Rock Art of the Eastern Desert



As discussed on the Rock Art Introduction page, I have chosen to isolate rock art from the other archaeological data discussed on this site on a page of its own, mainly because of two key problems overviewed on the Analysis page:  a) dating rock art depictions and sites and b) the related problem of tying in the rock art data with other scant forms of archaeological data from the Eastern Desert. 

I have not included the hieroglyphic and Graeco-Roman inscriptions or other depictions in this section.  Those are described on the Pharaonic and Graeco-Roman pages of the Archaeology section.  This page deals only with images, not scripts, and the main emphais is on isolating rock art from Pharaonic and Graeco-Roman images for purposes of analysis.  However, I agree with Morrow and Morrow when they say of their own research into Nilotic and desert images of the third and 4th millennia BC that “this approach restricts the attention later carvings merit, particularly those from the Graeco-Roman period onwards, and that these demand separate ananlysis” (2002, p.17).

The rock art of the Eastern Desert is based mainly in the wadis of the Red Sea Mountains.  These are vast, dried river beds, often very wide with steep, rocky sides which provide surfaces, many of which have been etched with engravings thought to date from the prehistoric period onwards.  Many of these “sites” contain large numbers of images (and sometimes texts), which have been added to over centuries. The EDS’s Site TJ-1 (Winkler’s 24b) in the Wadi Mineh is a good example, with images of boats thought to belong to the Predynastic period near to those dating to the Pharaonic periods and accompanied by Graeco-Roman texts. 

When the term “rock art” is mentioned, if often invokes images of Altamira and Lascaux where stunningly beautiful painted images of wild animal coat the walls of Franco-Cantabrian caves.  Saharan rock art is more modest both in its size and its artistic achievement, but still offers some remarkable examples of style, movement and colour. This is not to undervalue the Saharan rock art, but to put it into a slightly different context – it seems clear that the actual rendering of each image was less important, although the motifs and locations themselves may have had very particular significance to those that left the engravings.  It may be painted or engraved, and exhibits enormous technical and stylistic variation – probably because of the periods over which it was inscribed, but probably also partly due to cultural and traditional differences between contemporary groups.  Eastern Desert rock art is less visually impressive.  It consists of engravings, not paintings, and is usually made up of a series of highly stylized linear sketches, either in the form of isolated imags or in scenes. They are usually on rock faces and boulders.

There are a number of problems associated with the analysis of rock art in North Africa in general.  As already highlighted, dating is probably the most serious problem. These problems need to be understood, and have been listed in some detail on the Analysis page. This is not to under-value rock art studies – research projects are now dedicated to the study of rock art, and new approaches, methodologies and techniques are being applied with increasing levels of success.

Rock art in Egypt has been recognized and discussed for a long time.  I cannot improve on Dirk Huyge’s summary of the situation (2002, p.192):

    “Rock art in Upper Egypt and Lower Nubia (Northern Sudan) has been the subject of a vast body of literature, both ancient and modern (over 420 references in Hendrickx 1995).  As is the case for many rock art areas in the world, studies in this field have focused primarily on the recording of images, thematic and stylistic classification, and. to a much lesser degree, chronological seriation and cultural historical attribution.  Aspects of interpretation, that is, the meaning and motivation of the art, have for the most part been left unaddressed.”

A number of different survey teams have recorded new and re-recorded previously identified rock art findings in the Eastern Desert.  However, as Francis Lankester points out on the Wadi Barramiya page on his website, there are problems pulling all the data together to form a single cohesive record of different records, which is very unfortunate. 

There are also problems relating to the different naming conventions and site names adopted by different survey teams.  For this reason, where a site is mentioned, it is usually prefaced by the name of the survey team that gave it the name used in the text (e.g. E.D.S. TJ-1 is a collection of rock engravings named TJ-1 by the Eastern Desert Survey project lead by David Rohl). A “Site Decoder” in Appendix J, which aligns different naming systems with the identical sites that the different teams have labeled.

Rock art may be a useful support to archaeology when it can be isolated chronologically, and interpreted with some degree of confidence.  It is an archaeological anomaly that in the Eastern Desert more work has been devoted to rock art than any other form of archaeology, which may have given a rather unbalanced view of the archaeology (Majer 1992) – particularly for the prehistoric period, but also means that there is a useful, albeit incomplete and geographically limited corpus of survey data with which to work.  As Fuchs (1991, p.59) points out, rock art research is important, but it needs to be supported by other research including field survey and excavation to identify and examine other types of site.

Egyptian rock art studies are probably going to be at their most useful when considered in the light not only of other Egyptian data, but also in the context of other North East African and southern Levantine rock art and archaeology.  Smith (2005) has suggested that the lack of published academic research about North African rock art, and the sidelining of the work that has been completed, has lead to Saharan rock art being excluded from major comparative and global rock art studies, like that of Chippendale and Tacon (1998).  A short discussion of North East African rock art  can be found in Appendix E, but this is very superficial at the moment and needs to be expanded.

Dating Eastern Desert Rock Art

Introduction – Direct Dating vs. Relative techniques

Dating the rock art of the Eastern Desert has been done only using relative techniques.  None of the scientific (or direct)  dating techniques used in rock art projects elsewhere have been attempted here.  Studies have looked at a number of different ways of establishing which images might be older than others nearby.  Some of the studies will be discussed below.  For a general overview of the problems with dating rock art, see the Dating section on the Analysis and Issues page. 

One of the key problems in the Eastern Desert is that art and graffiti were clearly added to particular sites for 100s, even 1000s of years.  An example is Winkler’s Site 24b in the Wadi Mineh, where there are images dating from as early as the Predynastic (it is believed), through to the inscribed texts of the Graeco-Roman periods – and beyond. 

Another problem is that few organic remains survive in rock art engravings to provide the possibility of a direct date.

Sealed deposits

Unfortunately there are no Eastern Desert rock art faces sealed beneath earlier deposits.  The only site where this occurs in Egypt is at Abka, which is in the region of the 2nd Cataract.

Detailed Chronological schemes

The earliest attempt to impose some sort of order on the general chaos of rock art inscriptions was probably Winkler who suggested two schemes.  First, he divided the petroglyphs into four periods during which he believed they were carved:  Predynastic (the majority of engravings), Dynastic (few engravings), Blemyan (Graeco-Roman and Coptic) and Arab. He further divided the Predynastic into five levels of detail:

  • Wedge-shaped figures
  • Figures wearing wigs similar to those depicted in Naqada I
  • Boats with standards dating to Naqada II
  • Figures with plumed head-dresses shown with boats with high prows and sterns that he thought came from beyond the Red Sea
  • Figures with penis sheaths who used the Red Sea hills for cattle husbandry and herding

Winkler’s divisions were criticized, partly due to his methodology, and he himself revised his typology at least once.  Fuchs (1989) considered that his conclusions were largely invalid due to the small sample used (1989).

