The Dalai Lama

Past Lives and The Skeptic

Went to a lecture bySkeptic editor Professor Chris French recently (rather surprisingly at the SPR, but I won't go into that). He mentioned in passing that the current edition of the mag contains a 'devastating' refutation of Ian Stevenson's research on children's memories of a previous life. This sort of claim always gets my attention - I wonder whether a critic really could have demolished in single magazine article the life's work of a leading psychical researcher. But you never know, so I thought I'd have a look.

The 7-page article is by Leonard Angel, not someone I'd come across before, but who I learn is a Canadian philosopher, and author of Enlightenment East and West (1994). His book contains a chapter debunking parapsychology, and especially Stevenson's past life research, which is also the subject of a Skeptical Inquirer article by Angel at around the same time. In the earlier piece he focuses on methodological flaws. He attacks the Lebanese case of Imad Elawar in Twenty Cases Suggestive of Reincarnation, accusing Stevenson of gross failures in his methods of research, write-up of data, and analysis of hypotheses and claiming an 'unfortunate insensitivity to central issues in data-gathering, presentation, and interpretation'.

I don't have Twenty Cases to hand, so can't really say how well his arguments stands up. My feeling has always been that Stevenson's methods are open to criticism, but that this kind of sweeping assertion by a sceptic seldom turns out to be justified. It didn't help that Stevenson's rebuttal in the magazine, after complaining about 'grave omissions' and 'inappropriate emphases', only covered one of Angel's points (he said he wasn't given enough room to do more). However it was striking, as far as it went. 

Angel had criticised Stevenson for having in the Imad Elawar case relied on the testimony of one particular individual, who was the verifier for 28 items of information. The fact that one of these was contradicted by someone else, Angel thinks, is reason to consider him unreliable. Stevenson replies that he himself had felt the need to get more verification of this witness's statements and made a follow-up trip to the Lebanon the following year for this specific purpose - only five of the 28 items were made solely by him. This is all clearly stated in the book, and Stevenson considers Angel's failure to mention any of it to be 'ill-considered guile or carelessness'.

In this new piece Angel questions the degree of correspondence between the statements made by the child and the circumstances of the claimed previous life. He complains that Stevenson ignored one of the most fundamental tenets of science, which is to eliminate the 'null hypothesis' before plumping for a paranormal interpretation.

... I'm suggesting that if anyone got many groups of correspondences between one life and another earlier life, and abstracted them a bit so that they looked like claims of past life memories, and then randomly mixed them in with evidence about past life memories, similarly abstracted, and then had subjects - ordinary people, or scientists - make guesses as to which need special explanation - then there would be a mere chance distribution.
Angel is struck by a list of correspondences between his own biography and Stevenson's, as recorded in an obituary of him published in the NY Times. He considers that some people might call these correspondences remarkable, though in fact they would not be. The list includes that fact that both men were born in Montreal, got a degree at McGill University, married twice, had a mother interested in spirituality, once lived in a house with a big library, and so on. His point is that if you start looking for someone who fits these general sorts of characteristics you could succeed, but it wouldn't mean that the subject actually had once been that person.

To investigate this, Angel suggests an experiment aimed at finding out whether or not random batches of material about people will generate as many correspondences with possible past lives as they would with real supposed memories of past lives. A number of correspondences would be generated, some of which would not be plausibly be thought to be paranormally generated (the control group) and those which Stevenson and some other people might think are paranormal (the experimental group). Blind subjects would guess which correspondences came from which group. If there was no statistically significant correlation between the cases guessed as coming from the experimental group and those that actually came from it, as Angel suspects will turn out to be the case, then there is no reason to consider anything paranormal in such cases.

It's an interesting idea, and any experimental approach that can help to put this sort of field research onto a more solid scientific footing can only be a good thing. But I didn't really understand the relevance of it to Stevenson's work. The correspondences Angel lists between himself and Stevenson are very general, and many of them do not relate to other people, places or circumstances but merely to the two men's ideas. However this isn't typical of the best reincarnation-type cases, which are compelling precisely because some of the children's statements, which are subsequently found to match with the circumstances of a deceased person, are so very particular.

In Children Who Remember Previous Lives, Stevenson describes the Sri Lankan case of a three-year old girl who said she remembered having lived in a town called Kataragama, and that her father owned a flower shop, was bald and wore a sarong. If that were all, and was typical of such cases, then Angel might have a point.  But the child also said that in the previous life her father was called Rathu Herath, and that she had drowned after being pushed into a river by a boy, and not just any boy, but one who happened to be mute - a set of circumstances that surely could not reasonably be held to apply to more than one person. 

Or take the case of Sunil Dutt Saxena in Stevenson's Indian collection.  He said he used to live in a certain town in a big house with servants, and that he had owned a fridge, a radio and other luxury items - all pretty general stuff. He claimed to have owned a factory and to have been married, which narrows it down a bit. But he also said he had been married four times and that he had founded a college, which was named after him, and named the principal, who was one of his best friends. In the same volume is the case of Jagdish Chandra, who among other things mentioned the names of the previous personality his father and his brother, stated correctly that the brother had died of poisoning, and identified the location of a safe his father kept his cash in.

In these three cases, and others that were 'solved', families were found in which the life of a recently deceased person matched the child's statements, in circumstances that researchers say largely precluded any non-paranormal explanation. (In the Jagdish Chandra case, for instance, his father, a lawyer, wrote down all the relevant items of information before advertising for help in tracing a person they might apply to.)

