[Help] [Aide] [Up]

Science Tribune - Article - September 1999


Unrepeatability: Parapsychology's only finding

Susan J. Blackmore

Dept. of Psychology, University of the West of England, Bristol BS16 2JP, UK.

This paper was presented at the Parapsychology Foundation 32nd International Conference, San Antonio, Texas, October 1983, and published in the Repeatability Problem in Parapsychology. Ed. B Shapin and L. Coly, New York, Parapsychology Foundation, 1985, 183-206.


Rereading this article after 16 years is like travelling back in time to a person I had almost forgotten I ever was. I still remember the events, and I still agree with most of the ideas I put forward then, but I no longer feel the anguish of being a sceptical parapsychologist who has had to change her mind in the light of the evidence.

After writing this article I spent nearly 15 more years hunting for psi and failing to find it, investigating fascinating experiences and explaining them in 'normal' not 'paranormal' ways, but eventually I had had enough (Blackmore SJ (1996) In search of the light. Amherst, New York, Prometheus, ISBN 1-57392-061-4). I am not sure whether parapsychology counts it as a loss, but I took the final option and left the field.

Meanwhile parapsychology has continued the search for psi (a). Now, as then, I meet parapsychologists who believe that at last they have found the repeatable experiment - whether that is Ganzfeld, remote viewing, remote staring, or some other experimental paradigm. I do not know whether they are right or not. I do know that it would involve a great deal of hard work for me to find out - and I am not, any longer prepared to do that work - and since I will not do it, I must simply accept that I do not know.

My own opinion, after all these years, is that psi probably does not exist. But I may be wrong, and any substantial theoretical progress might force me to change my mind all over again. For now I am content to wait and see.

Parapsychology is facing a crisis and its worst symptom, indeed perhaps the key to its disease, is the unrepeatability of psi. Amidst all the arguments about how much repeatability and of what kind we need in parapsychology, there are few who doubt that it is a real problem.

I shall discuss three approaches to tackling this problem which can be, and have been, used. These are:

a) to ignore it and hope it will go away,
b) to argue that parapsychology can do without repeatability,
c) to take unrepeatability as a reason for rejecting the hypothesis of psi.

I hope to persuade you that the last of these is the only viable solution if we are to have a thriving science of parapsychology in the future.

Before discussing these three, I must make clear what kind of repeatability I see as most critical and why it is such a fundamental problem. I shall not discuss the nature of repeatability in detail, others (e.g. (1)) have done so much better than I could. But briefly I can rule out some kinds as being inessential for parapsychology.

Inessential repeatability in parapsychology

For example, it is obviously true that the spontaneous phenomena, which form an important part of parapsychology, are not and cannot be expected to be repeatable at will. Once a ghost has been seen by someone, a poltergeist event has occurred, or someone has had an out of body experience (OBE), that event has been and gone and is unique. We cannot bring it back. But this, of course, presents no problem. Many sciences study such unique phenomena, from earthquakes or the single case histories of psychology to the origin of species. However, the crucial factor which makes all the difference is whether in a group of such events there is a meaningful pattern to be found. When there is, the process of hypothesis testing and theorizing can go on quite happily without any kind of repeating of the individual events. The question is whether in parapsychology we can find such repeating patterns.

Another kind of repeatability we need not have is strict methodological repeatability. It is not necessary that when you carry out an experiment it should be possible to get the same result again and again by repeating it. Many other sciences, especially psychology, have to rely on experiments which are far from ideally repeatable. There are even phenomena which seem to be repeatable at one time, but then cease to be later on, like the "risky shift" (b) in social psychology. This has apparently become harder to find nowadays, but this does not hamper the progress of the subject because there are still meaningful patterns. It can be suggested why the phenomenon has ceased to be repeatable and the relevant factors can in turn be used for further experimental tests. ln most fields this kind of process yields comprehensible results and the original phenomena are either accepted as valid, if the reasons for the unrepeatability can be understood, or rejected as spurious. In parapsychology we never seem either to find the relevant factors so that the pieces fall into place or reject the whole hypothesis. This cannot be blamed on the lack of this sort of repeatability. One hundred percent methodological repeatability is a luxury other sciences have learned to do without. Parapsychology could certainly do the same if only it could find patterns of some kind in its phenomena.

