Who Inhabited Dakhleh Oasis ? Searching for an Oasis Identity in Pharaonic Egypt

A variety of textual and pictorial sources demonstrate that the inhabitants of Dakhleh Oasis were considered different from the people who lived in the Nile Valley. The archaeological profile of this oasis, however, displays nothing that is identifiable as non-Egyptian. This paper discusses why such an evidentiary inconsistency exists by examining contextual issues that contribute to the manifestation of identity in the archaeological record. Current theories on the identification of ancient identities are adapted to what is known of Dakhleh Oasis to demonstrate that the lack of non-Egyptian material culture does not necessarily equate to a population in Dakhleh Oasis that is homogeneous with the Nile Valley.

A recognised Egyptian identity was not based solely on ethnic homogeneity but an individual's adaptiveness to an Egyptian way of life and the degree to which they were willing to participate in society (Baines 1996: 343-344, 361-363;Baines and Yoffee 2000: 15;Assmann 1996: 80).Non-Egyptians were those who, by ancient Egyptian standards, lived outside Kmt and did not accept the social, cultural and religious ideologies of traditionally-defined Egyptian society (Baines 1996: 360;Panagiotopoulos 2005: 403).
The Western Desert of Egypt is approximately 681,000 km2 of arid, inhospitable desert.The five largest oases of the Western Desert are, from north to south, Siwa, Bahariya, Farafra, Dakhleh and Kharga (Fig. 1).Permanent habitation in this region is possible only in the oasis depressions which encompass thousands of square kilometres and can dip 100 m below the surrounding desert plateau (Ball 1939: 9-10).Dakhleh Oasis, positioned approximately 800 km south south-west of Cairo and measuring 60 km eastwest and 25 km north-south, has a total area of approximately 2000 square kms (Mills 1979: 166; Fig. 2).The chief Inspector for Middle Egypt and the Oases, Ahmed Fakhry, systematically explored the region in the 1930s to 1940s and 1960s to 1970s (Osing et al. 1982).Currently the Institut Français d'Archéologie Orientale and the Dakhleh Oasis Project excavate at multiple sites throughout the oasis.The material culture of Dakhleh oasis, as understood from the excavation of temple, settlement and cemetery sites, is homogeneous with that of the Nile Valley, displaying nothing identifiably non-Egyptian (Hope 2001: 29;Figs. 3a and 3b).Yet Egyptian textual and pictorial data present the oases of the Western Desert as foreign regions and the people who lived within them as different from the inhabitants of the Nile Valley.For example, a stela dated to Year 34 of the reign of the Middle Kingdom King Senwosret I was erected to commemorate the visit to Thebes of a man named Ikudidi (von Schäfer 1905).Although its provenance is unknown, Ikudidi's claim that he 'made this offering chapel at the terrace of the great god' may indicate an Abydene origin (Simpson 1974: 13).The crude inscription reveals that Ikudidi travelled 'to the land of the oasis-dwellers' (von Schäfer 1905: 124) and considering the trip likely originated at Abydos or Thebes (modern Luxor), this region was most likely the southern oases of Kharga and Dakhleh (Limme 1973: 44 & no. 23;Simpson 1974: 6, 13;von Schäfer 1905: 126-127).This is the earliest known reference to oasis-dwellers on a document from the Nile Valley and it is significant that the region is categorised by the specificallydifferentiated people who lived there rather than merely its geography.