It's true that we are making a subjective judgement about what kinds of correspondences can be judged to point to a paranormal process. This is one of the biggest issues in parapsychology, much of which involves just this kind of inexplicable matching - in statements by mediums, in telepathic and apparitional cases, in out-of-body perception, and so on. And in principle it's something that does need to be addressed, with the use of control groups for instance. But I suspect the reason researchers in all these categories don't give this much priority is because the information that gets matched is often far too specific for it to be a matter of chance coincidence. The reason sceptics press for it is because they don't read the research, so they don't know that. (I can't believe that the 'cold reading' argument relating to mediums could survive a detailed knowledge of the Leonora Piper material, for instance, but sceptics seem hardly aware of its existence.)

Despite this, Angel argues rather forcefully that Stevenson's failing even to consider an experiment of the kind he suggests is an indication that he didn't care about the scientific method. This allows him to criticise his credentials - he uses words like 'laxity' and 'sloppiness'. It's rather clear that he doesn't intend it as a constructive criticism - surprise surprise - so much as a way of diminishing Stevenson's work, to discourage other people from taking it seriously, and give cheer to sceptics like himself who are deeply discomforted by its implications.

Just to be sure we get the point Angel constructs the article in the form of  a dialogue, in which the researcher is portrayed as querulous and evasive, flailing under the onslaught of our hero's incisive questioning. Stevenson is made to flounce out whenever the questioning becomes too hot, returning sheepishly each time to face his tormentor. A sample:

LA: So you think that a succession of case investigations such as you conducted for about forty years is sufficiently scientific?
IS: Yes. That is what I just said.
LA. True, that is what you said. But it seems to me what you did may only look as though it is scientific. Perhaps you didn't follow the most basic rule for scientific investigation.
IS: Please! To suggest that I failed to do the most basic thing is absurd...This is nonsense... I'll be gone. Then you won't be able to talk to me any longer.

As always, this sort of thing leads me to speculate about the sceptic's own thought processes. He can't for one minute accept the claims, because they contradict knowledge and assumptions which he believes to be unassailable. So his mind starts to generate a defensive strategy. This involves fastening onto a particular idea, and selectively picking up elements that seem to support it, while screening out those that would quickly refute it. Once he has detached himself from the research, the counter-strategy that has taken root is now free to flower in its own environment, unchallenged by any restrictions. It only needs to be fleshed out, in a way that represents the researcher as a cretin, to restore a sense of emotional equilibrium in himself and in his like-minded readers. 

That's not to suggest that the work of Stevenson and other researchers in the field should go unchallenged, or that experimental approaches cannot be found that might strengthen this type of field research. It's hard for many people - and I include myself - to read the reports without constantly interrogating the statements and at least sometimes wishing for extra information and reassurance. They are just flat-out incredible. But we can readily distinguish between a constructive suggestion and one whose real purpose is just to make sceptical propaganda.


Feed You can follow this conversation by subscribing to the comment feed for this post.

It all smacks of a Trevor Hall style debunking of a dead pioneer. Crookes, Palladino, Harry Price and countless others have been victims of this kind of thing.

Stevenson's work is incredibly important but IMO not near the top of the list of strongest cases in psychical history, only because of a necessary lack of control involved in these cases.

Good that you mentioned Piper, in her case this effort falls flat every time. Prominent "skeptics" like Martin Gardner have made efforts to dismiss her and her abilities but because this is an impossible task their attacks are so confused and selective that one is left wondering if they actually bothered to consider the positive evidence or if they went into the enterprise just to allay the fears of their brethren, who need constant reassurance that there isn't anything in this whole business.

Expect the debunking efforts to go on forever. But don't expect them to be substantive.

Part of Gary Schwartz's experiments with mediums have included controls for the 'general similarities' argument made by Angel. They include having subjects rate various 'readings' to see how much the information applies to them. I'm too lazy to look it up right now, but the readings NOT intended for a specific subject were rated about 30% accurate. However, those rated by the INTENDED subjects were 70 to 80% accurate. Of course, I am sure Angel would dismiss Schwartz before reading his papers...

On another note, take a look at Return From Heaven, by Carol Bowman, which despite it's unpromising cover, is a very good book showing a researcher applying Stevenson's methods to American cases. The degree of specificity of the past live memories, and the researchers thoroughness, is extremely persuasive. Personally, it moved me from very skeptical to undecided about reincarnation.

Ah, yes, Tony I was going to bring up Gary's work as well. This criticism is nothing new. Gary has responded in detail why Hyman's accusations, based on the same type of an idea, are invalid:

I am especially entertained by this part:


At the two-day meeting I convened of seven highly experienced professional mentalist magicians and cold readers, they all agreed that they could not apply their conventional mentalist tricks under these strict experimental conditions. However, a vocal subset (Hyman was one of the three), made the claim that if they had a year or two to practice, they might be able to figure out a way how to fake what the mediums were doing. My response to this vocal subset was simple. It was "show me."


I used to be terribly interested in these 'skeptic' vs 'paranormal investigator' battles, but I don't really care anymore. Skeptics concocting 'just so' stories and passing them off as legitimate criticism: it has somehow gotten old to me - and not very skeptical.

Personally, I don't understand the point of Angel's "refutation." He hasn't contributed anything substantially new to this debate. Stevenson and Tucker were/are the first ones to admit that the research is far from conclusive and that it is only suggestive at best.


please, see the link

There are very good analysis about the case Imad Elawar, Leonard Angel and Ian Stevenson.

The comments to this entry are closed.