Nicholas Humphrey's Soul Searching: Human Nature and Supernatural Belief claims to be a "devastating critique" of the paranormal. The author's reputation as a psychologist, his racy style, and the superficially attractive nature of some of his arguments may lead those who do not examine the book too carefully to believe that the awkward subject of the paranormal has been disposed of. However, critical examination of the arguments reveals this not to be the case.

An apparently strong argument is one that asserts that paranormal powers are being tested all the time in experiments on normal abilities: "In literally thousands of experiments (psychologists) have established that there really are sensory stimuli that people cannot see, cannot hear . . .", and which therefore have "incidentally, but nonetheless conclusively, shown that ESP (extrasensory perception) does not occur". The argument is a spurious one. Supposing I do have ESP abilities and someone shows me smaller and smaller letters and keeps asking if I can still recognise them. Will my ESP abilities result in my going on indefinitely reporting that I can see the letters? Clearly not, any more than if a person were helping me by shouting out to me what the letters were.

The argument would be more valid if applied to experiments such as those where a person is asked to guess what pattern is being displayed when that pattern is made fainter and fainter, since ESP abilities might be expected to increase the probability that a person makes a correct guess. But here the problem is that as this is a statistical effect it will not show up in a definitive way unless large amounts of data are taken. Unless the situation is analysed properly we just cannot say from the absence of a determinable statistical effect that no influence from ESP is there.

Humphrey unwittingly makes this very point when he alludes to a survey of 300 students designed to discover if they were aware of the current phase of the moon that yielded only chance results: "Such ignorance . . . would seem remarkable by any standards. But if ESP is a reality it ought surely to be considered doubly remarkable." If in such an experiment a sample size of 300 is insufficient to demonstrate the fact that there are some people who know reliably the phase of the moon, why should anything at all be inferred from the fact that it did not reveal clairvoyant knowledge of the phase of the moon by some people either?

ÁñÁ«ÊÓƵ

Another section of the book deals with arguments dignified by the name The Argument of Unwarranted Design: "If a phenomenon shows signs of being unduly restricted in its form and manner of occurrence, so that our theory of its underlying cause provides us with no principled reason why it should take just the peculiar form it does, then we should suspect that the true cause of the phenomenon lies elsewhere." The argument might equally be called The Argument of Unreasonable Constraints, which Humphrey applies by deeming it unreasonable, and hence suspicious, that a person could perform one kind of psychical skill effectively but not another. But why should it be unreasonable? Humphrey writes books on psychology but not on physics; should we then think that suspicious, or ascribe it merely to his having practised activities relevant to the one skill but not to another: a kind of explanation surely equally applicable to the area of psychical skills?

Further along in the book we come to arguments against psi (paranormal faculties) which claim to be of a purely logical character, though in fact - the original fallacious argument published in Darwin College Magazine having been laid to rest and superseded here by a different one - they are in essence assertions of the unimaginability of mechanisms for particular manifestations of psi. It is suggested for example that since two different people's thoughts are encoded by neuronal firing patterns in such individual ways there is no imaginable mechanism by which two such people could communicate their thoughts to each other by ESP. The invalidity of this argument is demonstrated by the fact that people can to some extent communicate with each other by gestures even if they have never met before and do not know each other's languages. It simply is not the case that the complex patterns of neural activity are not translatable into a common code.

ÁñÁ«ÊÓƵ

The awkward matter of the experimental evidence in favour of psi is dealt with by means of what appears to be sleight of hand, suggesting that the author has profited well from his association with members of the Magic Circle. Experiments are divided into two kinds, those with "no independent confirmation", and those where the results "might possibly have been due to something other than ESP". The former category Humphrey "leave(s) on one side for the moment" (exactly where it is dealt with is not made clear), and the latter is disposed of by reference to a single case, relating to an experiment by Daryl Bem and Charles Honorton designed to test for ESP abilities using the ganzfeld technique. Analyses by Richard Wiseman and colleagues showed that, in principle, if the ESP sender shouted his approval every time he heard through his headphones that the receiver had said something related to the target, the experimenter could have faintly heard the sound and unconsciously cued the receiver. Humphrey focuses on this possibility, but does not mention the fact that Wiseman et al did a statistical analysis and concluded that the idea that the possible sound leakage helped the receiver get the correct answer was not supported by the data. He equally fails to mention that statistical analysis of experiments on psychic phenomena designed to test the hypothesis that design faults have been the cause of successful results have not supported that idea.

