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

Pottery - S. Wallace-Jones

General remarks
All of the ceramic material from the site, including both body sherds and diagnostics, was examined and as this represents a considerable corpus of material certain typical areas of the site have been selected for detailed analysis. In particular WG 32, WG 49, and WG 51 are discussed in depth here because they are very representative of other WG units on the site. Finds within other WG units are also discussed when they have particular significance. The overall chronological profile of the site is clear and uniform, dating to the 12th Dynasty with occasional indications of the early 13th Dynasty, especially in the development of ceramic technology. Many jars show a typical Middle Kingdom combination of turned upper sections with a hand or coil made base scraped on the outer surface. (Arnold and Bourriau 1993).
The profile of cups is generally rather shallow, suggesting an earlier rather than later Middle Kingdom date, whilst the technology used for bases shows evidence that wheel turning was becoming important. A number of cups have bases which were cut to shape on the exterior surface; one in particular from WG 47 showed clear internal rilling lines, whilst the exterior was rather clumsily cut to shape. There are also several round bases from small jars which show clear internal evidence of turning, having a spiral in center interior; these along with small plates also have evidence of scraping on the exterior. Of the ring bases at the site at least one in Marl A variant 3 is clearly wheel-made and applied to the vessel, and one in Nile B1 appears to have used some hand modelling. All of this evidence supports a date of 12th to early 13th Dynasties with the emphasis more on a 12th Dynasty date. The cupped rim usually found in Marl C fabric and very typical of Mersa/Wadi Gawasis is not well known at other sites in Egypt; however, it can be dated with confidence to the period between the mid-12th and early 13th Dynasties because it is always found in combination with a few other rim types, and in particular Bader type 46 corrugated rims and Bader zir rims of types 2 and 3 (Bader 2002).
The difference between WG 49 and areas such as WG 32 and WG 51 form an interesting comparison since WG 32 and WG 51 reflect one trend at the site, and WG 49 another. WG 32 and WG 51 illustrate areas where a wide range of forms and fabrics is present, representing an eclectic mix of vessels, technology and functions, whilst WG 49 represents areas containing only large storage vessels in limited numbers of shapes and mainly in Marl C and its variants. WG 51 may be compared to areas such as WG 32 and WG 33 and also WG 55 around the mouth of the caves, whilst WG 49 forms a parallel with WG 45, WG 46 and WG 47. This in turn would suggest that the area around WG 49 identified as the "harbor area" had a clearly defined storage function, whilst the areas defined by WG 32, WG 51, etc. have a much more complex and wide ranging set of functions (Figure 26).

WG 32, Body Sherds
Clays have been defined using the Vienna system. (Bourriau 1981; Arnold and Bourriau 1993):
Marl A variant 3: from several medium and large jars with clear evidence of rilling lines on the interior surface and hand making marks on pieces from lower sections, between 6 mm and 15 mm thick; also pieces from a number of different shallow profile cups varying in thickness between 2 mm and 5 mm, as well as several small plates. One piece of fine quality from the neck of a small bottle wheel-made and finely finished with a thick burnish on the whitish cream firing surface. This represents a finely made high quality vessel. There were also 3 pieces from the shoulder of a fine wheel-made jar about 5 mm thick, possibly from a small drop pot. Three pieces of this fabric from the body of a large jar were considerably coarser than the rest with a high proportion of sand inclusions, which could almost fit the profile for Marl A variant 4.
Nile B1: many body sherds from small fairly shallow cups and also flatter, more plate like profiles with interior evidence of rilling, but clear exterior scraping. Body thickness varies between 3 mm and 6 mm.
Nile B2: evidence of a number of medium and large closed vessels, as well as some open forms, including a large platter with a red inner coating and body thickness of ca. 9 mm. There are sherds from jars, including a large jar of about 19 mm in thickness with a red outer coating, as well as some body sherds from smaller bottle type vessels, both plain and red-coated, some with traces of burnishing. There is also evidence of some small cups of body thickness 4-6 mm.
Nile C: 8 pieces of very thick (20 mm) body material from one large jar, covered on the outer surface with a red wash. There are also pieces from a large platter.
Marl C: many sherds from a variety of large and medium jars with wall thicknesses between 6 mm and 15 mm.
Marl C compact: many sherds from a variety of large jars. As with other contexts this fabric appears quite regularly and in unusually large quantity.
Marl C2: many sherds from a variety of large jars.
Nile E: 1 piece of heavily smoke-stained body, probably from a cooking pot. This fabric seems to originate in the Eastern Delta and has particular prominence at Tell el-Daba.
Bread mold: a number of sherds of coarse, sandy, tubular bread molds were also found.
There was also a small quantity of non Egyptian material some of which seems likely to be Middle Nubian. This material has been studied by Andrea Manzo (see below).