Figure 2. Map of Dakhleh Oasis (by author).
The Royal Ballas Inscription, thought to have been created early in the reign of the Nebhetepre Montuhotep II (Callender 2000: 151;Fischer 1964: 105-106), also of the Middle Kingdom, similarly documents the oases as non-Egyptian.It was found re-used as part of a New Kingdom column-base at Deir el-Ballas and contains the following fragmentary inscription: '… Wawat and the Oasis.I annexed them to Upper Egypt.I drove out the re[bellious (?btkw?)]…' (Fischer 1964: 114).Little remains of the text and it is difficult to determine its context but the document does demonstrate that the oasis needed to be bought under Egyptian rule, either because the region was no longer administered by the Nile Valley, likely as a consequence of rebellion, or possibly because control over the region was practically and ideologically necessary for a functioning and powerful central regime.It is curious that the oasis is discussed in the same context as the region of Wawat, a situation also present in an Old Kingdom text, namely the Autobiography of Harkhuf (Edel 1955).That both needed to be annexed to Egypt illustrates that they were perceived in a similar manner: foreign and located outside what was considered traditional Egyptian territory.Two documents, the Banishment Stela and the Tale of Woe, depict the oases and their inhabitants in a decidedly negative light.The Banishment Stela, also known as the Maunier Stela, is likely to have originated from the Temple of Amun at Karnak and was probably produced during the Twenty-fifth regnal year of Pinodjem I (von Beckerath 1968: 7, ln.12) who assumed rule in Thebes towards the end of the reign of Smendes I in Dynasty 21 (Jansen-Winkeln 2006: 224;Kitchen 1986: 258-259).It documents the arrival in Thebes of the High Priest of Amun Menkheperre, and a series of subsequent events (von Beckerath 1968: 13, lns. 15-16).Of interest to this discussion are comments concerning people who had been forcibly sent to 'the oasis, where they are banished' (von Beckerath 1968: 12, ln. 11).The text details that the God Amun made a decree: 'You shall listen to my voice on this day and you shall relent toward the servants, who you have banished to the oasis, and they shall be bought back to Egypt' (von Beckerath 1968: 13, lns. 15-16).It is not apparent to which oasis, or oases, the text refers but the southern oases of Dakhleh and Kharga are likely candidates due to their proximity to Thebes (Kaper 1992: 119-122).In documenting that unwanted peoples were banished, this text indicates that the oases were clearly delineated as undesirable places that were located, geographically and socially, outside the traditional boundaries of Egypt.
The second document is Papyrus Pushkin 127, also known as the Tale of Woe.It is a fictional narrative in the form of a letter written by a man named Wermai to his friend Usimarenakhte and is dated to Dynasty 21 of the Third Intermediate Period, circa 1000 BC (Caminos 1977: 1-2, 78).The Tale of Woe is a lamentation made by Wermai about his unfair exile from his city and dispossession of all his belongings (Caminos 1977: 70-71, lns. 2.5-2.11).Wermai describes travelling throughout Egypt and west of the Nile Valley including the southern oases (Caminos 1977: 71, lns. 2.12-3.4).In the text the oasis-dwellers appear devoid of kindness and empathy in what may be a demonstration of the lower standing afforded the people who inhabited regions outside Egypt (Loprieno 2003: 47-49).The Banishment Stela (von Beckerath 1968: 13, lns. 15-16) demonstrates that the oases were once considered undesirable places for habitation by people of the Nile Valley, while the Tale of Woe (Caminos 1977, 71-72, lns. 3.5-5.6)reveals that although the southern oases were under the control of the Nile Valley administration early in the Third Intermediate Period, the regions were perceived as decidedly unpleasant and the inhabitants disagreeable in temperament.
In the New Kingdom pictorial scenes from the tombs of private individuals buried in Thebes show people, including oasis-dwellers, presenting to the king taxes in the form of tribute (Giddy 1980: 121-122;1987: 70;Newberry 1900: 35).Those identified as coming from the oases were depicted as different from the Egyptians and the other subjugated groups by the manner of their dishevelled and discoloured hair, unusual eyes and distinct, animal skin-like dress.A clear oasis-type is on display.For example, the figure from Tomb 131 of User has a kilt made of animal skin that is cut away from the front with a long white flap and a long tail falling down the back leg (Fig. 4; Giddy 1980: 124).The tomb of Huryw from the reign of Amunhotep III or IV is also particularly interesting; it has a scene showing what is described as 'women who were brought from the oasis for the raising of the Djed-pillar' (Fakhry 1943: 483, pl. XLVIb;Giddy 1987: 81).Their depiction, while not exceptional, is nonetheless interesting as the dancing women were shown with close-fitting caps, necklaces, double-layered skirts and bare chests crossed by two straps held at the waist.This iconography is particularly significant for it recalls the Dynasty 26 representations of Taneferbast and Bastetirdis in the tomb of Tjaty in Bahariya Oasis (Figs. 5 and 6;Fakhry 1942: 127, fig. 97, 130, fig. 101;Porter andMoss 1927-1999: VII, 303) and may be indicative of consistent oasis-dweller depictions.The function of the dancing women within the ceremonies is not known but they were purposely identified as being from a region apart from Egypt.
