These remote and otherwise quite obscure islands produced a good which was widely traded all over the ocean and beyond to Africa and China. There was an extensive trade in them to Yunnan from the ninth century, some done overland and some by sea via eastern India and southeast Asia, where in both places they were much in demand as currency. Again Ibn Battuta's account is revealing. The inhabitants of the Maldives sold them for the common currency, the dinar, or in exchange for rice from Bengal, or to Yemenites who used them instead of sand for ballast. Later in his travels he found them in Mali and Gao in West Africa.
75
The trade in slaves represents another extensive and high-value item of exchange. This trade, using 'product' from East Africa, began in earnest in the eighth and ninth centuries, though Zanj, that is African, slaves are first mentioned in Sassanian Persia, shortly before Islam in the early seventh century. The most important trade was from East Africa to the Abbasid capital of Baghdad from the eighth to the tenth centuries, where they were used to perform backbreaking work draining and controlling the marshes south of Baghdad in the Tigris-Euphrates delta. The trade expanded greatly, until the huge slave revolt of 868–883, which contributed to the decline of the Abbasid empire. However, the trade to the Middle East continued.
Other African slaves were found even further afield. Habshis, that is a corruption of the Arabic Habash, or Ethiopian, were being sent to India at least from the early thirteenth century, while Arab traders brought them to China in Tang and Sung times. Slave trading was widespread in southeast Asia also, though here using local people rather than those from far-away Africa.
76
We have been writing about luxury long-distance trade, and indeed this was important, but it was by no means dominant. As one example, much has been made of finds of Chinese porcelain in various Middle East sites in the period from the eighth to the fourteenth centuries. Yet these products from far away are very minor, representing less than 1 per cent of total finds. Nor do these rare finds show that Chinese traders came to the Middle East in any numbers: rather, the porcelain took part in the relay trade from port to port, proceeding by stages and passing through many hands.
77
These relay trades were very complex indeed, linking small production and exchange locations with the great port cities. One could study this theme in any area around the ocean, but East Africa can stand as a type case, to give some impression of the complexity of items traded, and of the cosmopolitan trading community in Malindi. The Muslim inhabitants
are great barterers, and deal in cloth, gold, ivory, and divers other wares with the Moors and Heathen of the great kingdom of Cambaya; and to their haven come every year many ships with cargoes of merchandise,
from which they get great store of gold, ivory and wax. In this traffic, the Cambay merchants make great profits and thus, on one side and the other, they earn much money. There is great plenty of food in this city, rice, millet, and some wheat which they bring from Cambaya.
78
Porcelain, precious stones, spices are the sorts of trade items which have left records or remains behind. However, much more basic things were traded. The two most essential were food and water. In the case of Hurmuz, Ibn Battuta wrote that 'On this island [of Jarun] water is an article of price; it has water-springs and artificial cisterns in which rain-water is collected, at some distance from the city. The inhabitants go there with waterskins, which they fill and carry on their backs to the sea [shore], load them on boats, and bring them to the city.'
79
Mozambique similarly had to 'import' its water.
There was a very extensive trade in foodstuffs, especially rice. Several of the great port cities produced almost no food for themselves. Melaka and Hurmuz in the fifteenth century both had very large populations, maybe up to 50,000. It is revealing that in the former there was no land tax, such was the lack of significance of agriculture. All of Melaka's rice came from Pegu, Java and Siam.
80
Hurmuz got its necessities from far afield: rice from Chaul and other places, grain from the Punjab via Sind, and grain also from the Persian mainland. Rope, iron and coconuts came from Kerala, wood from East Africa.
81
Aceh got supplies from Pegu, Bengal, Arakan and Sumatra. From the great inland state of Vijayanagar rice was exported to the coast, to Sri Lanka and the Gulf. Bengal and Pegu supplied rice to western Indo-China, Sumatra, Sri Lanka and the Maldives. There was even an extensive exchange of new varieties of food crops. African types of millet went to India, and southeast Asian crops like sativa rice and bananas to East Africa. A final necessity item is mangrove poles from East Africa. Mangrove is a very hard, dense wood, heavier than water and termite-resistant. It has always been cut to a standard length of 2.6 metres. It has been used in building in the Red Sea, southern Arabia and the Gulf from the tenth century: the Lamu area and the Rufiji delta were vast lumber yards.