WG 32, Diagnostics
As would be expected, the diagnostic sherds from WG 32 represent the range of styles indicated by the body sherds. Notable in this context are a large number of Nile B2 and Marl C rims of Bader bottle types C and D, and also type 7B dating to the late 12th and early 13th Dynasties, as well as large rims with diameters up to 27 cm of Bader type 2 and 3 zirs, dating to the mid-12th Dynasty (Bader 2002). Also in Marl C are some flat bases and examples of the cupped rim, which is very typical of Mersa/Wadi Gawasis but unusual elsewhere (Janine Bourriau personal communication). A number of very typical plates and plate rims occur from vessels with diameters around 16 cm. These are always of Marl A variant 3 and have characteristic rilling lines at the rim, but a scraped finish on the exterior of the base. Such vessels occur quite commonly at the site and may have been used as a standard size ration plate, and may also have had a function as lids. There are also many fine cup rims in both Marl A variant 3 and in Nile B1, and to some extent in B2. One Marl A variant 3 rim is of an interesting funnel shape, resembling a 12th Dynasty water jar but much smaller than the usual size of these vessels, having a diameter of only 6 cm and a vessel wall thickness of about 2-3 mm. Its function remains uncertain although it could be a miniature. It is clear from the material found in WG 32 that this area of the site was being used for a wide range of activities and that as a result the ceramic profile is very rich and varied. The same is also true of WG 51.

WG 51, Body Sherds
Marl A variant 3: from a number of different jars including one large vessel with wall of 13 mm thickness and one of 15 mm. Also one from a smaller jar with wall of 10 mm thickness, and from fine cups with thicknesses of 2-4 mm.
Nile B1: 3 sherds probably from the same vessel: a small cup with red slip inside and out and hand trimming on the base. 6 mm at base, tapering to 2 mm toward the rim.
Nile B2: body sherds from several large closed vessels; one up to 19 mm thick with some evidence of red slip on the outer surface. Other pieces are uncoated. A few pieces are probably coarse enough to classify as Nile C.
Marl C: a large quantity of body sherds from storage jars varying between 5 mm and 9 mm in thickness.
Marl C compact: a large quantity of body sherds from big jars up to 14 mm thick and one smaller jar up to 6 mm thick. There was a surprisingly high proportion of Marl C compact at the site, representing up to one-third of all Marl C storage vessels.
Marl C2: a small quantity of Marl C2 body sherds all from large closed vessels.
Nile E: 2 body sherds from a closed vessel and one probably from a large open form with coarse interior burnishing.
Five pieces of tubular bread mold in a coarse straw-tempered silt fabric without coating. This may be a local clay, but has close affinity with Nile Silt B2/C.
In addition, there were some foreign sherds, a number of which appear to be Middle Nubian, which are the subject of further study by Andrea Manzo.

WG 51, Diagnostics
Three folded Marl C zir rims from bag-shaped jars ca. 24 cm in diameter, wall thickness 12 mm, and a similar rim but in Marl C compact. These reflect types found at Tell el-Daba in the 12th Dynasty, Bader type 1 (Bader 2002). Three Marl C rims from vessels of Bader type 46 and of late 12th Dynasty date (Bader 2002). A small folded rim from a Marl C compact jar (8 cm in diameter) and also a flat base of Marl C from a large zir. Also one cup-shaped rim of Marl C, which is the most commonly found shape at Mersa/Wadi Gawasis. A Marl A variant 3 rim from a small water jar, 11 cm in diameter, of early-mid 12th Dynasty type, and at least 6 different Marl A variant 3 cup rims, between 2 mm and 5 mm thick, tapering to very fine rims, one of the rims being notched before firing. A large rim from a Nile B2 platter with a red coating inside and out, and one similar but uncoated. A Nile B1 cup rim with a shallow profile tapering to 2 mm at the rim, 2 pieces of Fine Nile B2 (almost B1 but with some large straw pieces) with red slip and burnishing inside. The form is a large open vessel, which is too wide and flat to be a cup.
Clearly, both units WG 32 and WG 51 contain a wide range of vessels in many different fabrics and forms, suggesting that these areas had a very mixed use that probably combined small-scale storage with food consumption and preparation. The collection and storage of water or other beverages may also have formed an important activity in this area as a number of bottle type vessels are attested there. The presence of foreign ceramics including Middle Nubian sherds and probably material from farther south and east is also of interest, and these sherds have been studied by Andrea Manzo.