Their novelty was due to their origin, their appearance or perhaps even their particular dancing style (Fakhry 1943: 483;Giddy 1987: 80-81).Panagiotopoulos (2005: 388) stresses that these scenes were created, not as graphic representations of the historical reality, but as decorative works of art.It is clear, however, that there was a desire to emphasise the distinctiveness and otherness of the oases and their inhabitants.Indeed, their appearance may have been exaggerated or even stereotyped to accentuate this difference.The people of the oases and all non-Egyptians functioned to demonstrate Egyptian superiority over all who were not of the Nile Valley (Giddy 1987: 76) and their depiction within these scenes as exotic subordinates was a way to signify their inferiority and lowly status in the hierarchical ideology of the Egyptian worldview.ancient Societies and the recognition of Identity Is it possible to reconcile the archaeological profile of Dakhleh Oasis that is homogenous with that of the Nile Valley with the Egyptian textual and pictorial sources that demonstrate the non-Egyptian nature of the oasis-dwellers?Indeed, with self-ascription a fundamental aspect of identity (Cordell and Yannie 1991: 97), is it even possible to distinguish between the people who lived in Dakhleh Oasis and those of the Nile Valley?
The evidence that proves significant in the structuring of identity may be contextually variable and dependent on the type and scale of social interaction (Jones 1996: 72) and so the identification of a distinguishable oasite community requires an understanding of the nature of archaeological preservation and the possible ways in which difference may be expressed.
As identity is sustained and reproduced as part of social processes, the pursuit of ancient identities as can be understood from material remains is not merely a facet of archaeological inquiry but an essential contribution to the overall purpose of archaeological enterprise.An appropriate and concise definition is provided by Diaz-Andreu and Lucy (2005: 1): 'identity will be understood as individuals' identification with broader groups on the basis of differences socially sanctioned as different.'Although a consensus of what cultural aspects contribute to the recognition of identity has yet to emerge, critical indicators seem to be very specific (McGuire 1982: 174).Household architecture serves as a tool through which aspects of the daily ritual and behaviour may be understood (Emberling 1997: 325).Individual and group language, presentation of dress, jewellery and hair, religious and domestic rituals, the preparation and disposal of food and the tools used to prepare it and the choice of funerary architecture are also possible indicators (Jones 1996: 68;Lucy 2005: 91-101;Meskell 2007: 24-25;Tyson Smith 2003: 7).Identity differentiation may also be maintained by territorial, behavioural and/or ideological markers (Burgess 1978: 270;Eriksen 1991: 127).Despite the possibility that these categories do help to recognise identity, its dynamic and multifarious nature means that any attempt to define an ancient identity requires a situation-specific and contextual understanding of the situation.
The discussion of identity in reference to archaeological enquiry is not a recent development (Jones 1996: 63-64).Traditional archaeological methods concerned with the identification of peoples have tended to focus on the distribution of material culture, which was seen as delineating the spatial distribution of a particular group (Lucy 2005: 86).It was assumed that peoples, be they designated as tribes, ethnic groups or races, correlated to uniform and identifiable bounded cultural entities (Jones 1996: 63-64).
In the 1960s and 1970s theories concerning cultural differentiation underwent a major shift.Terminology such as 'ethnicity' and 'ethnic group' replaced traditional markers of difference such as 'race' and 'tribe'.This signalled the increased emphasis that was to be placed on understanding the nature of groups, particularly self-identification and group interaction (Jones 1996: 66).The recognition of difference for members of a shared social system and its intentional maintenance saw Barth (1969: 14) in his seminal work Ethnic Groups and Boundaries suggest that it is 'the ethnic boundary that defines the group, not the cultural stuff it encloses' (his emphasis).With the advent of post-processualism in the 1980s the perception of ethnicity shifted towards the idea that it was part of a social process and was therefore relative, subjective and required an interpretative approach (Diaz-Andreu and Lucy 2005: 6; Jones 1996: 66-68).
In recent years, however, there has been movement away from the exclusive use of ethnicity as a means of discussing ancient identities.As a fundamentally subjective manner of categorisation defined by the relationship and opposition to other ethnic identities, this usage appears more concerned with self-identification than the observations of objective parties (Jones 1997: 60-61, 64;Tyson Smith 2003: 6).Such an approach fails to account for limitations in the construction of these definitions and requires the perpetual maintenance of the groups' boundaries through the articulation of similarity and difference.This is particularly problematic when a person's association with an identity is fluid, susceptible to change and can be contextually-dependent and, indeed, when the individual has an affinity with more than one identity (Diaz-Andreu and Lucy 2005: 11; Jones 1996: 91).
A primary assumption underlying the culture historical approach is that bounded uniform cultural identities correlate with particular peoples, ethnic groups, tribes or races.When discussing peoples in antiquity in terms of their identity it is perhaps more constructive that ethnicity is examined as part of a larger system (Lucy 2005: 95-97).This approach, which is attempted in this article, aims to encompass features such as age, gender, status and ethnicity, moving away from an individualistic examination of these aspects (Diaz-Andreu and Lucy 2005: 9).