Cervicek used existing data to provide a new scheme in 1974.  He suggested that there were three main phases of rock art to be observed, including the Predynastic:

  • From second half of the fourth millennium to the mid third millennium BC
  • From the mid 3rd millennium to 1000 BC
  • From 1000 BC to the 6th Century AD

He produced a revised chronology in 1986, based on comparison with Nile artefacts, which consisted of six different time periods for the engravings, but is now considered to be outdated and in need of revision (Morrow and Morrow 2002, p.16):

  • A – 4000 BC
  • B – 2100 BC
  • C – 1400 BC
  • D – 1050 BC
  • E – 250 AD
  • F – post 250 AD

The Redfords (1989) proposed a relative dating system based on the part of the rock face that was used by rock art creators.  They suggested that the best part of the rock face was used by the earliest artists.  However, this has not met with general acceptance.

Huyge (2002) devised a scheme consisting of seven horizons fitted into a framework of four general chronological periods for the region of Elkab, which might be useful to compare with Eastern Desert art where motifs, techniques and styles display, at first glance, a number of key similarities (although a key difference is the emphasis on Ethiopian type fauna – elephants, giraffes and other large species):

  • Phase 1
    • Horizons I and II – Middle and Late Predynastic, starting with Naqada I
  • Phase 2
    • Horizon III – the Terminal Predynastic and Early Dynastic
  • Phase 3
    • Horizon IV and V – The Old to New Kingdom
  • Phase 4
    • Horizon VI and VII – the Late Pharaonic to Post-Antique period

The most promising type of motif in the desert for cross-dating purposes is probably the boat, and this is the most commonly used theme of the various different schemes that have attempted to create a chronological framework for Egyptian Desert rock art.  However, Vinson points out that most of these schemes are conflicting (1987, p.124-162), highlighting the difficulties of developing a rock art typology for any collection of motifs, and there is evidently still a lot of work to be done before a framework can be agreed upon.  Both Wilkinson (2002; 2003) and Morrow and Morrow (2000; 2002) consider motifs shown in rock art versus Naqadan ceramics and other artefacts in the Nile Valley.  These will be discussed below.

Typological comparisons with Naqadan motifs

This initially promising line of work in the Eastern Desert has been proposed by two writers in particular, discussed below, and continues to be interesting but has not been acceptable to a number of other scholars. 

The method involves identifying close similarities between rock art motifs and motifs shown on closely dated artefacts from Nile Valley contexts. As Wilkinson puts it, “The trick is to match up some of these securely-dated decorated objects from graves with similar motifs in the Eastern Desert rock art (2003 p.63).

Cervicek (1974; 1986) developed a chronological system which involved a system of horizons, based on comparisons with Naqada iconography.  This approach attempts a) to establish which images are generally older than others, and b) to cross-date method distinctive rock art images with those in more securely dated contexts, like images on Naqada I and II ceramic vessels.  The raised loop-armed figure, often with twin plumes on its head, an image shared by both rock art images and Naqada II vessels, are examples of motifs used in cross-dating. Other schemes have been proposed by Dunbar (1974), Engelmayer (1965), Bassch and Gorbea (1968) and Hellstrom (1970).  Fuchs suggests that the work carried out by Cervicek (1974) is particularly useful for dating boats, but as mentioned above Vinson points out that the various different schemes conflict with each other, and that this is probably because the schemes are too fine-grained (1987).

Wilkinson (2000; 2003) suggests that comparison of motifs and styles might help to tie in rock art to other data.  Some forms, like boats, are quite distinctive in Pharaonic tombs and temples, and could be used to isolate friezes which contain these distinctive forms.  For the Predynastic period, Naqada I and II pottery have very distinctive themes painted onto their surfaces, and some of them seem to equate very closely with rock art motifs. A basically sound typology has been established for  Naqada period pottery. In Followers of Horus, Wilkinson (2000) highlighted the difficulties of using artistic traditions to date rock art:  “Little is known about the  artistic traditions of Egypt’s margins and we run the risk of miss-dating rock art based upon parallels with decorated artefacts from the Nile valley (where the sequence of artistic development is much better understood” (p.160). However, he also suggests that this line of enquiry offers the possibility of identifying cultural ties and “chronological clues” (p.160), and in a more recent assessment (2003) he offered an optimistic and detailed approach to this type of comparative study.  His main arguments are as follows, and will be discussed below.  Wilkinson’s main source of comparison for rock art motifs is the corpus of Naqada I and Naqada II decorated pottery (C-Ware and D-Ware respectively).  He sees a close correspondence between the themes and combined motifs of Naqada I C-Ware and rock art, including profile choice, style and nature-orientated subject matters.  He points to differences between boat motifs in Naqada II and rock art depictions, warning against drawing superficial comparisons based on subject matter without looking at details of the depictions involved. Wilkinson focuses, therefore, on similarities between Naqada I and rock art motifs.  He draws on three particular objects – a square hulled boat on a Naqada I vessel, a unique object in a grave at El Amra and the unique Gebelein cloth.  In all three cases he sees striking similarities.  Other items are not unique  and are therefore more useful, including the human-shaped forms with closely comparable fronds, and loop-armed stances. He concludes that “the evidence is overwhelming in indicating a date of Naqada I, or c.4000BC for much of the rock art” (Wilkinson 2003, p.82).

The conclusions that Wilkinson (2003) presented, particularly in Genesis of the Pharaohs, were not universally accepted.  One of the problems that I have with Wilkinson’s approach (2003) is that he some of the artefacts that he selects for comparison are one-off examples.  He is using items which are archaeologically unique to form generalizations about Naqada I motifs, and extrapolates these to encompass Eastern Desert rock art iconography.  However, the close correspondence between loop-armed figures and figures with twin plumes with Naqada II ceramics do appear to be convincing.

Morrow and Morrow (2000, p.183) discuss some of the problems using Naqadan motifs to draw analogous references with Eastern Desert rock art.  They pick out the main features of boats depicted on Petrie’s D-Ware pottery, which seem to follow fairly clear trends and repeat the same motifs (p.183) and compare them with depictions of the desert boats, which show much less repetition of, or commitment to specific motifs and combinations of motifs (p.183).  They conclude that comparison between D-Ware pottery and desert boats “shows that the latter possessed no such formalistic approach, and that even carvings from the same site offer a variety of styles”.   Their conclusions are summarized as follow (p.183):

D-Ware Pottery boats

Desert Rock Art

Usually sickled shaped boats

Boat types are mixed, and often show very similar patination at a single site

Most have two central cabins

Cabins are not always shown – there may be none, one or two.

Most have a pole and standard

Few boats have standards

All have a bow attachment, usually a palm frond

Prows and/or sterns usually have some form of decoration

Oars below the hull are usually separated by a mid hull gap

Oars are not always shown

No vertical above-hull lines are shown (oars or crew)

Oars and crew are often shown above the hull lines

It is common to have large figures, some dancing  or with arms looped above their heads (“loop-armed”).

Dancing and loop-armed figures do occur, but by no means always.

Rare occurrences only of stern steering oars

Steering oars are rare

Morrow and Morrow (2000) state on the basis of the above data that “these general assumptions may be unsafe as there is significant stylistic variations between the sample groups (p.183).

