Swimmers in the Sand


  • General view of Wadi Sura I – Cave of Swimmers (Photo Sandro Vannini)


It has been long supposed that some of the manifold roots of ancient Egyptian civilisation might be connected with the climatic depredation of the Western Desert, which led to the last of intensive sedentary waves in the Nile valley (Middle and Upper Egypt and Sudan) and to the emergence of the ancient Egyptian civilisation. Iconographic analysis of the rock-art motifs in the area of south-western Gilf Kebir in Egypt’s Western Desert provides indications that several concepts traditionally connected with ancient Egypt, and its civilisation thriving in the Nile valley, were possibly formulated much earlier and at a location other than the Nile valley itself. The sites of the Cave of Beasts (Wadi Sura II) and Cave of Swimmers (Wadi Sura I) indicate that the local prehistoric populations of hunters and gatherers probably formulated some notional outlines that later became key concepts of the ancient Egyptian world definition (for their complexity and different meanings see recently Tassie 2013, 133–6).

At the same time these ideas represented some of the founding elements of the Egyptian identity. It now becomes feasible to postulate that the Western Desert populations of the Sixth and early Fifth millennia BC may be considered as partial intellectual precursors of the ancient Egyptians who contributed to the specific character of this civilisation, transferring to the Nile valley their mundane experience and transcendental conceptions. These populations merged with older Nile valley inhabitants due to the harsh depredation of the Western Desert environment and considerably expanded the ancient Egyptian culture (Bárta 2010, passim).

Mythological landscape

The area of Gilf Kebir belongs to the archaeologically least known parts of contemporary Egypt. It is located in the south-western corner of the Egyptian Western Desert, close to the Libyan and Sudanese border and is famous for its rock art and painted shelters (Rhotert 1952). For our purposes, two of them are of eminent importance. One of the oldest examples of this kind, the Cave of Swimmers, was discovered by the famous Hungarian count, desert explorer and WW II spy, Lásló Almásy, in the 1930s (Almásy 1998). Its name is derived from the fact that the walls feature, among other scenes, many small human figures in attitudes that may loosely recall the act of swimming. The second cave, the nearby lying Cave of Beasts was discovered only in 2001 (Le Quellec, Flers, Flers and Grimal 2005; for the ultimate publication of its decoration see Kuper et al. 2013).

In 2009 a German team from Cologne University started a very detailed multi-scientific survey and documentation of the Cave of Beasts and the surrounding areas. Based on their preliminary conclusions, it may be said that most of the finds associated with the cave date to the Gilf B Period (6800/6600–4400/4300 calBC), followed by very scarce finds dating to the Gilf C Period (4400/4300–3500/3000 calBC) (Kuper, Riemer and Förster 2009, 8; Kuper, Leisen, Riemer and Förster et al. 2009, 16–19).

While there is sufficient evidence for the contribution of the Levant to the ”Neolithisation” of the Nile valley, the importance of the Western and Eastern Desert for the origin of these prehistoric cultures, especially in Middle and Upper Egypt, is meagre at best. (Wengrow 2006, 13–55). It was Farouk El-Baz who suggested that significant elements of ancient Egyptian culture were formed during the Predynastic period, when the people of the Western Desert started to withdraw from large, environmentally depredating areas and settle in the Nile valley (El-Baz 2003, 64–72). In a way he followed the argument presented many years earlier by S. Morenz (Morenz 1992, 233, the original publication in German dates back to 1960).

During Holocene periods with relatively humid climate the annual precipitation in Gilf Kebir was very likely below 100 mm. This development was the consequence of a gradual drying process of surface lakes whose remains are today called playas. The uranium-series technique used in the analysis of lacustrine carbonates from Bir Tarfawi, Bir Sahara East, Wadi Hussein, Oyo Depression and the Great Selima Sheet sites in northeast Africa shows that this was indeed so. The results obtained show convincingly that there were five palaeolake episode dates which had intervals between 320–250 000, 240–190 000, 155–120 000, 90–65 000 and 10–5 000 years ago (Szabo, Haynes, Jr., Maxwell 1995, 227–42). Four of these periods can also be directly linked to major interglacial substages. In this particular context it is interesting to note that one of them occurred during the early and middle Holocene when the Sahara experienced a longer period of humidity (El-Baz 2003, 67). Following the end of the last humid phase, around 5300 BC, the Sahara and the Western Desert as a part of it started to desiccate and local populations were forced to move east- and south-ward and gradually settle in the Nile valley.