a non-Egyptian Identity in Dakhleh Oasis
The definition and recognition of an identity from the archaeological record is a challenging task even when aided by textual data.Indeed, the search for an oasite identity is problematic as the relevant activities are not necessarily those that produce the most identifiable archaeological remains.For this examination of the identity of the oasis-dwellers of Dakhleh the possible indicators of difference between the oasis and Nile Valley populations, namely foodways, language and funerary behaviours, are investigated, as are the contextual circumstances that may have resulted in the absence of non-Egyptian indicators.The aim is to demonstrate the approaches that can be used to attempt to recognise identity as well as specific practical and methodological issues arising in the search for distinctively non-Egyptian or even oasite activity in Dakhleh Oasis.

The identification of non-Egyptian Language
As an intrinsic aspect of Egyptian identity, language can serve as an indicator of commonality as well as difference (Kemp 2006: 20-21).In Dakhleh Oasis the prospect of identifying aspects of the spoken language that may have differed from what was spoken in the Nile Valley, if indeed there was a difference, is extremely difficult.Many factors were likely to have contributed to the loss of linguistic differences.The ability to speak the language spoken by the elite of the oases, whether this was Egyptian or a regional dialect, may have been a strong identifier for acceptance within such a social group.The economic, political and social advantages afforded to those who spoke this language could even have influenced its use in Dakhleh Oasis.Texts from the oasis such as the Greater Dakhleh Stela (Gardiner 1933; Spiegelberg 1899), the Smaller Dakhleh Stela (Janssen 1968), Stela JE 52478 (van Zoest et al. 2006) and the Amhida Stela of Takelot III (Kaper and Demarée 2005), all of which date to the Third Intermediate Period, are available for examination.The non-elite oasis occupants not represented in the documentation may have spoken an oasis dialect; however, confirmation of the language(s) spoken by the oasis-dwellers is not possible from these texts because they were of the elite administration and not necessarily a reflection of contemporary oasis speech.While language has the potential to demonstrate and perhaps even differentiate the identity of the oasis-dwellers and the inhabitants of the Nile Valley, the Egyptian sources that provide our evidence have no way of preserving such distinctions.

The Identification of non-Egyptian funerary Ritual and Behaviour
The presentation of an individual upon death, the death ritual and associated architecture can be one of the more powerful indicators of identity, functioning as tools of differentiation through the maintenance of ancestral and familial ties (Emberling 1997: 325;Tyson Smith 2003: 7, 38, 40).These practices can be direct indicators of belief and a physical and deliberate expression of identity, which can differ markedly between different groups.
The burials in Dakhleh Oasis are consistent with the contemporary funerary customs of the Nile Valley (Baud 1997: 27).Excepting the cemeteries of Qal'a ed-Dabba and 'Ain Tirghi, from which over 50 graves and tombs from multiple periods have been excavated, and Tell Marqula with approximately 25 excavated burials (Frey 1986;Mills 1983: 128-129;Yamani 2002), the pharaonic cemeteries of Dakhleh Oasis have not been systematically examined or excavated to a significant degree (Mills 1980: 266;1981: 183).While there is a widespread selection of inhumations from a number of cemeteries, the number of graves and tombs excavated as a proportion of the total burials within a cemetery is generally extremely small.Although this is common archaeological practice it nevertheless reduces the likelihood of finding or identifying any non-Egyptian artefact, practice or enclave.To increase the probability of discovering difference in funerary practices such as a regional oasite variation a greater number of graves within the cemetery would need to be excavated.
The absence of non-Egyptian burial indicators in Dakhleh Oasis would be indicative of a population consistent with the Nile Valley if these peoples held a monopoly on the oasis funerary behaviours.It is possible, however, that the oasis-dwellers practiced Nile Valley rituals to such a degree that any non-Egyptian indicators were indistinguishable in the archaeological record.Particular social circumstances can result in the alteration or loss of identity indicators (Cordell and Yannie 1991: 107) and the adoption of Nile Valley burial practices may have functioned, in some advantageous manner, as a method of projecting a desired status, be it economic, political or cultural (Tyson Smith 2003: 156, 197).The wealthy and ruling elite effectively monopolised the use of monumental inscriptions and were the people with the means to create monuments and artefacts of preservable quality (Baines 1996: 353, 358;Tyson Smith 2003: 156).Those who were part of a formal cemetery may reflect the more wealthy members of the oasis population and if a non-Egyptian were to be part of this economic elite, a requirement of that membership may have been to display cultural practices consistent with the Nile Valley.

