Provisions for the pyramid builders : new evidence from the ancient site of Giza

Between about 2550 and 2470 BC the three Egyptian pharaohs of the fourth dynasty, Khufu, Kha­ fre and Menkaure, built their colossal pyramids and mortuary temples on the Giza plateau close to present-day Cairo. These enormous state projects would have employed thousands of conscripted labourers, whose settle­ ment near the pyramids is being studied by the Giza Plateau Mapping Project (GPMP). 1 The first aim of the GPMP was to locate the settlement, and then to investigate the social and economic infrastructure that supported the pyramid-building enter­ prise in what was one of the major urban centres in Egypt in the third millennium BC. Excavations to find the settlement of the pyramid builders eventually focused on an area of low desert south of the pyramid complex, about 400 m south-southeast of the Sphinx (Fig. 1 ) . Excavations began here in 1988 and continued sporadically until 1998. Then, in October 1999, we began a more intensive continuing phase of exca­ vation. The excavations have so far revealed a substantial well planned settlement that covers some 9 ha and can be divided into at least 30 distinct areas. They include the 200 m-longWall ofthe Crow, which bounds the site on the northwest, and a large royal administrative building with associated storage structures (silos) at the southeast­ ern edge of the site (Fig. 2) . An enclosure wall connects the Wall of the Crow and the royal administrative building, thus fram­ ing the main part of the site, which consists primarily of the ruins of four sets of long mudbrick galleries. Three streets cut west to east through the galleries. Scattered else­ where throughout the site are house-like structures, bakeries and craft workshops, as well as two major domestic settlements associated with the complex, known as the Eastern and Western Towns (Fig. 1 ) . Dating evidence from the pottery, the mud sealings (Fig. 3) and the detailed stratigraphic phasing of the site indicates that the settlement was in use during the reign of the pharaoh Menkaure (249024 72 BC) , the builder of the final Giza pyr­ amid. The evidence also indicates that the site was a royal facility for the production of food and other items used to support the builders of the pyramids. The tombs of these workmen are located just above the settlement in the hills to the west (Fig. 1 ) and are being excavated by Zahi Hawass, secretary-general of the Egyptian Supreme Council of Antiquities. A major aim of the project is to investi­ gate the daily life of the pyramid builders, including their food supply, by asking such questions as: what types of food did they have, did they all consume the same foods, did they have to produce it them­ selves or was it provided for them, and how does the Giza settlement compare in this respect with other Egyptian settlements? To address these questions, a comprehen­ sive programme of sampling plant and

Provisions fo r the pyramid builders: new evidence from the ancient site of Giza Mary Anne Murray The great pyramids of Giza are fa mous emblems of ancient Egyp tian civilization, but until recently little was known about where and how the pyramid builders lived.The site of their large settlement has now been fo und, and excavation is revealing its complex layout and providing evidence of the plants and animals on which the builders depended fo r their fo od supply.

