Introduction

Hypotheses on crop development in Africa are long on theory and short on fact. Let me present my own ideas. Harlan & Stemler (1976) have presented a theory that sorghum developed in the southern Sudan–Chad region. My problem with that is the answer to the question ‘how’? It is true that as soon as Man began to sow the seeds of wild grasses, selection and sowing over the years would result in cultivated types being developed. That assumes that Man somehow learnt the idea of agriculture. Perhaps he did, but it is not clear how this happened in the rainfed savannahs. There were lots of grasses there anyway, and lots of grass seed. Why sow more? How would man have learnt to sow seed? How would he have distinguished between the masses of grass seedlings coming up with the rains and those which he had put in? By clearing a separate plot of land for sowing? That presumes the idea of agriculture. Having worked in the tropics for many years, I find this altogether too difficult to imagine.

It is more likely, to my mind, that agriculture was discovered along rivers and that the discovery was a rather rare event. One should always look at the possibility of the spread of the idea of agriculture from elsewhere before concluding that it had been discovered all over again. My scenario for the discovery of agriculture in a situation of this kind is presented below.