What's the origin of the long-ago Swahili civilization? Genes offer a revealing answer

Chapurukha Kusimba was a young boy in Kenya in the 1960s just as many African nations were attaining independence from western European countries. The power and success of African nationalists impressed him. So did the archaeological discoveries of ancient humans by the Leakey family showing, as he saw it, that "to be human is to be African."

But he was confused by what he was learning in school. "There seemed to be very few incidences in history where African or Black people had actually attained anything or contributed much to global civilization," he says. "Something wasn't quite right."

Kusimba grew up in western Kenya and is not Swahili, but he became fascinated by the thousand-year-old culture of the Swahili people, who live along the coast of eastern Africa and early on built cosmopolitan city-states. Through his 40-year career as an archaeological anthropologist, including stints at museums in the U.S. and Kenya, professorships at several U.S. universities (he is now with the University of South Florida) and authorship of a book, The Rise and Fall of Swahili States, the question dogged him: Were the earliest Swahili from the African continent, or did they migrate from somewhere else?

And now he's the senior author of a research paper published in March in Nature that has the answer.

The research shows that the Swahili civilization British colonizers encountered was not primarily foreign, as they believed. Nor was it primarily African, as has been more recently thought. The study uses DNA analysis to show that it's a mix, as Swahili oral tradition has said for centuries.

In addition, Kusimba and his colleagues found that the ancestors of today's Swahili people were African women and foreign men. As they note in their study, ancestry narratives for the eastern African coast have a complex history, and the genetic findings add to the complexity.

The politics of ancestry

When British colonizers came into eastern Africa several hundred years ago, they credited the origins of Swahili civilization to foreign traders from India and Persia. "The good was always attributed to non-Africans," says Kusimba. Colonial archaeologists thought Africans lacked the initiative and agency to build Swahili culture. "And colonial archeology was very powerful," says Kusimba. "Any views that were divergent to what was the then-mainstream view of the Swahili would either be silenced by not getting published, or people just wouldn't believe it." Some nationalist politicians emphasized a foreign origin to whip up anti-Swahili sentiment and consolidate power.

But all along, Swahili people knew they were a mix. Their origin stories told them so.

To research the beginnings of Swahili people, Kusimba began working with an American biological archaeologist on bones from archaeological digs along the Swahili coast. The shape and size of bones can hint at origins. Wider bones in the nasal area for example, can suggest African heritage. The bones they were examining looked more African than Asian or European.

Other archaeologists and historians were coming to believe that the original Swahili people were mostly African because Kiswahili, the Swahili language, clearly has its origins in the Bantu languages known to be native to Africa.

The American archaeologist thought genetic science might help sort things out. So did Kusimba. He contacted geneticist David Reich at Harvard, an expert in ancient genes, who agreed to help.

What the genes reveal

Kusimba and colleagues in east Africa collected bones and teeth from seven burial sites, checking with local religious and community leaders for permission, and sent the samples off to Harvard. Once the bones arrived in Boston, the genetic work had its own complexities. The work had to be done in special "clean" labs to prevent contamination with modern DNA. And the group at Harvard couldn't get a clear read on all the samples, because some were degraded.

Ultimately, they were able to analyze and compare 80 samples from people buried between 1250 and 1800 C.E. in sites found mostly along the coast in modern-day Kenya and Tanzania.

They looked at bits of DNA known as single-nucleotide polymorphisms, DNA that varies from population to population.

The genetic analysis showed the mixing started out around 1000 C.E., about the same time that Islam, a hallmark of Swahili life, became widespread in the area. The new genes initially were Persian (essentially modern-day Iran), and eventually, Indian and southern Arabian.

Esther Brielle, a postdoctoral fellow in Reich's lab at the Department of Human Evolutionary Biology at Harvard, worked on the genetics. She and others calculated that from the people they studied (whom they surmise were elite because they were buried in cemeteries near the main mosques) the male ancestors of elite Swahili people were a mix of approximately 83% south Asian (about 90% of that Persian, and the rest Indian) and 17% African. The female ancestors were less of a mix – appearing to be 97% African, and 3% south Asian. (Female analyses were based on DNA from mitochondria, that are passed down by the mother, and for males it was DNA from Y chromosomes, which come from the father.)

Brielle loved working on the project. Oral tradition as well as some writings told of Persian princes coming across the Indian Ocean. "It was very satisfying to get results that verified what they say about their own history," she says. It was also a chance to even out a scientific imbalance. "Everyone focuses on European genetics and I think that we owe it to other places in the world to focus on them," she says.

Kusimba, for one, was pleased that the results backed up what Swahili people thought. "I have no reason not to believe in the authenticity of oral tradition as a legitimate source of historical knowledge," he says.

"This work is really exciting," says Peter Schmidt, an emeritus professor of anthropology at the University of Florida who was not involved in the Nature study. "It's the finest example thus far of how DNA can help to correct and rewrite African history." And, he says, it's a chance to reflect on how scholarship sometimes swings like a pendulum to satisfy social and political needs.

But what about that male/female imbalance – that in male Swahili forbearers, foreign genes predominated, and Swahili women were mostly African? In the U.S., white male genes in African Americans are markers of sexual exploitation of female slaves. Were African women raped by foreign invaders? Probably not, says Kusimba. The local culture was (and is) matriarchal — women hold the power. "The men likely weren't predators," he says. They came with all sorts of coveted goods, like spices and cotton from India, and silk and porcelain from China.

Foreign traders, it turns out, were probably marrying into powerful women-led families. "In this small sample that we have, we have a situation whereby African women are opportunistically making a decision that is going to be good for their own families," says Kusimba.

What's next for Swahili ancient DNA research? "The rich and famous often have the last say, because their remains are more often preserved," says Kusimba. Maybe not this time. He's now in Kenya, picking up artifacts and human remains he excavated in the mid-1990s. The remains come from common graves, and could show whether the Persian-African mix held for non-elite ancient Swahili people as well. They'll be incorporated into the Harvard study and Brielle is tremendously pleased. "We're trying to understand things that people want to know," she says.

Joanne Silberner is a freelance journalist and former health policy correspondent for NPR. She has covered global health issues since the outbreak of HIV.