In the same paper Morrow and Morrow also look at the specific artefacts which are sometimes cited in attempts to tie in securely dated motifs with rock art:  the Gebel el- Arak knife, the Gebelein painted linen fragment, and the Hierakonpolis Tomb 100 boat depictions.  Their analysis of each of these artefacts (2000, p.182-185). There are even doubts as to the authenticity of the Gebel el-Arak knife.  Morrow and Morrow state that “there are considerable stylistic differences between the selected artifacts; so much so that there must be doubt whether they were operated by the same culture, represent the same objects or were fabricated at a similar time”. My own reservations concern both the authenticity and proposed date for the Gebel el-Arak knife and the sheer number of interpretations of Tomb 100.  On the subject of Tomb 100, Steve Vinson raises some serious concerns:  “ Vinson says of T100 that “Because this painting was bade before any kind of decipherable writing was developed in Egypt, no-one has ever advanced a convincing explanation as to what the artist meant to portray (Vinson 1994, p.13).  Examples of different previous explanations that he provides are:  funerary voyage, a naval battle, daily life on the Nile, events in the life of the tomb owner and the representation of aspects of the Heb-Sed festival (1994, p.13-14) all of which go to show just how difficult it is to interpret Tomb 100 in any definitive way. 

In a later work, Morrow and Morrow (2002, p.18-20) take an independent look at possible connections between Naqadan artefacts from the Nile Valley and Eastern Desert rock art motifs, focusing mainly on ceramics:  “Dynamics insist that through the sheer quantity discovered, decorated pottery should form the main part of reference with the petroglyphs.  Of particular interest are the white-cross lined C-ware and buff D-ware” (2002, p.18).  They point to three particular motifs which form a useful point of reference:  linked human figures one of which is often shown in the loop-armed pose, plumed figures with maces or clubs and scenes of boat towing.  Morrow and Morrow look first at C-ware from Naqada I, which is characterized by vessels in a number of shapes and styles featuring numerous different motifs, which are far more varied than those in later traditions and include representational and geometric forms.  There are no parallels for the geometric designs in Eastern Desert rock art, and an analysis of the representational forms indicates that there are few parallels between boat forms.  The closest correspondence comes in the hunting scenes:  “Hunting scenes on specimens of C-ware pottery mirror those found in the rock art with hippopotami and antelopes being the most common quarry” (Morow and Morrow 2002, p.18).  Morrow and Morrow then consider D-ware pottery (p.19-20).  they suggest that Naqada II D-Ware was technologically superior and far more formalized in terms of style than Naqada I C-ware, but that it was very short lived:  “Standard D-ware pottery examples demonstrate such remarkably similar characteristics that it can be assumed a local ‘elite’ art style and manufacture were responsible” (Morrow and Morrow 2002, p.19). They conclude that D-ware motifs are quite unlike those featured in rock art in a number of significant ways:  “The petroglyphs  show no such formalised approach, and sickle, incurved sickle, square, incurved square and flared boats often appear on the same surface and bear a similar degree of patination” (Morrow and Morrow 2002, p.20).  They list another set of D-ware features which differ from those seen in rock art and conclude that D-ware, which is  found primarily in funerary contexts “was prepared for a (male) ruling elite by craftsmen attached to official workshops, while the desert images are almost certainly the work of non-specialist artisans for a variety of purposes and as such complex analogies based on perceived similarities or dissimilarities are difficult to sustain” (Morrow and Morrow 2002, p.20).

Another writer who doubts the viability of C-Ware and D-Ware pottery for making typological comparisons with Eastern Desert rock art is David Wengrow who says that although the approach is “appealing” (p.112), there are problems with the lack of precision between the proposed correspondences, and the difficulties of accounting for the discrepancies.  He gives as an example the linear C-Ware animals which would have been easy to reproduce on rock surfaces, but rarely are (p.112).  He also discusses discrepancies between boat image forms.  Finally, he points out that many decorative motifs on pottery vessels are simply never reproduced in the rock art examples.  He concludes: “It may be misleading to expect direct correspondences between painted designs on pottery and the often much larger images incised on rock.  However, the alternative method of asserting relationships purely on the basis of shared subject matter or general resemblance presents considerable pitfalls”.

Adopting a more narrowly focused geographical approach, Berger (1992), focusing on a narrow set of areas, sees close similarities between images of boats ceramics from Hierakonpolis and those of Wadi Hammamat and Wadi Mia:  “The boats from Hierakonpolis are clearly part of an artistic tradition in southern Egypt and lower Nubia” (1992, p.118).  He believes that the closest parallels are with the Wadi Mia:  “This shared style suggests that the Egyptian artists were conveying ideas common among these different regions” (p.118).  Wadi Mia is a north-eastern extension  of Wadi Abbad, which is the main route between Hierakonpolis and Ras Samadi.

Taking a completely different line, Marks (1997) suggests that the depictions of different types of boat actually represent the evolution of different boat types, constructed using different materials, but he does not attempt to impose a chronological scheme on the different boat types.

The best comparison between rock art boats and those represented in models, on vessels and as physical remains, is that produced by Steve Vinson’s thesis about boats before the Old Kingdom.  Instead of trying to develop typologies that have fine-grained divisions, Vinson chooses to use broader strokes, and his results are less detailed but they are more convincing.  I have summarized his findings about the changes in boat morphology throughout Egyptian prehistory and Prehistory in the table in Appendix L.  Although he can only work with explicit data, he also comments that “Egypt was inhabited as early as the Palaeolithic Period and it is difficult to believe that there was ever a time when humans failed to take advantage of the ubiquitous papyrus to build rafts or floats” (1994, p.11).  There is no actual evidence for the use of boats in Egypt, however until between 6000 and 5000BC, with model boats found at Merimde and later from a Badarian site (click here to go to the page on the Petrie website for an image) where a boat is shown with built up sides probably made from papyrus (Vinson 1994, p.11).  Data following the Predynastic includes actual boat remains, images on temples and in tombs, models and images on papyrii. Vinson’s work may well represent the best framework for tying in boat petroglyphs and Nile images.  This in turn may help to tell us not how the Eastern Desert was used, but perhaps where and how it was used.

Analysis of Faunal Assemblages

Attempts to date rock art by analysis of the animals depicted and comparing them with environmental conditions are fraught with problems.  The basic premise is that as climate changes, certain animals can no longer survive, and depictions of these animals must belong to periods when climate was more favourable to plant and animal survival.  However, no-one is suggesting that crocodile and hippo could have survived in the Eastern Desert since the end of the Middle Palaeolithic, and there certainly are depictions of these animals.  As Cherry points out (2000, p.188), it is quite likely that these animals could have been seen elsewhere and are recorded from memory rather than life. 

Muzzolini (2000, p.89) considers that the bovids shown with forward-pointing thick horns actually existed and are not merely the result of artistic flourishes.  He sees them as culturally differentiated “special cattle” (p.93) belonging to the early Predynastic.  These particular forms are not found in late Predynastic or Old Kingdom Egypt.  In the Old Kingdom, the bovids most represented are those with long thin horns (Bos africanus) and survive into the Horse and Camel periods elsewhere in the northern African Sahara (p.93).  He says that short-horned types emerge only late as domesticated forms after c.1000BC and that the rare depctions of humped cattle probably date to the early first millennium BC.


Distribution of Eastern Desert Rock Art

Bearing in mind the warnings made by Fuchs (1989), it is still possible to list the available data about rock art distribution and suggest some tentative explanations.  Wilkinson (2000) observes that “The topographical location of rock-art sites is a crucial key in helping us to understand the significance of the drawings themselves” (p.162).