B
etween about 2550 and 2470 BC the three Egyptian pharaohs of the fourth dynasty, Khufu, Kha fre and Menkaure, built their colossal pyramids and mortuary temples on the Giza plateau close to present-day Cairo.These enormous state projects would have employed thousands of conscripted labourers, whose settle ment near the pyramids is being studied by the Giza Plateau Mapping Project (GPMP).1 The first aim of the GPMP was to locate the settlement, and then to investigate the social and economic infrastructure that supported the pyramid-building enter prise in what was one of the major urban centres in Egypt in the third millennium BC.
Excavations to find the settlement of the pyramid builders eventually focused on an area of low desert south of the pyramid complex, about 400 m south-southeast of the Sphinx (Fig. 1).Excavations began here in 1988 and continued sporadically until 1998.Then, in October 1999, we began a more intensive continuing phase of exca vation.
The excavations have so far revealed a substantial well planned settlement that covers some 9 ha and can be divided into at least 30 distinct areas.They include the 200m-longWall ofthe Crow, which bounds the site on the northwest, and a large royal administrative building with associated storage structures (silos) at the southeast ern edge of the site (Fig. 2).An enclosure wall connects the Wall of the Crow and the royal administrative building, thus fram ing the main part of the site, which consists primarily of the ruins of four sets of long mudbrick galleries.Three streets cut west to east through the galleries.Scattered else where throughout the site are house-like structures, bakeries and craft workshops, as well as two major domestic settlements associated with the complex, known as the Eastern and Western Towns (Fig. 1).
Dating evidence from the pottery, the mud sealings (Fig. 3) and the detailed stratigraphic phasing of the site indicates that the settlement was in use during the reign of the pharaoh Menkaure (2490-24 72 BC), the builder of the fi nal Giza pyr amid.The evidence also indicates that the site was a royal facility for the production of food and other items used to support the builders of the pyramids.The tombs of these workmen are located just above the settlement in the hills to the west (Fig. 1) and are being excavated by Zahi Hawass, secretary-general of the Egyptian Supreme Council of Antiquities.
A major aim of the project is to investi gate the daily life of the pyramid builders, including their food supply, by asking such questions as: what types of food did they have, did they all consume the same fo ods, did they have to produce it them selves or was it provided for them, and how does the Giza settlement compare in this respect with other Egyptian settlements?To address these questions, a comprehen sive programme of sampling plant and Fi gure 1 Aerial view of th e Giza plateau, showing the three pyramids of(from right to left) the pharaohs Kh ufu, Kh afre and Menka ure, the Sphinx, the site of the pyramid builders ' settlement (righ t fo regro und) and their cemetery (left foregro und).Fi gure 2 Simplified plan of the settlemen t of the pyramid builders at Giza.Redrawn fr om a plan prepared by Mark Lehner, director of the Giza Plateau Mapping Project.
animal remains was undertaken.All ani mal bones from the excavation were retrieved (more than 152,000 items to date) and they are being analyzed by Richard Redding of the University of Michigan.My role as the archaeobotanist for the project is to retrieve and analyze the plant remains, all of which were preserved by charring.2 We recovered them by machine fl otation, which separates out the charred plant frag ments by washing excavated soil samples through sieves of 1 mm and 0.25 Jlm mesh Figure 3 Mud sealings were used to seal documents, jars, doors, etc.They often record inform ation in the form of hieroglyphs such as the names of kings (which provide usefu l dating eviden ce), the contents of the object being sealed, or the destination of the sealed product.
size.More than 30,000 litres of soil have been processed in this way from the nearly 4000 samples collected so far, and nearly 2000 of these have been analyzed.When the plant remains have been recovered and identified, they are analyzed using a series of quantifiable indices, calculated for several variables (Table 1).However, full interpretation of the plant and animal remains depends on understanding their provenance (i.e. from what parts of the site they came), so in the next section the main features of the site are brief!y described.
Table 1 The quantifiable indices and variables used in analyzing the plant remains recovered from the Giza site.o

