Children who remember dying: Dr. Ian Stevenson's reincarnation research

Of everything I have researched so far in this series, I think Dr. Ian Stevenson's work might be the hardest to explain away. Not because it is the most dramatic or the most emotionally moving, though parts of it certainly are. It is because Stevenson built his entire career on the least mystical approach imaginable to a deeply mystical question. He was a psychiatrist. He believed in evidence, documentation, and verification. And he spent nearly fifty years, until his death in 2007, doing something almost no one else in his position was willing to do: traveling the world to methodically investigate young children who claimed to remember dying before.

Stevenson chaired the Department of Psychiatry at the University of Virginia School of Medicine, a serious, mainstream academic post, and it was from that position of institutional credibility that he began what would become the largest systematic investigation of potential reincarnation ever conducted. His 1966 book, Twenty Cases Suggestive of Reincarnation, was just the beginning. By the time he retired from active fieldwork, he and his colleagues had documented and investigated more than 2,500 cases from around the world.

It is worth pausing on how unusual this career choice actually was. Stevenson was already an established, respected psychiatrist before he turned his attention to reincarnation research. He could have spent his career on more conventional psychiatric questions and been comfortably regarded within his field for the whole of his working life. Instead he chose a subject that most of his colleagues considered, at best, an embarrassing distraction from serious science. He pursued it anyway, funded largely through private donations rather than traditional academic grants, because he believed the phenomenon itself, regardless of how it was ultimately explained, deserved the same rigorous documentation any other unexplained pattern in medicine would receive.

What a typical case looked like

The pattern Stevenson documented was remarkably consistent. A child, usually between the ages of two and four, would begin speaking spontaneously about a previous life. Not in vague, dreamlike terms, but with specific, concrete details: a name, a town, the names of family members, details about how that person had died. In the strongest cases, the child's statements were later verified against a real deceased individual the family had no prior connection to and no plausible way of knowing about.

One of Stevenson's earlier well-documented cases involved a young girl named Swarnlata Mishra, who at three years old began singing unfamiliar songs and performing dance steps her own family had never taught her or seen her exposed to, from a region of India more than a hundred miles from where she lived. She went on to describe details of a previous life in that distant town with a level of specificity that was later checked against and matched a real deceased woman's life there. Cases like this followed what Stevenson came to recognize as an almost predictable pattern. The memories typically emerged around age two or three, often described with striking clarity for a child that age, and then gradually faded as the child grew older, usually disappearing entirely by six or seven.

Stevenson noticed something else that recurred across an enormous number of cases: the deceased individual being remembered had very often died violently or suddenly, in an accident, a murder, a drowning, a wartime death. He came to see this pattern as significant enough to investigate on its own terms, wondering whether a sudden, unresolved death might somehow be connected to the intensity or persistence of the memory that followed.

He also documented a related pattern that I find genuinely striking: many of these children displayed intense, specific phobias connected to how the previous personality had died, fears that seemed to have no origin in the child's own life. A child recalling a drowning death might show an inexplicable, severe fear of water from infancy, long before any incident in their current life could account for it. A child recalling death by a specific weapon might show a similarly specific fear reaction. Stevenson documented these phobias carefully because they represented something measurable and observable in the child's present behavior, not simply a verbal claim about the past. A family could not easily coach a toddler into a convincing, persistent phobia. Either the fear was genuinely present, often from the child's earliest days, or it was not.

The birthmarks that matched fatal wounds

Perhaps the most startling and, to me, the most scientifically provocative thread in Stevenson's research involves birthmarks and birth defects. Early in his fieldwork, he began noticing that some children who reported memories of a previous life also had a birthmark or birth defect in a location that corresponded to a fatal wound on the person whose life they claimed to remember.