In a recent article, statistician Jessica Utts commented that there seems at this time little point in doing experiments designed solely to prove the existence of psi, since if the critics do not accept the present evidence it is hard to imagine what one can realistically do further that would lead to them becoming convinced in the future. Soul Searching seems to illustrate the point very well.

Where not engaged in trying to show that belief in paranormal events is mistaken, Humphrey has some interesting ideas, and the book can be recommended as a comprehensive analysis of the factors that lead people to believe in things for which there is no real evidence. It is tempting, of course, to point out that these analyses might very well be applied to the sceptics instead.

More deserving the appellation "devastating critique" is Michel Schiff's The Memory of Water: Homoeopathy and the Battle of Ideas in the New Science. Technical in places but in general explained in such a way as to be accessible to the general reader, it details the struggles that new ideas in science have had and are still having to get a hearing, faced as they are with the variety of means, normally used in an unexceptionable manner, that editors, referees and review panels, and so on have at their disposal to prevent work that they consider unsatisfactory from being published or funded. The general directions of the author's critique may be indicated by a selection of his headings: "It is impossible a priori, hence it never happened", "Debunking as a substitute for scientific arguments", "Censorship as part of the normal scientific process", "Mock attempts to duplicate an experiment", and "A scientific exploration gets paralysed by the burden of proof".

As a historical example, Schiff cites the case of the Hungarian obstetrician Ignazius Semmelweis, who 20 years before the discovery of bacteria by Pasteur showed that deaths from puerperal fever could be reduced if the doctors were to wash their hands with antiseptic before attending their patients and was ridiculed for his proposals, and as a current parallel the suppression of evidence gained by Schiff's colleague Jacques Benveniste that particular kinds of saline solution might have had adverse effects on patients in whom they were injected.

ÁñÁ«ÊÓƵ

Much of the discussion relates to Benveniste's work on homeopathy and the "memory of water", which expressions, the author observes in his introduction, are "capable of turning a peaceful and intelligent person into a violently irrational one". Benveniste's in vitro experiments on homeopathically prepared samples met with a hostile response from Nature and its referees when he submitted the work for publication there, but since they could not point to any errors in it the editor eventually agreed to publication under the curious condition that after publication Benveniste would allow a team of investigators to carry out investigations at his laboratory.

Schiff lists a number of errors that he claims are present in the published investigators' report, as also in published reports of failure to confirm the Benveniste results by other scientists. Publication of a successful replication by Benveniste was refused on the basis of a referee's report which, according to Schiff, contained elementary mistakes such as confusing error and variance (i.e. error squared).

Psychologist Lawrence LeShan's The Medium, The Mystic and the Physicist has been republished by Penguin Arkana after being out of print for a number of years. In it, LeShan describes his attempts to answer the question "what is the nature of the paranormal?". Like Schiff, LeShan observes how he repeatedly encountered an attitude describable as "I have made up my mind; don't bother me with the facts". He decided to concentrate on the question "How does the world look to the psychic?", and found there were universal regularities in psychics' experience. For example, while in ordinary reality distance between two entities separates them so they can communicate only through an intervening medium, for the psychic two objects are part of a unity and distance cannot separate them, especially when there is love between the two entities.

ÁñÁ«ÊÓƵ

The author then decided to test the concepts he had arrived at by trying to teach himself the ability to heal psychically. Using symbols appropriate to the situation and other means he attempted to put himself into a state where he would understand the clairvoyant reality and "know" that the individual he was attempting to heal was not just an embodied being but also one that like himself existed in all of space and time. He felt that his successes at healing in this way were frequent and impressive enough to justify this as a method of exploring this alternative state. The theory could also be used to teach others both healing and clairvoyance skills. Later on in the book LeShan discusses matters such as the by now familiar parallels between mystical accounts of reality and those of 20th-century physics.

The clairvoyant reality as disclosed by this book is, from conventional points of view, strongly counterintuitive. We may need to appreciate this reality properly in order to make progress in understanding it. The original publication of this book made less of an impression than it might have done, possibly because it offered no very clear avenues for research. The situation has perhaps by now changed, making the republication of the book a potential catalyst for new scientific progress.

Brian D. Josephson, Nobel laureate, is professor of physics, University of Cambridge; he heads the Mind-Matter Unification Project at the Cavendish Laboratory.