Children Who Remember Past Lives: The Cases That Won't Close

The nightmares started when James Leininger was two years old. He would thrash in his bed in Lafayette, Louisiana, kicking his legs into the air as if he were trying to force something open, and he would scream the same words his parents would come to dread: "Airplane crash on fire. Little man can't get out." In daylight he was calmer, but no less insistent. He told his father the plane had been shot down by the Japanese. He said it had flown off a boat. When his father asked the boat's name, the boy said "Natoma." When he asked who the little man was, the boy said, without hesitation, "Me."

James was not the child of a family that wanted to hear any of this. His father, Bruce Leininger, was a devout Christian who found the whole subject of past lives faintly blasphemous and spent the next several years trying to prove his own son wrong. What he found instead was a real World War II escort carrier called the USS Natoma Bay, a real pilot who had died in the Pacific, and a set of details a preschooler in Louisiana had no ordinary way of knowing. His book about it is titled, tellingly, Soul Survivor.

The Leininger case is unusual for how far into the American mainstream it traveled, but it is not unique. It belongs to a body of research that spans more than half a century, thousands of documented cases, and a small university department that has quietly logged the accounts of very young children who claim, with startling specificity, to remember dying as someone else. No one has been able to fully account for what those records contain.

This article contains affiliate links. If you buy through them we may earn a small commission at no extra cost to you.

What You'll Learn

• •The Boy Who Dreamed of Fire
• •A Psychiatrist Who Took the Children Seriously
• •How the Research Actually Works
• •The Shape of a Typical Case
• •The Birthmarks That Match Old Wounds
• •Return to the Natoma Bay
• •The Boy Who Was a Hollywood Extra
• •The Conventional Explanations
• •What the Careful Skeptics Get Right
• •Why the Question Sits Outside Science
• •Frequently Asked Questions
• •Further Reading

The Boy Who Dreamed of Fire

What unsettles people about the Leininger case is not the nightmares. Small children have nightmares. It is the accumulation of correct detail, delivered flatly, over months, by a child who had barely learned to talk.

James named the boat. He said his plane had been a Corsair. He said he flew with a man named Jack Larsen. When his mother bought him toy planes, he pointed to something slung under the belly of one and corrected her: that was a bomb, he told her, and it was called a drop tank. He signed his drawings of dogfights "James 3." He would list, unprompted, the names of other pilots who he said had been there. And when Bruce Leininger began pulling naval records, veterans' rosters, and reunion attendee lists to disprove all of it, the names his toddler had been saying kept turning out to be real.

This is the quality that runs through the strongest cases in the literature. It is not vagueness that draws investigators in. It is precision, in the mouths of children too young to fake it and, in many cases, in families with every reason to want them to stop.

A Psychiatrist Who Took the Children Seriously

The reason these accounts exist as data rather than folklore is largely one man. Ian Stevenson, born in Montreal in 1918, was a respected academic psychiatrist and the chair of the Department of Psychiatry at the University of Virginia School of Medicine. He was, by every account of those who knew him, meticulous, reserved, and allergic to overstatement. And in the 1960s he did something that could have ended a lesser career: he began traveling the world to interview children who said they had lived before.

In 1967 Stevenson founded a small research unit at Virginia, first called the Division of Personality Studies and now known as the Division of Perceptual Studies. The funding came from an unlikely benefactor, Chester Carlson, the inventor of the Xerox photocopying process, who backed Stevenson's work and left a bequest that endowed his research chair. That endowment freed Stevenson to spend the next four decades doing something no one had done at that scale: collecting, cross-checking, and archiving cases of the reincarnation type, one child at a time.

His 1966 book Twenty Cases Suggestive of Reincarnation laid out the method and the caution that defined him. He did not claim the cases proved anything. He chose the word "suggestive" deliberately. By the time he died in 2007, he and his colleagues had gathered more than 2,500 cases from India, Sri Lanka, Lebanon, Turkey, Thailand, Myanmar, West Africa, and the tribal communities of Alaska, and had built the largest archive of its kind in existence.

How the Research Actually Works

The method matters, because the entire value of the archive rests on ruling out the ordinary ways a child could come by the information.

A typical investigation begins after a child, usually between the ages of two and five, starts making statements about a previous life: a name, a place, a family, a way of dying. Ideally the investigator arrives before the "previous personality" has been identified, so the child's statements can be written down while they are still unverified. Stevenson would record what the child said, then attempt to locate a real deceased person whose life matched those statements. He would interview the child's family and, separately, the family of the deceased, looking for any prior connection between them, any path by which the details could have leaked.

He was hunting for the ordinary explanation as hard as any critic. Could the two families have met? Could the child have overheard a story, a radio program, a village rumor? Was a parent feeding lines, consciously or not? He tallied statements, marked which were verified and which were wrong, and noted how many people had witnessed the child speak before verification took place. The cases he considered strongest were the ones where the two families were strangers, separated by distance, before the child's words led them together.