Parapsychology is based on a 'degenerating' research program

In finding repeatable patterns, it does not matter what form the patterns take, how complicated or simple they are, whether they involve obvious intervening variables or obscure ones, just so long as in the end they repeat themselves. It is to this kind of repeatability that I shall refer in my discussion.

Of course, what constitutes a pattern is relative to the theoretical structure being used. Theories change with changing findings. In fact, unrepeatability is a spur to theoretical advance because it forces the attempt to fit results into a pattern. To this extent the plea for repeating patterns is tantamount to a plea for a new theory of parapsychology. However, it is no good simply asking for a new theory - we have been trying, and failing, to find one for a hundred years. Rather, we should see that the problem lies deeper than this. Since its conception parapsychology has failed to make sense of its findings. I suggest this is because its whole research program has been fundamentally misguided.

Lakatos (2) provides a powerful philosophical framework within which parapsychology can be seen as based on a "degenerating research programme". He says (p. 112) :

"A research program is said to be progressing as long as its theoretical growth anticipates its empirical growth, that is as long as it keeps predicting novel facts with some success ('progressive problemshift'); it is stagnating if its theoretical growth lags behind its empirical growth, that is as long as it gives only post hoc explanations either of chance discoveries or of facts anticipated by, and discovered in, a rival programme ('degenerating problemshift'). If a research programme progressively explains more than a rival, it 'supersedes' it, and the rival can be eliminated (or, if you wish, 'shelved'). "

I would say that judged against these criteria parapsychology has long been based on a degenerating research program. Its theoretical growth has not anticipated its empirical growth. In fact, parapsychology abounds with data in search of a theory. Even the classic "repeatable" findings such as the sheep-goat effect or positional effects (c) remain isolated in weak theoretical structures applicable only to very limited domains. There are some isolated instances where theory has predicted novel facts; for example, the prediction of retroactive PK (d) within the observational theories. It is effects like this which might lead to progress, but so far we cannot claim it has been achieved.

If we truly have a degenerating research program, why has it not been superseded and abandoned already? I suggest that it is because there is no better rival at the moment. Other scientists have not been sufficiently interested in our strange phenomena and have not done enough work to provide more progressive alternatives. However, I am sure they soon will (as I discuss below). We shall not much longer be able to maintain our precarious position as a dominant, while degenerating, program. This makes the future look very bleak indeed for what I call traditional parapsychology. It highlights the depth and significance of the repeatability problem and makes an imminent solution imperative.

Three approaches to solving our problem

I shall now discuss three approaches to solving our problem.

Ignore it and hope it will go away.

To some extent this is what has happened for a long time. It is not quite true that repeatability has been totally ignored. It was mentioned by Rhine (3) and many later commentators such as Murphy (4) and Thouless (5) and more recently discussed in detail by Ejvegaard (6) and Sargent (1). However, on the whole parapsychology has simply gone on with its business without regard for the fact that psi appears to be unrepeatable. The big question is, what will happen if we continue to do this.

I do not think it is entirely unreasonable to hope that the problem will suddenly go away; that research in parapsychology will suddenly become progressive. It is quite possible that someone is going to find the theoretical advance which suddenly makes sense of all our messy findings; that suddenly everything will fall into place and we shall all wonder why we had never thought of it before; that some crucial variable will suddenly be located and at last meaningful patterns will emerge. We shall be able to repeat our findings and overhaul our research program.

It may happen. This has happened in other sciences and it may happen in parapsychology. Probably every new parapsychologist thinks he is going to be the one to find the key (I must admit I have been guilty of such pretensions). However, we cannot assess the chances of this happening and time is running out. It is simply not good enough just to go on and on hoping. At some point we have to say that we have given it long enough and it is time for a fundamental change. Is one hundred years of research long enough? For me, personally, it is.

I am not arguing that everyone should give up hoping, only that there are precious little grounds for continuing hope. We shall be very lucky indeed if this approach proves to be the solution to our problem.

Take a revolutionary stance and argue that parapsychology can do without repeatability.

Parapsychology is sometimes held to be a revolutionary science, and some parapsychologists have argued that we should start behaving like revolutionaries as befits our subject. Others, like Hövelmann (7), have roundly criticized this stance.

