Kelly Cheng Travel Photography / Getty

Earlier this year, while flying over northern Botswana, Vanessa Hayes looked out over the Makgadikgadi Pans—giant salt flats that stretch for more than 6,000 square miles. They are the remnants of what was once Africa’s largest lake. Hayes could see traces of the lake’s shoreline from the air. She glimpsed massive fault lines running across its former bed—signs of the tectonic activity that eventually broke it apart into a patchwork of lush wetlands. For Hayes, who is a geneticist at the Garvan Institute of Medical Research, the view carried a special significance. She thinks that this region was once humanity’s homeland—the place where the ancestors of modern humans got their start.

Hayes and her team analyzed the DNA of 1,217 people from southern Africa who represent a particularly important and poorly studied slice of human genetic diversity. By using that DNA to create a family tree, the team calculated that anatomically modern humans originated in the Makgadikgadi wetlands about 200,000 years ago. They then stayed put for about 70,000 years, before climatic changes allowed some of them to venture outward to other parts of Africa, and eventually to other continents.

But her claims are proving controversial, and other researchers I contacted were either skeptical or outright mad. The study, they noted, is based on just a sliver of DNA from living people, and doesn’t account for the rest of the genome, DNA from ancient human specimens, fossils, or stone tools and other cultural artifacts—all of which suggest that humans arose much earlier, and in a variety of locations. “It ignores a swath of evidence supporting an older origin for our species,” said Eleanor Scerri, an archaeologist who studies human origins at the Max Planck Institute for the Science of Human History.

“The conclusions are far-fetched and very much overstated,” added Carina Schlebusch, a geneticist at Uppsala University who specializes in southern Africa. “It tells us very little about human origins as a whole. It only tells us about the origin of a very small part of the human genome, and nothing more.”

That small part is the mitochondrial genome, or mitogenome, which sits apart from the main genome and is inherited only from mothers. The family tree of all human mitogenomes is rooted somewhere in Africa, and its base splits into two great branches. One, known as L0, is mostly found in southern Africans such as the Khoe and San peoples. The other, known as L1’6, includes roughly everyone else. Since genetic studies have focused on people in Western countries, much of the L0 branch is poorly studied, which means large swaths of human diversity—and human history—are still unknown. Hayes’s team tried to fill that gap by searching for people who represent the least studied subbranches of L0. They found about 200, and added their mitogenomes to an existing set of 1,000.

By comparing this set, and backtracking through time and space, Hayes’s colleague Eva Chan estimated that the L0 lineage emerged in Makgadikgadi about 200,000 years ago. Given how deeply rooted that lineage is, the team is using it as a proxy for humanity at large. By its reckoning, L0 changed very little between 200,000 and 130,000 years ago. “We see almost no diversification for almost 70,000 years,” Hayes says. Climate models show that during this time, the region was gripped by a mega-drought, and the wetlands would have provided a welcome but isolated oasis. But then increasing humidity created new green corridors to the northeast, and later the southwest, allowing humans to venture out of the homeland. As they migrated, their mitogenomes also diversified.

That’s interesting and valuable to know, said Chiara Barbieri, a geneticist and anthropologist at the University of Zurich. But inferring when and where the L0 lineage originated is tricky based only on the DNA of people who share that lineage today. Prehistory is long. People move. “The only way to precisely anchor genetic variation in time and space is with DNA from carbon-dated fossils,” Barbieri said. And Hayes’s team didn’t consider any sources of ancient DNA in its study.

What’s more, it looked only at present-day mitogenomes, which “simply don’t have the power to support narratives of this sort, because they represent a tiny fraction of human ancestry,” Scerri said. “Reconstructing deep ancestry from mitochondrial DNA is like trying to reconstruct a language from a handful of words.”