82
Some products were traded over very long distances indeed. In the thirteenth century date honey was produced in Bahrain and was much in demand in China by Buddhist pilgrims travelling to India. The great Chinese admiral Zheng He brought back to Beijing several giraffes, including one from Malindi and one from Bengal, the latter having apparently been given to the ruler of Bengal, Saifu'd-Din, by the ruler of Malindi.
83
Most extraordinary, and mysterious, was the discovery in 1944 by an Australian radar team of five Islamic copper coins from Kilwa on a beach in the remote Marchinbar Islands, part of the Wessell Islands off Australia's Northern Territory coast. None have dates, but from the inscriptions two may be tenth century, and three early fourteenth. We have no idea how they managed to travel clear across the whole Indian Ocean.
84
Long-distance trade was governed by the monsoons. One example was a route from the Gulf region to China around 1000 on the longest voyage sailed by any one
ship. The Arab geographers claimed that a passage from Oman to China took about three months and ten days, though one exceptional voyage was completed in 48 days. These sound extraordinarily rapid, but they are only sailing times. Several stops were necessary on the way, partly to trade, and partly to wait for the right monsoon, so that the actual time from leaving the Gulf to reaching Guangzhou (Canton) was at least six months. The dhows sailed down the Gulf before it became too rough, in September or October, and then went on to Malabar on the northeast monsoon, arriving in mid December. They stayed there while they traded, and waited for the cyclone season in the Bay of Bengal to end. In January they sailed to Malaya, and used the last of the northeast monsoon to get around the straits of Melaka and so catch the southern monsoon in the South China Sea and reach Guangzhou in April or May. The return voyage began in October to December when the northeast monsoon took them back to Melaka and over the Bay of Bengal to the west coast of India. The last stage, back to the Gulf, was sailed using the beginning of the southwest monsoon, reaching home around mid year.
85
Another example comes from five hundred years later. Very early in the sixteenth century Barbosa left us a compelling description of one of the major long-distance trade routes of this period, that is from Malabar, specifically Calicut, to the Red Sea. He wrote that the Muslim traders in Calicut from the Red Sea and Egypt:
took on board goods for every place, and every monsoon ten or fifteen of these ships sailed for the Red Sea, Aden and Meca, where they sold their goods at a profit, some to the Merchants of Juda, who took them on thence in small vessels to Toro, and from Toro they would go to Cairo, and from Cairo to Alexandria, and thence to Venice, whence they came to our regions. These goods were pepper (great store), ginger, cinnamon, cardamons, myrobalans, tamarinds, canafistula, precious stones of every kind, seed pearls, musk, ambergris, rhubarb, aloes-wood, great store of cotton cloths, porcelains, and some of them took on at Juda copper, quicksilver, vermilion, coral, saffron, coloured velvets, rosewater, knives, coloured camlets, gold, silver, and many other things which they brought back for sale in Calecut. They started in February, and returned from the middle of August up to the middle of October of the same year. In this trade they became extremely wealthy. And on their return voyages they would bring with them other foreign merchants who settled in the city, beginning to build ships and to trade, on which the King received heavy duties.
86
These two accounts point to a major change in the structure of long-distance sea trade in our period. Barbosa was describing a trade divided at south India, while the first account sketched a direct passage from the Gulf to China. What happened is that around the eleventh century the trade became segmented, with one merchant and ship doing the Arabian Sea part to south India, where the goods were exchanged, and then taken on by other ships and merchants to southeast Asia, where there was another exchange, and so to China. South India was always a place
where there was a halt, and exchange, but the difference is that in the earlier time the same merchant and ship kept going beyond there, while later they did not.
In the earlier period, from say the eighth century, the very long distance trade from the Gulf to China was handled by Persian merchants. In the Gulf Siraf, on the east bank, was the main centre, where were to be found goods from all over the Indian Ocean, including East Africa. Later Julfar, on the west coast up from Hurmuz, was important, and later still Hurmuz. Another old centre was Daybul, in present day Pakistan. Arabs also took part in this trade, and soon became more important than the Persians. Later some Chinese ships also, from the twelfth century and particularly in the fourteenth, traded into the Arabian Sea. However, from around the eleventh century the direct passage from Baghdad to Guangzhou declined, and we see the rise of emporia, that is shorter routes connecting the major port cities of Baghdad, Hurmuz, Cambay, Calicut, Melaka and Guangzhou, with many minor routes from, say, the Bay of Bengal feeding into this network. What evolved then was a basic change in the orientation of long-distance trade, which in the earlier period was on an east–west axis, from Baghdad to Guangzhou, and later was more north–south, that is Baghdad down to India, then an east–west segment to southeast Asia, and then north–south again up to China. We can even see here an early version of today's divide between north and south, for the north, India and China, provided manufactures like cloths and porcelain, and the south unprocessed tropical products such as ivory, slaves, gold and spices.
