Madagascar, Polynesia and colonisation against the flow

by Edward Pegler on 26 April, 2010

What the colonisation of Polynesia says about open ocean colonisations and particularly what it suggests about the Madagascar colonisation by South-east Asians.

people rowing an outrigger canoe with a view of the vast pacific  behind them

The Polynesian sailors who colonised the islands of the Pacific from 1000BC onward didn’t just have huge distances to tackle as they journeyed east. They were also sailing against both the ocean currents and the wind. This seems daft. Why not go the other way? But in Geoffrey Irwin’s 1992 book ‘The Prehistoric Exploration and Colonisation of the Pacific’ he shows why this is a good strategy. I’ll explain.

Rules of cycling for the lazy

As a bit of a pedal cyclist, I sometimes like to take a day trip out from Swindon to one of the nearby towns, such as Devizes or Marlborough, some 15 to 20 miles away. Generally, I find it’s best to pick a day when the wind is against you on the outward journey.

Although this is hard work, the cycle home is relatively straightforward due to my being blown home by the wind. Or if I find I’m tired on the way out I can just give up, turn round and catch the wind home again

Two Polynesian scenarios

It must have been similar with those Polynesian sailors, hoping to find yet another island out there. On setting out they must have been at their peak of fitness , boats packed with breadfruit, animals and any other provisions. Perhaps waiting for a lull in the wind, they would set off, rowing or tacking their way out toward a rolling eastern horizon full of promise.

Days or even weeks of working their way steadily east would take their toll on the sailors. Short of food and energy, they would be desperately hoping to sight land. But if no land appeared it was time to cut their losses. The boat would be turned around and guided into the strong ocean currents. The sails would be unfurled to catch the wind. It was the nearest they were going to get to a free ride home.

But play the scenario in reverse, with the boats starting out westward, and you have a potentially disastrous story. The sailors would journey comfortably west for hundreds or thousands of miles. However, after weeks at sea, with supplies low and no island sited, the crew would reluctantly turn the boat around and start to head home… and into the teeth of the wind. They would never be seen again.

Europeans finding America

This is not just true for Polynesian sailors, however. Before the advent, in late medieval Europe, of fast ships such as caravels and carracks, which could sail close to the wind and, in the case of the carrack, had massive storage space, everyone in the world would have been governed by the same practical rules.

To take one example, ancient Britons, as far as anyone can tell, did not invade or colonise Norway and Denmark, yet the wind was generally with them. Conversely there is plenty of evidence, both historically and genetically, for the repeated colonisation of Britain by peoples from Norway and Denmark, people who often had to fight the prevailing westerly winds all the way.

This example can be extended further. The first European people to reach the Americas were, again, Norse sailors. This they did against the prevailing north Atlantic westerlies, setting up bases wherever they could along the way. But Iberian sailors, who should have simply been able to catch the north-easterly trade winds to reach the Americas, apparently did not until Columbus’ crew reached central America in caravels and a carrack.

I say believed because it is just possible that some Iberians may have reached America before Columbus’ crews. However, we would never know because none ever came back. This is because on the return journey they would have faced fierce winds and there would have been few places to stop.

The colonisation of Madagascar

I believe all this has interesting implications for another great story, the colonisation of the island of Madagascar, off the East African coast, by speakers of a Malay language from south-east Asia (probably Borneo). To this day, the people of Madagascar speak a Malay language, Malagasy.

Archaeological, linguistic and genetic evidence from Madagascar implies that people, who we’ll call ‘Malays’, first reached Madagascar only during the mid first millennium AD (perhaps as early as the first century AD but with significant expansion around 800 AD). Linguistic evidence also indicates that they had lost contact with their homelands by the thirteenth century at latest.

This colonisation is sometimes presented in books as a simple case of the Malay sailors getting into their boats, unfurling their sails and catching the ocean currents and winds across the Indian Ocean straight to Madagascar. There, so the story goes, the sailors, stranded by unfavourable winds, ended up their days isolated from their South-east Asian homeland.

The Polynesian alternative

But the Polynesian experience suggests that the scenario above is unlikely. Far more likely is that these Malay sailors fought their way, against the wind, to India then continued on, using the switching monsoon winds, to the horn of Africa and southward to Madagascar.

This kind of journey, like that of the Polynesian sailors, is more likely to be the work of several generations. There would need to have been settlements of Malays along the way, in southern India, in southern Arabia perhaps, on the Horn of Africa and along Africa’s east coast.

In this scenario, those same Madagascar colonisers would be able to get home to Borneo by letting the wind take them back along the same routes they fought their way out along. This has all the hallmarks of the later Indian Ocean trading patterns of the Muslim world.

However, no archaeological evidence of these Malay colonising settlements has so far turned up around the margins of the Indian Ocean. In fact there is no current genetic evidence and little linguistic evidence either. So if Malays did set up settlements around the Indian Ocean these settlements have utterly vanished now.

Additionally, the absence of towns and the poor quality of pottery produced by the early settlers to Madagascar indicates that they themselves were not part of any ocean-wide trading system. So perhaps the Polynesian method is wrong.

But what about bananas?

But there are small shreds of evidence for the colonisation of East Africa and South India by South-east Asians. Bananas, yams and other crops of South-east Asian origin are widespread in tropical Africa and southern India today and have long been thought to have arrived before European influence.

Banana remains (phytoliths) from Nkang, Cameroon in West Africa, dated to the mid first millennium BC, indicate some south-east Asian interaction with coastal East Africa dating back to perhaps 1000 BC. Other evidence from Uganda even pushes back the dates of introduction to before 3000 BC and there is unconfirmed evidence taking the cultivation of bananas in the Harappan civilisation of Pakistan to between 2000 and 3000 BC.

So perhaps ‘Malay’ sailors arrival on Madagascar in the first millennium AD was the final, and rather ignominious end to well over a thousand years of involvement of South-east Asians in Indian Ocean colonisation and trade.

Maybe the fact that the Malagasy language seems to have been cut off from other Malay languages in the first millennium was because that was indeed when they were cut off, effectively when all the other ‘Malay’ colonisers around the fringe of the Indian Ocean packed their bags and went home. If such a story is real it’s largely lost to history.


Fuller, D.Q. and Madella, M. 2009 Banana Cultivation in South Asia and East Asia: A review of the evidence from archaeology and linguistics. Ethnobotany 7, 333-351.

Irwin, G. 1992 The Prehistoric Exploration and Colonisation of the Pacific. Cambridge, pp260.

Lejjua, B.J., Robertshaw, P., Taylor, D. 2006 Africa’s earliest bananas? Journal of Archaeological Science 33, 102-113.

Mbida, Ch. et. al. 2006 Phytolith Evidence for the Early Presence of Domesticated Banana (Musa) in Africa, 68-81 in Zeder, M. A. 2006 Documenting domestication: new genetic and archaeological paradigms, California, pp375.

Mitchell, P. 2005 African Connections: Archaeological Perspectives on Africa and the Wider World, AltaMira, pp307.

Photo: / CC BY-NC-ND 2.0

Additional References

Gunn, B.F., Baudouin, L. & Olsen, K.M. 2011 Independent Origins of Cultivated Coconut (Cocos nucifera L.) in the Old World Tropics, PLoS One (online)

Provides an interesting and persuasive counter to the argument above, based on the occurrence of Pacific Ocean coconut genetics in Madagascar. If the post above was right, the genetics of the coconuts would more likely be Indian Ocean-type, as they are for East Africa. However, the claim in the paper that the trade route had to be south of the Seychelles, rather than through it, doesn’t consider the possibility that the trade route went completely west of the Seychelles. One to think about, I suspect.

unread reference for future checking

Allibert, C. 2008 Austronesian migration and the establishment of the Malagasy civilisation: Contrasted readings in linguistics, archaeology, genetics and cultural anthropology, Diogenes55 p7-16.

{ 4 comments… read them below or add one }

InterestedInHistory August 3, 2013 at 7:01 am

The presence of bananas and other south-east asian fruits and vegetables in India does not automatically mean polynesian sailors were bringing them there.

Indian merchants and traders had been visiting South-East asia for millenia, owing to favourable trade winds linked to the monsoons. The directions of the winds change with the time of the year, making it convenient to travel back-and-forth. The same winds also brought chinese traders to South East Asia, making it a trading point for many different cultures.

These Indian merchants, followed later by Arab traders who regularly visited India in a cross-Arabian Sea trade, could easily have brought back bananas and other such south-east asian fruits and vegetables back.


Edward Pegler August 11, 2013 at 4:07 pm

Dear X

This appears to be a reasonable point.



Dr Luc Baudouin July 25, 2011 at 3:02 pm

Firstly I would like to make clear that the arrows on the map of the Gunn Baudouin Olsen paper are not meant to be realistic nautical routes. We have no competence in this domain. They simply connect introduced population to their source. The fact is that the genes of human populations in Madagascar associate Austonesian and Bantu genes but no Indian contribution.

As a cyclist I understand your point very well (and I do the same). But I would make a difference between exploratory travels and trade route. In the first case, it is a good idea *not* to take the easiest route. But once trade and/or migration route are established, it is more efficient to choose the right season and the right latitude to secure the easiest journey in both direction. The austronesians who colonized Madagascar in the 6th century and later in large number were not the first ones to visit the region. It is thought that there were already Austronesian mearchants in East Africa 2000 years ago. Maybe they navigated along the coasts but migration didn’t begin until a short and direct route was found.


Edward Pegler August 3, 2011 at 10:28 am

Dear Dr Baudouin

My apologies for misreading your paper. I think, as you rightly point out, the evidence of lack of genetic admixture in the Madagascar populations is an important factor, suggesting a direct sea link without a south Asian connection, at least toward the latter part of the period in question. Could you point me toward a paper that I can reference for that?


Leave a Comment

Previous post:

Next post: