tell el-Dab'a and Levantine Middle Bronze Age Chronology: A Rejoin- der to Manfred Bietak." Buletin of the American Schools of Oriental Re- search 281: 73-79. DIXON, D. M. 1972 "The Disposal of Certain Personal, Household and Town Waste in Ancient Egypt." In Man, Settlement and Urbanism, edited by P. J. UCKO, R. TRINGHAM and G. W. DIMBLEBY. Society Proceedings of a Meeting of the Research Seminar in Archaeology and Related Subjects held at the Institute of Archaeology, London University. Cambridge, MA: Schenkman, 647-50. DUNHAM, D., and J. M. A. JANSSEN 1960 Second Cataract Forts. VoL I: Semna, Kumma. Excavated by George Andrew Reisner Boston: Museum of Fine Arts. EMERY, W. B., H. S. SMITH, and A. MILLARD 1979 The Fortress of Buhen: The Archaeological Report. Excavations at Buhen 1. Egypt Exploration Society Excavation Memoir 49. London: Egypt Exploration Society. ENGELBACH, R. E. 1923 Harageh. British School of Archaeology in Egypt and Egyptian Research Account Publication 28. London: British School of Archaeology in Egypt and Bernard Quaritch. FRANKFORT, H., and J. D. S. PENDLEBURY 1933 The City of Akhenaton. Part 2: The North Suburb and the Desert Altars: The Excavations at TeU el-Amarna during the Seasons 1926-1932. Egypt Exploration Society Memoir 40. London: Egypt Exploration Society. GRATIEN, B. 1978 Les Cultures Kerma: Essai de Classification. Publications de l'Universite de Lille 3. Centre National de la Recherche Scientifique, avec la collaboration de la Direction Generale des Relations Culturelles, Scientifiques et Techniques. Villeneuve-d'Ascq: Publications de l'Universite de Lille. 1985 "Le village fortifie du groupe C ai Ouadi es Seboua Est, typologie de la Pots and Politics 77 ceramique." Cahier de Recherches de l'Institut de Papyrologie et d'tgyptologie de Lille 7 (Societes Urbaines en IEgypte et au Soudan): 39-70. 1986 Sai I: La Necropole Kerma. Centre National de la Recherche Scientifique, Mission Archeologique Francaise au Soudan, Institut de Papyrologie et d'Egyptologie de l'Universite de Lille 3. Paris: Editions du Centre National de la Recherche Scientifique. HOFFMAN, M. A. 1974 "The Social Context of Trash Disposal in an Early Dynastic Egyptian Town."American Antiquity 39: 35-50. HOLSCHER, U. 1939 The Excavations of Medinet Habu. Vol. 2: The Temples of the Eigh- teenth Dynasty. Oriental Institute Publications 41. Chicago: University of Chicago Press. HOLTHOER, R. 1977 New Kingdom Pharaonic Sites: The Pottery. The Scandinavian Joint Expedition to Sudanese Nubia Publications 5: 1. Scandinavian University Books. Lund: The Scandinavian Joint Expedition to Sudanese Nubia. HOPE, C. A. 1987 "Innovation in the Decoration of Ceramics in the Mid 18th Dynasty." Cahiers de la Ceramique Agyptienne 1: 97-122. 1989 Pottery of the Egyptian New Kingdom: Three Studies. Victoria College Archaeology Research Unit Occasional Paper 2. Australia: Victoria College. KEMP, B. J. 1989 Ancient Egypt: Anatomy of a Civilization. Cambridge: Cambridge University Press. MAYSTRE, C. 1980 Akasha I. Mission archeologique de la Fondation Henry M. Blackmer et du Centre d'etudes orientales de l'Universite de Geneve sous la direction de Charles Maystre 1. Geneva: Georg. MERRILLEES, R. S. 1968 The Cypriot Bronze Age Pottery Found in Egypt. Studies in Mediterra- nean Archaeology 18. Lund: Studies in Mediterranean Archaeology. MORKOT, R. 1987 "Studies in New Kingdom Nubia 1. Politics, Economics and Ideology: Egyptian Imperialism in Nubia." Wepwawet 3: 29 MOUNTJOY, P. A. 1986 Mycenaean Decorated Pottery: A Guide to Identification. Studies in Mediterranean Archaeology 73. Goteborg: P. Astrom. NICHOLSON, P. T., and H. L. PATlTERSON 1989 "Ceramic Technology in Upper Egypt: A Study of Pottery Firing." World Archaeology 21: 71-86. NORDSTROM, H. A. 1985 "Ton." Lexikon der Agyptologie 6: 629-34. NORDSTROM, H. A., and