former years the number of slaves imported into the three Portuguese settlements . . . averaged from 250 to 300 per annum’. But this trade appears to have suffered a decline, so that by the late 1830s the annual import amounted to only thirty-five slaves. The small dimensions of the East African slave trade to India are confirmed by the size of the African population in Indian territories with closest contact with the East African coast. In 1837 the total African population in the province of Kathiawar was only 550, and the Portuguese port of Diu had an African population of between 200 and 225 in 1841. Various censuses and other reports for the whole Presidency of Bombay between 1848 and 1881 give an African (‘Sidi’) population of only 700 to 1,200.19
If we allow for an annual total of about 1,000 slaves from East Africa for the Red Sea and the Indian markets, we shall arrive at an aggregate of about 3,000 slaves being annually exported north from East Africa during the first half of the nineteenth century. This estimate comes fairly close to that of E. B. Martin and T.C.I. Ryan, based on slightly different data. They estimate an average of 2,500 for the period 1770–1829, and they suggest an increase to 3,500 for the 1830s. Austen’s estimate is also fairly modest for this period, giving an average of 2,250 per annum for the period 1700–1815.20
What is contentious, however, is the rising trend of the export of slaves to the north during the nineteenth century which both Martin and Ryan as well as Austen attempt to project. The former suggest a rise to an annual export of 4,000 during the 1840s, and to 6,500 during 1850–73. Austen, on the other hand, suggests an annual average of about 6,625 for the whole period 1815–75.21 I find little evidence for any increase in the prosperity of the importing region, and a lot suggesting stagnation and even decline in mercantile prosperity especially in Oman after the Napoleonic Wars. On the other hand, there is quite substantial warming up of British anti-slavery sentiments under Palmerston from the 1840s onwards, and with it the bloating of slave trade estimates. For example, the actual figure of 1,217 slaves who passed through Kharg Island in 1840–1 is successively blown up to 4,000 to 5,000. Martin and Ryan quote the British Consul in Zanzibar, C.P. Rigby, writing in 1861, giving his estimate of 4,500 for the 1840–1 period. Both Austen, and Martin and Ryan indiscriminately use all these figures in their statistical exercises.22 The use of more numerous bloated estimates, compared with the single actual count, naturally has the effect of shifting the average towards the inflated figures. Curiously, neither Martin and Ryan nor Austen take account of the evidence presented by slave captures during the fairly comprehensive anti-slavery campaign launched by the British navy in the late 1860s (see Table 6.1). What is remarkable is that the number of slaves liberated annually during three consecutive years remained fairly constant at about a thousand. These figures suggest, if anything, a declining curve for the northern slave trade compared with the early part of the nineteenth century.23
The northern slave trade, therefore, appears to have reached its plateau during the eighteenth century when the demand for slaves in the productive sector was at its peak. In the unproductive domestic and military sectors, the demand may have peaked during the commercial boom that the Omani merchant class experienced during the Napoleonic Wars at the end of the eighteenth and the beginning of the nineteenth century. With the bursting of that bubble by 1810, and the migration of a considerable section of the Omani merchant class to Zanzibar, conditions were hardly auspicious for the northern slave trade to thrive.24
The French slave trade and the re-subjugation of Kilwa, 1770–1822
The northern slave trade which had developed within the pre-capitalist modes of production in south-western Asia had a fairly limited potential for expansion beyond the dimensions attained during the eighteenth century. Communities of people of African origin are scattered along the non-African parts of the littoral of the western Indian Ocean, but they do not constitute substantial national minorities in any of the countries of this coast as they so conspicuously do in the Americas.
On the other hand, from the 1770s eastern Africa was drawn into the vortex of the genesis of capitalism, and experienced what Marx has described as ‘the rosy dawn of the era of capitalist production’.25 The development of sugar plantations in the French Mascarene islands of Ile de France (Mauritius) and Bourbon (Réunion) was an extension of similar developments in the West Indies which fed into the infamous Atlantic triangular trade. These developments sapped the vitality of East Africa’s trade with the north, initially sharpening the contradiction between the northern and the French slave trade. The northern slave trade could not ultimately withstand the tide of nascent capitalism, and the Arab merchant class therefore had to accept the new compradorial role being assigned to them in the new global capitalist system.
French demand for slaves was modest in the first decades after the 1730s and was largely met by Madagascar, and intermittently by Mozambique and the East African coast. It was not until the 1770s that the supply from the African coast reached significant proportions. However, with the decreasing profitability of the West African slave trade a tremendous demand for slaves was added in the 1770s and 1780s for the American market as well. That market became even more attractive as the small Mascarene islands became ‘so well stocked with blacks’ by the mid-1770s. An annual average of about 3,000 slaves was then being traded by the French from Mozambique.26
The rapidly rising French demand for slaves began to cause a major dislocation of the Arab trade along the East African coast. Previously, during the middle of the eighteenth century, Omani demand for slaves had been responsible for the penetration of Swahili traders of Mombasa and Pate into the ports of southern Tanzania, and they even encroached on the Portuguese sphere along the northern coast of Mozambique. By the mid-1770s this northward flow of slaves had been partly reversed. It was stated that Swahili traders were taking slaves to the south ‘when they do not have to sell to the Arabs’.27
The French also began to encroach directly into the Omani sphere north of Cape Delgado. Mongalo, or Mgao Mwanya along the southern coast of Tanzania, was located close to the Makonde plateau where, it was alleged, the Makua, Makonde, Ndonde and Yao were ‘continually at war, solely to make each other prisoners, whom they sell’. By the 1750s it had become a regular port of call for Swahili traders in search of ivory and slaves.28 To secure their supply of slaves, the Omanis had imposed their suzerainty over Mongalo in 1776, although they had no permanent political representative there, and no tribute was exacted. The Omanis were more interested in economic control over the trade which was channelled through Kilwa. They were seasonal traders who deposited their merchandise with the local ruler who collected slaves and ivory for them while they continued their voyage to Mozambique or Zanzibar according to season. The French sought to encroach upon this Omani sphere through various schemes to establish a factory there by forming an alliance with the local ruler. In 1786 an agreement was concluded with local authorities to supply slaves at 25 piastres each, a price far below that prevailing elsewhere along the coast. But in view of the acute demand, this low price could not be maintained, and the following year the price was reported to be equal to that at Kilwa, though the tax was kept to 3 piastres to attract French traders there. However, Mongalo had a fairly restricted hinterland, able to supply a smaller number of slaves than Kilwa. Most of the French schemes, however, failed to attract French governmental support and appear to have all fizzled out.29
A more serious encroachment was attempted at Kilwa which was ‘the entrepôt of the slave trade for all the coast of Zanzibar.’ The French slave trader, Morice, had made two voyages to Zanzibar in 1775, taking off 1,625 slaves, most of whom probably came from Kilwa. The following year he shifted his trade to the source, buying 700 slaves, and negotiating a 100-year treaty with the Sultan of Kilwa to buy 1,000 slaves a year at a fixed price of 22 piastres each, including tax, and regardless of sex.30
This was not merely a commercial transaction but also a conscious entry into the local politico-economic struggle along the East African coast between the various Swahili polities attempting to maintain their independence, and the expanding Omani hegemony. Previously the Omanis had exercised a rather loose control over the Swahili
