The Skeptic Who Moved Two Points

TL;DR: Psychologist Jesse Bering went into Ian Stevenson’s reincarnation case files intending to supply the psychological explanations Stevenson had missed. He came out having written The Incredible Afterlives of Dr. Stevenson (2026), moved from the negative end of the scale to roughly a one or two out of ten on the afterlife, and said out loud that a rational person could believe in reincarnation on this evidence. That last sentence is the one being quoted. It is also the one doing the least work. The load-bearing admission is elsewhere.

I host a show about what got recorded, by whom, and whether anybody followed up. So let me put this file in order before anybody tells me what it means.

What is actually being claimed?

The Daily Grail’s piece reports on Bering, a psychologist trained in cognitive science and evolutionary theory, and a science communicator with a long public record on the wrong side of this question from the believers’ point of view. He wrote a book about Stevenson published this year. Going in, per his own account, the plan was ordinary: read the reports carefully and find the psychological mechanisms Stevenson had failed to rule out.

That is a legitimate methodology and it is the correct first move. Stevenson spent decades collecting cases, overwhelmingly of small children who made statements about a previous life that were then checked against a real deceased person. The standard skeptical account is that these are produced by ordinary things: parental prompting, coincidence, retrospective matching, cultural expectation, memory contamination between the first interview and the written record, and investigator bias in the selection of which cases get written up at all. Every one of those is real and every one of those has explained cases before.

What Bering reports is that it did not go the way he expected. His words: “explaining away the data proved to be a more difficult task than than I anticipated going in.”

Now the arithmetic

Here is where I want to slow down, because this story is going to get flattened in both directions within a week.

What Bering conceded, precisely: he moved from the negative end of a ten-point scale to “a one or a two out of 10.” And when asked whether a rational person could believe in reincarnation on the strength of this evidence, which is Stevenson’s own 1974 formulation, he said yes.

Read the scale carefully. A one or a two out of ten is not a conversion. It is a man saying he is still overwhelmingly unconvinced. If you handed me a witness who rated their own confidence at fifteen percent, I would not build a case on them and neither would you.

What he did not concede: that reincarnation occurs. That Stevenson’s methodology was sound. That any specific case is unexplainable. None of that is in the reporting.

What he did concede that actually matters: that these are “legitimate empirical research questions.”

That is the finding. Not the number. The number is a mood. The claim that a question is empirically legitimate is a claim about whether evidence could in principle bear on it, and that is a much harder thing to say and a much harder thing to take back. It moves reincarnation research out of the category of things that are dismissed on their face and into the category of things that are answered badly or well.

And the second concession, the rational-person one, is narrower than it sounds too. Stevenson’s 1974 phrasing was always careful. He did not claim proof. He claimed that a reasonable person, having examined this specific body of material, could hold the belief without having abandoned reason. Bering agreeing to that proposition is Bering agreeing that the material is not junk. It is not Bering agreeing that the material is correct.

What is missing from the record

I do not have the book in front of me and the article does not supply what I would need to close this file. Three gaps, named plainly:

Which cases. Stevenson’s corpus runs to thousands, and their quality is not uniform. The strong ones, the ones that get cited, typically involve a written record of the child’s statements made before anyone located the deceased person, which is the single control that matters and the one most cases lack. Whether Bering’s difficulty was with the strong subset or with the corpus generally is the entire question, and this article does not say.

Which mechanisms failed. “Harder than anticipated” is a summary, not a finding. A skeptic reporting that a specific explanation did not fit a specific case is a data point. A skeptic reporting a general feeling of difficulty is a memoir. The book presumably distinguishes these. The coverage does not.

Whether anyone has replied. Stevenson’s work has an active critical literature, and it did not stop when he died in 2007. A shift by one prominent skeptic is interesting. Whether the people who have spent careers on the methodological objections think Bering engaged with their strongest version is a different question, and it is the one that determines whether this holds up.

Why I would still keep the file open

Because the direction of travel is the unusual part.

The ordinary career arc for a researcher who spends years inside anomalous material is toward belief, and that arc is so common it is nearly worthless as evidence. Somebody who goes looking for wonder finds it. What is rarer, and therefore more informative, is somebody who arrives with a specific debunking apparatus, applies it, and reports that the apparatus underperformed.

That is not proof of anything. Bering could be wrong. He could be a psychologist who spent four years in a corpus and got acclimated, which is a documented thing that happens to investigators in every field including mine. Two out of ten is also exactly where you would expect a careful person to land after prolonged exposure to material that is strange but not conclusive, which is what I suspect this material actually is.

But he did the work before he had the opinion, and he published the number rather than a vibe. That is more procedural discipline than most of the people who will now quote him have ever shown.

The question is legitimate. Nobody has answered it. Both of those can be true at once, and the fact that almost nobody wants them to be is the most interesting thing in this story.

Frequently asked questions

Did Jesse Bering say he now believes in reincarnation?

No, and the reporting is careful about this even where the headlines will not be. He describes moving to roughly a one or two out of ten on the afterlife, which is a statement of continued disbelief with a small opening in it, and he separately agreed that a rational person could hold the belief given Stevenson’s evidence. Those are two modest claims that get merged into a conversion narrative the moment they leave the article.

What made Stevenson’s cases hard to dismiss in the first place?

The strongest ones share a structure: a very young child makes specific, checkable statements about a life they could not plausibly have learned about, those statements are written down, and only afterward is a matching deceased person identified. When that order of operations is genuinely preserved, the ordinary explanations get harder, because retrospective matching and family coaching both require the answer to exist before the claim. The persistent argument is over how many cases actually preserved that order, which is a documentation question rather than a metaphysical one.

Is a one or two out of ten a meaningful shift or just a rounding error?

On its own it is close to noise, and treating it as the story is the mistake. What carries weight is the accompanying statement that these are legitimate empirical research questions, because that is a position about method rather than about probability, and it is the position that determines whether anyone bothers to run better studies. A number can drift. A concession about what counts as a researchable question tends to stay conceded.