By 1997, he had documented 210 such cases, and in 49 of them, he was able to obtain actual autopsy or medical records confirming the precise location of the fatal wound on the deceased individual, records he compared directly against the child's birthmark. One of his most frequently cited individual cases involved a Turkish boy named Süleyman Çaper, whose birthmarks on his ear corresponded to the entry and exit wounds of a shotgun blast that had killed the person whose life he described remembering. This was not a vague resemblance. Stevenson was comparing documented medical evidence of a specific wound against a specific birthmark on a child born afterward with no biological explanation connecting the two.

He compiled this research into a two-volume work, Reincarnation and Biology, spanning roughly thirty years of case investigation. I find this particular strand of his research the hardest to categorize as simple coincidence or cultural storytelling, because a birthmark is not something a child can be coached into producing. It is either there, in the corresponding location, matching a documented wound, or it is not.

What makes this evidence especially compelling to me is the standard Stevenson held himself to. He did not simply accept a family's word that a birthmark matched a remembered wound. Wherever possible, he sought out independent, contemporaneous documentation, medical records, autopsy reports, death certificates, created before anyone had connected that death to a living child, and compared them directly against the physical birthmark years later. This sequence matters. The documentation of the wound existed first, entirely separate from and prior to any claim about the child. Only afterward did Stevenson bring the two pieces of evidence together for comparison. That order of operations rules out one of the most common explanations skeptics raise about this kind of evidence, the idea that a story gets shaped backward to fit a birthmark that was already there. In the cases with prior medical documentation, the wound was recorded first, without reference to any child, and the birthmark was compared against it only afterward.

Stevenson's caution, and why it matters

I want to highlight something about Stevenson's own temperament, because I think it is essential to understanding why his work holds up as well as it does. He was, by every account from colleagues and critics alike, an unusually careful and methodical investigator. One psychiatrist reviewing his work in the Journal of Nervous and Mental Disease in 1977 described him as methodical, careful, even cautious, noting that either Stevenson was making a colossal mistake or would eventually be recognized for uncovering something genuinely important.

Stevenson himself never claimed that any single case proved reincarnation. He was explicit about this throughout his career. What he argued was that the strongest cases, the ones with verified birthmarks matched against autopsy records, or with detailed statements confirmed against a real deceased person entirely unknown to the child's family beforehand, made reincarnation the best available explanation for that specific evidence, even while acknowledging that alternative explanations could not be fully ruled out in every instance.

This kind of restraint matters enormously in a field this easy to dismiss. Stevenson was not interested in believers nodding along. He was interested in building a case sturdy enough that a skeptical scientist would have to engage with the actual evidence rather than dismiss the entire subject out of hand. His research included careful attention to cultural context too. He specifically studied whether cases appeared more frequently in cultures that already believed in reincarnation, reasoning that if belief alone explained the phenomenon, cases should cluster heavily in reincarnation-friendly cultures and appear rarely elsewhere. What he found complicated that simple explanation. Some regions with strong cultural belief in reincarnation, like South India, actually produced fewer documented cases than expected, while cases also turned up in cultures without any strong prior belief in reincarnation at all, including a notable set of cases among the Tlingit people of Alaska and, later, a growing number of cases investigated in the United States and Europe.

This cross-cultural spread matters because it undercuts the simplest dismissal of Stevenson's research, the idea that children in reincarnation-believing cultures are simply absorbing and repeating a story their community expects to hear. If that were the whole explanation, cases should track cleanly with cultural belief, appearing everywhere belief is strong and nowhere it is absent. Instead, the pattern is messier and, I think, more interesting than that. Something is generating these cases that does not map neatly onto what a given culture already expects or encourages a child to say.

Why this connects to everything else in this research

I keep noticing the same underlying thread running through every source I study for this series, even when the sources have almost nothing else in common. Edgar Cayce's trance readings described the soul choosing its next step after a period of rest and reflection, framed as an act of free will guided by wisdom rather than obligation. Swedenborg described souls settling into an afterlife shaped by what they had already become. Raymond Moody's near-death interview subjects described a life review conducted with compassion rather than judgment.

I also want to be fair about where Stevenson's work has drawn real criticism, because I think that honesty makes the strongest cases more credible, not less. Skeptics have pointed out that much of the fieldwork relied on retrospective interviews conducted after a family had already connected a child's statements to a specific deceased person, raising the possibility that memories of exactly what the child said early on could have been unconsciously reshaped over time to fit better. Critics have also noted that verifying a child's claims often depended on the same family's own account of what was said and when, which is harder to fully separate from the family's own beliefs and hopes. Stevenson was aware of these limitations and tried to address them by seeking cases investigated as early as possible after the child began speaking, and by prioritizing cases with independent documentation, like the autopsy-verified birthmark cases, that did not rely solely on family testimony. Even so, the criticism is a fair one to hold alongside the strongest cases, and it is part of why Stevenson himself never claimed certainty, only that reincarnation was, in his judgment, the best available explanation for his strongest documented cases.

Stevenson's work adds something genuinely different to this picture, because it approaches the question from the opposite direction. Rather than asking what happens right after death, his research asks what might happen before a new life begins, whether the soul carries something forward into its next incarnation, sometimes even a mark on the body itself, testifying to what it experienced in the one before. If Cayce and Swedenborg and Moody are describing the same terrain from the vantage point of someone leaving a life, Stevenson may be describing that same terrain from the vantage point of someone arriving at a new one.

I do not think any one of these sources should be treated as final proof. But taken together, spanning a trance medium, an eighteenth-century scientist, a philosopher turned doctor interviewing cardiac arrest survivors, and a university psychiatrist investigating children on nearly every continent, the consistency is hard to dismiss as coincidence. Something keeps showing up across every method of inquiry humans have tried, whether through altered states, near-death interviews, or rigorous medical documentation of small children who have not yet learned to filter what they say.

What Stevenson's work leaves us with

Stevenson's research continues today, carried forward by his successor at the University of Virginia's Division of Perceptual Studies, Dr. Jim Tucker, whose ongoing work includes a growing number of American cases. I plan to write about his research separately, because it deserves its own space, particularly the ways it challenges the assumption that this phenomenon only appears in cultures already primed to expect it.

For now, I keep coming back to what Stevenson himself seemed most interested in throughout his career, not proving an afterlife exists in some final, unassailable way, but simply documenting, with as much rigor as any serious scientist could bring to the question, what actually happens when a small child insists, with startling specificity, that they remember a life that was not their own. Whatever conclusion you draw from that, the documentation itself, thousands of cases, birthmarks matched against autopsy records, statements verified against real deceased individuals, is not something that can be waved away as simple wishful thinking. It took a rigorous, skeptical, credentialed scientist most of his working life to compile, and he did it precisely because the pattern would not leave him alone.

I think about what this means for how we understand a soul's continuity more broadly. So much of what I write about in this series focuses on what happens in the moments and years after someone dies, the welcome, the healing, the life review, the eventual choice about what comes next. Stevenson's work fills in a piece of the picture that the other sources mostly leave implicit. If a soul really does choose a next step, as Cayce's readings describe, or settle naturally into a state shaped by its own inner nature, as Swedenborg describes, Stevenson's cases suggest that choice is not always a clean, unmarked beginning. Sometimes, in these documented cases at least, something carries over. A fear. A fragment of memory. Even, in the most striking cases, a mark on the body itself. That continuity, if it is real, reframes a new life not as a blank slate but as one chapter in something considerably longer, with threads still connecting it to whatever came before.

Ready to experience a reading?

If you're feeling drawn to connect with a loved one in spirit, I'd be honored to walk that path with you. You can book a mediumship reading at aperfectsoul.com.

Want more like this in your inbox? Get one story a week on grief, mediumship, and what I've learned from years of sitting with people who are healing, straight from my writing here on the blog. Join the list ➞