Read our COVID-19 research and news.

Rice paddies

Malagasy rice fields hint at the Asian origin of some of the island's early settlers.

Image courtesy of Alison Crowther

Ancient crop remains record epic migration to Madagascar

The settlement of the Indian Ocean’s largest island is one of the great mysteries in humanity’s colonization of the globe. Madagascar lies just 400 kilometers off the East African coast. Yet the Malagasy people’s cuisine, rituals, and religious beliefs resemble those of Borneo, some 9000 kilometers to the east. Their language is more closely related to Hawaiian than to Bantu, and about half their genes can be traced to Austronesia—that is, Indonesia and the islands of the Pacific. Archaeological evidence of this distant connection was lacking, however.

Now, new studies—recent or soon-to-be-published—trace a wave of Austronesian colonization between 700 C.E. and 1200 C.E. The telltale evidence is, in effect, breadcrumbs: crops distinctive to Austronesia, sprinkled across Madagascar and neighboring islands. “We finally have a signal of this Austronesian expansion,” said Nicole Boivin, an archaeologist and director of the Max Planck Institute for the Science of Human History in Jena, Germany, who discussed the findings at the recent Society for American Archaeology (SAA) meeting here.

The study by Boivin and her colleagues, published today in the Proceedings of the National Academy of Sciences, found that these voyagers did not stop at Madagascar. Some also settled the Comoro Islands, scattered between Madagascar and the African coast. “The discovery of an Austronesian connection for the Comoros is surprising,” says David Burney, a paleobiologist at the National Tropical Botanical Garden in Koloa, Hawaii, who has studied the region. Yet the Austronesians stopped short of the African coast. “There was a culinary frontier,” says Alison Crowther, an archaeologist at the University of Queensland, St. Lucia, in Brisbane, Australia, who led the study.

Her team collected more than 2400 samples of botanical remains at 20 sites on the African mainland and on offshore islands and Madagascar, and obtained 43 radiocarbon dates from crop seeds. Between 700 C.E. and 1200 C.E., the researchers found a clear boundary between sites dominated by African crops like pearl millet, cowpea, and sorghum, and those with Asian rice, mung bean, and cotton. The Asian crops were common on the Comoros and on Madagascar, but rare on the East African coast.

A. Cuadra/Science; Adapted from Crowther et al., PNAS

The line traced by the study shows that the two regions, although close geographically, were worlds apart in their way of life, suggesting a wholesale colonization of Madagascar and the Comoros. A thousand years ago and more, Arab and Indian sailors conducted a bustling trade between East Africa to India. But the crops indicate that the settlers came from even farther east. Although Asian rice and mung bean are common on the Indian subcontinent, other common Indian crops like horse gram and urd (two legume varieties) are absent from the Madagascar and Comoros samples.

The genetic studies that support the group’s conclusion about an Asian colonization also link roughly half the genome in modern Malagasy to Africans. When the Africans arrived is a further mystery. “People have speculated that there might well have been a number of transitory arrivals—most likely from Africa—before settlement on the island really took off” with the arrival of Austronesians, says Peter Forster, an archaeologist at the University of Cambridge in the United Kingdom.

The late Yale University archaeologist Robert Dewar claimed in 2013 that people reached the island around 2000 B.C.E., millennia earlier than had been thought, based on radiocarbon dates of organic matter found with stone artifacts in a rock shelter on the north coast. But even members of Dewar’s team were not ready to rewrite Madagascar’s history. Burney, meanwhile, says that environmental data, such as signs of widespread burning, suggest the first humans more likely arrived around 400 B.C.E.

Based on studies of the unique megafauna Madagascar once hosted, including giant flightless birds and huge lemurs, paleoecologist Simon Haberle of the Australian National University in Canberra argues that the arrival was likely more recent. He reported at the SAA meeting that radiocarbon dating of fungi from the dung of megafauna at five sites in southwest Madagascar suggests that larger animals, like a 150-kilogram lemur, began to decline about 500 C.E., presumably because human hunters had reached the island by then. Megafauna extinctions gathered speed between 700 C.E. and 1000 C.E., coinciding with the wave of Austronesians.

The identity of the earlier settlers, whenever they made landfall, remains obscure. Madagascar still has hunter-gatherers such as the Mikea, who have an oral tradition asserting that they were the island’s original inhabitants, driven by later migrants into the dense forests of the island’s southwest. A 2013 study, however, found that the group is genetically similar to the Malagasy: a mixture of African Bantu and Austronesian stock. The team suggested that the Mikea were farmers and became hunter-gatherers, rather than a remnant of a pre-Austronesian wave of African colonists.

The latest evidence leaves little doubt about Austronesian settlement, but their predecessors remain in the shadows. “What is clear is that the island has a complex settlement history involving multiple colonizations by different populations at different times,” Crowther says. Her new study, she adds, contributes “a small piece to the puzzle.”