Phase
The layout of the settlement The most conspicuous feature is the Wall ofthe Crow, which probably controlled the flow of people and materials into and out of the four sets of mud brick galleries (Fig. 2).
Each set is about 35 m long north to south and about 52 m wide (or 100 ancient Egyptian cubits).There are eight galleries in each set.Like many ancient Egyptian houses, the galleries have baking and cook ing facilities in the back (southern end) and an open living area in the front, thus resem bling an elongated version of what is a common Egyptian house plan.There are also low-angle platforms at regular inter vals within the galleries.The function of the gallery sets is not clear, but one inter pretation is that the angled platforms are bed platforms and that the galleries may have been used as a type of barracks for rotating shifts of pyramid builders.
Evidence from ancient Egyptian texts indicates that groups of unskilled workers on pyramid-building projects were tempo rarily conscripted, and worked in overlap ping rotation in and out of these long-term royal projects.For example, graffiti from Giza depict the rivalry between two gangs of this period, the "Friends of Menkaure" and the "Drunkards of Menkaure".3 Egyp tologists infer that these gangs were More than a dozen bakeries have also been identified, most of which are in the backs of the galleries or attached to them, with others scattered throughout the set tlement.If a rotating labour force was housed in the galleries, these bakeries pre sumably produced their daily bread.The byproducts of the two types of cereal found on site, emmer wheat and hulled barley, are bread and beer, which are always found together in ancient Egypt and were likely  CD7 bowls, a type unique to Giza, which is likely to have been used for the consump tion of both food and drink.
The textual and artistic evidence asso ciated with Old Kingdom bakeries was also studied, and it was decided that some baking experiments were needed to help understand the archaeological evidence.A standard type of Giza bakery, which had most of the major elements found in other bakeries at the site, was chosen for exper imental reconstruction (Fig. 6).During the experiment, it was found that all parts of the bakery had a function when we actu ally tried to bake bread, which involved experimenting with different mixtures of emmer wheat and barley, heating and stacking the bread moulds, pouring the dough and ensuring that the fire stayed at an even temperature.The bread from the first experiment proved to be barely edible (Fig. 7), but we plan further experiments and hope to learn how to improve the prod uct.
There are also areas that were clearly used for storage.South of the galleries, for example, we found a large double-walled building.The evidence from the mud seal ings found there indicates that it was some type of royal administrative building (see Fig. 2).Further excavation uncovered a huge storage facility, including eight large mudbrick silos (each one about 2.8m or 5 ancient Egyptian cubits in diameter, Fig. 8).It is like!y that grain was dispensed from this central storehouse, protected behind the double walls of the administra tive building.Ultimately, before the end of the fourth dynasty c. 2500 BC, the silos were destroyed and some 20 tonnes of limestone and granite were dumped into them.
Associated with the gallery sets there is a roofed colonnade (a hypostyle hall; see Fig. 2), which has a series of troughs, benches and column bases at regular inter vals.The remains of similar troughs and benches excavated elsewhere, for example at the site ofT ell Karrana in Mesopotamia,4 show that these constructions often served as elevated platforms for the drying of grain and other foods, and in this case a concentration of fish remains from the 25 m-long colonnade points to the drying of fish.It has also been suggested that the hall may have been a common area or pub lic space, and a concentration of CD7 bowls (used for food and drink), together with the fish and other animal remains, suggests that it may have been used as a communal eating area.
What are believed to be domestic houses are another prominent feature of the Giza settlement.They include, for example, a row of dwellings along the west side of the hypostyle hall and the so-called gate houses at the beginning of each of the three west-east streets, which were perhaps used to monitor people and goods going into and out of each set of galleries.In all, more than three dozen house units are recognizable within the settlement area.Unlike the planned layout of the main area, the Eastern and Western Towns (see Fig. 2) are self-organized domestic settlements that more closely resemble other ancient Egyptian town sites, and were probably home to longer-term residents.Small silos and ash-filled chambers suggest that resi dents, perhaps families, stored their own food and cooked for themselves.Excava tions have shown that the Eastern Town extends beneath the modern town nearby.The mud sealings, pottery and stratigraphy all suggest that the eventual and probably intentional abandonment of the site took place at the end of the fourth dynasty, c. 2500 BC, soon after the reign of the pharaoh Menkaure.

Feeding the pyramid builders
So what did the pyramid builders live on?What types of food did they consume?The plant remains consist of relatively few spe cies, primarily two cereals (emmer wheat and hulled barley), several pulses (small seed legumes such as lentil, grass pea and bitter vetch), and fruits such as common fig, sycamore fig, Christ's thorn and possi bly grape and date.The animal remains are limited to four domesticates (cattle, sheep, goats and pigs) and six species offish (Nile perch, catfish, tilapia, tiger fi sh, grey mul let and puffer fi sh).This is a surprisingly short list and it is clear that the plant and animal assemblage from Giza lacks the diversity of food species commonly found at other ancient Egyptian settlements.
The plant samples are dominated by weeds, primarily large wild grasses and legumes, which were probably weedy con taminants of the cereals coming into Giza as offerings, tributes or tax, whether from local sources or more distant provinces.There is little evidence of nuts or roots and tubers , and, as yet, no evidence of plants providing oil or fibre; nor do the animal bone remains contain any evidence of hunting, which is most unusual for an ancient Egyptian settlement.
Another striking feature of the plant assemblage is that, on average, the site has relatively few plants (only 3.6 plant items per litre of soil overall), which is much lower than at other settlements, such as New Kingdom Memphis (96.5) and First Intermediate Abydos (144.4).5 The low density of plant remains and the few spe cies represented may indicate that food preparation, fuel use and the disposal of fuel were specialized and routine activities.
When the number of items per litre is assessed by area, however, those with the highest relative density are found to be the so-called houses or domestic areas.They also generally have a higher diversity of food plants than other areas and are more similar to domestic areas in other ancient Egyptian settlements.Most of the cereal, pulse, fruit remains and other food items have been found in the houses, probably because most of the food preparation took place in these dwellings.In the gallery sets, both the density of plant remains and the diversity of species are very low, which is to be expected if they were used as sleeping areas.The composition of the plant remains from the storage areas and the bakeries is similar.As yet, no storage facilities with their original contents in situ have been discovered, the plant material found there having been redeposited from elsewhere.The low density of items in the bakeries may be attributable to the abundance of fine ash, where most plant remains have been burned away.The hypostyle hall has the lowest plant density and diversity, but this does not exclude its having been used as an area for drying food or for communal eating, because it is improbable that plants would have become charred and been preserved in the course of either of these activities.
The evidence from the animal remains suggests clearer contrasts between differ ent areas.As with the plants, the greatest diversity of animal species occurs in the houses.For example, the three gate houses and the royal administrative building (see Fig. 2) contained the remains of what would have been the best cuts of beef, as well as very large Nile perch.Certain houses in the Eastern and Western Towns also show this pattern, although with more sheep, goats and pigs.In contrast, the galleries and the hypostyle hall contain poorer cuts of meat from cattle, sheep and goats, as well as large amounts of small bony fish.
The plant and animal remains so far analyzed indicate that the pyramid build ers themselves subsisted mainly on bread, beer, pulses, cattle, sheep, goats and cat fish, whereas the inhabitants of the gate houses and the royal administrative build ing, who were presumably the overseers of the workmen in the galleries, had, in addi tion to these fo ods, much better cuts of meat, especially of beef and pork, and large Nile perch.The longer-term residents in the Eastern and Western Towns (where we are still analyzing individual houses in more detail) appear to have lived on a more varied diet consisting of all these foods.
There is also good evidence that the settlement was provisioned with food and fuel from elsewhere.The large amount of meat indicated by the bone remains is likely to have come into the settlement in the form of live animals, which were butchered there, an inference supported both by the finding that the cattle were most] y young males ( 1-2 years old) and by the lack of evidence of hunted animals.The cereals, too, probably arrived partly processed, apart from the large weeds we find, which would have been removed by hand prior to grinding the grain for baking or brewing.Wood was the main fuel used in the bakery ovens, and the analysis of wood charcoal from the site carried out so far shows that over 99 per cent of it came from acacia trees.Acacia wood is a high quality slow-burning fuel that would have been an ideal choice for firing the bakery ovens.Unlike most other ancient Egyptian settlements, animal dung was apparently not a major fuel at Giza, and may not have been used at all.

