2007 Field Note 3 - More From the HK24B Predynastic Industrial Site
Excavating at HK24B
Overview of the complex, which evidence suggests is a large-scale food-production site, at the end of the excavations.
After New Years, the time available for the excavations at HK24B flew by in a flash. Although frustrating at time, the piles of potsherds and burnt mud eventually gave up their secrets and hidden surprises, but only after a good deal of earth shifting and detective work.
The excavated area, which ultimately extended to encompass an area about 7.5 m long and more than 5 m wide, yielded a structure with an exposed interior approximately two meters in width and almost six meters in length. Our working hypothesis is that this is a large-scale food production site, perhaps a brewery, but further scientific analysis particularly of the botanical remains is necessary for accurate identification of its function.
The complex is enclosed by walls on the north, west (Feature 1), and south made of stacked up pot sherds cemented together with mud and coated with a plastering of mud on the interior surfaces. In places, finger marks of the ancient builders can still be discerned in the mud plaster, which was made more durable by cycles of heating in the use of the structure. We presume the structure interior is a firebox, originally topped by a makeshift vault of sherds and mud. Indeed, we excavated through great quantities of jumbled, re-used mud-encrusted sherds.
The wall called Feature 1 bounded the structure on the north and west and probably continued to the south.
"Feature 1" is a distinct stretch of wall, constructed of pottery fragments and mud. The intriguing line of mud-coated, vertically set sherds visible on the surface of the desert was what led us to dig here in the first place. This wall is interrupted by three--originally, possibly originally five (the southern part is poorly preserved)--apertures or "doors" that are 20 to 30 cm in width and set at more or less regular intervals. There is no doubt these apertures are deliberately made, as they are neatly and continuously plastered on the interior and across the extant thickness of the wall segment or what we may call the door jambs. In some, vertically set sherds, oriented to capture the north wind, partially blocked the opening, indicating that they could be closed as needed at the end of the firing process or should the weather take a turn for the worse.
The site is sufficiently disturbed that we cannot tell the full thickness of the wall, as the exterior face does not survive intact. Nor can we be sure of the relationship of the wall (and the entire structure) to the compacted pavement found on the west side, which contains postholes. Perhaps the postholes supported a wattle-and-daub screen deliberately situated close to the structure and meant to baffle the wind, which tends to be northwesterly at Hierakonpolis. However, the space between the postholes and the structure is so narrow that passing or working in between would have been difficult. It is possible that the pavement, which shows no evidence of burning or exposure to fire, is a pre-existing structure which made a convenient building site. Having so few examples industrial architecture in Predynastic Egypt and no exact parallels for this structure at all, it is hard to know for sure.
We noted in the last entry that we had found a northeastern end of the structure. Further excavation defined this end more neatly (well, as much as possible). Here we found the third aperture placed at the rounded northern corner of the structure and the continuation of the wall, which connects with a remarkable construction we called Feature 3. An apparently an angular pillar, created again from stacked up potsherds and mud, Feature 3 was plastered on at least its two interior sides. We did not excavate the exterior, so we are not sure whether it was plastered on all four sides, but surface indications suggest it is roughly rectangular in shape, ca. 70x100 cm in dimensions, and as we eventually discovered, preserved to a height of 65 cm above the semi-subterranean base level of the structure's interior.
||Left, At the north end of the structure, the wall of Feature 1 curves around and connects with Feature 3, a pillar or pilaster of potsherd and mud construction. Feature 4, nestled between them is just emerging. Right, the plastered face of Feature 3
Feature 2 from above
Feature 2 bisected showing its internal construction
At the corresponding south end of the structure a feature (Feature 2) of similar construction and shape was also defined beneath a dense tumble of sherds and burnt debris. To ascertain that it was not some sort of oven, we bisected Feature 2 and found it to be entirely sherd and mud filled--a seemingly crude but obviously effective and durable construction technique. The rectangular shape of Feature 2 was clear, with three sides coated with a thick mud plaster retaining the ridges of the mason's fingers. The exterior face may also have been coated with mud, but as it was not in direct contract with the heat on the interior, the mud appears to have eroded away, leaving only indistinct traces.
It took us a while to figure out the function of these features as a sand filled trench, possibly cut in modern times, nearly severed all the relationships between features 1, 2, and 3. However, at the north end, the trench stopped just in time to preserve Feature 4--the emplacement for a large conical vat, nestled in the corners created by the wall of Feature 1 and the pillar of Feature 2.
We all got a little bit excited as we came down on it, thinking we had finally found a vat in situ, unfortunately this was not the case, but why make it easy? Finding intact archaeological features is really just a matter of luck, figuring out what the disturbed remains represent is where the skill and imagination comes in. Although some vat fragments were still held in place by the reddened mud plaster, what we had found was essentially the impression left by a vat in its surroundings of plaster, ash and charcoal, remnants of the concentration of fuel and heat adjacent to the vat. The impression indicated a somewhat conical vat with a maximum diameter of 80 cm, a base diameter of 50 cm (we found the base fragments) and a height of about 65 cm, commensurate with known examples and the vat sherds we had found reused throughout the site. Indeed, vat body sherds, some up to 3 cm thick, were in abundance in Feature 4, with clean breaks and mud only on the exterior (convex) surface. We recovered a considerable quantity of black organic residue from beneath and adhering to the interior surface of these sherds, giving us a good indication of their function in food production. This residue is similar in appearance to that found at the brewery site at HK24A in the 1980s, but provides new and fresh samples for study in order to determine the product involved. While not collected in the best of all possible contexts--from the sides of an intact vat--there can be no question that the residue was present in the vat of Feature 4. Although the disturbed state of this context was a slight disappointment, it is remarkable that the material survives at all after nearly 6,000 years.
||Residue of the foodstuff produced in the vats was found on and beneath the sherds.