In their article in Science 2006, Rudolph Kuper and Stefan Kröpelin have published to date the best summary of the climatic variations observed in the Sahara during the Holocene (Kuper, Kröpelin 2006, 803–7). According to their research, during the Early Holocene occupation phase (8500–7000 BC), the number of rapid monsoon rains increased, thus the Sahara turned into a savannah-like environment suitable for occupation and became resettled by populations from the, at that time, inhospitable Nile valley and from the south (today’s Sudan). These newcomers were hunter-gatherers practicing limited animal husbandry. During this period most of the Nile valley was not occupied, probably due to harsh and unpredictable Nile fluctuations. During the Mid-Holocene (7000–5300 BC) this vivid traffic in the Sahara continued. It was also marked by the domestication of cattle, sheep and goat. Whereas cattle were domesticated locally, sheep and goat were introduced from the Levant (Bradley, Cunningham, Loftus 1996, 5131–5). By around 5300 BC cattle and sheep-goat herding had become a widespread phenomenon.

In the final period of Mid-Holocene regionalization (5300–3500 BC) we can observe clear proof of the gradual desiccation of the Western Desert. Populations started to withdraw southward and eastward and retreat to the area of Gilf Kebir and Gebel Uweinat. This was probably due to a change in the seasonality of rainfalls. The gradual withdrawal of transhumant populations from the Eastern Sahara was synchronized with the beginnings of sedentary life and the first Neolithic communities in the Nile valley. These concentrated in the areas of Fayum and Merimde in the northern part of the Nile valley and Badari in the south (Upper Egypt). These settlements appeared around 5000 BC and used the Levantine strategy of growing wheat and barley crops. The Neolithic communities of the Badari culture built settlements that formally resembled African livestock enclosures (Brunton and Caton- Thompson 1928). Similarly, the cattle burials as practiced by these early communities are also indicative of the fact that quite a few social and religious aspects of ancient Egyptian civilization may have been introduced by communities coming from the Western Desert. The period around 3500 BC represents a definite turning point in the occupation history of the Egyptian Western Desert when permanent occupations definitely ceased and only limited transhumant populations managed to survive for another few centuries.



Ancient Egyptian Iconography?

At this stage of research of the subject, it is possible to attempt some interpretations of several major or, in my view, most significant scenes. In this context it is important to note that most of the rock art preserved in the Cave was dated by the German team to the Gilf B phase (Wadi Sura style) (Kuper, Leisen, Riemer, Förster et al. 2010, 20, fig. 21), as was the playa in front of the Cave (Kuper, Leisen, Riemer, Bubenzer et al. 2011, 11). What seems to be almost certain is that despite the unparalleled richness of the decoration, the individual scenes may not be necessarily interconnected. The Cave of Beasts features at least two compositions that are of essential importance for our understanding of the mythological concepts of their creators and which, at the same time, are highly important for the ancient Egyptian civilisation.

The painted scenes were most frequently made with red ochre in combination with white pigment. The compositions are positioned rather irregularly within the whole complex of other scenes and correspond with them stylistically, yet they represent self-contained independent units. Most importantly, some of the scenes seem to display close resemblances to much later ancient Egyptian mythological motifs.

  • Wadi Sura II– Cave of Beasts – Detail of a scene showing chieftain smiting an enemy (Photo Sandro Vannini)

As an example, the extreme left part of the cave features a relatively small scene which depicts a chieftain with a mace and two rows of what seem to be slaughtered enemies to his right. The enemies are lying on their backs and are separated by a rock fissure that intentionally divides the upper and lower row. The upper row of enemies contains ten robust figures with their arms lifted above their heads. The lower row consists of twenty-three figures almost half the size of those in the upper row and whose bodies are slender (females?) with their heads down. In addition, their arms are arranged differently: while one arm is always hanging along the body, the other is raised above the figure´s head. The difference between the two rows in terms of their robustness may be ascribed, hypothetically, to the different sexes of the figures as well as to their different heights. One more fallen enemy can be seen to the left of the chieftain.