The Eastern Desert Survey team (Rohl 2000) observed that a considerable number of the sites are to be found “along the two principal transverse routes between the Red Sea and Nile valley” (p.4), both of which terminated at the Nile end at the important Predynastic centres of Hierakonpolis and Naqada, and on routes to the Red Sea.  Rohl describes a network of wadi routes “crisscrossing the desert” (p.4), with rock art marking some of the main ones. 

Rohl also suggests (2000) that rock art has been left in strategic hunting locations – where wadis narrow and could be used to trap migrating herds, and near possible watering holes.

Cervicek (1974) lists a number of additional distributional observations:

    Up to a 40% of carvings are in the Wadi Hammamat and the neighbouring vicinity

    There are no more than 20% of carvings in the Nile Valley between the First and Second Cataracts

    There are fewer than 10% of rock art carvings north of the First Cataract in the Nile Valley

    There are fewer than 1% of carvings in the Nile Valley in the neighbourhood of the Second Cataract.

Fuchs (1989; 1991) made a number of points regarding what he refers to as the “non-radom distribution” (1991, p.61) of major rock art sites in the Wadi Barramiya and elsewhere in the Egyptian Desert.  His observations are summarized as follows:

    Distribution is largely determined by the availability of suitable surfaces for engravings (1991)

    Petroglyphs are usually (in 85-90% of cases) located on Nubian sandstone.  “It would be plausible, therefore, to propose that the distribution of the petroglyphs depends primarily upon the availability of suitable rock.  However, while the geological structure clearly has some influence on the distribution of sites, it does not determine the overall pattern, nor is it responsible for their eastern limit” (1989, p.149)

    There are only rare sites in the Red Sea area.  Fuchs (1989, p.149) has suggested that this may be a real distribution pattern and may indicate a watershed marking a former cultural or territorial boundary.  However, lack of suitable rock surfaces or Nubian sandstone may be an alternative explanation.

    90% of sites are confined to wadis which drain into the Nile Valley, and the Nile Valley itself (1989), the greater percentage of which are located within wadi bends (1991)

    Main wadis are preferred to tributaries, probably partly due to their use as through-routes (1989; 1991) for moving people, herds and goods, but also due to their suitability for habitation, particularly during the more humid Neolithic (1989, p.150).

    There are fewer sites in the eastern parts of the wadi drainage system (1991)

    Most sites are located near ground level (1991)

    Rock shelters, overhangs and otherwise shaded positions tend to be preferred (1991)

    Convex  contours of main wadis are preferred “as shortcuts serve to gain a minimum travelling distance” (1991 p.61)

    The number of boat petroglyphs decrease in number and density from north to south (1989) -although this has been challenged by recent finds in Wadi Atwani.

Wilkinson (2000) adds to these conclusions:

    Many were in wadis that served as routes to the Nile, the Eastern Desert mines and the Red Sea. Examples include the route from Qift on the Nile to the Red Sea at Quseir(Wadi Hammamat), the route from Qift to Berenike on the Red Sea (via Wadi el Qash, Wadi Mineh and Wadi Mikab el-nes) and routes to the Pharaonic gold mines (Wadi Barramiya). He points out that the narrow north-west orientated Wadi el Atwani was not a linkage between areas of any industrial, commercial or residential importance, and that its use may have been provided an attractive area for wild herds to congregate, providing hunting opportunities.

On the basis of their survey work in 2000 and 2001, Morrow and Morrow are also able to add to the overall picture: “It became apparent during site recording that certain topographical conditions encouraged the carving of petroglyphs” (2002, p.13).  They add the following information (2000, p.13):

    The wadis preferred are those that provide a through-route rather than side wadis or dead ends

    Any area offering “significant protection from the sun, typically on the south face of a steep-sided wadi or under a rock overhang, such locations being ideal for camp sites” (p.13)

    Promontories and inner bends are preferred, probably because they were the shortest through-routes.

    Smooth and high quality rock faces near to the wadi floor were preferred

    Some of the petroglyphs were in a prominent position

    In spite of a relatively even distribution of animal motifs the boat depictions lie to the east of Naqada, Abydos and Hierakonpolis, the main centres of power during the Predynastic period (p.15)

    East-west routes were clearly important, but diagonal routes were also clearly important.  Examples are those from Laqeita well running south east towards Berenice and the series of wadis heading north of Barramiya to the north east (p.13)

Fuchs (1989, p.150-151) concluded that there is no reported site without boat images.  In the Wadi Hammamat area, the boat images do not extend into the Nile Valley, although boat images do occur, to a lesser degree, in the Nile Valley around the First and Second cataracts.  Hellstrom (1970) also observed that boat images fall off  south of the  First Cataract.  Of 1051 sites surveyed in the region between the First and Second Cataracts, only 91 included boat depictions (Hellstrom 1970, p.55).  Hoffman (1979, p.245) points out that there are no known drawings of boats along or even near the Red Sea coast, with the closest being over 50 miles inland and all in valleys that drain not into the Red Sea, but into the Nile

In the Fourth Cataract area, early observations were accompanied by suggestions that rock art was always located near to roads/routes.  However, more recent observations have suggested that this may have been due to bias in surveys, which favoured surveying areas near to such routes.  More recent surveys have found that often sites are hidden from view, often in very awkward corners and locations, and are only visible at certain times of day.  In the Eastern Desert, the greater percentage of sites do appear to occur within wadis which were probably used as through-routes, although not all images are clearly visible and some are very difficult to locate.

The University of Illinois team’s survey in May and June of 2006 worked on the theory that art would be more abundant on specific soil types that could produce grasses in the rather better environmental conditions in prehistoric times, and this proved to be effective for locating rock art sites. 

Further surveys will produce more information about which areas were favoured and what these had in common to make them attractive to the hunters and herders who probably made the bulk of the early rock art.


Themes in Eastern Desert Rock Art

The themes mainly describes the subject matter – boats, hunting, animal herding, etc.  It is somewhat artificial to divide them off into individual themes, because many of them occur in context with other types of motif in deliberate scenes. 


The individual images described below are often incorporated into scenes where a clear relationship exists between the different components.  Hunting scenes, which include human figures, dogs and ostriches, are one type of structured scene.  Another activity frequently depicted appears to be the herding of bovids.  The close association between relatively huge figures with upright hairstyles or head-dresses and boats in which they are standing are less easy to define as specific themes.  Suggestions that they represent some type of hierarchical element or a religious idea are interesting but purely speculative.  Similarly, figures in boats with arms raised in loops over their heads may have some special symbolic significance, but it is impossible to say what.  One of the difficulties with arrangements of images is determining whether they are simple representations, narratives, or structured arrangements of symbols.  the re-use of some surfaces over long periods of time is clear at some locations.  This can be interpreted in a number of ways – for example, a particular location may have a particular value for subsistence activities, or may offer shelter on a favoured route between places.

Some rock art faces are covered in images, but what, if any relationship existed between the images is impossible to detect.  An attempt by Tilley (1991) in Scandinavia with similar rock art faces was unable to reach any helpful conclusion, but it may be that new approaches will be developed to assess this type of rock art where it appears in such profusion.

Individual Components

Although it is somewhat artificial to isolate individual components breaking down some of the individual components and listing some of the associations.


