In nineteenth century Africa a substantial proportion of the population was nomadic.
Hunter gatherers, fishers and pastoralists all had production systems based on
exploitation of seasonally and spatially fluctuating natural resources. Hence people lived
in housing that could be packed up and moved in order to follow this seasonal availability
across space. Although some forms of agriculture also require mobility (swidden, slash
and burn agriculture) such movements are usually every few years compared to the year
round mobility practised by hunter-gatherers, fishers and pastoralists. The widespread
nomadic lifestyle enabled the exploitation of arid and semi-arid lands that could not easily
support a sedentary population.
Throughout the twentieth century the proportions of the population who were nomadic
declined, along with a decrease in actual numbers of nomads. In the case of pastoralists
this was often the result of catastrophic disease, such as the east African rinderpest
epidemic in the late nineteenth century (Homewood 2008) and or drought which caused
livestock and human deaths; more general erosion of viable pastoral livelihoods caused
individuals and households to drop out of their traditional production system and take up
agriculture or urban occupations. Occasionally people managed to acquire enough
resources to re-enter pastoralist production but this seems to have been quite rare
(Bonfiglioli 1990, Cisse 1981, Little & Leslie 1999). Beyond dramatic catastrophes the
rapid population growth of all African populations, the expansion of agriculture into former
pastoral zones and increasing restrictions on movement across national boundaries have
had a major impact on reducing the viability of mobile extensive livestock raising.
It has frequently been put forward that nomadic pastoralist populations have lower natural
population growth rates than sedentary farming populations (Henin 1968, 1969, Swift
1977, Roth 1994). In fact the evidence for this seems to have emerged partly from
colonial bias and problems in managing nomadic populations (Randall 2009) and does
not take into account the fact that data on pastoralist demographic dynamics are often
based on very small samples and data quality is often very poor. A review of all available
studies of pastoralist demographic dynamics could not identify any systematic pattern of
lower (or higher) fertility or systematic mortality differences between mobile pastoralists
and sedentary agriculturalists (Randall 2008). It is true that in the Sahelian belt of West
Africa, some of the pastoralist populations, particularly the former slave owning Tuareg
and Maures, did have somewhat lower fertility than neighbouring agricultural populations;
this is almost entirely due to the monogamous marriage pattern (Randall 1984).
8
Polygamous east African nomadic pastoralists do not show similarly low fertility, and
neither do mobile Fulani in northern Burkina Faso (Hampshire and Randall 2000).