WG 49, Body Sherds
This unit contains only a huge number of rim and body sherds of Marl C compact, Marl C, and Marl C2 all of which come from large jars. Marl C is the most commonly occurring fabric accounting for about 70% of the Marl C material; about 25% is of Marl C compact and 5% Marl C2, which is often very coarse. There is also a significant quantity of Marl A variant 3 in use for large storage jars. This material is comparable with types found in the Nile Valley and the fabric is known to be of Upper Egyptian origin. Therefore, it must have been brought from the Nile Valley to Mersa/Wadi Gawasis. In addition, there is a small but significant quantity of large jars, both plain and red coated, made in Nile B2, again likely to have been imported from the Nile Valley.

WG 49, Diagnostics:
Rims of Marl A variant 3 from bag-shaped jars, Bader rim type 2 (Bader 2002).
A large number of Marl C type 47 rims with diameters around 12 cm.
Many rims of cup type typical at Mersa/Wadi Gawasis in Marl C.
Many rolled type jars, rims in Marl C and C1, of Bader types 1 and 2 (Bader 2002).
There is no variation in the material from this context and judging from the size and quantity of both body sherds and rims from this area, it is quite clear that it was the location of a significant number of storage jars, mainly of Marl C and its variants, but including small but important quantities of Marl A variant 3 and Nile B2. There are no other vessel types in this unit and it seems that there is also a limited number of types of jar rims present, confined to Bader rim types 1 and 2 of rolled type, corrugated rims of type 46/47 (Bader 2002), and the most commonly found cupped type, which is common at Mersa/Wadi Gawasis but unusual elsewhere. Therefore, it would appear that the range of jar shapes was limited to a few types and that these had a specific storage function. The few bases discovered in these areas are large and flat and the technology of these jars is typical of the Middle Kingdom, having wheel-made upper sections with clear rilling lines joined to coiled and hand-made lower sections with the outer surface scraped and trimmed to shape (Arnold and Bourriau 1993).
Units WG 45, WG 46 and WG 47 are very similar in their composition and closely resemble WG 49, although WG 47 was also found to contain a number of bottle type rims with a kettle-shaped profile Bader type 7b (Bader 2002), dating to the early 13th Dynasty. These were quite uniform, being made from Nile B2, some uncoated and some with red exterior wash and all of about 9 cm in diameter. However, these could also be consistent with a storage function in this area. Also from WG 47 were a number of foreign sherds, possibly both Middle Nubian and South Arabian, which are being studied by Andrea Manzo. One handle from this area appears to be from a Canaanite jar. The dark brown paste is clearly different from both Nile and Marl clays, being extremely coarse and very gritty with a considerable quantity of black particles. The paste, form and technology all suggest a Canaanite jar (Janine Bourriau personal communication). This would not necessarily be inconsistent with a storage area, since the jar may well have been imported as a container for its contents and could also have been re-used. It is the first Canaanite jar known from the site but the presence of wood such as oak and cedar at the site seems to confirm the likelihood of contact with this area[23]. A further interesting rim from WG 47 is that of a large cooking pot made from Nile E fabric which is very typical of those found at Tell el-Daba. It is one of the only sherds from this area which is not a storage jar. However, other pieces of Nile E have been found at the site and its presence is not, therefore, a particular surprise as a single rim could easily have been deposited in this area without changing its essential storage function in any way. What is more interesting is the presence of a fabric known to originate in the Eastern Delta, alongside the handle of a Canaanite amphora.
Amongst other items of significance at the site are three very small sherds which may, in some way represent miniatures or model vessels. One of these, the water jar form, from WG 32 has already been described above; another, from WG 55/WG 56 is a tiny Marl A variant 3 rim, 1-2 mm thick and with a diameter of 45 mm. It was found in the entrance to Cave 7 and appears to be similar in from to the ridged rims of Bader types 46 and 47 (Bader 2002), although it is significantly smaller. It was clearly wheel-made and finished very carefully and finely. Its size makes it likely that it is a miniature jar, although its function is at present unclear. The third example, of Marl C fabric, was excavated in the 2006-07 season in WG 38 and may be a model of a zir. It is 9cm in diameter with a typical folded rim and has every appearance except its size of being a large storage jar. It is without parallel at the site. Miniatures are known from early in ancient Egyptian history and occur at many sites where they were used as offerings, in foundation deposits and similar contexts. It is impossible to be sure that this is occurring at Mersa/Wadi Gawasis, but the presence of three potential miniatures raises some questions about the use of miniatures at the site. From these areas it is also quite clear that storage was the major function.
A large number of pot marks was identified, almost exclusively on jars of Marl C and related fabrics with a few from Marl A variant 3. Some of these appear to be related to quantity or counting, whilst others may represent different and as yet unclear meanings. They were all made before firing. An ostracon of Marl A variant 3, clearly from a large jar, was also found to have a hieratic inscription in ink. This material will be studied by the epigrapher and is mentioned here as an adjunct to the ceramic report.
A number of wooden and pottery discs have been found at Mersa/Wadi Gawasis and had tentatively been identified as gaming pieces, net floats etc. However, the discs are not pierced, although they do have a distinctive groove in mid-circumference and in many examples this groove has had string fitted into it. One of the discs also appears to have been marked with a geometric shape. It was noted during the 2007-08 season that these discs may have been used as jar stoppers, with the string acting as a kind of seal whilst the disc fitted into the top of a jar or bottle to prevent the sealing mud from falling into the contents. Many of the bottle and jar rims have a cupped internal ridge and a fairly standard diameter, and this is especially true of the cupped rim seen most frequently at Mersa/Wadi Gawasis where the diameter is often around 15 cm. Bottle rims also seem to have diameters which are fairly consistently sized around 9 cm. There are two sizes of discs at Mersa/Wadi Gawasis: a smaller one often made of a cut down potsherd and a larger size which seems to occur mostly in wood. When tried, the smaller size fitted well into a variety of bottle necks, whilst the larger size fitted the cupped rim very well. Therefore, it is proposed that these discs were used in the necks of jars and bottles to give a better and more effective seal, which might have been of great significance if goods were being provided for long sea voyages.
The foreign material known from previous seasons continued to be well represented and it is clear that Middle Nubian pottery forms a small but significant part of the assemblage along with material from Eritrea, Ethiopia and Yemen. Although the study of this material is not within the scope of this report, it is worth noting that WG 33 produced a large and unusually high concentration of foreign material found together in association with a large number of pieces of Nile B2 platter, Marl C jars, Marl A variant 3 and Nile B1 cups, Nile B bottle necks of Bader type 6 and 7 and rims of Bader type 46 (Bader 2002), in other words, a mixed assemblage typical of the material from this area of the site with a date of late 12th to early 13th Dynasties.
One foreign sherd which is of especial interest was found in WG 55 in association with a mixed range of fabrics and forms similar to the assemblages from WG 32 and WG 51 and dated to the mid-12th to early 13th Dynasties. The tiny, very fine piece of ceramic is coated inside and out with a dark slip and has a fine burnish. The vessel wall is no more than 1 mm thick and the visible rilling lines are also extremely fine. The paste is hard and has very little in the way of inclusions. It was immediately clear that this was not Egyptian and it has now been confirmed that it is from a Minoan cup of pre-Palatial Kamares ware from Crete.[24] Further study may be necessary in order to understand this sherd fully, but it is an exciting addition to the foreign material from Mersa/Wadi Gawasis.


Conclusions
Work this field season has shown that different areas of the site were used for different functions and that the range in the ceramic material found at different areas of the site reflects these varied functions. It is clear that storage was an important activity at the site and that the range of jars used for this was limited to a few types in Marl A and Marl C. It is also clear that a whole range of Middle Kingdom ceramics is present at other areas of the site, which were more likely to be associated with living activities. These include a range of plates, platters bottles, jars and cups, a few of which appear to be of very fine quality. A considerable quantity of the material must have been brought from the Nile Valley, especially the Nile fabric material and that in Marl A variant 3, which came from Upper Egypt. Other material was brought from further afield, such as the Eastern Delta in the case of Nile E, and also from outside Egypt.
It also seems likely that the cupped rim which occurs very frequently at Mersa/Wadi Gawasis was made in a particular workshop which supplied the site. No doubt the site was supplied by different workshops allied to the state and pharaoh, but this rim form is common at Mersa/Wadi Gawasis and is rare elsewhere, suggesting that it may be the product of one particular workshop. However, the range of rims present suggests that this would not have been to only source of supply. The presence of some pieces of large low fired platters in a friable buff clay also continues to support the theory that some ceramic material was produced locally. Ongoing study of the range and distribution of ceramics will continue to play an important role in understanding the workings of the site.

Pleases find attached .pdf files with the full report and the related figures