The Identification of non-Egyptian Foodways
Aspects of food preparation, culinary equipment, the type of food eaten and its quality can be clear social markers for group identification as well as indicators of ancestry and wealth (Bresciani 1997: 226;Hamilakis 1999;Lucy 2005: 105;Tyson Smith 2003: 43-52, 189-193).Ancient food production, consumption and disposal, or 'foodways,' are determined through the examination of floral and animal remains, food residue and the tools and containers employed during food use in both ritual and domestic settings (Tyson Smith 2003: 50, 52-53).
Domestic ceramics have been recovered from sites such as Mut el-Kharab, Amhida, 'Ain Aseel and 'Ain el-Azizi (Hope et al. 2000: 192;Hope et al. 2006: 35-36;Marchand 1997: 5;Pyke 2005: 18-20) but, as has been noted, no non-Egyptian material has been revealed.Botanical evidence from Mut el-Kharab is under preliminary investigation by Dr Thanheiser from the Department of Systematic and Evolutionary Biology of the University Vienna and in recent years she has been assisted by Amy Veller of Monash University.The 2001-02 excavations garnered samples which revealed cereals in the form of rice, barley and wheat, as well as pulses, flax and fruits including grapes, olives, dom-palms and nuts (Mills 2002: 15).The organic material of the 2008 season is mostly burnt and desiccated and difficult to examine comprehensively but it has been identified as predominantly wheat-based (A. Veller, pers. comm. 2008).Analysis of the archaeobotanical samples from Area 4.1, the location of the pharaonic temple dedicated to Thoth at Amhida, has also revealed the remains of wheat, barley, grape, olive tree, date palm, fig, lentil, flax, cotton, safflower, coriander, rosemary and a variety of grass weeds (Bagnall 2006: 25-26).
There is nothing unusual in these remains that would warrant the identification of a non-Egyptian food pattern.Such recognition could only occur if the floral and faunal food remains and the culinary tools and cooking vessels are demonstrably different from those of the Nile Valley, a task necessitated by a concerted research programme designed to obtain as much foodways data as possible for the express purpose of examining culinary practice.Nevertheless it may be possible to identify differences in the quantities and types of food consumed by the oasis-dwellers and the Nile Valley inhabitants, for example less fish in the diet of the former, and local environmental pressures were more likely a contributing factor to the foodways of the oasis than specific or traditional culinary provisions.Unlike textual sources and elite burial remains, the manner of food preparation is less likely to be compromised by any deliberate agenda to distort the depiction of an identity.If it is the intent of the archaeologist to determine specific foodways, there is the real potential to reveal differences that may be seen in the preferred manner of preparation and consumption.

conclusions
The dynamic and situational nature of identity means that cultural units rarely correlate directly to recognisable boundaries (Jones 1996: 68) and the reliance on a straightforward interpretation of the textual and archaeological evidence can often mask the recognition of past identities whose complexity and flexibility are rarely easily identified (Insoll 2007: 13-14;Tyson Smith 2003: 101).The expression of identity in the archaeological record could be manifested in a myriad of ways (Tyson Smith 2003: 32), with the identification of non-Egyptian foodways, language and funerary behaviours is a task made difficult by several factors.The most important of these concern the lack of complete cemetery excavation, the unavoidable reliance on classical and Egyptian texts that have the potential to present distorted or biased views of the oasis-dwellers and, lastly, the possibility that status and wealth contributed to the manifestation of wholly Egyptian language and funerary behaviours in the oasis.
Egyptian ideological influence on the use and representation of language, burial and religious evidence is unavoidable but not prohibitive, for a contextual analysis of the social, political, religious and economic situation will allow for an understanding of the factors that contribute to ancient identities.The material culture of the oases, while homogeneous with that of the Nile Valley, does not necessarily contradict the Egyptian textual and pictorial sources that demonstrate the perception of difference because the very systems that contribute to identity -language, cultural and funerary traditions -are often unrecognisable in the archaeological record.

Figure 1 .
Figure 1.Map of Egypt including the oases of the Western Desert (by author).

Figure 3b .
Figure 3b.Faience figurine of Ta-weret from Trench 16, Mut el-Kharab.Another example of the oasis material culture which is clearly identifiable as Egyptian (Hope et al. 2005: fig.11a).

Figure 3a .
Figure 3a.A fragment of a funerary inscription of Sa-Igai from Mut el-Kharab, Dakhleh Oasis.This item exemplifies the Egyptian nature of the material culture of Dakhleh Oasis (Hope et al. 2008: fig.6).