There is an interesting paradox here. In some ways parapsychology is potentially revolutionary. At least, if some of its claims are valid, then many accepted tenets of science would have to be overthrown and this would certainly mean a revolution in science. However, the actual process of parapsychology seems to be almost the opposite of revolutionary. It uses the most conservative of research methods, sticks to traditional types of research design and, while having made great advances in standards of method, has not stopped outside the normal methodologies of science.

It can be and has been argued that we have to stick to such methods because of fear of ridicule and because we need to fight so hard to retain a place in acceptable science. I do not believe this is necessary for a moment. Science is extraordinarily flexible. If we could find any kind of repeatable patterns in psi, I am sure that we could use any sort of revolutionary methods and be acceptable as a science, if these methods led to recognizable progress.

Here, of course, we must face the question of what is to count as progress. I don't wish to argue the merits of different philosophies of science, but Lakatos does at least provide a prescription for what is to count as progress. As we have seen, judged against his criteria, parapsychology has made very little (if any) progress in its study of psi. I am convinced that if we did make this kind of progress, then it would not matter to anyone else how it was achieved. Parapsychology would be accepted as a science if it progressed as a science and, if that means using revolutionary methods, then let us use them.

I have, therefore, tried to explore the possibility that a subject like ours needs to forge a way ahead without relying on the old fashioned notion of repeatability. However, I have failed.

Imagine a parapsychology which admitted that none of its findings were expected to be repeatable (in the sense I have discussed). There could still be investigations of spontaneous cases, of poltergeists, hauntings, reported precognitions and so on, and one could still make collections of cases, but they could not be used to find repeating patterns and so to relate to theories. Experiments could still be carried out, but there would be little point in them. The whole basis of experimental science is that the results are assumed to be generalizable. Just how far they can be generalized is always dependent on the way in which the experiments were carried out and the theoretical framework in which they were designed. In some cases one has to be content with very little generalizability, but some is essential if the results are to have a bearing on the theory they were designed to test or the question they were designed to answer. Theorizing would, of course, continue, but in the absence of any claim for repeatability it would be totally divorced from experimental and other research findings. I believe that this would condemn parapsychology to be a pseudoscience. Like much of astrology, palmistry and other pseudosciences, its theories would proliferate away from the constraints of nature's answers.

I would argue that parapsychology has come perilously close to this situation, but not yet reached it. We have never given up the idea that we carry out experiments in order to find out the truth, or some improved version of it. Indeed parapsychology's main strength has been its adherence to the experimental paradigm with its basis in generalizability and repeatability. I submit that if we abandoned this we could no longer have a science of parapsychology and this is sufficient reason for me to reject this approach.

Take unrepeatability as a reason for rejecting the whole research program built on psi.

This may sound a drastic and even depressing prospect, but I hope to persuade you that it is not and, in fact, that this route provides the only hope for the future of parapsychology.

To do this I want to explain the reasons and the research which lead me to arrive at this conclusion.

A personal story: Looking for psi and not finding it

I began my research in parapsychology ten years ago, with an idea which I was sure was right. It is a very exciting feeling, being certain that you are on to something really good, something that will change the face of science, or at least of parapsychology. I have had this same arrogant feeling since. However, there is now a difference. l know that in the past I have had ideas that I was sure were right, only to find they were wrong.

One such idea was my memory theory of extra-sensory perception (ESP). This seemed to be so promising. It seemed to explain so much, which at the time I mistakenly took for an advantage. It seemed to account for all those manifestations of psi which I had heard about, thought I had experienced for myself and assumed to be true. It also fitted with the fact that then, as now, psychology did not understand the physiological basis of memory. It seemed perfectly feasible that memory was stored in some way outside of the individual brain.

Of course, I wanted to test my theory. I was convinced that it was right, or at least on the right track, so I set about designing experiments (8) (9) (10) (11). Over several years these gave me the unequivocal answer - the theory was wrong. Whichever way I tried to make it work, it would not. I did not find the predicted effects and could not make sense out of the data my experiments produced. Very reluctantly I had to admit my error. It was a painful process, but obviously the only possible one. There is no point doing experimental science unless you believe the results of your experiments.