Very close parallels to this composition seem to occur much later in ancient Egyptian sources where the standard elements of the scene show the king (in earliest scenes a chieftain) with a raised mace above his enemies, about to smash their heads. This ideological feature of a victorious king successfully protecting his territory and people from evil forces and enemies permeated the whole ancient Egyptian civilisation. The king was, from the very earliest stages of the ancient Egyptian state, considered to be a superior force whose task was, among many others, to maintain order, drive the forces of Chaos out of Egypt and protect his subjects from malevolent forces including enemies from different territories. The first attestations of this motif are dated to the Predynastic Period and can be found in the same form more than three thousand years later on the walls of Ptolemaic temples (Arnold 1999). One of the earliest examples of this iconographic element is attested from the late Predynastic tomb L 100 at Hierakonpolis belonging to one of the rulers of a local chiefdom. On the wall of his tomb he is depicted smiting the heads of three bound captives with his war mace (Quibell, Green 1902, pl. LXXVI).

One of the most important scenes in the Cave of Beasts features a large figure of a composite body painted white. It is a combination of beast´s legs (resembling a panther) and a female torso with a clearly visible breast. The head is missing. The figure is leaning against the ground in a way similar to later depictions of the goddess Nut in ancient Egypt. Her unique white colour even allows one to suppose that she may have been considered to be an anthropomorphic form of the Milky Way, which was later on in the Nile valley associated with the legend of the birth of the sun god Ra (Wells 1992). As for the earth, it is rendered as a red figure of a male, which seems to support the body of the white creature, reclining on his right elbow and with his left arm touching/supporting her body. His legs are unnaturally long and nine men are depicted walking on them from the right side. In their hands they carry large elongated items resembling joints of meat carried in a similar fashion as the later offering bearers attested in Egyptian tombs from the Old Kingdom (27th cent. BC) onwards. Based on the scheme of the scene, it may be said that the largest and thinnest figures in the composition represent one species of creature in human shape, most likely beings of different, perhaps divine nature.

The scene is completed by two more large and thin figures, one to the left of the White Goddess and another one below her trunk, with both arms outstretched to her breast. They may also have been divine beings, known in later Egyptian tradition as Shu and Tefnut. Two considerably smaller human figures complement the scene on the right. They (male?) are shown just in place of the missing head of the goddess, are wearing white kilts and appear to be venerating the White goddess.

The whole composition shows that a clear difference can be made between ‘normal/standard’ humans and hypothetically divine beings, rendered as oversized thin figures and the White Goddess respectively. The overall significance and interpretation of the scene can be understood and explained only with the help of much later evidence from ancient Egypt. One of the basic mythological concepts attested from at least the 24th century BC uses the female body of the goddess Nut as a personification of the sky which is supported by the earth god Geb (Bonnet 1952, 536-9). Nut is depicted in a mortuary context, guaranteeing the deceased an eternal existence in the sky (Kurth 1982, 535-41). Later (13th cent. BC), in the reigns of Sethi I and Ramesses IV, the celestial cow (the sky goddess Nut) is depicted as being separated from the earth by the god Shu (Hawass, Vannini 2007, 284–93).

Iconography and evidence – Cave of Swimmers (Wadi Sura I)

In connection with ancient Egyptian mythology and symbolism, it may be of some use to have a look at some iconographic elements still preserved on the walls of the Cave of Swimmers (Wadi Sura I). One of the more common yet unique features of the Cave of Swimmers and Cave of Beasts is that the scenes frequently show small human figures rendered in attitudes on the back or on the belly which may formally remind one of swimmers. It was the author of the most comprehensive publication on the Cave of Swimmers, Hans Rhotert, who first considered them to be actually dead persons (Rhotert 1952, 105). Jean-Loïc Le Quellec also interpreted the swimmers in a similar fashion. He hinted at later parallels which survived in the so-called Coffin Texts where the ‘swimmers’ are considered to be deceased souls floating in the waters of Nun (Le Quellec 2008, 31–33). The fact that swimmers occur in different caves (including Wadi Sura II) and the fact that they seem to form a continuous line spanning most of the cave’s interior[1] make it even more probable that the cave art may indeed feature some incipient concepts that were converted into relevant compositions in the Nile valley some centuries later.