The Shape of a Typical Case

Read enough of the files and a pattern emerges that is almost eerier than any single story, because the pattern itself is what resists a simple answer.

The children begin to speak of the previous life very young, often as soon as they can form sentences, and they usually stop somewhere between ages six and eight, as if a window closes. The memories are frequently tied to violent or sudden death, and they arrive with real emotion, not the flat recitation of a party trick. A striking number of the children have phobias that match the manner of death the "previous person" suffered: a fear of water in a child who describes drowning, a terror of loud bangs in one who describes being shot. Some show unusual skills, tastes, or habits that fit the claimed former life and not their present one. And a subset display something harder to wave away.

The Birthmarks That Match Old Wounds

Stevenson's final major work was not about words at all. It was about bodies.

Published in 1997, Reincarnation and Biology: A Contribution to the Etiology of Birthmarks and Birth Defects ran to two volumes and well over two thousand pages, and it documented more than two hundred cases in which a child bore a birthmark or a birth defect that corresponded to a wound, very often a fatal one, on the body of the person they claimed to have been. A child who remembered being shot in the head with a mark on the scalp where the bullet was said to have entered, and sometimes a second mark, larger and more ragged, where it left. In the cases Stevenson prized most, he was able to obtain a medical or autopsy record for the deceased and compare it against the location of the living child's mark.

This is the part of the research that even sympathetic readers find hardest to metabolize, and the part that is hardest to dismiss out of hand. A story can be absorbed from a village. It is far less obvious how a shared cultural belief in rebirth produces a puckered scar on a newborn in the precise place a stranger was stabbed a decade before.

Return to the Natoma Bay

Bruce Leininger's search led him, eventually, to a squadron reunion and a name: James M. Huston Jr., a young Navy pilot from Pennsylvania who flew off the USS Natoma Bay and was killed in action on March 3, 1945, during the fighting in the Pacific near Iwo Jima, shot down over water. Jack Larsen, the friend James had named, was a real Natoma Bay pilot who survived the war. The boat, the fellow flier, the theater of death, the shootdown by Japanese fire: the frame his son had been describing since before he could read matched a man who had died half a century earlier.

Honesty requires naming the friction in the case, because the site does not exist to sand mysteries smooth. James called his plane a Corsair, while the aircraft Huston is recorded as flying off the Natoma Bay was a Wildcat, and critics have pressed that discrepancy hard. Some details were interpreted generously by a father who badly wanted an answer, and memory in a household that has told a story many times is not a sealed vault. All of that is fair. What it does not dissolve is the core: a two-year-old naming a real warship, a real pilot's first name, a real comrade, and a real manner of death, in a family that started out wanting none of it to be true. The plane may be wrong. The knot at the center has not come loose.

The Boy Who Was a Hollywood Extra

If the Leininger case reads like wartime tragedy, another American case documented by Stevenson's successor reads like a lost film.

Jim Tucker, the child psychiatrist who now carries the research at the University of Virginia, investigated an Oklahoma boy named Ryan who, around the age of four, began telling his mother he "used to be somebody else." He spoke of Hollywood, of big houses and travel, of an agency that changed its name, of dancing and of a man he had known. One night his mother brought home library books about old Hollywood, and Ryan stopped on a still photograph from a 1932 film, pointed to a man in the frame, and said that was him.

The man in the photograph turned out to be Marty Martyn, a dancer who became a Hollywood agent and died in 1964, decades before Ryan was born. Tucker documented dozens of the boy's statements and, working with records and a genealogist, found that a long list of them lined up with the details of Martyn's life, including specifics the family could not plausibly have gathered on their own. Tucker lays the case out in his book Return to Life, and like Stevenson, he declines to shout a conclusion. He records what the boy said, and what was found, and lets the gap between them sit.

The Conventional Explanations

Every one of these cases has been met with a conventional account, and each account explains part of the record while leaving a piece uncovered.

The first is cryptomnesia: the child absorbed the information normally, overheard it, saw it, forgot the source, and later reproduced it as memory. This is a genuine phenomenon and it surely accounts for some weak cases. It strains against the ones where investigators wrote the statements down before any matching person had been found, and where the two families lived far apart and had never met.

The second is family shaping. Parents, especially in cultures that expect rebirth, notice a stray remark, encourage it, ask leading questions, and unconsciously build a story around a child who is happy to please. That is real, and it is why Stevenson weighted cultural expectation heavily. It reads differently in a home like the Leiningers', where the parents fought the interpretation for years and found it disturbing rather than delightful.

The third is coincidence stretched by loose matching: a child makes many vague statements, and someone later finds a dead person who fits a few of them. Stevenson's answer was to count. He tracked how many statements were specific, how many were verified, and how many were flatly wrong, and the cases he ranked highest were not the vague ones. They were the ones dense with correct, checkable particulars.

