On the worldwide scale Africa is the least-known continent demographically. Until the mid-twentieth century not even the size of the population was precisely known in many areas of sub-Saharan Africa. The major problems in African historical demography have either been the almost total lack of relevant sources or, if some have been available, they have been fragmentary and non-systematic. The reliability of the most commonly-used sources in African historical demography—population counts and early censuses—remained questionable until the 1960s. However, fairly far-reaching conclusions and estimations based on these sources using indirect methods have been drawn. Despite the development of methods in historical demography, the questionable source materials have naturally provided serious grounds for argumentation.An excellent example is the debate between the natalistic and antinatalistic school over changes in fertility and mortality in sub-Saharan African societies during the precolonial and early colonial period. The fragmentary nature of the available sources has offered a firm basis for the disagreement.The objective of this paper is to discuss limitations, pitfalls, and opportunities related to sources used in African historical demography. The paper first reviews the conventional sources—population counts, censuses, and surveys—and then presents an old but seldom-used group of sources, Christian parish registers. The usability of parish registers is discussed through a concrete research project based on data produced since the late nineteenth century in the parishes of the Evangelical Lutheran Church in Namibia (ELCIN). Finally, attention is paid on widening the range of disciplines where African parish registers could be utilized.