Careful excavation also revealed details of the construction. We found that the plastered faces of Feature 2 and the curving north end of Feature 1 extended all the way down to the base of the vat, creating a mud-plastered basin into which the vat had been cemented. Fuel and wood, which had burnt down to white ash, was then placed around the sides of the vat. To maintain the heat, wood was fed into the structure through the apertures, which were actually still filled with a thick deposit of acacia wood charcoal.
Careful excavation revealed the construction sequence of Feature 4.
The stratigraphic section along the sandy trench allowed us to detect the presence of other vat emplacements, evident from the white ashy semicircular lens against the wall segments of Feature 1 flanked by thick deposits of charcoal coming in through the doors.
With the lessons learned from the excavation of Feature 4, it was then possible to recognize the location of other vats, no longer preserved, in the section conveniently cut for us by the sandy trench. The telltale signs were the concave lens of white ash atop blacker ashy sediment mixed with the rubble of burnt mud cement located against each wall segment of Feature 1, and the thick deposits of black charcoal placed to either side where the apertures allowed access to stoke the fire. Excavation to these white ash lenses revealed the same curved ghost impressions of the vat walls as were found in Feature 4. Although no where were vats or vat fragments found in the limited excavation, there can be little doubt that the structure originally contained four large vats, which were probably all heated up at the same time to produce a large amount of foodstuff.
While we can feel fairly confident about the configuration and usage of the northwest side of the structure, the east side remains an enigma. We expected a wall running parallel to Feature 1 along the east side, but instead we found a broad area of black ash resting on a fairly solid and dense base of potsherds at about the same level as the pavement on the exterior of the structure, or in essence at floor level.
Within these black ash-filled areas was a concentration of vegetal material. Soil samples were collected and the botanical material was separated by the technique called flotation. Once the samples are placed in water, the heavier sands and soils sink while the lighter organic remains float to the surface, where they can be collected with a tea strainer. Once they had dried, Dr. Ahmed Fahmy examined them, identifying high quality emmer wheat that was preserved in a charred state. These grains had been dehusked and were ready for consumption, so they appear to have been accidentally lost during either the loading of the vats or the decanting of the product into more mobile receptacle for distribution.
Floating the soil samples to collect the botanic remains
Apart from basket after basket of potsherds and our prized botanical remains, we recovered 14 pottery discs. When perforated at the center, they are generally understood to be spindle whorls (flywheel weights for spinning), and when not perforated, as jar lids. I believe that at a production site where ingredients, fuel, and product change hands these discs might be counters or tokens representing quantities of material. Such counters are recognized in archaeological regions adjacent to Egypt, but rarely in the Egyptian literature.
While this may remain a conjecture, HK24B has yielded an industrial structure that is most likely a food production site for a cereal product, possibly beer. Analysis of the residue and other organics should confirm or refute the brewery function of the site, and might shed new light on the step in the brewing process that vats and residue represent, e.g., malting, mashing, or fermentation.
The plastered pillars made of mud and sherds are fascinating additions to our understanding of Predynastic pyrotechnology as similar features have been unearthed in the pottery kilns at HK11C (back in the wadi) during excavations currently being undertaken by Masahiro Baba of Waseda University. Caches of pottery-making tools and other features there leave no question about the function of that site, thus similar technology was used in the manufacture of a variety of products.
In an experimental pottery firing, Jane Smythe demonstrated how a kiln made of (modern) sherds and mud is constructed, and how part of the structure may remain intact following firing and subsequent removal of the fired product. She, as an experimental archaeologist who seeks to gain a hands-on understanding of ancient technologies and lifeways, emphasized that with experience this process may be accomplished rapidly and is not at all as onerous as it might seem. Jane demonstrated how the heat and atmosphere can be controlled by maintaining vents and how the fire may be stoked through them, opening and closing them as necessary with mud and sherds. This suggests a function for the apertures in the excavated structures, and more to the point suggests a purpose for the vertical sherds found in the path of air flowing through those apertures.
||Left, The pottery kiln at HK11C has a wall of similar sherd and mud construction pierced by doors. Right, the experimental firing kiln created of mud and potsherds
Whatever the product at HK24B--whether beer or perhaps a porridge--it was not being made on a domestic scale. This level of industrial food production may indicate a pooling of agricultural resources or in fact the control thereof. Many questions still remain. Was this simply an effective way with regard to fuel resources and human effort, of processing food for a large family unit (a very large family)? However, given the industrial character of the general area around HK24B and the lack of obvious domestic habitations nearby, was this the central catering area for the funerary feast? Or does it indicate the control and redistribution of agricultural produce by a higher authority and the incipient bureaucracy that would entail?
Looking at the broader picture, in conjunction with the excavations at the cemetery of the Predynastic elite (it is more than just an elephant--update coming soon!) and the temple at HK29A, our excavations have provided further food for thought on social complexity at Predynastic Hierakonpolis at about 3600 B.C. Together, these excavations show ever more clearly what a surprisingly sophisticated place it was and that the Egyptian state and dynastic Egyptian civilization did not emerge suddenly from a vacuum but was the end product of almost a millennium of developments. Remarkable things are no doubt still lurking under other pink rags scattered across the site.