A number of typologies have been proposed for boats of the Eastern Desert.  However, as detailed by Vinson (1987. p.126), they conflict and there is no general agreement in favour of any one of the schemes.

In the EDS Survey (Rohls 2000), 240 high prowed ships were identified, described in three main groups, based on Winkler (1958)

  • With square-shaped hulls
  • With sickle-shaped hulls
  • With in-curved sickle-shaped hulls
  • (and those with a combination of profiles)

The boats appear with different furnishings, tools, animals and people (Rohls 2000).  Vertical marks within the hull are interpreted as crew, marks below the level of the hull are interpreted as oars.  Some boats are shown without either crew or oars.  In some cases, the crew are shown outside the boat, apparently pulling it along with ropes.  In a number of instances figures within the boat may be vast, some with weapons or staffs, some with distinctive head-dresses, and are interpreted as shamans, gods or kings. Animals appear as standards – particularly horned bovids and Horus-type falcons.  Palm fronds and pennants may appear on the prows of some of the boats.  Various types of cabin and structure can appear within the boat.

David Rohls believes that the boat representations are intended to be accurate images of boats that were used during Predynastic (2000 p.5).  Depictions of figures dragging boats are also not uncommon, but are more difficult to date.  Although there have been suggestions that some are Predynastic, scenes of boat dragging are known from the New Kingdom tomb of Merenptah, and the Amduat. 

Interpretations of the boats vary from the religious to the military.  David Rohls sees the boats as military, partly on the grounds of the number of crew members, which can reach 70 individuals (2000 p.5), but others see continuity with Old Kingdom concepts of solar barques (Wilkinson 2003). 

Animals and Birds

Rohls (2000) suggests that the climate seems to have influenced the choice or rock art subjects, with wetter conditions sustaining herds of wild animals which are no longer present today:  elephant, giraffe, antelope, ibex, even hippopotamus and crocodile, although Cherry, in the same volume (2000a, p.188) has highlighted problems with this approach (see above in Dating).

The identification of species of animal in the Eastern Desert rock art is not always easy (Cherry 2000a, p.188).  Whereas painting allows for subtleties, rock engraving does not allow for reproduction of the same detail. Weathering and superimposition by other engravings can also cause problems.  Finally, some sub-species are actually very difficult to tell apart, although horn shapes and lengths can be a useful indicator.  Cherry offers the following list of animals represented (2000a p.189):

  • Bovid
  • Addax
    • Long spiraling horns
  • Oryx
    • Long straight horns
  • Ibex
    • Exaggerated semi-curved horns extending as far as rump
    • Beard
    • Very short tail
  • Barbary goat
    • Horns shown as diverging from each other rather than as parallel, when the head is viewed from the front
  • Hartebeest
    • Short thick lyre-shaped horns
  • Ambiguous
    • Kob
    • Nile lechwe
    • Gerenuk
    • Dibatag
    • Various gazelle, including Dorcas
    • Animals with long backward curving horns
      • Nubian ibex
      • Scimitar
      • Roan antelope
  • Rare
    • Sheep
    • Goat
  • Missing
    • Zebra
    • Wildebeest
    • Pigs
    • Ducks
    • Cats

Bovids feature frequently, either on their own or in hunting scenes with dogs and humans.

Dogs are represented hunting bovids and ostriches.

Some of the engravings show people hunting ostrich or ibex, sometimes with the assistance of dogs.  Ostrich were known from the Predynastic period and ibex were known in Eastern Desert contexts from the Predynastic  through to the Old Kingdom.

Francis Lankester has reviewed the animal species recorded in the Eastern Desert Survey (2000) and the Rock Art Topographical Survey (2002) and found the following frequencies in some of these species:  Ibex 66%, cattle 40%, Elephant 13%, Hippopotamus 6% and Crocodile 5%.  The reliability of these figures obviously depends on the reliability of identification – elephants are fairly unambiguous but many fo the ungulates can look quite similar, in spite of distinctive horn arrangements.

Many of the themes described above are unique to the Western Desert. 

Human Figures

Although some of the human figures are associated with boats, as described above, others are associated with other scenes, including hunting activities.  Again, dating causes problems.  Some of the engravings showing people hunting ostrich and ibex, sometimes with the assistance of dogs, may be early.  Ostrich was known from the Predynastic.  Ibex were known in Eastern Desert contexts from the Predynastic  through to the Old Kingdom. The use of bow and arrow is well known throughout the Pharaonic period, with high numbers found in the tomb of Tutankhamun.

One of the figure depictions most commonly thought to be associated with Predynastic populations, on the basis of comparisons with Nile artefacts, is the figure which has arms stretched in a curve over its head.  These figures are sometimes referred to as “orant”, which means that the figures have their arms raised towards the skies in prayer.  As this preempts an interpretation (prayer), I have called them “loop-armed”, which is far less elegant, but hopefully avoids preconception. These are found both in Eastern Desert rock art and in Naqada II ceramic decoration and as statuettes accompanying burials during Naqada II. 

Other human images often thought to be Predynastic, are the giant figures standing in boats.  They are distinguished by head-dresses or vertically arranged hair styles.


A small number of Eastern Desert rock art images look like abstract designs.  They are curvilinear and, at first glance, appear to be non-representational. However, some of them  may be maps of the surrounding wadis.  Cherry says that “They look very much more like  rudimentary maps that we have all sketched from memory to assist others in navigating through unfamiliar territory” (2000b, p.166) .

Cherry describes three possible examples from the Eastern Desert.  One of them is at Site 26, where a large fallen rock has been engraved with a pattern of curved lines.  The lines appear to resemble the wadi system from Laqeita well in the north, to Barramiya well in the south.  Another of the maps, and to my eyes the most convincing of Cherry’s examples, is EDS site MG-1 in Wadi Umm Salam, where the main line could show Wadi el Ibatur to Wadi Shagab, with three smaller wadis off to the east (Wadi Shalul, Wadi Abu Mu Awwad and Wadi Miya).  A photograph of the “map”, an illustration of it, and a map of the wadi are all shown side by side for comparison on Francis Lankester’s Desert Boats website at:  It actually looks fairly convincing to me, but as Francis warns:  “Caution is advised here. What is the smaller squiggle to the right?” (Lankester WR). 

Cherry also offers a warning, saying that “it is all to easy to be seduced into finding patterns and shapes which may be nothing more than coincidences” (Cherry 2000b, p.168).

Abstract Images

Some of the engravings are truly abstract, at least to modern eyes.  It is possible that they were representational of concepts or ideologies understood by the people who engraved them, but it is impossible to guess what they were supposed to represent.


Technique in Eastern Desert Rock Art

The subject of technique, or the way in which rock art is executed,   is used in identifying similarities between rock art depictions and in attempting to build typologies that form the basis of dating schemes. 

In the Sahara art is either painted or engraved.  Engraving may involve a number of different techniques including incision, pecking etc.  However, in the Eastern Desert, there is only one known piece of painted art (in the Wadi Hammamat).  The rest are engraved into rock surfaces using a variety of different techniques, including pecking, and incision.

Style in Eastern Desert Rock Art

Style, or the overall design concept, of an item can be defined as a specific aspect of approach, or a combination of elements that render something immediately recognizable across a whole range of items or backgrounds. This may be, for example, use of symbols, colours, methods of manufacturing, or highly distinctive shape. It  is potentially useful for determining similarities and differences between rock art canvases, and for creating typological frameworks.  Unfortunately, the images of rock art in the Eastern Desert are very simple linear representations which defy any attempt to define them in stylistic terms.  Images are rendered with such simple strokes and with such little detail that it is impossible to pick out any remarkable stylistic elements which a viewer could detect across a range of depictions.  As Cherry points out (2000): “The technique of pecking or incising on a hard rock surface does not permit subtle differences to be represented in highest detail and subsequent weathering and over-carving during the intervening millennia can further degrade the image which the artist was trying to portray” (2000, p.188).

Origins of Predynastic Rock Art

There has been a lot of speculation about which people were responsible for the earliest examples of Eastern Desert rock art. 

Frankfort  (1924) was the first to suggest that much of the Predynastic archaeology could be explained by incomers from Mesopotamia across the Eastern Desert.  This view was supported by Kantor (1942; 1952) who supported a Mesopotamian origin for various Egyptian artefacts and motifs.  Winkler (1938) also considered the rock engravings of certain boats to be the work of Eastern people who entered the Upper Egyptian Nile from the Red Sea – he believed that the straight hull and high prow and sterns were the criteria identifying the boats as Mesopotamian. Hoffman, however, suggested that a close examination of Winkler’s drawings showed that the distinctions he drew between boat types were not as clear cut as he considered due to differences both in the boat size and the curvature of the prows and sterns.  Other supporters of an invasion theory were Baumgartel (1955) and Emery (1961).  Emery said that “it would seem probable that the principal cause was the incursion of a new people into the Nile valley, who brought with them the foundation of what, for want of a better designation, we may call Pharaonic civilization.  Whether this incursion took the form of gradual infiltration or horde invasions in uncertain, the the balance of evidence, principally supplied by the carvings on the ivory knife-handle from Gebel-el-Arak and by paintings on the the walls of a late Predynastic tomb at Hierakonpolis, strongly suggests the latter” (1961, p.38).  However, as I mentioned above, both the Hierakonpolis Tomb 100 and the Gebel Arak knife are somewhat convtraversial forms of evidence.

Invasion theories are largely out of favour, although David Rohls (2000) has revisited the concept in recent years.  Invasionist theories have generally been replaced by a view of Pharaonic Egypt as a home-grown civilization, with rock art the products of indigenous groups. 

Morrow and Morrow (2000, p.185) describe another possibility, which sits in the middle of these two extremes:  “a more speculative scenario can be postulated in which foreigners came through the wadis, established (or captured) trading towns near the wadi exists on the Nile, e.g. Naqada and Nekhen, and competed with each other for lucrative commercial opportunities”.  Competition between the towns could have accounted for the legends of Seth (Naqada) and Horus (Nekhen). 

Rohls (2000) believes that a rock art image from the Eastern Desert site EDS site JCB-1 shows a standard with a “crescent-moon-with-rising-sun-motif” (p.8). He suggests that there are distinct similarities between this motif and those representing the Sumerian sun deity Utu (and which also occur in Pharaonic times).  This motif is also shown on the Gebel el Arak knife, where it is show in the context of a boat (but note that the Gebel el Arak knife is unprovenanced and its authenticity has been disputed).  Rice (2003) also picks up on these symbols and buys into the idea that they “suggest a notable Mesopotamian contact with southern Upper Egypt in late Predynastic times (p.44).

Marks (1997) does not disagree that there were strong connections with Mesopotamia from a very early date (p.14), with an intensification of contact during Naqda II c/d (p.20).  However, his view, based on a very detailed analysis of different types of artefact and their distribution throughout Egypt, Palestine, and Syria is that the relationship was based upon trade (not invasion) and that the route via which trade was conducted was a northern one, through the Nile Delta, and not at any time across the Eastern Desert from the Red Sea.  His arguments for a northern route are very convincing.  His arguments for ruling out a route from the Red Sea to the Nile are based on his analysis of the boat motifs, which are summarized as follows.  The Eastern Desert boat motifs that are of interest in this context are those with high ended sterns and prows, with long flat hulls.  Marks includes a detailed analysis of the Gebel el Arak knife  and the Hierakonpolis Tomb 100.  His main objection against the Gebel el Arak knife is that it is completely unprovenanced.  His objections to Hierakonpolis Tomb 100, which has itself been dated only on the basis of the motifs to Naqada IID, are that it has been interpreted in numerous ways and there are only two features in the entire composition that have Mesopotamian affinities, one of which, the black boat, has no exact parallels in any Mesopotamian depictions of the Late Uruk period (p.77-79).  Marks also makes the observation that none of the drawings are associated with the Red Sea – the nearest is 50 miles inland, and most are closest to the Nile (p.83, referencing Hoffman 1991 p.245).  Marks’s conclusions about the similarity between Wadi Hammamat and Nile boat types are partially supported by an earlier work (Berger 1992, p.107-120) which suggests that the Hierakonpolis boats “are clearly part of an artistic tradition in southern Egypt and lower Nubia.”  Details are shared between boats from the Wadi Hammamat region, the Wadi Mia, Hierakonpolis and elsewhere.

Interpreting Eastern Desert Rock Art

It has already been observed that the themes of the Eastern Desert are

  • Boats
  • Wild animals
  • Bovids
  • Game birds
  • Figures
    • Loop-armed figures
    • Twin-plumed figures
    • Hunters
    • Big figures with head-dresses
  • Abstract images

Scenes include:

  • Boats with crew and animals
  • Hunting scenes
  • Pastoral activities

Although some images in the Eastern Desert are show in in clear relationships with one another (e.g. hunting scenes combining humans, hunting dogs and prey, crew pulling boats etc) others are shown in complete isolation.  Some appear together without any sense of underlying composition, and may precisely contemporary, broadly contemporary or of vastly different date – more often than not it is very difficult to tell.

Some of the boat forms are unique to the Eastern Desert and southern Nile, which makes them particularly interesting in terms of trying to understand movements throughout the eastern Sahara, and the role in which the Nile may have played as a barrier between different groups and lifestyles.

In the case of rock art shelters, Wilkinson (2000) suggests that “Natural rock shelters not only offered protection from the sun and over for hunters lying in wait, but they also provide a degree of seclusion – an environment where people can commune with the supernatural world” (p.162).

Whereas context gives some hint of meaning, like the abstract art on Irish Boyne tombs that are associated exclusively with funerary contexts, the only clearly identifiable context for the rock art of the Eastern Desert is the wadi-carved landscape itself.  The engravings are laid out on open rock faces and boulders. Even though it is sometimes etched into somewhat obscure crevices and corners, the rock art was not hidden and private in the way that, for example, French cave art is.  Either its creators intended for it to be seen, or didn’t mind whether or not it was visible to other people.  Having said that, it is fair to assume that the engravings are part of a humanly defined landscape, part of a mental map that incorporates geographical landmarks, subsistence information and cultural associations.  Comparisons with Bedouin tribes throughout the Eastern Desert of Egypt and the Sudan certainly confirm this association between people and the landscape in a very intimate and knowledgeable format.

