Architectural and Stratigraphic Overview

The fortification at Bi’r Samut is part of an archaeological zone that lies approximately 130 kilometers by road from the city of Edfu in the Nile Valley and almost an equal distance to the Red Sea coast, as the crow flies. Difficult to reach today and presumably even more so in antiquity, the Samut area lies deep in the remote heart of the Eastern, or Arabian, Desert. This hyper-arid region was intensively explored and exploited from the late fourth century BCE until the fifth century CE.⁴

The Ptolemaic kings, like their Pharaonic predecessors, were interested in the desert’s abundant mineral resources—particularly its rich veins of gold-bearing quartz and precious stones—and exploited it as a source of wealth and as a conduit to distant lands. Beginning in the late fourth century BCE, an archipelago of mining installations,⁵ fortified wells, subsidiary settlements, and small shelters were constructed along the route from Edfu, ancient Apollonopolis Magna, to Berenike Troglodytika on the coast, where a fortified outpost and harbor were constructed to facilitate the transport of battle elephants from East Africa under the supervision of the Ptolemaic army.⁶

Bi’r Samut is one of the largest Ptolemaic fortifications (with El-Kanaïs) along this road and indeed in all of Egypt, and the best documented (figure 12.2). As mentioned above, excavations at Bi’r Samut took place between 2014 and 2016 under the auspices of the Mission Archéologique Française du Désert Oriental and uncovered a large and well-preserved stone-built fortress surrounding a cistern and a destroyed well.⁷ Two large middens were also excavated outside the main gate and postern gate, which produced some thirteen hundred Greek and demotic Egyptian ostraka dating to the third century BCE.⁸ From these documents and other remains, we have learned the fort’s likely foundation date is sometime in the last years of the first half of the third century BCE, probably during the 250s BCE, like other forts or fortified wells of the area.⁹

There are not many well-documented houses in Upper Egypt that date securely to the third century BCE and can therefore offer direct comparanda for Bi’r Samut’s architectural plan (although see below for discussion of a second-century example). However, well-excavated examples of third-century domestic architecture are known from Tebtynis.¹⁰ No two houses at Tebtynis are identical, but the excavators have noticed a few patterns in the house plans. Houses in this period were generally square or slightly rectangular, with one or two rooms incorporating purpose-built installations for various stages of food preparation and storage.¹¹ All had a second story, but it is unclear how the upper spaces of the house were used. Not every house had an unroofed space incorporated into the architectural block of the house, but if there was outdoor space, it was located at one end (see, for example, house 5200-I). Marouard has suggested that some cooking and household production activities occurred outside the architectural frame of the house in communal exterior spaces.¹² The lower floor required linear movement through the house since, unlike at Bi’r Samut, there was no central space from which to access many rooms.

At Bi’r Samut, we see a vastly different type of spatial configuration. The overall plan of the fort was largely visible on the surface before work began in January 2014 (figure 12.2). Exterior walls were clearly defined in most areas, indicating a large, almost square fortification with three intact square corner towers and a series of rooms built against the fort’s original outer wall. Damage to the eastern half of the fort was due to wadi wash over the centuries, as well as damage to the site by looters in 2012 and 2013.

The fortress had two gates: a central gate in the northern wall and a postern gate in the southern wall. Large middens were visible outside the two gates of the fort—a main gate in the north (North Midden) and a postern gate in the West (West Midden). Midden deposits were also documented inside the fort’s postern gate (especially in rooms 20 and 21), overlying some architectural features; these suggest that trips outside the fort to deposit refuse were no longer practical in the years or months immediately prior to the fort’s abandonment sometime in the last decade of the third century BCE.

The fortress itself comprised more than fifty interior spaces. Excavations in 2015 and 2016 revealed that the outer walls and series of rooms were original to the fort’s construction, and a series of additional rooms abutting the outer line of chambers was added later to provide smaller rooms in the fort’s interior. A large cistern was also uncovered, along with the outline of a damaged well and several other ambiguous features. Much of the functional space of the courtyard was absorbed into these large water features. Although third century BCE houses in the Fayum took full advantage of outdoor spaces, the lack of preserved activity outside of the fort and the limited available workspace in the central courtyard indicate that outdoor work areas would have been at a premium in the fort, particularly after the construction of the smaller courtyard rooms.

At a minimum, the small courtyard rooms, each of which was provided with a cooking and storage installation, increased the number of lodgings available for the fort’s occupants. This increase suggests that there was pressure on the fort to accommodate more people and to do so in semiprivate rooms with individual facilities. It also indicates that shelter and security were of more importance than the open courtyard space that might be required for communal cooking, assembly, or other activities.

Distribution of Activities at Bi’r Samut

Another aspect of Bi’r Samut is the diversity of functions attested both within and between rooms. In January 2015, excavation began on the southwestern corner of the fort, which had been damaged by a glancing blow from a bulldozer. This round corner tower contained a largely intact Ptolemaic bathing complex (figure 12.3), including a central round plastered room with small basins for the first ablutions, a second room with a hip bathtub and an immersion bathtub for the cleansing operations, and an intact heating system with a furnace still retaining evidence of its last use.¹³ The inclusion of a bathing facility in the fortification is not unprecedented,¹⁴ although this is the first documented instance of such a facility in a Hellenistic fortification in Egypt. The suite of tubs and the heating system are integral to the fort’s design, indicating that the provision of this facility was part of the fort’s original rationale, despite the water-poor environment. These baths likely served only a select few in the fort—for example, high-ranking soldiers or privileged travelers; only two people could bathe at the same time.

