The 'Ella' stone platforms in Mursiland,
Lower Omo Valley, south-western Ethiopia

Timothy Clack & Marcus Brittain

Figure 1
Figure 1. Map of the Lower Omo Valley, Ethiopia.
Click to enlarge.

Investigation into the archaeology of south-western Ethiopia has been dominated by human evolutionary interests, in spite of the identification of potential sites from later periods (e.g. Brown 1975). This project aims to redress this imbalance by means of an interdisciplinary study into the correlation between the construction of memory and the experience of past and present landscapes. The broad geographic focus of this project is the Lower Omo Valley (Figure 1), inhabited by eight large, distinct ethnic groups speaking six different languages, together representing an unusual cultural and linguistic diversity within a compressed area. The core of the present study is the landscape of one of these groups, the Mursi, whose understandings of the past are dependent upon oral histories recounting the group's migration into the locale, their expectations and realisations on arrival and their enactment of ownership through diverse practices of place-making (Turton 2005). Oral histories also confirm the Mursi's nomadic heritage and their aggressive colonisation of the landscape approximately 200 years ago. Unfortunately very little is known about the previous inhabitants of the locale, although it is generally supposed that most rangeland cultures from the period exhibited fluid identities with highly permeable social boundaries (Sobania 1991). In the present age, most transhumant groups demonstrate minimal architecture of permanence or substance, and carry few endurable or inorganic artefacts. Thus for these reasons certain features found during survey of the Mursi landscape provide impetus for further study.

Survey and findings

The project was initiated in December 2008 with an archival and literature review. The first field season was conducted between May and July 2009 and incorporated periods of foot survey and site testing along with the collection of ethnohistories from local groups. Although lithic and ceramic surface scatters were found throughout the survey area, it was the knowledge and way-finding of indigenous informants which led us to the most significant of our finds: a series of stone platforms (Figures 2, 3 and 4) near the remote Mursi village of Dirikoro about 45km south-west of Jinka, the nearest urban centre. To date, a total of 15 platforms, features known locally as the 'Ella', have been identified. Perhaps significantly, the platforms are nestled within a landscape considered sacred by the contemporary indigenes.

Figure 2
Figure 2. Photograph of Platform 3.
Click to enlarge.
Figure 3
Figure 3. Plan of Platform 3.
Click to enlarge.

Figure 4a
Figure 4a. Platform 15 facing west.
Click to enlarge.
Figure 4b
Figure 4b. Plan of Platform 15.
Click to enlarge.

The platforms so far identified indicate that the Ella are sited in a general north-westerly alignment. It is interesting to note that certain features in the landscape, considered by the contemporary inhabitants to possess supernatural potency, also appear to fit into this linear pattern of orientation. Such features include: a pit, where dark soil is extracted for numerous ritual usages; a large sacred tree, under which important meetings and ceremonial activities are performed; and an area of high relief (Figure 5). The precise correspondence between the Ella and these other features still remains to be determined. Eight of the platforms were cleared during the field season, each varying between 2.5m and 13m in diameter, with multiple concentric rings of stones and a gulley or blocked entrance almost exclusively oriented to the west. Cleaning between the stones exposed large quantities of struck lithic debitage in addition to faunal remains, much of which had been burnt at a high temperature, possibly cremated. A quarter section was opened on one of the smaller outlying platforms from which some of the stones had already been displaced (Figure 6). The quarter section revealed a large central pit beneath the platform, from which a single struck stone flake was recovered. On the outer periphery of the platform was a small pit containing a small quantity of cremated bone fragments. It is possible that this once held an upright stone or post.

Figure 5
Figure 5. Photograph of the landscape after clearance giving a sense of linear orientation with the sacred tree and Dirikoro peak. Platforms 1, 2 (foreground) and 3 are exposed.
Click to enlarge.
Figure 6
Figure 6. Plan of two of the smaller outlying platforms, 1 (bottom) and 2 (top, with quarter trial section).
Click to enlarge.

Provisional assessment
Figure 7
Figure 7. Two Mursi women deliberate an Ella platform recently cleared of vegetation.
Click to enlarge.

Many details concerning the Ella stone platforms remain unknown. Whilst we can be fairly confident that the features predate the arrival of the Mursi colonialists, the important possible issue of re-use notwithstanding, their antiquity, purpose and cultural origins need further investigation. The Ella platforms seem to be exceptional monumental forms. Nonetheless, they share certain qualities with other proximal archaeological features. For example, stone monuments, energised by the interplay of human and bovid remains, are known from the Pastoral Neolithic in the Turkana Basin (Robbins 2006); numerous lengthy traditions of monumental stelae are reported throughout Ethiopia, Eritrea and Somalia (Joussaume 2007); and linear monumental construction throughout the Sahara and Eastern Africa (Paris 1996). At the same time the significance of the platforms as components in the Mursi worldview must be documented (Figure 7). According to oral accounts the Ella platforms were constructed by the previous inhabitants of the locale, and used as house floors elevated from the ground at a time when the climate was significantly wetter. The recent findings make it hard to validate this interpretation, and offer a potential insight to the way that memory impacts upon the Mursi experience of the landscape.


Fieldwork funds were gratefully received from the John Fell Fund, University of Oxford, the McDonald Institute for Archaeological Research, University of Cambridge, and the British Institute in Eastern Africa. We would like to thank the ARCCH for facilitating our ongoing work in the locale, Miguel Sevilla-Callejo and Alberto Arzoz for permission to use Figures 1 and 5 respectively, and David Anderson, Marco Bassi, Graciela Gil-Romera, Tadele Solomon, Serge Tornay and David Turton for their magnanimous assistance.


  • BROWN, F.H. 1975. Barbed bone points from the Lower Omo Valley, Ethiopia. Azania 10: 144-8.
  • JOUSSAUME, R. 2007. Tuto Fela et Les Stèles du Sud de l'Éthiopie. Paris: Ministere des Affaires Étrangères.
  • PARIS, F. 1996. Les Sépulcheres du Sahara Nigérien du Néolithique à l'Islamisation. Paris: Orstom.
  • ROBBINS, L.H. 2006. Lake Turkana archaeology: the Holocene. Ethnohistory 53(1): 71-93.
  • SOBANIA, N.W. 1991. Feasts, famines and friends: nineteenth-century exchange and ethnicity in the eastern Lake Turkana regional system, in P. Bonte & J. Galaty (ed.) Herders, warriors and traders: pastoralism in Africa. Boulder (CO): Westview.
  • TURTON, D. 2005. The meaning of place in a world of movement. Journal of Refugee Studies 18(3): 258-80.


* Author for correspondence

  • Timothy Clack*
    Institute of Archaeology, University of Oxford, 36 Beaumont Street, Oxford OX1 2PG, UK (Email:
  • Marcus Brittain
    Cambridge Archaeological Unit, University of Cambridge, Downing Street, Cambridge CB2 3DZ, UK (Email: