A 15-year Smithsonian research project has produced what scientists call "extraordinary proof" that Thomas Jefferson sired children with his slave Sally Hemings, and the same genomic techniques helped convict two of America's most notorious serial killers.
Researchers sequenced Jefferson's entire genome from a lock of hair clipped from his head as he lay on his deathbed at Monticello in 1826, then compared the results against DNA samples from 13 people believed to be living descendants of Jefferson and Hemings. Biologists ran 20,000 simulations of each paternity scenario, testing whether Jefferson, his brother Randolph, or Randolph's sons could have been the father. The odds that Jefferson did not father either of Hemings' sons, Madison or Eston, came back at less than one in 1,000, the New York Post reported.
The findings close a question that has swirled around the third president since the early 1800s, and they arrive with a level of scientific rigor that prior investigations could not match.
Smithsonian scholar Richard Kurin launched the project in 2011 after stumbling on three strands of hair while previewing a documentary about Jefferson's personal Bible. Jefferson had famously created that Bible by cutting out passages containing Jesus' moral teachings and pasting them into a new book. The hairs were tucked inside.
Kurin's team, archaeologists, genealogists, molecular geneticists, and historians, first tested those strands against DNA from a female-line descendant of Jefferson's mother. No match. But a separate lock of hair, taken from Jefferson on his deathbed at Monticello in 1826, proved genuine. From that lock, using what the researchers described as "extraordinary new techniques," the team built a full genome sequence.
They then collected samples from 13 believed living descendants of both Jefferson and Hemings. David Miller, a sixth-generation descendant of Jefferson, said his mother, Sarah Coachman Miller, provided one of those samples. She died before the results came in.
Miller told the New York Times the findings were overdue.
"It's positive the family now hopefully gets the recognition and acknowledgement and acceptance they deserve."
The study did not go without friction. One Hemings descendant who gave a DNA sample in 2015 asked this month to withdraw her participation, frustrated that Kurin's book about the project was appearing before a related scientific journal article had been published. The Smithsonian confirmed it destroyed her sample and all related data, then reran the analysis without it. The conclusions held.
Speculation about Jefferson and Hemings traces back to his own era. Kurin pointed out that the question predates modern politics by two centuries.
"This is a topic of interest to the American people. I didn't invent this question, and it wasn't invented by the Smithsonian. It goes back to [Alexander] Hamilton and Jefferson and [John] Adams."
A landmark 1998 DNA study had already provided scientific evidence that Jefferson fathered at least six children with Hemings. The Thomas Jefferson Foundation later concluded, after reviewing documentary, scientific, statistical, and oral history evidence, that Jefferson most likely fathered Hemings' children. Historians have long noted one corroborating detail: Jefferson freed all of Sally Hemings' children but did not grant freedom to any other enslaved family unit at Monticello.
The new Smithsonian work goes further than the 1998 study by sequencing Jefferson's full genome and running thousands of statistical simulations that specifically tested alternative paternity candidates. Randolph Jefferson, Thomas's brother, had long been floated as one of the most likely other candidates. The 20,000 simulations effectively ruled him out, along with his sons.
Kurin acknowledged the Smithsonian has no formal institutional position on the conclusion, but he was direct about what the data shows.
"Other people may have other ideas, and I don't think there's necessarily an institution position, but I think it's compelling. It's very clear."
He framed the work as a matter of obligation.
"If we have evidence, scholarship and science that helps elucidate American history, I think we have a duty to pursue it."
The research did more than settle a historical argument. The genomic techniques developed over 15 years fed directly into two of the highest-profile criminal cases in recent American history.
Rex Heuermann, convicted in connection with the Gilgo Beach murders on Long Island, was caught in part because of DNA recovered from a pizza crust he discarded outside his Manhattan architecture office. Joseph DeAngelo, the Golden State Killer, was identified with the help of family genealogy websites. Both investigations relied on advanced DNA methods that grew out of the same scientific ecosystem as the Jefferson project.
Beth Shapiro, an ancient DNA expert at the University of California, Santa Cruz, drew a straight line between the two efforts.
"It's hard to say we ever would have gotten to this place, where all these tools were used to convict the Gilgo Beach serial killer, if it hadn't been for starting this project."
That connection is worth pausing on. A project that began with three hairs in a hand-cut Bible helped produce forensic tools that put serial killers behind bars. The science moved from the 18th century to the 21st, from a Founding Father's deathbed to a pizza crust on a Manhattan sidewalk.
Kurin is promoting a new book about the project, but the related scientific journal article has not yet been published. That gap is the reason one Hemings descendant pulled out. Peer review matters, and the full scientific community has not yet weighed in on the methodology or the statistical models behind the less-than-one-in-1,000 figure.
Several questions also remain unanswered. The specific "extraordinary new techniques" used to sequence a full genome from nearly 200-year-old hair have not been publicly detailed in a peer-reviewed venue. Whether all 13 descendants' samples were retained and used in the final analysis, beyond the one that was destroyed, is unclear. And the Smithsonian has not stated whether it plans to take a formal institutional position on the findings.
None of that erases what the numbers say. Twenty thousand simulations. Less than one in 1,000. A genome pulled from a lock of hair taken from a dying man in 1826, matched against living Americans nearly two centuries later.
History does not care about comfort. It cares about evidence, and for the descendants of both Jefferson and Hemings, the evidence now speaks with a clarity that politics never could.