The function of certain other rooms in the fort were also suggested by purpose-built installations. For example, Rooms 12 and 13 in the southeastern tower appear to have functioned as storage facilities (figure 12.4). Room 12 contained twelve unusually large Egyptian amphorae with their necks removed; these amphorae were embedded in the floor for reuse as storage containers.¹⁵ Analysis of the sediment in these vessels indicates that at least two of them stored charcoal, while others contained organic debris, although it is unclear whether this debris was introduced after the room’s abandonment.¹⁶ Small groups of storage containers were embedded in floors elsewhere in the fort, but this is the most condensed large-scale storage area. Room 13, adjacent to 12, had two small stone-lined, somewhat shallow storage pits dug into the floor.

Other items recovered from the floors of these rooms include a handmade cook pot, likely originating outside of Egypt—perhaps from somewhere in East Africa—and an iron pick head, along with many discarded and fragmentary ceramic vessel fragments. In addition, a faience amulet was also found on the floor of Room 12, alongside several grindstones and installations that likely served as work surfaces. A complete lamp and several coarsely formed alluvial lids for closing the reused amphorae round out the assemblage of materials abandoned on the floor of these adjoining rooms. The mixture of different functional artifact types—iron tools, storage vessels, lamps, cooking pots—is entirely typical of the floor groups from Bi’r Samut and suggests that spaces were not strictly divided into zones dedicated to specific activities.¹⁷ In other words, there was no strict division into spaces dedicated simply to “storage” or food preparation, for example.

This conclusion is supported by the presence of artifacts suggesting specialized activities alongside more mundane tasks, such as cooking or the sheltering of animals. In Room 45, twenty-seven loom weights were found spread on the floor in groups of one, two, and five, probably where the loom stood when it was abandoned (figure 12.5). On the same floor, a fragment of gaming board in limestone and two sets of seven stone gaming pieces lay together; these were probably used by the inhabitants to pass the time. Although only fragmentary traces of wood among the weights suggest the loom itself, these objects are clearly associated with cloth production, and they appear alongside traces of leisure activities.

Adjacent to this room, Room 28 seems to have functioned exclusively as a venue for food preparation and storage. The room contains several stone-built basins and grain-grinding installations, as well as two ovens constructed from repurposed storage jars. The ceramic vessels from the floor include a mix of amphorae, jars, bowls, and cooking vessels, along with several specialized objects related to food preparation. These include three Egyptian-made bread stamps, two with geometric sunburst motifs and the third with a Silenus face (figure 12.6).

Room 25, in the same suite of rooms, produced a remarkable floor assemblage containing more than thirty-five complete or restorable vessels. These lay in no discernible order, jumbled among Greek and demotic Egyptian ostraka, writing implements, and organic remains. All lay beneath the carbonized remains of the fort’s palm frond roofing.¹⁸ This group of vessels exhibits several of the key characteristics of the third-century ceramic assemblage as a whole; the vessels from this room are very diverse in terms of their function (figure 12.7). There are jars, cook pots, perfume vessels, alabastra, lekythoi, aryballoi, flasks, amphoriskoi, and several types derived from older Egyptian forms, including high-shouldered jars and pomegranate-shaped vessels. The room also contained several large utilitarian vessels, including dinoi and kraters, as well as at least nine complete Egyptian Type 1 bead-rim amphorae.

In addition to this functional diversity, the vessels from Room 25 represent a wide range of production sources from within Egypt. Ptolemaic black ware, likely produced in the Delta, and vessels from the Mediterranean coast and the Theban region are all represented. Nubian eggshell wares, handmade desert wares, and Aswan-produced wares are also part of the fort’s third-century assemblage. Given the fort’s isolation and restricted access to supply routes—presumably, all materials coming to Bi’r Samut would have been carried by donkey or camel via the Edfu road—the range of supply sources represented is remarkable. It is tempting to see this as evidence for the centralized network of suppliers involved in provisioning Bi’r Samut. Perhaps the state’s network of suppliers reached beyond the Thebaid because of the infrastructure chain that supported the state’s expanding economic power over the third century. In any case, this diversity of resources is atypical for contemporary household assemblages, which are primarily regional even when small numbers of foreign imports are present.¹⁹

Many of the smaller rooms in the fort’s interior (for example, Room 34) also yielded floor deposits that included plates, bowls, and jars containing the remains of foodstuffs (figure 12.8). Sheep/goat bone, antelope horn, fish bones, and murex shells appeared in several rooms, as did personal utensils fashioned from mollusks.²⁰ In situ cooking installations made of repurposed jars or amphorae were also present in most rooms, usually in pairs. Many rooms provided a place for individuals or small groups to prepare food, a consideration that might suggest that groups of travelers or other residents of the fort were assigned space within its walls and expected to prepare their own food during their tenure at the site.

In any case, the division of the fort’s interior into many small rooms, most of which appear to have duplicate functions, supports this view. The densely packed arrangement of these spaces, and the lack of any subsidiary constructions outside the fort, also suggest that security issues constrained the availability of areas outside the fort for these activities. This constraint resulted in an intensity of use that is reflected in the distribution of materials within the fort’s abandonment levels. The repetition of functional spaces, especially for cooking, grain processing, and animal husbandry, at a large and small scale indicates that differences in inclusivity were part of the system of social membership in the fort. The small “studio” units, such as Rooms 33 and 34, which were equipped with a kitchenette and small living space, provide a setting for a different scale of activity from the large-scale food preparation and storage areas in Rooms 25 and 28, and presumably also afford different kinds of social groupings.