This hypothesis may be confirmed by yet another enigmatic detail featured in the left part of the Cave of Swimmers. Here, a running human figure is seen. The figure has outstretched arms and is rendered in a very vivid attitude of an expressive run. This male (?) figure holds a small, adze-shaped tool in one hand and is wearing an elongated, whitish, cone-like helmet on the head. This object particularly resembles in shape the Upper Egyptian white crown worn by the Upper Egyptian chieftains, later a symbol of the southern part of unified Egypt (Wilkinson 1999, 194–5). The running figure is almost naked except for a loincloth and has a long flying bandage under one knee. The scene seems to be imbued with symbolic/ritual meaning as there is no mundane activity depicted around the figure which would relate to the running person. It was Rhotert who first noticed a similarity between this representation and those of the later pharaohs (Rhotert 1952, 55, 62 and fig. 87 and pl. XXIX, 4-5). The Ancient Egyptian civilisation put significant stress on the ritual renewal of the king (former chieftains) and his physical prowess. The ruler had to demonstrate his ability to maintain his superiority and protect his subjects. Later on, a deeper symbolical meaning was associated with the ritual which was to guarantee the king rejuvenation of his physical and magical power and the renewal of the social contract between him and his people on one side and the gods on the other side. For this reason, the so-called sed-festival was maintained throughout the whole ancient Egyptian civilisation down to the Ptolemaic period. An indispensable part of this ceremony -which we still understand only partially – was a ritual run of the king, holding a baton-like mekes-sceptre in one hand and flails in the other hand and wearing an Upper Egyptian white crown (Hornung, Staehelin 1974; Wilkinson 1999, 212–15). One of the earliest iconographic scenes of the sed-feast has survived on several limestone panels from the underground corridors in the complex of king Djoser in Saqqara (27th century BC) (Friedman 1995, passim and espec. 3 and figs. 2a,b; 23, fig. 14; 33, fig. 19a). Thus it is very tempting to consider the Wadi Sura I representation to be a token of an early concept of the sed-festival.

Expanding the framework

Despite the general acknowledgement that prehistoric populations of the Western Desert played an important role in the creation of the Egyptian civilisation, primary attention has always been paid rather to the Egyptian – Near East connection (Wengrow 2006, 21-29). A direct connection that would link the populations of the Western Desert living in the area of Gilf Kebir and the early inhabitants of the Nile valley has been, however, missing. Only recently the Combined Prehistoric Expedition led by F. Wendorf and R. Schild has indicated that there might be a connection between the Sahara Neolithic and the Neolithic in Upper Egypt (Wendorf, Schild et al. 2006).

Above all, it was the discovery of the Nabta Playa late Neolithic megalithic culture dated to the second half of the fifth millennium BC that enabled us to assume a connection between the Egyptian Western Desert Neolithic and the rise of the Predynastic cultures in Upper Egypt (Wendorf, Close, Schild 1993, 7–16). The Nabta Playa settlement area also featured scientifically spectacular tumuli with cattle burials (Applegate, Gautier, Duncan 2001, 468–88). The latter prove clearly that these prehistoric cattle keepers practised a cult of sacred cows which later became one of the dominant features of ancient Egyptian religion (Hassan 1998, 98–112).

It may be suggested that there are quite a few indications supporting the notion of the cultural transfer of several concepts originally developed by the local populations of the Western Desert and Gilf Kebir areas. We may be correct in looking for the roots of some religious concepts originally thought of as being purely ancient Egyptian to the west of the Nile valley constraints. Some of the scenes of Wadi Sura I and II have the potential to cast a completely new light on several cornerstones of the ancient Egyptian concept of the world and their state (Shafer 1991). The scenes in the caves prove that there was significant social complexity existent in the society that produced them. At the same time, these local communities possessed significant intellectual capability to comprehend their surrounding environment within the framework of complex etiological compositions that later on became a characteristic part of ancient Egyptian culture, mythology and world-view.