J. BOURRIAU 1993 "Ceramic Technology: Clays and Fabrics." Fasc. 2 inAn Introduction to Ancient Egyptian Pottery, edited by Do. ARNOLD and J. BOURRIAU. Mainz am Rhein: Philipp von Zabemr, 144-90. O'CONNOR, D. 1991 "Early States along the Nubian Nile." In Egypt and Africa: Nubia from Prehistory to Islam, edited by W. V. DAVIES. London: British Museum Press in association with the Egypt Exploration Society, 145-65. PEET, T. E., and C. L. WOOLLEY 1923 City of Akhenaten, PartlI: Excavations of 1921 and 1922 at El-'Amarneh. Egypt Exploration Society Memoir 38. London: Egypt Exploration Society. 78 Egyptian Pottery PETRIE, W. M. F. [1891] 1974 IUahun, Kahun and Gurob. London: David Nutt. Reprint, Warminster: Aris and Phillips, with Joel L. Malter, Encino CA, U.S.A. 1907 Giza and Rifeh. British School of Archaeology in Egypt and Egyptian Research Account Publication 13. London: Brititsh School of Archaeology in Egypt and Bernard Quaritch. RICE, P. M. 1987 Pottery Analysis: A Sourcebook. Chicago: University of Chicago Press. RYE, 0. S. 1981 Pottery Technology: Principles and Reconstruction. Manuals on Archaeology 4. Washington, D.C.: Taraxcum. SADR, K. 1988 The Development of Nomadism: The View from Ancient Northeast Africa. Ph.D. dissertation, Southern Methodist University. Ann Arbor: Univer- sity Microfilms. SAVE -SODERBERGH, T. 1949 "A Buhen Stela (Khartum No. 18)." Journal of Egyptian Archaeology 35: 50-8. SCHIFFER, M. B. 1987 Formation Processes of the Archaeological Record. Albuquerque: University of New Mexico Press. SMITH, H. S. 1972 "Society and Settlement in Ancient Egypt." In Man, Settlement and Urbanism, edited by P. J. UCKO, R.TRINGHAM and G. W. DIMBLEBY. Society Proceedings of a Meeting of the Research Seminar in Archaeology and Related Subjects held at the Institute of Archaeology, London University. Cambridge, MA: Schenkman Publishing, 705-19. 1976 The Fortress of Buhen: The Inscriptions. Excavations at Buhen 2. Forty- eighth Excavation Memoir. London: Egypt Exploration Society. SMITH, S. T. 1990 'The Administration of Egypt's Southern Frontier: Middle Kingdom Sealing Practice at Uronarti and Askut Forts." InAegean Seals, Sealings and Administration: Proceedings of the NEH-Dickson Conference of the Pro- gram in Aegean Scripts and Prehistory of the Department of Classics, University of Texas at Austin, January 11-13, 1989, edited by T. G. PALAIMA, 197-216. Aegaeum 5. Liege: Universite de Liege, Histoire de l'art et archeologie de la Grece antique. 1991a "A Model for Egyptian Imperialism in Nubia." Gottinger Miszellen 122: 77-102. 1991b "Askut and the Role of the Second Cataract Forts." Journal of the American Research Center in Egypt 28: 107-132. 1993 "The House of Meryka at Askut and the Beginning of the New Kingdom in Nubia." In Atti del Sesto Congresso Internazionale di Egittologia, VoL 2. Turin: International Association of Egyptologists, 497-509. STEINDORFF, GEORG 1937 Aniba. VoL 2: Plates. Mission archeologique de Nubie 1929-1934. Service des Antiquites de l'Egypte. Gluckstadt: J. J. Augustin. VERCOUTI'ER, J. 1957 "Upper Egyptian Settlers in Middle Kingdom Nubia (Stelae Khartoum Mus. 11778, 372A and 2647, Statue Khartoum Mus. 5516)." Kush 5: 61-9. 1970 Mirgissa I. Mission Archeologique Fransaise au Soudan sous la direction de Jean Vercoutter 1. Ouvrage publie par la Direction Generale des Relations Culturelles, Scientifiques et Techniques, Ministere des Affaires Etrangeres avec le concours du Centre National de la Recherche Scientifique. Paris: Mission Pots and Politics 79 Archeologique Francaise au Soudan. 1976 Mirgissa III: Les Necropoles. Mission Archeologique Francaise au Soudan. Paris: Mission Archeologique Francaise au Soudan. WILLIAMS, B. 1987 "The Nubian Publication Project." In The Oriental Institute 1986- 1987Annual Report. Chicago: Oriental Institute, 57-60.