The trouble was that I had to apply the same principle to my belief in psi. It was something which I had originally taken for granted and not questioned in all my theorizing. However, the results of my experiments were telling me that I was getting no psi. At least, any significant results I got were unrepeatable. At one point in my research I calculated that I had performed 34 independent significance tests for psi effects and obtained two significant at the 0.05 level. In the face of these results I began to question the existence of psi. In other words, I became a skeptic.

I did not "give in" suddenly or easily. As I described in my paper ("Prospects for a psi-inhibitory experimenter") presented at the Parapsychological Association Convention (Cambridge, Aug. 1982) (12), I tried out many routes. I varied my experimental conditions and tried different subjects under different circumstances. I used young children as subjects (10) and tried Ganzfeld conditions (e). None of these changed the results I was getting. Most promising seemed the idea that it was an experimenter effect and I set about examining this possibilitv. After all, it was possible that other parapsychologists were actually getting psi and I was not. Indeed there were only two possibilities :
either others were getting psi
or there was misinterpretation, incompetence, error, self-delusion or deceit on a scale which was, to me, unimaginable.
I was, therefore, compelled to try to find out which of these alternatives was correct.

I did not visit the major parapsychology laboratories in the United States and therefore what I have to say does not refer to them. I cannot judge fairly whether the experiments of Schmidt, Honorton, Stanford, Targ or Jahn or many others actually show psi. I have only the published reports to go on, and these will of course be given different interpretations bv the skeptic and the believer. Now that I have become a skeptic I can easily convince myself (whether rightly or wrongly) that there are sufficient loopholes or omissions in any report.

What I did do was to look in some detail into the research being carried out at that time in England. Ernesto Spinelli was getting remarkable results with young children and since we were both at Surrey University I had the opportunity to follow his work in some detail. I also took part as a subject in Serena Roney-Dougal's Ganzfeld research and was involved in some aspects of her experimental design. And in 1979 I visited the laboratory of Carl Sargent in Cambridge with the intention of finding out why it was that he could get results using the Ganzfeld (13), while I could not (8). I subsequently wrote a report of this visit (14).

To cut a long story short, none of these persuaded me that my doubt had been misplaced. I did not conclude that an experimenter effect was responsible for the differences between my results and those of others. My skepticism deepened.

Now, it is possible to argue that by then I had become so biased and so hardened in my skepticism that I would not have admitted it if I had looked a psi source in the face. However, I can only plead that we have one criterion by which to judge and that is whether we make any scientific progress by taking one view or the other. This is the criterion I have tried to follow and it has led me inexorably to one conclusion.

I submit that after one hundred years the only remarkable thing about parapsychology is how little progress we have made. As far as psi is concerned we have only one really repeatable finding - its unrepeatability. We should take this finding seriously and admit that our research program based on the psi hypothesis has failed.

A parapsychology with or without psi

The question then becomes "Where do we go from here?" I apologize for saying the same as I said last year at the Parapsychology Foundation's 1982 conference, Parapsychology's Second Century, but I believe that parapsychology faces a stark choice and we do not have all that long in which to make up our minds. The choice is between a parapsychology with or without psi. I hope to persuade you that we could have a bright future without psi.

Let us compare the two possible futures, based on two fundamentally different research programs :

1. We may carry on as before.

Since its inception parapsychology has been based on the study of paranormal phenomena, with their negative definition. This has long been considered problematic (e.g.,(15)), but I believe it is actually the root of the whole problem.

One can look at different sciences as being either hypothesis-driven or phenomenon-driven. Most traditional areas of science are phenomenon-driven. By that I mean that they define themselves by the phenomena they study and stick to those phenomena while the theories and hypotheses they use change with time. Examples might be chemistry, biology or entomology. Others are rooted in particular theories, such as quantum mechanics or thermodynamics and the phenomena they study change as the applications of the theory change. But these theories are of a sophistication and application of quite a different order to anything based on psi. I don't wish to draw an artificially clear line here, but only to point out that parapsychology has always been hypothesis-driven and perhaps prematurely so. We have defined our subject matter as the study of psi phenomena when psi itself is hypothetical and negatively defined. The hypothesis may be useless and the class of psi phenomena empty.

