Merchants from abroad married into powerful local families on the Swahili coast
The Swahili coast, stretching more than 3000 kilometers from southern Ethiopia to Tanzania, was a hub of medieval trade, exporting ivory and other resources from the African interior to South Asia, the Arab world, and Persia. Its cultural legacy remains potent: Swahili is now spoken across large parts of Africa, and the ruins of ancient towns, many with mosques and other buildings cut from shoreline coral deposits, record the coast’s heyday. But whether Swahili culture was indigenous to Africa or arrived from overseas has been an ongoing debate.
One seemingly fanciful account dates from the 1500s, when Arab chroniclers recorded the stories Swahili people told about their origins. According to one version, known as the Kilwa Chronicle, seven Persian princes fleeing persecution set sail from the trading hub of Shiraz. After washing up on the coast of Africa, they founded a dynasty that ruled the Swahili coast for centuries.
An analysis of 54 genomes from people buried in Swahili coastal towns between 1250 and 1800 C.E. now gives that tale scientific support—while showing much of Swahili culture was derived from local African ancestors. The DNA of medieval people buried in elite Swahili cemeteries around 1200 C.E. shows their male forebears were closely related to people in modern-day Iran. Their female ancestors, meanwhile, were almost entirely local, with genomes resembling Bantu groups living in the region today.
University of South Florida archaeologist Chapurukha Kusimba, who led the study, published today in Nature, believes it finally resolves the mysterious history of the Swahili coast. “This long-standing question has been answered,” he says.
Gathered at seven sites in modern-day Kenya and Tanzania, the data represent the largest ancient DNA study yet from an African context. Combined with archaeological evidence from towns all along the Swahili coast and genetic evidence from people living there today, “It’s really an extraordinary piece of scholarship,” says Peter Schmidt, an archaeologist at the University of Florida who was not involved in the research.
The study does not support the simple picture that colonial-era British archaeologists favored. “The dominant paradigm was that this was a foreign civilization, with African involvement,” Schmidt says. “The idea was Persians or Arabs brought civilization with them to benighted, primitive Africans,” adds Mark Horton, an archaeologist at the Royal Agricultural University.
A postcolonial backlash posited the opposite, arguing that the medieval Swahili culture was entirely African in origin. The architecture of the “stone towns” was distinct from foreign styles, and Swahili was clearly a Bantu language, with loan words from Arabic, Persian, Portuguese, and other languages from overseas.
When Kusimba started to excavate the cemeteries of Swahili towns in the mid-1990s, what he found supported that picture. Of the artifacts he recovered, 95% were of local origin, he says, with only a few imported trade goods. But Kusimba decided to search for more direct evidence about the origins of the Swahili founders. “The people who lived and died in these towns were still there—why don’t we dig them out and examine them?” Kusimba, who is originally from Kenya, worked with local communities to excavate the human remains, analyze them, and rebury them.
The skeletons were similar in build to people buried farther inland, a clue that they were local in origin. Kusimba thought DNA extracted from the bones might tell a clearer story. But 20 years ago ancient DNA techniques were in their infancy; many researchers thought warm regions such as eastern Africa would never yield useful data because hot weather degrades genetic material. “The only way to answer these questions was to do archaeogenetics, but its time had not come,” Kusimba says.
Improvements in DNA sampling techniques and more powerful analytical capabilities changed that. In the new study, DNA from medieval cemeteries used by the Swahili elite revealed a genetic influx from Persia that was “overwhelmingly male,” according to co-author Esther Brielle, a geneticist at Harvard University. Some of the individuals derived more than 70% of their male-line ancestry from outside Africa, in contrast to their African female ancestors. “The sex bias was a surprise,” she says. “With results like that, we often think there must have been male marauders coming in for conquest.”
To collaborators familiar with Swahili culture past and present, however, a violent takeover seemed implausible. The society has been and remains matriarchal and matrilocal, with husbands moving in with their wives’ families. “Houses are owned by women, and women were the foundation of households,” says co-author Stephanie Wynne-Jones, an archaeologist at the University of York.
Seen through that lens, the genetic results look very different—and put the “Persian princes” chronicle back into play. “Archaeologists had been debating the origins of the Swahili people,” Brielle says, “and the whole time they had their own story—which, it turns out, might not be mythology.”
Merchants from Persia, the authors argue, sailed south across the Arabian Sea on monsoon breezes. After landing on the Swahili coast, they married into powerful local families before setting out to sea once more. “They would stay in the places they traded, sometimes for years,” Wynne-Jones says.
The unions were win-win: Local elites gained blood ties to far-off trading networks and the prestige of being related to people in Persia, an important center of the medieval Muslim world. Merchants, meanwhile, gained a foothold in local markets along with trusted partners to run their business during long overseas voyages. “It’s a cunning strategy,” Schmidt says. “They’re buying into existing infrastructure and networks.”
Based on the rate at which genes combined over generations, the team estimated the African-Persian mixing began by 1000 C.E. That timing suggests the stone towns themselves were a home-grown phenomenon. Recent excavations by Kusimba, Horton, and others have shown the coast’s distinctive architecture, built of carved coral blocks, evolved from wattle-and-daub buildings beginning around 700 C.E., long before the now-documented genetic influx from abroad began. “We have 300 years of Swahili civilization preceding this,” Horton says. “What we’re seeing is an event where Persians arrive into a well-formed culture or civilization and very rapidly get entangled.”
New waves of migration and settlement followed. The researchers saw an increase in Arabian-related DNA in the burials around 1500 C.E., as trade shifted from Persia and India to the Arabian Peninsula. India, too, was a source of migration to the region, contributing a small but measurable signature to the medieval DNA samples.
Today, many Swahili people have little genetic relationship with the medieval individuals in the study. Samples from almost 200 modern people who identified as Swahili showed that only those with ancestral ties to coastal towns retained large amounts of Persian ancestry. “These results highlight an important lesson,” says David Reich, a geneticist at Harvard and co-author of the study. “While we can learn about the past with genetics, it does not define identity.”
It is in the oral histories, often overlooked or neglected by modern scholars, that the Persian princes live on. “Sometimes we’ve been prone to dismiss local chronicles as made-up,” Kusimba says. “Probably we ought to take oral tradition more seriously.”