In attempting to interpret Eastern Desert rock art, several writers have seen a connection with rock engravings that they believe to be broadly Predynastic with religious traditions on the Old Kingdom.  Cyril Aldred proposed that nomads “imagined a destiny among the stars, beliefs which survive in the later eschatology of the historic Egyptians” (1998, p.66).  More recently Wilkinson saw a connection between rock art boat motifs and the solar barques of the Old Kingdom and later mythology (2003).  Michael Rice has also proposed a link between the two periods, based on an emphasis on animal images:  “The animal cults which are so powerful a part of later Egyptian ritual and belief obviously had their origins here”.  The main problem with all of these proposals comes back to twin problems of dating and interpretation – it is necessary to establish that the images concerned are indeed Predynastic, and that their meaning was in some way religious before such suggestions can have real merit.

It is very hard to interpret rock art of the Eastern Desert in terms of grammatical narratives or coherent compositions intended to impart ideas, and to illustrate this it may be worth drawing a comparison with the abstract art of the Irish Boyne tombs, described and discussed by Michael Herity (1974).  In many ways it is easier to see a design and believe in an underlying grammar in the purely abstract compositions of the Boyne tombs in Ireland, where huge stones are incorporated into tomb constructions and were used as vast canvases, often with the entire surface employed in the creation of a single coherent composition.  Even where isolated symbols are shown, or where compositions show no obvious coherence, motifs are familiar, instantly recognizable and stylistically distinctive.  Herity points to individual preference in the way in which these motifs were combined at different sites:  “The builders of each cemetery develop their own preferences in choice of motif and in style, but everywhere the central corpus of motifs is religiously displayed . . . . Everywhere there is an ideal of combining the devices in a larger artistic design, often covering the whole face or slab or orthostat, always in tune with the scale of the tomb” (p.97).  The designs and motifs are so distinctive that it is possible to identify regional variations.  Herity describes the range of symbols as “the vocabulary of this language” and he examines the “recurring combination of symbols into motifs or pictorial designs in the hope of proceeding towards an understanding of its syntax and meaning” (p.103).  I may have laboured the point a little, but the case of the Boyne tombs illustrates how difficult it is to extract meaning from the Eastern Desert petroglyphs which follow neither formal arrangement nor display repeated combinations of distinctive symbols in consistent  combinations.  

Huyge (2002), although working with rock art in Elkab in the Nile Valley, has offered the most interesting attempt to interpret rock drawings of a very similar type to those in the Eastern Desert. Huyge takes four of the traditional explanations for Egyptian rock art (magic, totemism, religion and ideology) and takes them in turn.  Sympathetic magic, by which hunters believe they can control successful access to game animals and influence their numbers, is rejected on the grounds that within the chronological framework within which Huyge believes that the art was made, hunted animals were only supplemental to a diet based on domesticates, of which very few are shown in rock art.  Totemism is also rejected.  the intimate association between animals and humans, involving rituals and taboos, does not seem to be consistent with the evolution of a religion which is not totemic in character and which employs different types of animals and iconography than those depicted in Elkab.  Religion is seen to be a more likely explanation for the drawings, partly due to the fact that Late Predynastic and Early Dynastic religious development had to have its origins in the past, and expression of that religious devotion could be expected:  “Even though this approach, often carried to extremes, is subject to serious criticism, the religious interpretation of rock art is attractive (to the point of being self-evident) because of the apparent permeation of Ancient Egyptian society with religion” (Huyge 2002, p.193).  Ideology is also seen as a plausible explanation for some later rock art, but Huyge suggests that if this is the case “it is logical to assume that it was only to a limited extent part of iconographical constructs that were mainly conceived by and for the intellectual elite” (Huyge 2002, p.193-194).  Huyge concludes that totemism and magic can be rejected as an explanation for rock art in Egypt, and that religion, and to a lesser extent ideology, offer the best avenues for interpretative explanations.

Huyge goes on to attempt to examine the rock art within both a diachronic framework and an interpretative context.  His approach is bold and contains much of merit but two main questions need to be asked of it.  The first is to question the validity of his chronological framework.  Like others before him, Huyge appears to have developed a system that is plausible but not provable, and if there are questions about the chronological scheme, those make interpretations based on the framework questionable.   The second question assumes that the chronological scheme is accepted, and looks only at the interpretations.  The question here is whether or not Huyge’s methodology, whereby he examines, then provisionally accepts a proposal and then extrapolates from it to reach a conclusion is really acceptable.  It is the same issue that worried me with Wilkinson’s (2003) attempts to extrapolate from unique artefacts to cross date rock art – the extrapolation to a conclusion is based on data that has not, in all cases, been thoroughly validated.  An example is the process by which images of giraffes become solar symbols in Phase 1 (2002, p.199-200) and are therefore part of a religious iconography – it involves accepting a number of suggestions, all of which are well argued, but remain speculative.  The theme of religious iconography is then developed in terms of this solar cosmology (giraffes, asses and Type VII boats), and conclusions link this with a funerary cult based on funerary barques  (Type I boats, human figures with raised arms and ibexes) in a framework of renewal and rejuvenation.  Huyge’s ideas are well worth examining, and considering in relation to rock art from other Egyptian areas.  Perhaps his most important conclusion is that “Both content (themes) and formal peculiarities (orientation and lateralisation) appear intricately interwoven in a complex pictography.  This language of imagery is clearly symbolic rather than narrative” (2002, p.204).

Morrow and Morrow (2002) point to the fact that those who have analyzed Egyptian rock art can be divided into two schools of thought:  religious interpretations, and interpretations that focus on socio-economics, culture and climate.  For Morrow and Morrow “zoomorphic illustrations combined with enigmatic boat depictions appear to reinforce primarily religious association albeit with cultural overtones (2002, p.21).

Comparisons with neighbouring rock art

I’m not entirely convinced of the value of this sub-section, but I’ve left it in for the sake of completeness.  The main concern that I have with trying to draw comparisons between Eastern Desert rock art and that of other areas is that highly stylized and simple engravings look much the same the world over.  I was looking at images of Scandinavian and Iberian rock art recently, and so much of it seemed, superficially, so similar to the rock art that I have been looking at in the Eastern Desert, that it somewhat undermined my belief in the value of looking for comparisons.  I have therefore kept this section very brief.

It is certainly the case that long distance contacts existed.  As an example, Smith (2005) points to ethnographic data, which shows how vast areas are covered by modern nomads.  He cites the Tuarag and the Fulani.  He also (2005) points to long distance connections between other areas in prehistoric times and proposes connections between Egypt and Niger:

  • The Khartoum Mesolithic has parallels with the Tenere Desert (Niger) in terms of pottery decoration and bone harpoons
  • The Khartoum Neolithic has parallels with the Tenerian, in terms of pottery motifs
  • The Nubian Desert burials with cattle horns c.5000BP have parallels with flat-topped tomb types of Late Pastoral Niger c.4500BP.

Egypt’s Western Desert

It is certainly worth considering what relationship, if any, is visible between the rock art images of the Eastern and Western Deserts.  It has become clear that in the Eastern Sahara pastoral nomads covered huge areas during the Early and Mid Holocene, making skillful use of different resources as they became available on a seasonal basis.

The Western Desert rock art locations are Dakhleh, Farafra, routes within the Qena bend and the route from the Qena Curve to Kharga Oasis, Gilf Kebir and Gebel Uweinat. 

