The slave trade in the Indian Ocean^201the comparative profitability of free and slave labour. Such studies were
conducted in the West Indies as well as in the Mascarene Islands and often
show slave labour to be the less economical. However, many assiduous inquiries
are still needed before it can be proved that the slave's output was 'unecono-
mical '. Having read the work by R. W. Fogel and S. L. Engerman,^7 one would be
tempted today, on the contrary, to look into the profitable aspects of slavery,
even in the mid-nineteenth century. Another point that needs clarification
has to do with the financing of the slave trade, and the profits made by the
exporter and the carrier. Summary evaluations do exist. I quoted Misson's
sally about the price of slaves in Madagascar and the value that was put upon
them in Barbados. In the mid-nineteenth century, a slave exchanged for two
goats near Lake Tanganyika was sold for $20 in Zanzibar and over $60 in
Muscat. These two towns, which were relay points along the slave routes and
centres for financing and redistribution, owed their wealth to the slave traffic.
The constitution of commercial networks, the profit-and-loss balance, and the
share of the slave trade in commerce as a whole are studies which still need to
be undertaken in most of the regions in the Indian Ocean. A remaining task
would be to study the slaves' descendants—freed slaves or fully fledged
citizens. This is not only a demographic problem but also one of integration in
society or rejection by it. In what generation is restitution made for the rob-
bery of Africa? At what point is a balance struck between what the slaves'
descendants have received from the host country and what they have given it?
The major difficulty is still that of counting the slaves. This is the key
both to an accurate answer regarding the enrichment of external economic
systems and to an evaluation of the demographic consequences for Africa.
E. A. Alpers has criticized Coupland's views on the connections between the
slave trade and the low population density in East Africa (1967). The matter
does not seem to have been fully settled. While plenty of estimates have been
made for the eastern part of the continent from the middle of the eighteenth
century onwards, there are virtually none for the previous periods. The contri-
butions made by the slave trade in the islands, many of which were 'population
traps ', are beginning to be assessed more accurately. But losses in the trans-
portation over land and sea are often still unknown and many generalizations
seem to be rash. Internal displacements—a sort of slave trade in a vacuum—are
still certainly being minimized, as are the volume and duration of the clan-
destine traffic in the European colonies and Muslim countries. It is on these
last points that the overall evaluations made by Curtin and Deschamps will
probably have to be modified most.
A great deal is expected from G. I. Inokori's work on population and the
impact of the slave trade on societies and powers in Africa. In the eastern
part of the continent, I would suggest only a few avenues of research : up to
about 1750, the political and social balance seems to have been very little