The consequences of this are dire. A parapsychology based on the paranormal is an ever shrinking field. We long ago lost hypnosis and certain aspects of trance states. Some may say "Good riddance," but we are now in danger of losing far more crucial parts. Early psychical research included as an important part of its subject matter all the phenomena surrounding death, especially death-bed and near-death experiences (NDEs). Since then these have been almost exclusively studied by parapsychologists, but now there is a new society, founded to study just these areas, and a new journal full of articles about it. They are making progress away from parapsychology. There is also another recently started journal on lucid dreams and the new journal Imagery invites papers on lucid dreams and related states. There, too, progress seems to be being made in topics traditionally a part of parapsychology.

Papers on OBEs are increasingly being published in places other than parapsychological publications. It cannot be long before these topics are lost to us, and how much longer will we have any claim over apparitions, hauntings or poltergeists? If other fields can take up these topics and make progress with them, without using the psi hypothesis, then a parapsychology based on psi will have lost them.

There is another important source of loss. As I have previously argued, there is precious little room within parapsychology for the "psi-inhibitory experimenter." Having tried all the other routes open to him, he is all too often forced to take the last and leave the field. If there is no psi, these experimenters can only be seen as victims of their own honesty and their departure should be counted another loss.

This emphasizes the point that it is a very dangerous game to rely exclusively on one hypothesis, especially if that hypothesis is as shaky as that of psi. It forces one into more and more defensive maneuvers to protect the core hypothesis. lt forces an artificial demarcation between the in-group (of believers) and the out-group (of skeptics) and a vast gulf to be crossed if you want to pass between one and the other. In fact, it makes scientific principles very hard to maintain.

Traditional parapsychology is in a dangerous position now. It can probably go on shrinking and defending itself for some time to come, but not forever. Ultimately it will end up with nothing but the psi hypothesis and that may still produce nothing but unrepeatability. If this is the route parapsychology decides to take, I don't want to be there when the end comes.

2. Now, let us consider a different scenario.

Let us redefine parapsychology.

Redefining parapsychology as a phenomenon-driven science

Instead of basing parapsychology on a hypothesis which only leads to unrepeatability, let us base it on some phenomena which we know occur; in other words, admit that it is an undeveloped science and make it phenomenon-driven. There need be no final list of phenomena, but we should certainly include the familiar topics of our field, claims of telepathy, clairvoyance and precognition, apparent PK, poltergeists, hauntings, ghosts and apparitions, NDEs, OBEs, mystical experiences and altered states of consciousness which seem to suggest other planes of mental functioning.

There are at least four good reasons for supposing that these could provide the basis for an acceptable and progressive research program for the science of parapsychology.

They are what we have been studying all along and so no perceptible change in our area of study is necessary. There would be a great change in our underlying assumptions, the basis of our research program, but little change in what we are studying.

It would certainly satisfy the founders of psychical research and parapsychology (were they still around to judge us). The founders of the Society for Psychical Research defined their subject matter as "those faculties of man, real or supposed, which appear to be inexplicable on any generally recognized hypothesis." They never listed what those faculties were, but they certainly intended the phenomena I have listed to be included, as can be judged from great works like Human Personality (16) or Phantasms of the Living (17). Also, they explicitly stated that these were to be a part of psychical research whether they were "real or supposed". Moreover, they were to be examined "without prejudice or prepossession and in a scientific spirit".

I don't think we have done justice to their intentions. We have concentrated too much on the inexplicability and too little on the phenomena themselves. Many of the phenomena may be real, but the interpretations we have given them are all too often "supposed". I think we have been guilty of studying with prejudice and prepossession and of losing sight of that scientific spirit. We could do worse than base our subject on the original intentions of Myers, Sidgwick, Gurney and others.

Remember also what the motivations of J. B. Rhine were. He spent the greater part of his life working on ESP, but his real motivation was to study the mysteries of the mind (18) and when he founded his independent laboratory it was called the Foundation for Research on the Nature of Man. Although he would, of course, disagree, I think we shall be closer to studying the nature of man if we abandon the psi hypothesis.

III. It is clear that there are still whole tracts of human nature which nobody else wants to study, even though there is now a vast new range of theoretical and experimental tools to apply which our founders did not have. A recent book on functional states of the brain (19) quotes Professor Laycock who, in 1859, proclaimed that "cerebral physiology was the only sound basis for a mental science from which all mysticism will be as effectively excluded as it is now from physical sciences". Venables adds "... a sentiment to which all contributors to this volume would undoubtedly subscribe". Whether or not he spoke for them all, it is clear that many psychologists do not want anything to do with mysticism or any of the fascinating topics which I have listed. If we can make real progress with any of them, we still have time to establish parapsychologv as a worthwhile science with its own subject matter and genuine bridges to other sciences. If we could do this, there would no longer be any question of being acceptable to "Science", whatever our subject matter. But we don't have much time and we shall never do this by basing everything on psi.

IV. The fourth, and final, reason is simple and crucial. The repeatability problem suddenly disappears. If you stop looking for psi and patterns derived from the psi hypothesis and look instead at the phenomena themselves, there is plenty of repeatability. Claims for paranormal events tend to take on certain patterns and certain types of people report them under similar kinds of conditions. The historical collections of cases such as that of Louisa Rhine (3) show this. Poltergeists, hauntings and ghosts also provide patterns which can tell us a lot about human nature if we cease looking for the paranormal in them. And, of course, NDEs, OBEs and mystical experiences have striking similarities across times and cultures: similarities which, I believe, we shall only understand by taking off the blinkers of thinking they are paranormal. If only we would follow the repeatable instead of the paranormal, I believe we should have a fascinating task ahead.

Now, many of you may feel that this isn't what you want of parapsychology; that this is not what it is all about. If so, then, of course, you will ignore my plea, but I would like to end by explaining a little of my own interest and how I think this new parapsychology can contribute to understanding some particularly interesting aspects of human nature.

The new parapsychology: a personal viewpoint

I am currently working on the OBE. This is mainly because in 1970 I had a long and profoundly challenging OBE. This was not my only reason for becoming involved in parapsychology. I was already running the Oxford University SPR at the time, but it certainly contributed to my continuing motivation.

I did not immediately start working on the OBE because I could see no way to do so. Working on ESP seemed far more acceptable and initially feasible. When I did finally start work on OBEs it was from the point of view of traditional paranormal explanations. I was also fascinated with the occult traditions of astral projection and the concepts of the different planes, the lower and higher astral and the mental, causal and other planes said to be beyond. I seemed to have glimpsed some of these planes and knew what the occultists were talking about, but these did not seem amenable to any kind of scientific investigation. Instead, I struggled with traditional parapsychological approaches. However, both my own experience and my later research guided me towards the conclusion that these planes, and the whole of the out-of-body world, are created by the imagination. There seemed to be sense behind the occult notions of the planes, though not behind a paranormal interpretation.

There is nothing which now infuriates me more than people who accuse me of saying that the OBE is "nothing but imagination". For if I have discovered anything, it is that the imagination holds a good many surprises for us yet. I am sure that it holds far more fascination than any paranormal hypothesis could.

I would now suggest that a productive way of looking at OBEs in particular and altered states of consciousness in general, is in terms of the model of "reality" which is being used at the time.

We all continuously construct models of ourselves and the world around us. In fact, the whole process of perception is one of model building and the final result is a model of a stable self perceiving a stable world outside. This model is built up from sensory input with a great deal of additional information from memory and from the cognitive map.

Most of the time we build the sort of model of self which is going to be most useful for interacting with the world; that is, based on the physical body. However, we may build all sorts of other models if we are able or if the conditions either require or allow. From this viewpoint the OBE can be seen as the natural consequence of conditions (such as partial sensory deprivation) in which the usual model cannot be maintained. As a result it may be replaced by a model which is partially based on information from the senses, but distorted as to the position of the observer. This can be seen as the very first step away from a perceptually tied and restricted model. On the other hand, one may create models which are much further away and much freer from sensory restriction. This applies to certain states near death, to lucid dreaming and to certain states achieved through meditation and prayer. In mystical states the model of self may be as of the whole world or as an insignificant pin-point compared with the power of God. It is not surprising that such visions of oneself can have profound effects on one's beliefs and attitudes towards others.