From the twelfth century or slightly later we have three segments: the Arabian Sea, the Bay of Bengal, and the South China Sea. Chinese and Indians went to Melaka, Persians and Arabs only to India. It is significant that the account by Wang Dayuan, who travelled extensively in the 1330s, finds a western ocean and an eastern one, with the division at the Straits of Singapore.
87
This important move towards segmentation may have been a result of traders realising that the direct passage in the same ship was inefficient, given that they had to wait for monsoons at several places, but it was probably also a result of the rise of important Indian trading communities in south India associated with the powerful Cola dynasty. We will turn to the influence of politics on trade presently, but we can remember here that the wealth and stability of the Abbasid empire from 750 CE, and of T'ang China, 618–907, certainly fostered this long-distance and quite perilous trade. The effects of the rise of the Cola empire in South India from the late ninth century has been less investigated, but it may be that the Colas, and the powerful merchant organisations, akin to guilds and associated intricately with state power, had two results. First, the stability provided by this state had the same effect as the equivalent in Baghdad and Guangzhou, that is a wealthy and stable state which had a large demand for foreign luxuries, and second merchants based in this state could trade both east and west, and especially to the east, to southeast Asia, where they met up with the powerful Sumatran-based trading empire of Srivijaya, which benefited from controlling the Melaka Straits up to the thirteenth century. South India seems to act as a fulcrum in this very long-distance connection. Later in our period other Indians joined in, this time Muslims based in the many emporia on the west coast, and in the
major Islamic state of Gujarat from the thirteenth century. Increasingly the trade beyond India was controlled by Indian Muslims, while Arabs, and a few Persians, were restricted to the Arabian Sea.
We can start our survey of routes, trade and ports in the east, in China. We have noticed that Chinese products, especially porcelain, were traded all around the Indian Ocean from very early times. We have already quoted Ibn Battuta's valuable description of the ships he saw in Calicut (see pages 70–1). His account dates from the early fourteenth century, but Chinese products have been found in the Arabian Sea from much earlier. Chinese pottery has been found on the Swahili coast from at least the eighth century, and a little later in Mauritius also. These goods were transshipped many times in a relay fashion, and some no doubt came overland to the Gulf and then were sent on by sea. An actual Chinese trading presence seems to date only from the twelfth century.
Many of the vast Chinese ships had both economic and political functions. We refer to the famous tribute system. Ostensibly this was a matter of foreign rulers accepting the superiority of the Chinese emperor, and sending tribute to signify this. However, much of the tribute was actually trade items, and the system then was a method of fostering exchange as much as a matter of political dominance. In the later thirteenth century the new Mongol dynasty, the Yuan, was keen to expand trade. In 1286 either the sons or younger brothers of the rulers of ten kingdoms ranging from Malabar to Sumatra came to pay tribute.
88
Marco Polo got part of the way back home accompanying one of these politico-trade missions. Around 1290 a Mongol princess was sent by sea to Persia to become the consort of the local ruler, Arghun Khan, and the Polos went with her. She travelled with 600 sailors and officials, in a fleet of fourteen ships. They left from Zaiton, of which more in a minute, and touched at Champa and the Malay peninsula. Reaching Sumatra, they were forced to wait for five months to avoid monsoon storms. They then travelled near the Nicobar Islands to Sri Lanka, the west coast of India, and so to Hurmuz. However, Arghun had died by this time, and the princess was handed over to his son, Mahmud Ghazan, instead. This sort of voyage has been described in Chinese sources also. They said that it took forty days to get from China to Sumatra. One spent the 'winter' there and then took thirty days to get to the Malabar coast.
89
This information again points to the good sense of the rise of the emporia trade, which meant that ships travelled shorter distances and did not have to wait for a change in the monsoon. Rather they could sell their goods and return home.