Elsevier

Quaternary International

Volume 343, 1 September 2014, Pages 69-84
Quaternary International

The Hargeisan revisited: Lithic industries from shelter 7 of Laas Geel, Somaliland and the transition between the Middle and Late Stone Age in the Horn of Africa

https://doi.org/10.1016/j.quaint.2014.04.038 Get rights and content

Abstract

This paper presents the lithic analysis from the stratified deposits of one of the Laas Geel rock shelters in Somaliland (Shelter 7). This set of sites identified in 2002 is composed of around 20 rock shelters, most featuring polychromic paintings attributed to the Holocene. Moreover, Shelter 7 yielded numerous remains that are divided into two major chono-cultural phases. Microlithic industries from the upper part of the deposit are ascribed to the Late Stone Age (LSA). In contrast, the industries from the lower part of the deposit feature an association of LSA characteristics (but quite different from the ones from the upper part of the deposit) with technical traits that belong to the MSA, and especially the presence of a Levallois technology. This association reintroduces the question of the existence of an industry (or facies) once labeled “Hargeisan” and characterized by mixed features from the MSA and the LSA in this part of Africa at the end of the Pleistocene.

Our analysis of these industries culminates in a reconsideration of the definitions proposed by Clark in his seminal book, “The Prehistory of the Horn of Africa” (1954). This study provides new typo-technological data from the upper part of the Laas Geel Shelter 7 deposit that contribute to refining Clark's definition of the ‘Somaliland Wilton’ industry. In addition, the association of blade/bladelet production with typical Levallois production in the lower part of this site's deposits supports a comparison with the “Hargeisan Industry”. The industries recovered from Laas Geel Shelter 7 reintroduce debates on the existence of the “Hargeisan industry” and provide new data on the technological transition between the MSA and the LSA in the Horn of Africa.

Introduction

Clark (1954) produced the first synthesis concerning the prehistory of the Horn of Africa. He characterized industries from the Early Stone Age (Upper Acheulian, which was barely documented in Somalia at the time) as well as the Middle Stone Age (Acheulio-Levalloisian, Levalloisian, Somaliland Stillbay) and the Late Stone Age (Doian, Somaliland Wilton). He described the archaeological potential of Somalia and put it in a broader perspective, by making comparisons with Eastern Africa and the African continent as a whole. He also stressed the particularities of the prehistoric record of this region, notably with the description of some industries that combine Late Stone Age (LSA) and Middle Stone Age (MSA) features (Somaliland Magosian, Hargeisan), which he considered a transitional facies: “Transional between the Middle and the Late Stone Age is the Somaliland variant of the Magosian and the intrusive blade and burin industries of the Hargeisan.” (Clark, 1954, p. 156).

This issue was examined again by S.A. Brandt and T.H. Gresham when they resumed research in Somalia (Brandt and Brook, 1984, Gresham, 1984, Brandt, 1986, Brandt and Gresham, 1990). The material collected during their fieldwork and a revision of the existing data allowed these authors to clarify some of the definitions proposed by Clark for the LSA industries. They gave a more detailed description of the Eibian (called Doian by Clark), introduced the Baardale and its different phases and strongly criticized the Somaliland Wilton. Finally, they were skeptical about the existence of a transitional facies between the MSA and the LSA – and of the Hargeisan in particular, because of the paucity of excavated and properly dated contexts.

We report new field research that renews previous data concerning the prehistory of the region, and particularly as a result of new data available from the site of Laas Geel in Somaliland (Fig. 1). This complex of rock shelters is located 80 km from Hargeisa (Somaliland) and is already known for its numerous rock paintings, which were described in 2002 (Gutherz et al., 2003a, Gutherz et al., 2003b, Gutherz and Jallot, 2011). Test excavations carried out in Shelter 7 allowed us to identify an important sedimentary deposit with stratified archaeological remains. The analysis of these deposits was first initiated to provide contextual data for the site's remarkable rock art. However, this analysis has resulted in new issues which are developed in this paper, and which are mainly based on examination and interpretation of the excavated lithic industries.

Our analysis of the lithic assemblage from excavations of Sondage 2 in Shelter 7 at Laas Geel enabled us to group the stratigraphic units (hereafter SUs) into two main archaeological phases: 1) the upper part of the sequence includes industries that we can undoubtedly assign to the LSA (SU 700–708); 2) by way of contrast, the stratigraphic units from the lower part of the sequence (SU 709–711) feature an industry combining technological components that are usually identified as both MSA and LSA. The lithic artifacts from Laas Geel therefore enable us to cross-examine some of the definitions proposed by Clark, namely the pertinence of “Hargeisan” and “Somaliland Wilton” labels, and to revisit the origins and developments of the LSA in the Horn of Africa.

Section snippets

The complex of Laas Geel

Laas Geel is one of the few rocky hills within a large eroded granitic peneplain halfway between Hargeisa and Berbera (Republic of Somaliland). The hill, which has an approximate elevation of 1025 m asl, overlooks the confluence of two wadis, whose underground flow is exploited by nomads who dig wells to water their herds. The hill features twenty or so natural rock shelters, most of which have wall paintings. Two shelters yielded preserved sedimentary deposits that were tested (shelters 7 and

The lithic industries of Laas Geel

Here, we present the results of our analysis of the excavated assemblage from Sondage 2A alone, because the material from Sondage 2B and 2C is currently being processed. However, the analysis of the lithic artifacts from Sondage 2A alone confirms the importance and the potential of the Shelter 7 deposits and contributes to our understanding of the technological changes that occurred between the end of the Pleistocene and the beginning of the Holocene in the Horn of Africa.

The richness of the

Raiders of the lost Hargeisan

As stated above, in his synthesis on the prehistory of the Horn of Africa, Clark (1954) proposed the existence of a transitional facies (the Hargeisan) between typically MSA (Somaliland Stillbay) and typically LSA (Somaliland Wilton) industries. The Hargeisan facies, which was distributed within the borders of the British Somalilands (nowadays the Republic of Somaliland) was characterized by Levallois productions (mostly oriented toward the production of unifacially retouched points) associated

Conclusion

The analysis conducted and discussed here, reinforces the existence of a very specific industry in the Horn of Africa at the end of the Pleistocene, which was originally identified and coined Hargeisan by Clark (1954). While Clark's Hargeisan facies is problematic given the taphonomic and chronological issues present on type-sites of this industry, his descriptions remain the closest example we have to date of the industry from the lower sequence of Shelter 7 at Laas Geel and pertain to a

Acknowledgements

The missions conducted at Laas Geel between 2002 and 2011 were made possible thanks to a warm and attentive welcome from the Republic of Somaliland's higher authorities. The site of Laas Geel and the Somaliland Archaeological Heritage were managed successively by the Ministry of Culture, the Ministry of Commerce, Industry and Tourism and are currently under the authority of the Ministry of Youth and Sports. We are thankful to the different Ministers and their teams who have been supportive of

References (24)

  • J.D. Clark

    The Prehistoric Cultures of the Horn of Africa

    (1954)
  • T.H. Gresham

    An Investigation of an Upper Pleistocene Archaeological Site in Northern Somalia

    (1984)
  • Cited by (26)

    • Improved ostrich eggshell and ungulate tooth enamel radiocarbon dating methods reveal Later Stone Age occupation in arid MIS 2 southern Somalia

      2021, Journal of Archaeological Science: Reports
      Citation Excerpt :

      Evidence of MIS 2 occupation in Somalia/Somaliland is restricted to three sites. At Laas Geel rock shelter, a radiometric date gap exists from ~ 41 to ~ 13 cal ka (Gutherz et al., 2014). A date of ~ 19–18.5 uncal BP has been recovered from Midhishi 2 rock shelter (Brandt, 1986).

    • Hunter-gatherer reliance on inselbergs, big game, and dwarf antelope at the Rifle Range Site, Buur Hakaba, southern Somalia ∼20,000−5,000 BP

      2018, Quaternary International
      Citation Excerpt :

      However, OES samples are not always accurate, as is potentially the case for any material submitted for radiocarbon dating. A recent example is the late Pleistocene/Holocene site of Laas Geel in Somaliland (Gutherz et al., 2014; Lesur pers. comm., 2016) where OES dates were found to be inconsistent throughout the sequence.

    • Engraved ostrich eggshell from the Middle Stone Age contexts of Goda Buticha, Ethiopia

      2018, Journal of Archaeological Science: Reports
      Citation Excerpt :

      Certain LSA components (e.g., backed microliths) are more common in the upper levels, as would be expected, but MSA features (e.g., unifacial and bifacial points, Levallois products) persist throughout, across the long occupational hiatus and into the Holocene (Leplongeon et al., 2014; Leplongeon, 2014; Pleurdeau et al., 2014; Tribolo et al., 2017). This absence of a clear technological delineation between MSA and LSA levels is consistent with observations at sites in other parts of Ethiopia (e.g., Mochena Borago; Brandt et al., 2012; Brandt et al., 2017) and across eastern Africa (Tryon and Faith, 2013; Gutherz et al., 2014). Over two hundred piece-plotted ostrich eggshell fragments were recovered from Goda Buticha, all from Complex II.

    View all citing articles on Scopus
    View full text