These spatial and artifactual patterns beg the question of exactly who was in residence in the fort. The ostraka (especially those written in demotic Egyptian) indicate that members of the Ptolemaic military, phylakites, and even women lived at and passed through Bi’r Samut.²¹ Although gender roles are a central topic within household archaeology, the role of women at Bi’r Samut remains to be studied. The ostraka also record a range of purposes for people who lived at the fort. Most of them were traveling through to other locations in the desert or along the Red Sea coast, sometimes accompanied by companions, and were involved in the logistics of the desert trade and provisioning.²²

The material evidence on its own points to a spectrum of activities that were appropriate to any domestic settlement, but here they were organized rather differently, because they were housed in spaces embedded in an architectural frame that was inherently supradomestic and defensive in purpose. Given these differences and similarities, what concept of a forthold can we outline at Bi’r Samut given the domestic character of the fort’s remains, architecture notwithstanding? How do the patterns briefly outlined here synchronize with what is known about third century houses in Egypt, for example? Do they represent a completely different way of utilizing space for activities that can, when viewed on their own, broadly be characterized as domestic? How do these patterns compare to another fortification, albeit of a slightly earlier period and in a different environment?

A Comparative View: Houses in Syene (Aswan)

Examples of second century BCE domestic architecture at Syene (Aswan) in Upper Egypt provide a general basis for comparison with Bi’r Samut, alongside Tebtynis. In Area 15 at Syene, the relationship between activity spaces and types of socioeconomic connectivity reflects activity patterns and spatial organizations of households that go beyond the architectural configuration of a fortification, presenting a view of Upper Egyptian domestic structures and deposits in the city. A comparison between Bi’r Samut and the households of Ptolemaic Syene demonstrates a much different pattern of spatial organization and domestic activity at the two sites.

Late Period Syene is characterized by an internally oriented set of buildings on a highly regulated plan, which conforms to the historical evidence for its organization as a military settlement at this time. During these Late Period phases, the standard house types represented are typical tower houses.²³ The area was reconfigured in the early Ptolemaic Period, most notably with the addition of a relatively luxurious bathhouse in the second century BCE. This addition suggests that new connections with the Mediterranean koine of social practice reached Upper Egypt at this time.²⁴ The architectural remains also reveal that by the early second century BCE, the population of Syene had shifted from a military community to a domestic one, since houses were built and modified in a more organic way than in previous centuries.

By the late second century BCE, houses in Area 15 were being adapted into smaller, self-sufficient units that were likely more private and increasingly densely packed (figure 12.9). For example, the previously shared courtyard of Houses 1 and 2 shrank and seems to have only been accessible to the occupants of House 2.²⁵ The incomplete preservation of Houses 1, 2, and 5 in this period make it difficult to fully assess the relationship between households, but the nature of the structures and their continual modification and subdivision suggest a premium on space in the urban landscape of the town. The excavators characterize these houses as “typically Egyptian” and see them as evidence for continuity of traditional domestic practices even into the early Roman Period (Stratum C).²⁶

Over this same period, the capacity for food production also increased. Two ovens in the courtyard continued to be the main location of food production for House 2. Sometime after, however, two additional bread ovens were installed in Room 2 of this house, and Room 10 acted as an indoor kitchen.²⁷ Based on comparanda from the Fayum, Delta, and Sinai, units with more than one bread oven usually provided bread for individuals who were not household members.²⁸ Could House 2 have provided the necessary bread beyond the immediate household, while other houses in the area were responsible for the production of different products? Or is the increase in capacity simply a reflection of the increasing density of occupation in the insula?

Partially finished faience vessels and large quantities of faience beads in Stratum D in House 5 at Syene suggest that specialized craft production was also a component of domestic life at Syene. Certain specialized activity spaces were also sometimes duplicated between households. For example, House 2 and House 5 both contained a dedicated dining room (andron). The increase in duplicated activity spaces suggests that households could operate independently.

The published ceramic assemblage from this domestic group consists primarily of vessels associated with Greco-Macedonian feasting practices.²⁹ One deposit from the substructures of House 1 reveals a valuable cross-section of ceramic vessels employed in this period. Tablewares suggest that even in Upper Egypt, the Mediterranean koine dictated drinking and dining habits. Red-slipped silt echinus bowls, plates with thickened rims, and cups with vertical rims imitate similar forms in the contemporary Mediterranean. Pouring vessels include a miniature amphora, a double-handled jug, and a fragment decorated with vegetal decoration similar to contemporary Hadra vases.³⁰ The cooking wares, however, are local Egyptian products and types; the angled-rim cookpot and the tall ledge rim cookpot are most common here, as they are at Bi’r Samut.

Although most ceramic vessels found at the site were produced in Egypt, a small number of the vessels found in domestic contexts at Syene were produced elsewhere in the Mediterranean. Fourth- to third-century imports include an Attic plate with incurving rim,³¹ Attic west slope or Gnathian amphorae, two fragments of the ivy-platter group, skyphoi, and three lamps.³² These vessels indicate connections with the Eastern Mediterranean, Greece, and southern Italy, though the vessels are not evident in large quantities, and their relative rarity could suggest that they arrived via individuals rather than through an established trade route. Sabine Ladstätter argues that these vessels demonstrate a connection to the Delta, rather than to long-distance locales, and Müller likewise sees these patterns as evidence for a light footprint left by foreign influence in the town.³³

In the domestic contexts, the number of imports declines by the late third century BCE, but tableware and cooking wares still engage with the Mediterranean koine. The biconical cookpots and casseroles at Syene are part of the contemporary widespread adoption of Aegean and eastern Mediterranean cooking forms, if not practices, in Egypt at this time. Imported transport amphorae are still found in the second century, but Egyptian amphorae in Nile silt represent most of the assemblage. These patterns are broadly similar to those of the vessels recovered at Bi’r Samut.

The similarities in the ceramic assemblages at Syene and Bi’r Samut suggests that both sites are integrated into the Upper Egyptian economic sphere—hardly a surprise—but that Bi’r Samut benefits from a broader range of supply sources in the third century BCE than Syene. The differences in the architectural configuration of the domestic zone at Syene, when compared with Bi’r Samut, are also much more striking. The duplication of functional spaces and installations at Bi’r Samut is notably different, and the overall configuration of space is also radically dissimilar. This comparison supports a view of the fort at Bi’r Samut as something qualitatively different from a compact house with many rooms, inhabited by permanent residents; it is instead a completely different spatial animal designed to house multiple, transient social units that shared certain spaces but also maintained discrete, semiprivate quarters. The setting of Bi’r Samut in an expansive wadi rather than the confines of a walled settlement must also be considered.

A Forthold in the Delta: Tell el-Herr

In order to understand the nature of the forthold at Bi’r Samut, an examination of the Late Period fortification at Tell el-Herr in the Sinai is also highly useful, since it contextualizes the patterns observed at Bi’r Samut through comparison with another complex that nominally shares the same functional designation. The types of architectural integration, permanent installations, and extant ceramic assemblages from the late fifth century to the second quarter of the fourth century BCE at Tell el-Herr provide comparanda for the replication of domestic activity spaces, food production areas, and industrial installations at Bi’r Samut.

The discrete occupation from the end of the fifth century through the first quarter of the fourth century is historically defined by a change in control in Egypt from Persian kings to the Thirtieth Dynasty. Unlike at Bi’r Samut, no definitive answer can be given for who controlled Tell el-Herr in this period. The fortification was most likely constructed in the fifth century BCE to protect trade routes and the territorial power of either the Saite or Persian rulers.³⁴ The metal vessels, transport amphorae, and imported mortars preserved at Tell el-Herr illustrate exchange connections with Greece, Cyprus, the Persian heartland, and elsewhere in the Levant. Homes typically included cylindrical cups and Achaemenid-style bowls made in local fabrics.

An increase in population in the third quarter of the fifth century BCE led to an expansion and restructuring of the Iron Age fortification and settlement at Tell el-Herr. The following three decades (period VA) were years of (primarily, but not exclusively) military occupation at Tell el-Herr, associated with a centralized power structure and standardized, replicated domestic and industrial units in the fort and its associated structures. The settlement at the site was much larger in scale and longer in duration than at Bi’r Samut, but it nevertheless provides a calibrated measure for assessing qualitative differences between the household and the forthold.

The western and northern sectors of the site are well-preserved for this phase of its history. Orthogonal roadways shaped the residential blocks and created connections between domestic neighborhoods and the garrison buildings. Houses were arranged around major places of assembly, such as gateways, administrative buildings, and sanctuaries.³⁵ The structures of the northern sector cluster around the “palatial building.” The houses in this area were strictly residential, with small installations or archaeological deposits that indicate food production. One building (47) was identified as a storage hub because of the enclosed floor plan and the presence of storage vessels. This building might have stored supplies used by the one extant tower house (44) or the insula of houses to the south.³⁶

The western sector is composed of many regularly arranged insulae. These complexes incorporated not only tripartite houses (described below) of identical form, but also religious spaces.³⁷ Several mixed units (79 and 81–83), in which craft production and domestic activities occurred, were more accessible from the fort’s gateway, but tripartite houses used for strictly domestic activities (including dwelling, food production, and storage) were part of the same insula.³⁸

Bathing facilities and cisterns were not found in Tell el-Herr, but Séverine Marchi suggests that rainwater could have been collected through gutters or in unpreserved receptacles.³⁹ A centralized system might require the population to use one cistern, but in the case of Tell el-Herr, individual occupants or neighborhood representatives would likely need to make their own journeys outside the fortification to collect water.