Conclusion
The pyramid builders clearly did not live on bread and beer alone.Our investi gations -which continue -have already demonstrated the great scale and complex ity of this state-organized project, which depended on a major investment of natural resources and man hours (also, no doubt, woman hours), as well as many workmen's lives, as is suggested by the size of their cemetery above the settlement.Much work remains to be done as we excavate and ana lyze the cultural remains from this unique settlement and endeavour to understand more fully not only how it was organized but also the small details of the daily lives of the pyramid builders of Giza.

o
Area type: bakery, house, etc.

Figure 4 l
Figure 4 Pictorial diagram of how the pyramid-building workforce at Giza may have been organized into gangs and smaller work groups (phyles}; from Lehner

Figure 5 Figure 6
Figure5An excavator at Giza holding an ancient bread mould fr om the site.tohave been the main staples in the diet of the pyramid builders -as for most Egyp tians at the time.This conclusion is sup ported by the fact that nearly 60 per cent of the over half a million pottery sherds from the site are from bread moulds (Fig.5), and

Fi gure 7
Exp erimentally baked bread (using various mixtures of emmer wh eat and barley) in locally made reconstruc tions of the typ e of mould used at ancient Giza.remains of beer jars are the second most common type of pottery.The third most common type are what are referred to as

Figure 8
Figure 8 Remains of large mudbrick storage silos wi thin the royal administrative build ing at the Giza settlement, with the pyramids of the pharaohs Kh ufu and Khafre in the backgro und.
Notes 1.The GPMP is directed by Dr Mark Lehner who is affiliated to both Harvard Univer sity and the University of Chicago and who has been surveying and mapping the Giza plateau for more than 30 years.For a recent comprehensive review of the site, see M. Lehner, "The pyramid age settle ment of the Southern Mount at Giza", Jo urnal of the American Research Cen ter in Cairo, 39, 27-74, 2002.Dr Murray is the project archaeobotanist and assistant director in charge of archaeological science.2. Dr Wilma Wetterstrom ofHarvard Univer sity was the project archaeobotanist from 1988 to 1997.3. See p. 128 in A. M. Roth, Egyp tian phyles in the Old Kingdom : the evolution of a sys tem of social organization (Studies in Ancient Oriental Civilization 48, Oriental Institute, University of Chicago, 1991).4. See pp.224-5 in M. Lehner, Th e complete pyramids (London: Thames & Hudson,