In view of these observations, it may be claimed that we are now confronted with evidence indicating that quite a few key concepts of ancient Egyptian civilisation such as etiological myths and the idea of sacred kingship could have had their origins in the Western Desert and were only adopted, but not created by the ancient Egyptian civilisation (for similar line of argument see, for instance, Malville, Wendorf, Mazar and Schild 1998, Le Quellec 2008 and D’Huy 2009). Last but not least, can we consider these caves to be the early temples of the predecessors of the ancient Egyptians?

Click here to watch a full lecture by Prof Bárta at the All Russia Festival.



Almásy, L. E. (1998) Schwimmer in der Wüste: Auf der Suche nach der Oase Zarzura. Munich, Deutscher Taschenbuch Verlag.

Applegate, A., Gautier, A., Duncan, S. (2001) ‘The North Tumuli of the Nabta Late Neolithic Ceremonial Complex’, in Wendorf, F., Schild, R., and Associates, Holocene Settlement of the Egyptian Sahara, Volume I, The Archaeology of Nabta Playa, 468–88, New York, Springer.

Arnold, D. (1999) Temples of the Last Pharaohs, Oxford, New York, Oxford University Press.

El-Baz, F. (2003) ‘Geoarchaeological Evidence of the Relationships between the Terminal Drought in North Africa and the Rise of Ancient Egypt’, in Hawass, Z., Pinch Brock, L., eds., Egyptology at the Dawn of the Twenty-first Century. Proceedings of the Eighth International Congress of Egyptologists, Cairo, 2000. Volume 1. Archaeology , Cairo, American University in Cairo Press, 64–72.

Bárta, M. (2010) Swimmers in the Sand. On the Neolithic Origins of Ancient Egyptian Mythology and Symbolism, Prague, Dryada.

Bonnet, H. (1952) Reallexikon der ägyptischen Religionsgeschichte, 536–39, Berlin, De Gruyter.

Bradley, D. G., M. D. E., P. Cunningham, R. T. Loftus (1996) Mitochondrial diversity and the origins of African and European cattle. Proceedings of the National Academy of Sciences of the United States of America, 93.10 (May 14), 5131–5.

Brunton, Guy, and Gertrude Caton-Thompson. 1928. The Badarian civilisation and predynastic remains near Badari. London: British school of archaeology in Egypt [etc.].

Bubenzer, O., Bolten, A., Darius, F., eds. (2008) Atlas of Cultural and Environmental Change in Arid Africa, 33–7, Cologne, H. Barth Institute.

Friedman, F. D. (1995) ‘The Underground Relief Panels of King Djoser at the Step Pyramid Complex’, Journal of the American research Center in Egypt 32, 1–42.

Hassan, F. A. (1998) ‘The Earliest Goddesses of Egypt: Divine Mothers and Cosmic Bodies’, in Goodison, L., Morris, C., eds., Ancient Goddesses: Myths and Evidence, 98–112, Madison, WI, University of Wisconsin, Press.

Hawass, Z. and S. Vannini (2007) The Royal Tombs of Egypt. The Art of Thebes Revealed, 284–93, Cairo, Cairo University Press.

Hoffman, M. A. (1979) Egypt before the Pharaohs: the prehistoric foundations of Egyptian civilization, New York, Knopf.

Hornung, E. and E. Staehelin (1974) Studien Zum Sedfest, Ægyptiaca Helvetica. Genève: Edition de Belles-Lettres.

D’Huy, J. (2009) ‘New evidence for a closeness between the Abû Râ’s shelter (Eastern Sahara) and Egyptian beliefs’, Sahara 20, 125–26.

Kuper, R., H. Riemer and F. Förster (2009) ‘Wadi Sura 1: Report on the first field season of the Wadi Sura Project (Gilf Kebir, SW Egypt), spring 2009,’ online report: http://wadisura.phil-fak.uni-koeln.de/uploads/media/WADI_SURA_report_1_2009-1.pdf.

Kuper, R., H. Leisen, H. Riemer, F. Förster et al. (2009) ‘Report on the second field season of the Wadi Sura Project (Gilf Kebir, SW Egypt) in spring 2010,’ online report: http://wadisura.phil-fak.uni-koeln.de/uploads/media/WADI_SURA_report_2_2009-2.pdf.