The rock art of the Western Desert at Gebel Uweinat and Gilf Kebir is quite unlike that of the Eastern Desert.  Images are painted and detailed, often very beautiful.  The best resource for this rock art is Andras Zboray’s wonderful CD-ROM containing a vast collection of images, but also see his website

In Farafra oasis a karstic solution cave with three adjoining chambers was found by Barbara Barich’s team.  It features both engravings and paintings.  The engravings include goat, giraffe and boat images, and the paintings show hand marks, and lion paw prints in cavities.  A large hearth was found in the cave. 

At for south west of the Qena bend, along a Western desert route far from the Nile, Darnell (2002, p.157) describes boat engravings in the so-called Cave of Pegs, accompanied by quadruped and bird images. Within and in the immediate vicinity of the cave are abundant amounts of Predynastic pottery of Nile Valley and Libo-Nubian types and flint assemblages, with large quantities of ash, burned material and organic matter. As I have stated earlier, the presence of artefacts in contexts where rock art is also found does not provide a date unless the engravings are sealed beneath the archaeological contexts – which is not the case here.  However, it is certainly interesting that after Naqada III the cave was not again occupied until the Coptic period.

Libya and Chad

The rock art of Tibesti and Ennedi comprises the main non-Egyptian rock art of the Eastern Sahara.  It can be compared with the petroglyphs of the Western Desert, but has no obvious similarity to those of the Eastern Desert.


As above, there have been suggestions (Rohls 2000; Rice 2003) that motifs located in the Eastern Desert are explicitly Mesopotamian, but these refer to comparisons with motifs on other types of artefact and are not related to any type of landscape art. 

Western Saudi Arabia

Winkler (1938) suggested that Eastern Desert rock art is similar to examples in western Saudi Arabia, where examples are found near former lakes which existed between c.8000 and 2000 BC when conditions were warmer globally.  Themes are, however, different and do not include elephant, giraffe or rhino.  They do include long horned male bovids, ibex/wild goat, dogs and men (Majer 1992, p.230).

Sinai and Neghev

The Sinai and Neghev are interesting because they serve as land routes through to the Near and Middle East, and are the most likely route via which domesticated sheep, goat and later pig and domesticated cereals, legumes and flax were introduced into Egypt.  The possible connection even very early on with Mesopotamia (Mark 1997) also points to a Sinai route out of and into Egypt.  From earlier times, the presence of lithics of  Near Eastern types – like the Emmireh points of the Middle Palaeolithic at Sodmein Cave and Helwan points corresponding to the Epiplalaeolithic near Cairo – indicate again that Sinai was sometimes used as a land bridge. This could be part of a tradition of movement between the two land masses over 1000s of years.

Rock art of the Neghev has similarities to that of the Eastern Desert, at least at a superficial level.  Anati’s paper about the rock art of Hor Karkom (1999) discusses the rock art of an area that is particularly rich in both archaeology (everything from sanctuaries and altars to stone circles) and rock art.  The mountain and its valley s have provided a vast concentration of rock art, with this one project discovering 200 sites with over 40,000 images (1999, p.22).  One of the most striking things about the paper is the aerial view of Hor Karkom, which has a remarkable resemblance to the Eastern desert as seen from the air.  Unlike the Eastern Desert, Hor Karkom has been thoroughly surveyed and the archaeology has been well documented.  

The problems with dating are quite clearly the same here as elsewhere, with a number of different strands being brought together to provide a chronological sequence – superimposition, patination and association with particular sorts of structure (1999, p.23).

Similarities with the Eastern Desert rock are superficial and are probably something of an illusion created, I believe, by the limitations of the technique applied – the same type of engraved image on rock surfaces, without paint.  The same sort of comparisons could be made with petroglyphs in many other parts of the world.

The Main similarities are:

  • Preference for smooth surfaces with dark patina
  • Themes of hunting and herding
  • Bows
  • People with raised (but not looped) arms
  • Abstract curvilinear images

But there are significant differences as well, including easily identifiable religious iconography, clear archaeological associations and geothemes.



Mike and Maggie Morrow highlight some of the serious difficulties still confronting any analysis of Eastern Desert rock art (2002):

    “(a) The number of sites properly provenanced, recorded and published remains pitifully small
    (b) Conclusions reached on such evidence are thus to a large extent speculative
    (c) Further investigatory work is needed to provide a structured international database for detailed analysis
    (d) Specifically, generally accepted conclusions based on comparisons with Naqada II pottery and the Tomb 100 painting are possibly unsafe
    (e) The evidence in situ is under severe threat and further surveys are needed before irreplaceable data is destroyed.”

The lack of properly recorded and published sites means that it is impossible to make reliable observations about rock art distribution and any underlying variations in meaning connected to it, and that comparative studies cannot be completed with any degree of confidence.  Another problem is that surveys to attempt to locate other archaeological data are also pitifully short on the ground. Dating, is as I keep mentioning, an ongoing problem.

At the most basic level of interpretation, the depictions show that the Eastern Desert was used by people, probably from the early Predynastic onwards.  Initially the desert was probably used for its vegetational and faunal resources – for hunting ibex, gazelle, wild bovid (Bos primigenius) and ostrich, and for pasturing domesticated bovids (Bos taurus).  The boat images and some of the more distinctive figures suggest thought processes which were not necessarily concerned with the immediate task in hand (hunting or herding) but of something more complex.

Reading Vinson’s M.A. thesis (1987) has given me the hope that a chronology can be developed for the boat images, and that some of these actually can be tied into the Nile’s artefactual datasets, but the detailed research necessary to develop that line of work has yet to be completed.

In the final analysis, nothing that I have seen or read about Eastern Desert rock art gives me a sense of what I am looking at.  I find it intriguing and profoundly attractive, but I can see no route through to it.  I cannot see how it can be used as a dataset in its own right.  I find this desperately frustrating because it leaves me with a sense that there is a vast body of data sitting out there which, in spite of excellent efforts, continues to defy attempts to blend it with other archaeological data in any very useful way.  If any sort of chronological system can be agreed upon, I think that the most immediate way of using the engravings will be to take Bradley’s type of approach (1994) and use it to explore the idea of mobile groups’ use of the Eastern Desert and their perception of it as a human and natural landscape.

In any future rock art work that takes place in the Eastern Desert, it would be useful to define a set of standards about what sort of data should be collected, how it should be recorded (on survey forms, in photographs and illustrations and GIS software in GIS integrated database format), and how it should be stored. In an ideal world it would be good to see a centralized research project with a database at its core, meeting the requirements of both rock art surveyors and archaeologists, and built by a professional database designer, designed to meet not only Eastern Desert requirements but incorporating requirements for including Western desert and Northern Saharan data at a later date.  The resulting database, published online, would be a resource for archaeologists attempting to pull together Eastern Desert and other north Saharan rock art and archaeological data.  Databases can, of course, include GIS data papers and articles, and can provide comment and interpretation. 

Obviously, these pages offer a very superficial look at the Eastern Desert rock art, and I am very much looking forward to seeing the PhD research produced by Tony Judd (University of Liverpool) and Francis Lankester (University of Durham), both of which focus on Eastern Desert rock art.





Copyright Andie Byrnes 2007