Ian Stevenson was chair of the Department of Psychiatry at the University of Virginia School of Medicine. That is the detail that makes the rest of it awkward, because he was not a fringe figure who drifted into strange territory. He was a department head with a conventional publication record who decided to spend his career on something his colleagues considered a career-ending subject, and then did it for four decades.
He made his first research trip to Asia in 1961 and found something he had not expected. The cases were about more than memory.
What did Stevenson study?
Small children, usually between two and five, who began making statements about a life that was not theirs.
The pattern repeats across cultures with unsettling consistency. A child starts talking at two or three. The statements are specific: a name, a village, a house, family members, an occupation, and very often the manner of death. The memories fade as the child gets older and are usually gone by seven or eight. And the person described, in the cases that resolve, died violently.
Stevenson published Twenty Cases Suggestive of Reincarnation in 1966, drawn mostly from India, Sri Lanka, and Lebanon. Four volumes of Cases of the Reincarnation Type followed between 1975 and 1983. A volume of European cases came in 2003. He died in 2007, and the work continues at Virginia under Jim Tucker.
The Swarnlata Mishra case is a fair example of the strong end. At three she began singing songs her family did not know and performing dances from a region more than a hundred miles away, in a style nobody around her could have taught her.
Why do the birthmarks matter more than the memories?
Because a memory is a claim and a birthmark is on the body.
Stevenson noticed early that many of these children had something physically distinctive, and that it corresponded to how the remembered person died. By 1997 he had documented 210 cases where a birthmark or birth defect matched a wound. In 49 of them he obtained the dead person’s autopsy report and could compare the mark to the recorded wound location.
Reincarnation and Biology, published that year in two volumes, contains 225 such case reports. The physical anomalies are not vague. Hypopigmented scar-like marks. Port-wine stains. Moles in places moles are almost never found, including the soles of the feet. Malformed fingers. Missing lower legs. A Turkish boy with the right side of his face congenitally underdeveloped, who described the life of a man killed by a shotgun at point-blank range.
One case Stevenson worked shows both the strength and the problem. A boy named Mounzer remembered being shot in the 1958 Lebanese civil war and had a birthmark on his abdomen, to the right of the navel and slightly above it, which Stevenson sketched. He later asked the mother of the man in question where her son had been shot, and she pointed to the right side of the abdomen. He then showed her the sketch and she confirmed it.
The critic D. Scott Rogo made the obvious objection: Stevenson should have had the woman mark the location herself before showing her anything.
What are the serious objections?
They are real and a piece like this is worthless without them.
Contamination. This is the strongest one. In the great majority of cases, the two families had been in contact before any researcher arrived. Sometimes they had met, compared notes, and agreed on an identification months earlier. By the time Stevenson recorded what the child said, the child had often already been told who he was supposed to be.
Leading questions. Champe Ransom, who examined the case files at Virginia, documented instances where the investigation shaped the answer, the Mounzer sketch among them.
Depth. Ransom also argued the investigations were often thinner than the write-ups imply, with too much weight resting on interviews conducted years after the events through interpreters.
Presentation. Stevenson discussed the weaknesses of his cases, which is to his credit, but frequently did so in a separate section instead of inside the case report itself. A reader gets the case in full force and the caveats somewhere else.
Selection. The cases come overwhelmingly from cultures where reincarnation is an accepted belief. That is either evidence that such cultures notice and report what others dismiss, or evidence that belief produces the phenomenon. Both readings fit.
What is the honest position?
That Stevenson collected a body of cases that are genuinely difficult to explain, and that difficulty does not amount to evidence for reincarnation.
The reviewer in JAMA said of the first volume that he had painstakingly and unemotionally collected a series of cases in which the evidence is difficult to explain on any other grounds, and had placed on record a large amount of data that cannot be ignored. It was, mostly, ignored.
Two things can be true at once. The strongest cases resist the easy dismissals, and no alternative mechanism has been demonstrated either. Coincidence, unconscious cueing, family construction, and memory drift each explain some cases and none explains all of them well.
Stevenson’s own framing was careful to the point of stubbornness. He titled the founding book Twenty Cases Suggestive of Reincarnation, and he meant the word. He argued that reincarnation was the best available explanation for a subset of his cases, not that it was proven, and he spent his life on the position that the data deserved investigation instead of a verdict.
What about hypnotic regression?
Worth separating out, because most people meet the subject through it and it is the weaker end.
Past-life regression under hypnosis produces vivid, detailed narratives, and the mechanism is well understood: hypnosis increases suggestibility and confidence without increasing accuracy. The classic explanation is cryptomnesia, where something read or overheard long ago returns without its source attached and arrives feeling like memory.
Stevenson had no time for it. His entire method depended on spontaneous statements from very young children, made before anyone was trying to elicit anything, and ideally recorded before the families met. The regression material and the child cases are not the same phenomenon, and treating them as one field is how the serious work got buried under the unserious.
Why write it as a novel?
Reincarnation did not come out of Stevenson’s files. It came out of something closer.
I have landed on Iwo Jima. I have died in the mud of the Western Front. I know that the way you know something you cannot unlearn, not as belief exactly, but as memory that does not belong to this life and will not leave.
I wrote God Is Everything to set out what I understand about consciousness and what happens when we die. That book works as philosophy. What it cannot do is show what any of it looks like from the inside, and that is what a novel is for.
The structure runs on three tracks: historical chapters, regression sessions in a New Haven therapy office in 1971, and the soul sequences between them. Two souls finding each other across eighty-two thousand years, a connection with no beginning either of them can locate, surfacing again in different bodies and centuries.
What I wanted from the historical chapters is the part nobody writes. Not grand history, which is easy and is not what a life feels like from within. The toothache in seventh century Basra. The bad light on a Tuesday in fifteenth century Florence. An arm rising toward something far away for a reason the body cannot explain.
Eighty-two thousand years is a deliberate span. It reaches back past agriculture, past the Toba winter, into a period where the same two people would have had nothing in common with their later selves except each other.
Writing about contested evidence
There is a craft problem here that comes up whenever fiction is built on a disputed claim, and both easy solutions are bad.
Make the phenomenon real and unambiguous inside the book, and you have written a tract. The reader is being told what to conclude and can feel it.
Make it ambiguous all the way down, hedged so the events might always be delusion, and you have written a book that refuses to commit. Ambiguity has to cost something or it is just evasion.
The workable answer is to be certain about the experience and silent about the mechanism. The characters know what happened to them. The book never explains how, never supplies a system, and never has anyone deliver the theory. That is also, incidentally, what Stevenson’s best case reports read like. A child said this, these things checked out, and here is what could not be ruled out. No mechanism offered.
What the record supports
Set the metaphysics aside and something remains that is worth stating plainly.
Several thousand cases were collected over four decades by a credentialed researcher using a consistent method. A subset of them involve information a young child had no ordinary route to. A smaller subset involve physical marks that match wounds recorded in autopsy reports. Nobody has produced an account of that subset that satisfies the people who have read the files closely.
That is not proof of anything. It is an open question that a small number of people are still working on and most of science has agreed not to look at, and the reason it stays open is that the strongest cases have not been explained, not that they have been believed.
The novel is Reincarnation. The rest of the research behind the fiction is collected in the writing hub.
The Guides That Get Your Book Written, Published, and Sold
Four short, practical guides on writing, publishing, and selling your book, plus the occasional note when there's something worth your time. No fluff, no daily inbox clutter. Drop your email and they're yours.
We use MailerLite to manage our list and send these emails. Your address is used only to send you what you signed up for. We will not sell it, share it, or use it for anything else, and you can unsubscribe anytime.
Frequently Asked Questions