If you look at different words—that is, different parts of the genome—conflicting narratives emerge. The family tree for the Y chromosome, for example, is rooted somewhere in Cameroon, 2,000 miles northwest of the supposed homeland. A study of several complete genomes, including several ancient ones, suggests that the ancestors of today’s Khoe and San diverged from the ancestors of other African peoples 350,000 to 260,000 years ago—well before the origin Hayes suggests. Another analysis of complete genomes hints at even deeper splits. “What really struck me is that there was not reference or discussion” of these other studies in Hayes’s new paper, said Katerina Harvati, who studies human evolution at the University of Tübingen.

These seemingly conflicting results hint that humanity’s history in Africa is far from simple, and that our species has been mixing, diversifying, and moving for a very long time. Fossils tell the same story. In 2017, researchers described 315,000-year-old bones from a Moroccan cave called Jebel Irhoud that are the oldest Homo sapiens fossils ever found. Shortly after, a 180,000-year-old jawbone from the Misliya cave, in Israel, showed that humans had ventured out of Africa far earlier than suggested. Just this year, Harvati announced that a 210,000-year-old skull from Apidima Cave, in Greece, also belonged to Homo sapiens. All of this suggests that Homo sapiens not only existed but spread far and wide before the homeland period that Hayes defines. Complex stone tools support this idea: They’ve been found at sites that are about 300,000 years old, in locations as diverse as Morocco, Kenya, and South Africa.

Based on such finds, many scientists have abandoned the simple idea that humanity originated in any one part of Africa. Instead, they think that the entire continent was our homeland. This idea, known as African multiregionalism, says the features that characterize modern humans arose in a patchwork fashion, among ancestors who were spread throughout Africa and who gradually mingled. That’s why human fossils and advanced tools crop up all over the place at roughly the same time. It’s also why family trees based on different parts of the genome root in different times and places.

Though Hayes’s paper doesn’t mention the fossils or tools, she told me that it “doesn’t refute those archaeological findings.” Many of the early fossils, such as those from Jebel Irhoud and Apidima, include a mix of both archaic and modern features, and though classified as Homo sapiens, they wouldn’t look exactly like people today. And Hayes said that she’s focused specifically on anatomically modern humans, whom she calls Homo sapiens sapiens. It’s entirely possible that other humans were spread throughout Africa, but Hayes sees these as “break-off lineages [that] came and died off.” Only the ones in the homeland gave rise to people today.

“This is a convenient response,” Scerri said, but not a convincing one. The features that define humans today don’t appear together in any single individual until 100,000 to 40,000 years ago, well after our species arose, and well after Hayes’s purported out-of-homeland migrations. Again, a patchwork origin provides a better explanation. The new study tells the story of one group of ancestors. We likely had many.

Hayes is no stranger to controversy. In 2010, she published the full genomes of four San elders from Namibia, in a bid to increase representation of southern Africans in genetic research. But shortly after, an NGO called the Working Group of Indigenous Minorities in Southern Africa (WIMSA) accused her team of “absolute arrogance, ignorance, and cultural myopia.” While Hayes had verbally acquired informed consent from the four men, San leaders questioned the effectiveness of that consent, and criticized Hayes for not also consulting and engaging with them. They were angry that the paper used the pejorative term bushman and discussed intimate and personal information about traits such as hearing, taste, and skin-cancer susceptibility.

Spurred by this incident, the South African San Council recently drafted an ethical code of conduct for researchers who want to work with the San. Among other things, the code asks that the San be involved in designing research projects, that their culture be respected, and that they be told about results before they are published.

In her defense, Hayes said that she worked with the communities in her study for 10 years. She speaks their language, visits them regularly, and took care to ensure that they truly understood what they were giving their blood samples for. As for the need to engage with San leaders, “every community was asked if they knew WIMSA and recognized them, and said no,” she said. “They represent themselves.”

This time around, Hayes is set to talk with the San Council’s directors next week (they did not respond to a request for comment). And two weeks ago, she went back to the communities whose mitogenomes feature in the new study to share her results. “They were the first to hear,” she said. “I found it easy—their stories already talk about being the first people.”

But if the reaction to the study is anything to go by, the scientific community may prove to be a tougher audience.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.