Iron production in second millennium AD pastoralist contexts on the Laikipia Plateau, Kenya

Abstract

Iron has played an important role within East African pastoralist societies for many hundreds of years, yet the means by which iron was produced or obtained by these communities has not been thoroughly documented. The bulk of our understanding is presently based on a limited number of ethnographic and artefact studies, which have tended to focus on the functional and symbolic nature of iron objects themselves. We argue that the research presented here provides the first opportunity to add to this narrow knowledge base by reconstructing the iron production technologies of pastoralist communities in Laikipia, Kenya, using an archaeometallurgical approach. Seven furnaces and one iron-production refuse area were excavated at two discrete workshop sites in Laikipia, central Kenya, that date to the second half of the second millennium AD. The recovered archaeometallurgical materials were analysed using optical microscopy, SEM-EDS and ED-XRF. These techniques revealed that the smelting technologies in question were complex and sophisticated and that they utilised titania-rich black sands and lime-rich charcoal. Whereas the technical approach and raw materials were found to be similar at both sites studied, there was striking stylistic variation in furnace design for no apparent functional reason, which might suggest nuanced differences in the socio-cultural affiliations of the smelters who worked at these sites. This paper explores some of the possible reasons for these differences. In particular, by integrating archaeological data with existing ethnographic and ethnohistoric research from the region, we discuss the technological choices of the smelters and what this might tell us about their identities, as well as considering how future research should best be targeted in order to develop a greater understanding of the organisation of production within pastoralist central Kenya.

Le fer a joué un rôle important au sein des sociétés pastorales de l'Afrique orientale depuis plusieurs centaines d'années, mais nous savons peu sur la façon dont le fer fut produit ou obtenu par ces communautés. La majeure partie de notre compréhension se base actuellement sur un nombre limité d'études ethnographiques et d’études d'artefacts, et celles-ci ont eu tendance à se concentrer sur le caractère fonctionnel et symbolique des objets de fer eux-mêmes. La recherche présentée ici offre une première opportunité d'ajouter à cette base de connaissances assez étroite, en reconstituant les technologies de production du fer des communautés pastorales à Laikipia, au centre du Kenya, par le biais d'une approche archéométallurgiques. Sept fourneaux et une zone de déchets de production du fer ont été fouillés à Laikipia sur deux ateliers distincts datant de la seconde moitié du deuxième millénaire de notre ère. Les matériaux archéométallurgiques recueillis ont été analysés par microscopie optique, SEM-EDS et ED-XRF. Ces techniques ont révélé que les technologies de fonte étaient complexes et sophistiquées et qu'elles employaient des sables noirs riches en dioxyde de titane et des charbons riches en chaux. Si l'approche technique et les matières premières paraissent similaires sur les deux sites étudiés, il existait une différence stylistique frappante dans la conception du fourneau, non explicable par des raisons fonctionnelles. Ceci pourrait indiquer une différence dans les affiliations socio-culturelles des forgerons qui ont travaillé sur ces sites, et le présent article explore quelques-unes des possibles raisons pour ces différences. En particulier, en intégrant les données archéologiques avec les recherches ethnographiques et ethnohistoriques conduites dans la région, on examine la question des choix technologiques des forgerons et ce que cela pourrait nous dire de leurs identités. L'article considère également la façon dont les futures recherches devraient être ciblées afin de développer une meilleure compréhension de l'organisation de la production parmi les sociétés pastorales du centre du Kenya.

Keywords:

• iron
• pastoralism
• Africa
• Kenya
• archaeometallurgy

Acknowledgements

We should like to thank all those who made this research possible, most especially the generosity of Robert Wells throughout the fieldwork on Lolldaiga Hills Ranch, and the assistance of his staff. The research was undertaken under the auspices of Kenya Research Permit MOEST 13/014 issued by the Ministry of Education to Paul Lane. The archaeological surveys between 2002 and 2005 were generously funded by the British Academy, under the auspices of the British Institute in Eastern Africa's project: Landscape and Environmental Change in Semi-Arid Regions of Eastern and Southern Africa, Developing Interdisciplinary Approaches. Considerable thanks must go to all those who worked in Laikipia in 2004 and 2006. The Arts and Humanities Research Council (AHRC) and the Institute of Archaeometallurgical Studies sponsored the 2006 fieldwork, with additional support from the British Institute in Eastern Africa. This paper is based on the MSc dissertation undertaken by Louise Iles at the Institute of Archaeology, University College London, funded by the AHRC, and supervised by Marcos Martinón-Torres and Andrew Reid. We wish to thank John Sutton, John Galaty, Pete Robertshaw and two anonymous referees for their comments on earlier drafts of this paper, as well as Oli Boles for preparing Figure 3 from various digital files. A preliminary version of this paper was presented at the Métallurgie du Fer et Sociétés Africaines workshop at the Université de Provence, Aix-en-Provence, in April 2010. Our thanks are extended to the organisers, and especially Caroline Robion-Brunner and Bruno Martinelli, for the invitation to attend and to various participants for their comments on this version of the paper. We are responsible for any remaining errors.

Notes on contributors

Louise Iles is a Marie Curie Research Fellow at the Department of Archaeology, University of York. She is currently examining the pre-colonial metal technologies of Pare, northern Tanzania. She has worked on metallurgical remains throughout eastern Africa and her main research interests include technological change and the interactions between technology, environment and culture.

Paul Lane’s main interests are in the historical ecology of African landscapes, the archaeology of colonial encounters, the materialisation of memory, the organisation and use of space and time in pre-industrial societies, maritime archaeology and the transition to farming in Africa. A former Director of the British Institute in Eastern Africa (1998–2006) and President of the Society of Africanist Archaeologists (2008–2010), he is currently Professor of Global Archaeology at Uppsala University and Honorary Research Fellow of GAES, University of the Witwatersrand.

Notes

1. Although there is debate about the suitability of the terminology ‘PN’ and ‘PIA’ (Robertshaw and Collett 1983), these terms are used here to provide a broad socio-economic context of the sites discussed (see Lane 2013).

2. Based on two dates obtained on charcoal, 1070 ± 110 BP (N-652) and 1300 ± 140 BP (GX-5543) (Ambrose et al. 1984: 83), calibrated to 690–1185 cal. AD (IntCal13) / 768–1224 cal. AD (SHCal13), and 431–1017 cal. AD (IntCal13) / 500–1127 cal. AD (SHCal13) respectively. Other radiocarbon dates on bone apatite and collagen are also available for the site. All dates in the text have been calibrated using OxCal v 4.2, with either IntCal13 and/or SHCal13 depending on site location relative to the Equator (Bronk Ramsey 2009; Hogg et al. 2013; Reimer et al. 2013) and are cited at 95.4% probability (2 sigma). All uncalibrated radiocarbon ages are presented in the text as years BP; calibrated dates are reported as cal. AD.

3. While not identical, this bears some similarity to the iron-working craft castes found in many parts of West Africa (Tamari 1991, 2012; Baroin 2012) and the Horn of Africa (Finneran 2003; Haaland et al. 2004).

4. Such as the Mukogodo, who are known historically to have been specialist hunter-gatherers (Mutundu 1999), and ethnographically to have supplied Samburu with pottery that is quite similar to Kisima ware (Brown 1989: 77–8). For an account of recent pottery manufacture by ‘Dorobo’ groups on the adjacent Leroghi Plateau, see Grillo (2012).