What the Careful Skeptics Get Right

The site does not debunk, but it also does not pretend the strongest counterarguments do not exist, because a mystery is only worth your time if it survives an honest look.

The most serious pattern the research itself reveals is that cases cluster in cultures that already believe in reincarnation. India, Sri Lanka, and the communities of the Middle East and Southeast Asia produce them in abundance. That correlation is exactly what you would expect if belief were manufacturing the phenomenon rather than the phenomenon supporting the belief. It is the single hardest fact for the reincarnation reading to absorb.

Two other honest points belong here. Many cases were investigated only after the two families had already met, which lets ordinary information flow before the record is made. And the best cases are unrepeatable: you cannot run a controlled trial on a spontaneous memory in a three-year-old. What Stevenson built is a vast catalogue of anecdotes gathered with unusual rigor, and a catalogue of anecdotes, however careful, is not the same thing as an experiment.

And yet the culture argument does not close the file either, because the Western cases exist. James Leininger was born into a country that does not raise its children on rebirth, into a family that actively resisted the idea, and the details still landed. If belief were doing all the work, the boy in Louisiana should not have had the boat's name.

Why the Question Sits Outside Science

Part of the reason this research never resolves is that it collides with an assumption most of science treats as settled: that the mind is what the brain does, and that when the brain stops, the person is gone. Nothing in a materialist model of consciousness has room for a memory that outlives the tissue that formed it, then reappears, intact and specific, in an unrelated newborn.

Stevenson knew this, and it is why he chose his words so carefully to the end. He never said the children proved reincarnation. He said the cases were suggestive of it, and that a serious account of the mind should be able to explain them, and that the standard account could not. That is a narrower and more durable claim than the headlines his work sometimes attracted. It is also the reason the archive at Virginia keeps growing while the debate stays exactly where he left it. The data is strange, the conventional explanations each cover part of the ground, and the part they leave uncovered is precisely the part that made a father in Louisiana go looking for a boat his son could not have named.

Frequently Asked Questions

How old are the children when this happens, and do they grow out of it? Almost always very young. The statements typically begin between ages two and five, when a child can barely be coached, and fade somewhere between six and eight. Parents often describe the memories switching off as suddenly as they switched on, as if the child crossed back over a line and forgot they had ever been on the other side of it. Whether something is lost at that threshold, or only a story a child grows tired of, is the question that lingers.

Did Ian Stevenson believe in reincarnation? He was careful never to say so in those words. He believed the cases were real, that the children were not lying, and that the strongest of them could not be explained by any normal path he could find. He also believed the honest scientific position was to say "suggestive" and stop there. That restraint is a large part of why his archive is still taken seriously.

Is the James Leininger case actually verified, or just a good story? The core facts about the USS Natoma Bay, the pilot James M. Huston Jr., his death near Iwo Jima, and the survivor Jack Larsen are matters of naval record. What remains contested is how a Louisiana toddler came to speak those names, and whether specific details drifted as the family retold the story. No investigation has produced an ordinary channel by which the boy could have learned what he said. The record stands with the question still open inside it.

Why do most cases come from Asia? Because the phenomenon shows up most where people are willing to notice it. In cultures that expect rebirth, a child's odd statement gets taken seriously and written down. In the West it gets dismissed as imagination and lost. Whether that means belief is creating the cases, or simply that the West is failing to record its own, is exactly the fault line the research runs along.

Can any of this be tested in a laboratory? Not in the usual sense. You cannot summon a spontaneous memory in a specific child on demand, which is why this work lives as fieldwork and case documentation rather than controlled experiment. Its defenders argue that some real phenomena, like earthquakes or supernovae, can only be studied where and when they occur. Its critics argue that anecdotes never add up to proof no matter how many you collect. Both are right, which is why the argument never ends.

Further Reading

If you want to weigh the evidence yourself rather than take anyone's summary of it, start with the people who did the work.

• •Twenty Cases Suggestive of Reincarnation by Ian Stevenson, the foundational study that established the method and the caution behind it. Find it on Amazon.
• •Life Before Life: A Scientific Investigation of Children's Memories of Previous Lives by Jim B. Tucker, the most readable overview of the University of Virginia research. Find it on Amazon.
• •Soul Survivor: The Reincarnation of a World War II Fighter Pilot by Bruce and Andrea Leininger, the family's own account of the case that reached a country that does not believe in it. Find it on Amazon.

The Question That Stays

Set the explanations side by side and none of them covers the whole field. Cryptomnesia handles the leaky cases and not the sealed ones. Cultural expectation handles Asia and not Louisiana. Coincidence handles the vague statements and not the dense, checkable ones. The birthmarks handle none of them at all.

What is left is a small department in Virginia holding more than two thousand files, each one a very young child who said, calmly and too early to know better, that they had been someone else, and who was sometimes right in ways no one has been able to explain. The children forget by the time they start school. The rest of us are left holding the record, staring at the boat's name, wondering how it got into the mouth of a two-year-old.