On the Origins and Dissemination of Domesticated Sorghum and Pearl Millet across Africa and into India: a View from the Butana Group of the Far Eastern Sahel

Abstract

Four decades have passed since Harlan and Stemler (1976) proposed the eastern Sahelian zone as the most likely center of Sorghum bicolor domestication. Recently, new data on seed impressions on Butana Group pottery, from the fourth millennium BC in the southern Atbai region of the far eastern Sahelian Belt in Africa, show evidence for cultivation activities of sorghum displaying some domestication traits. Pennisetum glaucum may have been undergoing domestication shortly thereafter in the western Sahel, as finds of fully domesticated pearl millet are present in southeastern Mali by the second half of the third millennium BC, and present in eastern Sudan by the early second millennium BC. The dispersal of the latter to India took less than 1000 years according to present data. Here, we review the middle Holocene Sudanese archaeological data for the first time, to situate the origins and spread of these two native summer rainfall cereals in what is proposed to be their eastern Sahelian Sudan gateway to the Red Sea and the Indian Ocean trade.

Keywords: Archaeobotany; Butana Group; Nubia; Origins of agriculture; Pennisetum glaucum; Sorghum bicolor.

Fig. 1

Map of the archaeological distribution of sorghum and pearl millet finds in northern Africa up to AD 350. The median age of the earliest phase of occurrence for each site is represented. Wild forms are indicated in gray, domesticated morphology indicated in black. Sites: 1. Waladé; 2. Cubalel; 3. Dhar Tichitt; 4. Dhar Oualata; 5. Djiganyai; 6. Oued Chebbi; 7. Oued Bou Khzama; 8. Dia Shoma; 9. Ounjougou (Varves West); 10. Windé Koroji; 11. Karkarichinkat; 12. Boase B5C; 13. Birimi; 14. Oursi; 15. Ti-n-Akof; 16. Nok sites; 17. Bwambe-Sommet; 18. Aband Minko’o; 19. Ganjigana; 20. Mege; 21. Kursakata; 22. Boso-Njafo; 23. Ti-n-Torha/Two Caves; 24. Uan Tabu; 25. Takarkori; 26. Uan Muhuggiag; 27. Tinda B; 28. Zinchechra; 29. Jarma; 30. Farafra Oasis; 31. Abu Ballas; 32. Dakhleh Oasis; 33. Kharga Oasis; 34. Wadi Qitna; 35. Qasr Ibrim; 36. Nabta Playa; 37. Kawa; 38. Umm Muri; 39. Berenike; 40. Shenshef; 41. Dangeil; 42. Kasala; 43. Jebel Tomat; 44. Abu Geili; 45. Shaheinab; 46. El Kadada; 47. Meroe; 48. Hamadab; 49. El Zakiab; 50. Kadero; El Zakiab 51. El Mahalab; 52. Umm Direiwa; 53. Sheikh el Amin; 54. Sheikh Mustafa; 55. Shaqadud; 56. Jebel Qeili. (map: D.Q. Fuller)

Fig. 2

Map of the Sudan, relating regions and sites discussed. The presence of sorghum in sites of various period is indicated. Sites: 1. Sai Island; 2. Kawa; 3. Umm Muri; 4. Dangeil; 5. Es Sour; 6. Meroe; 7. Hamadab; 8. El Kadada; 9. Sheheinab; 10. Geili; 11. El Zakiab; 12. Kadero; 13. Umm Direiwa; 14. El Mahalab; 15. Sheikh Mustafa; 16. Sheikh el Amin; 17. Jebel Qeli; 18. Jebel Tomat; 19. Rabak; 20. Jebel Moya; 21. Abu Geili

Fig. 3

Butana Group sites across the southern Atbai (from Winchell , Fig. 1.4)

Fig. 4

Rim sherds of Classic Gash rim banded bowls and cups (from Manzo , Fig. 83)

Fig. 5

Horizontally combed and grooved Jebel Mokram Group sherds (from Manzo , Fig. 85)

Fig. 6

Quantitative summary of archaeological spikelet base data for early sorghum from eastern Saharan/Sahel regions, based on macro-remains from Nabta Playa (Wasylikowa and Dahlberg 1999) and impressions from Sudanese sites (Stemler , ; Abdel-Magid ; Winchell et al. ; Beldados et al. 2018). These sites are selected based on the presence of quantitative data on spikelet base morphology

Fig. 7

Chipped stone pick of the Butana Group (from Mbutu , Fig. 2.12; reproduced with permission)

Fig. 8

a and b Seed-chaff tempered ceramics (Khordhag Plain type) of the Butana Group. ([b] from Winchell , Fig. A.37)