The most common configuration of dwellings at Tell el-Herr was a tripartite structure, which was the most frequent residential unit in the fort (figure 12.10).⁴⁰ Replication of the overall form, shared dividing walls, and compact agglomeration into insulae throughout the fortification suggest centralized planning that required long-term cooperation among the occupants.⁴¹ There is little evidence for utility spaces within these houses, suggesting that only small-scale food preparation and cooking would occur. The entry vestibule (A), in most instances,⁴² served as the kitchen, such as it was, with a masonry bread oven opposite the door; but this zone could be also used for craft production (as in Houses 68 and 69) or as the location for a staircase to an upper floor (House 48). The larger main room (B) was always the center of domestic life, evidenced by the presence of serving vessels, tablewares, and lamps, and was generally not used for cooking and only rarely for the processing of grain. The back room (C) was not well-ventilated or lit, and it was used as storage for items such as weaving implements, jewelry, tools, ceramic and stone vessels, and foodstuffs.⁴³

House 78 illustrates another variation on this house type, characterized by an open or covered courtyard (figure 12.11).⁴⁴ In this particular example, the location for domestic activities is not restricted in the same way. A courtyard (A), equipped with a large oven and hearth, a basin, and a storage area, was situated at the front of the house, while a second round bread oven was located at the back of the house (C), and these two areas were separated by a storage unit (B). Aside from these installations, very little material of daily domestic life was recovered from House 78. Did House 78 serve as a bakery, with its operators living elsewhere in the fort? This type of building suggests that some structures housed food production outside the household, an arrangement that would complement the smaller food production installations in some of the tripartite homes. At Bi’r Samut there were also both small and larger-scale bread production spheres operating simultaneously.

Larger structures like House 78 were not the only spaces at Tell el-Herr for large-scale or specialized production. Many tripartite homes also accommodated craft production, such as metallurgy (Houses 72 and 79–83). These smaller houses sometimes also had internally or externally attached bread ovens (Houses 77, 78, 84, and 93). Other tripartite houses included duplicate spaces for food production, such as House 94, which featured a food production space in the entry room but also a grinding installation in the main room.⁴⁵ Overall, there is no clear or consistent pattern of zones of food production at Tell el-Herr, but as at Bi’r Samut, there is duplication of production across a range of spaces and house types in the community, with some larger than others.

The division of codependent activity spaces into smaller units is again demonstrated by the distribution of bread ovens. Not every home had a bread oven, but every insula had at least one or more.⁴⁶ Pooling of necessities for everyday life, therefore, did not need to extend beyond the neighborhood. It is possible, however, that other tasks without archaeological traces required access to centralized facilities. The wide thoroughfares leading to the administrative hubs of the fortification suggest that these spaces were frequented, perhaps to purchase or retrieve goods from storage facilities. The presence of weights and measures in craft-producing houses suggests that consumers could access goods in more localized settings as well. Occupants of the northern, eastern, and southeastern sectors would need to visit the western Sector, where both craft production and imported vessels were clustered, or the administrative storage hubs, to purchase products. The distinct zones for activities other than food production suggests that these communities still potentially participated in a centralized system that facilitated access to necessary goods.

The sharing of production space and household integration, either throughout the fort or within smaller communities, does not necessarily reveal connections via kinship groups. Marchi rightly points to the difficulty in assigning certain tasks to specific ethnicities or genders.⁴⁷ The duplication of house types at two different sizes suggests heterarchically dependent groups within a larger hierarchical social system. An insula or groups of insulae contributed to the larger community of the forthold, but smaller communities within an insula could operate independently. The members of the forthold at Tell el-Herr represent a contingent of individual households engaged in craft production, trade, and domestic necessities at a scale and range that were atypical for domestic settings elsewhere. In this respect, Tell el-Herr is a clear precursor for Bi’r Samut, which has many of the same conceptual features, but a radically different plan, smaller overall scale, and much shorter lifespan.

Transience and Domesticity at Bi’r Samut

The types of activities taking place at Bi’r Samut in the third century BCE, including the production of cloth, metalworking, animal husbandry, and—above all—the distribution and storage of food and water resources, transcend a mere military function for the site, and in fact, are comparable to or even exceed the range of activities within contemporary houses. These activities were packed into the fort’s interior spaces, which seemingly necessitated close contact between, and regulation of, the occupants of the fort. The abandonment assemblages from the fortification of Bi’r Samut are also marked by a range of functional vessel types sourced from within Egypt, including the coastal Mediterranean and Delta, as well as Nubia. This finding highlights the diversity of supply sources available to the fort’s inhabitants and also suggests a different economic network than that typical of a contemporary household, especially in Upper Egypt. The presence of imported amphorae in relatively small quantities is similar to import patterns in Syene, as discussed above.

The archaeological evidence, in the end, allows for a few conclusions about the social relationships among the transient population of the fort and its shared system of membership. The differentiation in the size of functional space, especially cooking and grain processing, suggests that occupants did not have equal access to methods of food preparation. This inequality could extend to variation in dwelling arrangements more broadly (i.e., sleeping spaces), but the archaeological evidence does not confirm this. As we might expect from a kinship-based household, shared space in the courtyard was available, with a communal water supply and workspace. However, in this context, some of the communal arrangements might reflect the isolated desert environment, in which exclusivity of the water supply would be practically infeasible. Although most rooms incorporated a range of functions, and it is unclear who had access to more specialized spaces, the bathing installations and weaving equipment each occupied separate spaces, suggesting that these were not activities in which the general population of the fort would expect to take part.

It is also possible that certain aspects of the fort’s spaces and provisions were centralized or partitioned in service of permanent or semipermanent residents, while other facilities were geared toward the needs (and different social configurations) of itinerant travelers. The fort certainly served both groups simultaneously. Since part of the fort’s architecture was lost to the wadi, and part of the eastern wing remains unexcavated (and now destroyed), it is possible that some of the common facilities were simply not documented. Travelers may have camped outside the fort, as they did at ‘Abbad, another fort in the Eastern Desert, thereby leaving no archaeological traces.⁴⁸

So how should we understand these patterns? Is the fort a household on a grand scale, organized around a hierarchy determined by military (or other) role and status? Is it an agglomeration of spaces designed to house subsidiary, discrete social units, each with its own provisioned space? In the end, given the scale and specificity of the site’s overarching function, how useful is the analogy of the household when interrogating the material world of Bi’r Samut or other extradomestic spaces? At this stage of our analysis, the fragmentation of space and repetition of functions seems to be the most marked pattern in Bi’r Samut’s architecture and abandonment assemblages. This finding suggests that we might approach the fort not as a single household, but as many smaller social units housed in a single, complex structure. We might draw an analogy to a ship’s crew and passengers to help us conceptualize a space that is both domestic and transient, and designed to offer temporary quarter to social units whose relationships are multiscalar.

Notes

1The French mission is funded by the French Institute of Oriental Archaeology, the French Ministry of Foreign Affairs, and the French National Center for Scientific Research. The fort of Bi’r Samut was excavated under the direction of Jean-Pierre Brun in 2014 and 2015, and by Jean-Pierre Brun, Thomas Faucher, and Bérangère Redon in 2016. Annual reports are published in the Supplement to the BIFAO, accessible online (https://www.ifao.egnet.net/recherche/rapports-activites/). For the three campaigns at Bi’r Samut, see Redon and Faucher 2015; 2016; 2017. A first overview of the results of the excavations is published in Redon 2016; 2018. Gates-Foster and Redon are members of the ERC Desert Networks project, and this project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement no. 759078).

2See Boozer, chapter 8, this volume.

3Cuvigny 2020, 2022; Chaufray 2023.

4Gates-Foster 2012a; Brun et al. 2018; Sidebotham and Gates-Foster 2019.

5Faucher 2018.

6Gates-Foster 2012b; Redon 2018.

7For an overview of bibliography related to the site of Bi’r Samut, see Sidebotham and Gates-Foster 2019, 194–99; and Redon and Faucher 2014, 2015, 2016.

8Chaufray and Redon 2020; Cuvigny 2022; Chaufray 2022, 2023. The Greek ostraka are being studied by Hélène Cuvigny, with Laura Aguer and the participation of Bérangère Redon; Marie-Pierre Chaufray is in charge of the publication of the demotic ostraka.

9Redon 2018; Sidebotham and Gates-Foster 2019.

10Hadji-Minaglou 2007.

11See Marouard, chapter 3, this volume.

12Marouard 2008.

13Brun et al. 2018.

14In Egypt, mainly for the Roman Period, see Redon 2010.

15They are all inscribed with jar labels, except one; see Chaufray 2019; Chaufray and Redon 2020. For discussion of the amphora assemblage at Bi’r Samut, see Gates-Foster 2022.

16The archaeobotanical study is conducted by Charlène Bouchaud.

17An obvious difficulty lies in whether or not the materials from the various floors and fills of these rooms constitute a representative sample of the actual activities housed in these rooms. The relationship between abandonment deposits and actual use deposits is complex and not to be assumed. In the case of Bi’r Samut, the fort was abandoned in the last decade of the third century BCE and only selectively reoccupied in the Roman Period. The rooms discussed in this article were not reoccupied, with the exception of the bath.

18The roofing system of the fort was likely composed of a mix of materials, probably including rough cloth and palm frond ceilings.

19See, for example, the domestic deposits at Tebtynis, discussed below, which show a reliance on locally produced products for table and coarseware types. See also Gill 2016 for comparable patterns from Dakhla and the Western Desert, where ceramic production was well established.

20The faunal remains are studied by Martine Leguilloux.

21Inv. nos. 990, 1101, 1102, 1186; Chaufray 2023. The mentions of soldiers are notably variable in the ostraka. The demotic ostraka mention members of the military and policemen, while in the Greek ostraka the army is almost entirely absent. See Cuvigny 2020; and Chaufray 2023 for the first published discussion of the ostraka from this point of view.

22Chaufray 2019, 2022, 2023; Cuvigny 2020, 2022.

23See Marouard, chapter 3, this volume; Müller 2008, 2010; Hepa 2021.

24Berlin 2013; Müller and Hepa 2017.

25Müller 2010, 435–36.

26Müller (2010) dates the abandonment of the bath to the early Imperial Period. The relationship between the later Ptolemaic houses and the early Imperial iterations of these structures is not at all clear. Müller sees the area as developing from a more open plan in the Ptolemaic Period to one more densely packed, with increasingly subdivided and smaller living units.

27Müller 2010, see discussion on 436.

28Hadji-Minaglou 2007; Marchand 2011; Marchi 2014, especially 190; Hudson 2016.

29See Lynch 2018 for a discussion of these vessel types and their functions in that setting.

30Ladstätter 2010, 451.

31Ladstätter 2015, 138, no. 15. For fourth-century parallels found elsewhere in Egypt and the Levant see Ladstätter 2015, nn. 38–41.

32Ladstätter 2015, 139, nos. 19–20 (Gnathian amphorae); Ladstätter 2015, 142, no. 26 (ivy-platter group); Ladstätter 2015, 139–40, nos. 21–25 (skyphoi). Stratigraphically, vessel 22 (skyphos) originates from a deposit dated to the first half of the third century BCE. For discussion of the lamps, see Ladstätter 2015, 142–43, nos. 27–29.

33Ladstätter 2010, 452; Müller 2010, 438.