Kuper, R., H. Leisen, H. Riemer and F. Förster et al. (2010) ‘Report on the third field season of the Wadi Sura Project (Gilf Kebir, SW Egypt) in autumn 2009,’ online report: http://wadisura.phil-fak.uni-koeln.de/uploads/media/WADI_SURA_report_3_2010-1.pdf.

Kuper, R., H. Leisen, H. Riemer, O. Bubenzer et al. (2011) ‘Report on the fourth field season of the Wadi Sura Project (Gilf Kebir, SW Egypt), spring 2011,’ online report: http://wadisura.phil-fak.uni-koeln.de/uploads/media/WADI_SURA_report_4_2011-1.pdf.

Kuper, R., H. Leisen, H. Riemer, S. Krause et al. (2011) ‘Report on the fifth field season of the Wadi Sura Project (Gilf Kebir, SW Egypt), autumn 2011,’ online report:


Kuper, R., Kröpelin, S. (2006) ‘Climate-Controlled Holocene Occupation in the Sahara: Motor of Africa’s Evolution,’ SCIENCE vol. 313, 11 August 2006, 803–7.

Kuper, R. (2013). Wadi Sura - The Cave of Beasts, Africa Praehistorica 26, Cologne: Heinrich-Barth-Institut.

Kurth, D. (1982) ‘Nut’, in W. Helck, W. Westendorf, eds., Lexikon der Ägyptologie IV, 535–41, Wiesbaden, Harrassowitz.

Malville, M., Wendorf, F., Mazar, A.A. & R. Schild, C., 1998. Megaliths and Neolithic astronomy in southern Egypt. Nature 392, 488-491.

Morenz, S. (1992) Ancient Egyptian religion, Ithaca, NY, Cornell University Press.

Le Quellec, J.-L. (2008) ‘Can one 'read' rock art? An Egyptian Example’, in P. Taylor, ed. Iconography without Texts, London, Warburg Institute (Warburg Institute Colloquia 12), 25–42.

Le Quellec, J.-L., P. d. Flers, P. d. Flers, N.-C. Grimal (2005). Du Sahara au Nil, peintures et gravures d'avant les pharaons, Paris, Fayard.

Quibell, J. E. 1900. Hierakonpolis I, Part I, London, B. Quaritch.

Quibell, J. E., Green, F. W. (1902) Hierakonpolis. Part II, London, B. Quaritch.

Rhotert, H. (1952) Libysche Felsbilder. Ergebnisse der XI. und XII. deutschen inner-afrikanischen Forschungs-Expedition (DIAFE) 1933/1934/1935, Darmstadt, L. C. Wittich Verlag.

Szabo, B. J., C. V. Haynes, Jr., T. A. Maxwell (1995) ‘Ages of Quaternary pluvial episodes determined by uranium-series and radiocarbon dating of lacustrine deposits of Eastern Sahara,’ Paleography, Paleoclimatology, Paleoecology 113, 227–42.

Tassie, G. J. (2014). Prehistoric Egypt. Socioeconomic Transformation in North-east Africa from the Last Glacial Maximum to the Neolithic. 24,000 to 6,000 cal BP. London: Golden House Publications.

Wells, R. A. (1992) ‘The Mythology of Nut and the Birth of Ra,’ Studien zur altägyptischen Kultur 19, 305–21.

Wendorf, F., A. Close, R. Schild (1993) ‘Megaliths in the Egyptian Sahara’, Sahara 5, 7–16.

Wendorf, F., R. Schild and Associates (2006) Holocene settlement of the Egyptian Sahara. Volume I. The Archaeology of Nabta Playa, New York, Springer.

Wengrow, D. 2006. The archaeology of Early Egypt. Social transformations in North-East Africa, 10,000 to 2650 BC, Cambridge: Cambridge University Press.

Wilkinson, Toby A. H. (1999) Early Dynastic Egypt. London, New York, Routledge.

[1] Personal observation of Veronika Dulíková in both Wadi Sura I and II.


Dr. Miroslava Bárta is Professor of Egyptology at Charles University in Prague, Czech Republic, and Director of the Czech Institute of Egyptology.
The views expressed in this article are the author's own and do not necessarily reflect Dr Hawass' views.
All information and media on this website are © Laboratoriorosso and cannot be used or distributed without permission.

Comments are closed.