34Marchi 2014, 5.

35Marchi 2014, 169.

36Marchi 2014, 182.

37The absence of religious spaces at Bi’r Samut is surprising, given that most comparable structures either of the Late Period or the Roman Period included religious facilities.

38Marchi 2014, 163–64.

39Marchi 2014, 186. Other Late Period military sites include bathing facilities, such as Elephantine (Kaiser et al. 1990, 214–21) and Syene (Müller 2008, 315, fig. 3).

40Marchi 2014, 40.

41Marchi 2014, 186.

42Tell el-Herr Type III houses with food preparation installations in the entry room include Houses 55 and 67–70 in the North Sector, 77, 84, and 93 in the West Sector, and 28 in the East Sector.

43Marchi 2014, 169–71.

44Marchi 2014, 41, 170–71. This category of house (Type V) does not present a standard plan, but always includes a covered or open courtyard at either the front and/or rear of the unit.

45Marchi 2014, 170.

46Marchi 2014, 190.

47Marchi 2014, 204–5.

48Cuvigny 2017.

Works Cited

Berlin, A. 2013. “Something Old, Something New: Native Cultures under Ptolemaic Rule.” In Networks in the Hellenistic World: According to the Pottery in the Eastern Mediterranean and Beyond, BAR International Series 2539, edited by N. Fenn and C. Römer-Strehl, 229–37. Oxford: Archaeopress.

Brun, J.-P., T. Faucher, and B. Redon. 2017. “An Early Ptolemaic Bath in the Fortress of Bi’r Samut (Eastern Desert).” In Collective Baths in Egypt 2: New Discoveries and Perspectives, EtUrb 10, edited by B. Redon, 13–23. Le Caire: Institut français d’archéologie orientale.

Brun, J.-P., T. Faucher, B. Redon, and S.E. Sidebotham, eds. 2018. The Eastern Desert of Egypt during the Graeco-Roman Periods: Archaeological Reports. Paris: College de France. doi:10.4000/books.cdf.5230.

Chaufray, M.-P. 2019. “Tituli ptolémaïques du désert oriental égyptien.” In New Approaches in Demotic Studies. Acts of the 13th International Conference of Demotic Studies in Leipzig, 4–8 September 2017, edited by F. Naether, 39–92. Beihefte zur Zeitschrift für Ägyptische Sprache und Altertumskunde No. 10. Berlin: De Gruyter.

Chaufray, M.-P. 2022. “Les Blemmyes dans le désert Oriental égyptien à l’époque ptolémaïque (O.Blem. 1‑16).” In Blemmyes. New Documents and New Perspectives, edited by H. Cuvigny, 75–103. Le Caire: Institut français d’archéologie orientale.

Chaufray, M.-P. 2023. “Multilinguisme à Samut.” In Le multilinguisme dans la Méditerranée antique, edited by R. Roure. Pessac, Ausonius éditions, collection PrimaLun@. https://una-editions.fr/multilinguisme-a-samut/

Chaufray M.-P., and B. Redon. 2020. “Ostraka and Tituli Picti of Samut North and Bi’r Samut (Eastern Desert of Egypt). Some Reflections on Find Location.” In Using Ostraca in the Ancient World: New Discoveries and Methodologies, edited by C. Caputo and J. Lougovaya, 165–82. Berlin: De Gruyter.

Cuvigny, H. 2017. “Quand Lichas plantait sa tente à Abbad. Un dossier de distribution d’eau sur la route d’Edfou à Bérénice (c. 240–210a).” Chronique d’Égypte 92 : 111–28.

Cuvigny, H. 2020. “L’élevage des chameaux sur la route d’Edfou à Bérénice d’après une lettre trouvée à Bi’r Samut (IIIe siècle av. J.-C.).” In Les aisseaux du désert et des steppes: Les camélidés dans l’Antiquité (Camelus dromedarius et Camelus bactrianus), edited by D. Agut-Labordère and B. Redon, 171–80. Lyon: Presses de la MOM.

Cuvigny, H., ed. 2022. Blemmyes. New Documents and New Perspectives. Le Caire: Institut français d’archéologie orientale.

Faucher, T. 2018. “Ptolemaic Gold: The Exploitation of Gold in the Eastern Desert.” In The Eastern Desert of Egypt during the Graeco-Roman Periods: Archaeological Reports, edited by J.-P. Brun, T. Faucher, B. Redon, and S.E. Sidebotham, 1–16. Paris: College de France. doi:10.4000/books.cdf.5230.

Gates-Foster, J. 2012a. “The Eastern Desert.” In The Oxford Handbook of Roman Egypt, edited by C. Riggs, 736–47. Oxford: Oxford University Press.

Gates-Foster, J. 2012b. “The Eastern Desert in the Ptolemaic Period: A Developing Picture.” In The Peoples of the Eastern Desert, edited by H. Barnard and K. Duistermaat, 190–203. Los Angeles: University of California Press.

Gates-Foster, J. 2022. “Third Century BCE Supply Networks and Ptolemaic Transport Amphoras from ‘Abbad and Bi’r Samut in Egypt’s Eastern Desert.” In Networked Spaces: The Spatiality of Networks in the Red Sea and the Western Indian Ocean, edited by B. Redon, 347–63. Lyon: Maison de l’Orient et de la Méditerranée.

Gill, J.C.R. 2016. Dakhleh Oasis and the Western Desert of Egypt Under the Ptolemies. Oxford: Oxbow.

Hadji-Minaglou, G. 2007. Tebtynis IV. Les habitations à l’est du temple de Soknebtynis. FIFAO 56. Le Caire: Institut français d’archéologie orientale.

Hepa, M. 2021.Syene V. Zwei Rettungsgrabungen im Norden der antiken Stadt (Areal 62 und 90). Keramik und Befunde. Beiträge zur ägyptischen Bauforschung und Altertumskunde, Band 23. Darmstadt: Philipp von Zabern.

Hudson, N. 2016. “A Hellenistic Household Ceramic Assemblage from Tell el-Timai (Thmuis), Egypt: A Contextual View.” Bulletin of the American Schools of Oriental Research 376: 199–244.

Kaiser, W., et al. 1990. Stadt und Tempel von Elephantine. 17./18. Grabungsbericht. Mitteilungendes Deutschen Archäologischen Instituts Kairo 46. Mainz am Rhein: Philipp von Zabern.

Ladstätter, S. 2010. “Keramische Fundkomplexe aus Areal 15 der Stadtgrabung in Syene/Aswan.” In Städtisches Wohnen im östlichen Mittelmeerraum 4. Jh. v. Chr.—1. Jh. n. Chr., edited by S. Ladstätter and V. Scheiblreiter, 449–73. Wien: Österreicheschen Akademie der Wissenschaften.

Ladstätter, S. 2015. “Greek Pottery from Syene.” In From the Delta to the Cataract: Studies Dedicated to Mohamed El-Bialy, edited by A. Jiménez-Serrano and C. von Pilgrim, 132–49. Leiden: Brill.

Lynch, K. 2018. “The Hellenistic Symposium as Feast.” In Feasting and Polis Institutions, edited by F. van den Eijnde, J.H. Blok, and R. Strootman, 233–56. Leiden: Brill.

Marchand, S. 2011. “La dernière occupation d’une maison d’époque ptolémaïque du village de Tebtynis (Fayoum). Une céramique de transition tardo-hellénistique.” Cahiers de la céramique égyptienne 9: 269–310.

Marchi, S. 2014. L’habitat dans les forteresses de Migdol (Tell el-Herr) durant les Ve et IVe siècles av. J.-C. Étude archéologique. Paris: Presses de l’Université Paris-Sorbonne.

Marouard, G. 2008. “Rues et habitats dans les villages de la chora égyptienne à la période gréco-romaine (IIIe s. av.–IVe s. apr. J.-C.): Quelques exemples du Fayoum (nome arsinoïte).” In La rue dans l’Antiquité: Définition, aménagement et devenir de l’Orient méditerranéen à la Gaule, edited by P. Ballet, N. Dieudonné-Glad, and C. Saliou, 117–28. Rennes: Presses universitaires de Rennes.

Müller, W. 2008. “Excavations in the Town Centre from the Late Period to Late Roman Times (Area 15).” In The Town of Syene: Report on the 5th and 6th Season in Aswan, edited by C. Von Pilgrim et al., 314–38. Mitteilungen des Deutschen Archaologischen Instituts Abteilung Kairo 64. Mainz am Rhein: Philipp von Zabern.

Müller, W. 2010. “Domestic Structures in Graeco-Roman Syene (Modern Aswan).” In Städtisches Wohnen im östlichen Mittelmeerraum 4. Jh. v. Chr.—1. Jh. n. Chr., edited by S. Ladstätter and V. Scheiblreiter, 429–48. Wien: Österreicheschen Akademie der Wissenschaften.

Müller, W., and M. Hepa. 2017. “Two Baths from Syene.” In Collective Baths in Egypt 2: New Discoveries and Perspectives, EtUrb 10, edited by B. Redon, 13–23. Le Caire: Institut français d’archéologie orientale.

Redon, B. 2010. “Les bains et l’armée en Égypte ptolémaïque et romaine.” Le Bulletin de l’Institut français d’archéologie orientale 109: 407–50.

Redon, B. 2018. “The Control of the Eastern Desert by the Ptolemies: New Archaeological Data.” In The Eastern Desert of Egypt during the Graeco-Roman Periods: Archaeological Reports, edited by J.-P. Brun, T. Faucher, B. Redon, and S.E. Sidebotham, 1–21. Paris: College de France. doi:10.4000/books.cdf.5230.

Redon, B., and T. Faucher. 2014. “Désert Oriental, district minier de Samut.” Rapport d’activité de l’IFAO 2013–2014, Le Bulletin de l’Institut français d’archéologie orientale 113: 12–22.

Redon, B., and T. Faucher. 2015. “Désert Oriental, district minier de Samut.” Rapport d’activité de l’IFAO 2014–2015, Le Bulletin de l’Institut français d’archéologie orientale 114: 24–33.

Redon, B., and T. Faucher. 2016. “Désert Oriental, district minier de Samut.” Rapport d’activité de l’IFAO 2015–2016, Le Bulletin de l’Institut français d’archéologie orientale 115: 10–24.

Sidebotham, S.E., and J. Gates-Foster, eds. 2019. The Archaeological Survey of the Desert Roads Between Berenike and the Nile Valley: Expeditions by the University of Michigan and the University of Delaware to the Eastern Desert of Egypt, 1987–2015. Archaeological Reports Series Vol. 26. Boston: American Schools of Oriental Research.