If we assume that these are all models constructed by the brain, then of course the kinds of models which are possible will be constrained by the structure of the brain. In this way the possible range of altered states of consciousness is ultimately constrained by brain processes. This explains why different people in different times and cultures have nevertheless experienced comparable states of consciousness and are able to communicate them. It explains also why we should expect to find things like the lower and higher astral and the other planes beyond. They are all reflections of the vast, but finite capacity of the human mind to model itself.

I should add that I am not advocating trying to understand these models in terms of brain processes. Rather, I suggest we shall progress by working at the level of the models of "reality", while assuming that ultimately constraints are applied by the physical processes underlying their construction.

There are few psychologists who would want to see psychology interesting itself in this kind of area, especially where it relates to occult notions. Perhaps there are few parapsychologists, too. But I do hope there will at least be some, for I deeply believe that the explorations of this aspect of man's nature will lead us to real insights. It may even lead to a rapprochement with occult, mystical and religious traditions which have been rejected as mere superstition in the past and even to the solution of some of the puzzles about consciousness.

This whole endeavor may turn out to be totally misguided, but at least it provides a new framework within which to explore some traditionally "psychic" experiences. It raises a mass of new questions to be asked and new experiments to be carried out. What we have to ascertain now is whether this theoretical progress will anticipate empirical growth or not. Even if it does not and has to be rejected, I believe it provides hope for a research program which is more progressive than that which we have had for a hundred years.

If there is a revolution imminent, I suggest that it will not be the overthrow of the rest of science by parapsychology, but the overthrow of psi-based parapsychology by a new parapsychology. Parapsychology has only one real choice and I think we must take it soon. Psi is unrepeatable - we need a new hypothesis.


a Psi: A general term covering all paranormal phenomena including telepathy, clairvoyance, precognition and psychokinesis, or the supposed mechanism underlying them.

b Risky shift: The finding that people in groups make more risky decisions than they do individually.

c Sheep-goat effect: Believers in psi (sheep) score higher in extrasensory perception (ESP) tests than disbelievers (goats). Positional effects: Effects that depend on the position of a guess in a run of guesses, e.g. the decline effect (scores are better at the start of a run and then decline).

d Retroactive PK: Psychokinesis operating backward in time.

e Ganzfeld: A technique for producing partial sensory deprivation. The participant wears half ping-pong balls over the eyes, hears noise through headphones and relaxes on a comfortable chair or bed..


1. . Sargent CL. The repeatability of significance and the significance of repeatability. Eur J Parapsychol 3, 423-443, 1981.

2. . Lakatos I. The methodology of scientific research programmes. Philosophical Papers, vol 1. Cambridge University Press, 1978.

3. . Rhine LE. The invisible picture : A study of psychic experiences. McFarland, Jefferson NC, 1981.

4. . Murphy G. Challenge of psychical research. Harper, New York, 1961.

5. . Thouless RH. From anecdote to experiment in psychical research. Routledge and Kegan Paul, London, 1972 .

6. . Ejvegaard R. Parapsychology and repeatability. Eur J Parapsychol 3, 409-422, 1981.

7. . Hövelmann G. Seven recommendations for the future practice of parapsychology. Zetetic Scholar, 11, 123-138, 1983.

8. . Blackmore SJ. Extrasensory perception as a cognitive process. Unpublished Ph.D. thesis. University of Surrey, 1980.

9. . Blackmore SJ. Correlations between ESP and memory. Eur J Parapsychol 3, 127-147, 1980.

10. . Blackmore SJ. A study of memory and ESP in young children. J Soc Psychical Res 50, 501-520, 1980.

11. . Blackmore SJ. Errors and confusions in ESP. Eur J Parapsychol 4, 49-70, 1981.

12 Blackmore SJ. Prospects for a psi-inhibitory experimenter. Research in Parapsychology (Eds WG Roll, J Beloff, RA White). Metuchen, New Jersey, Scarecrow, 17-20, 1982.

13. . Sargent CL. Exploring psy in the Ganzfeld. Parapsychology Foundation, New York, 1980.

14. . Blackmore SJ. A report of a visit to Carl Sargent's laboratory. Journal of the Society for Psychical Research, 54, 186-198, 1987.

15. . Boring EG. Introduction to Hansel CEM. ESP: A scientific evaluation. Scribner's, New York, 1966.

16. . Myers FWH. Human personality and its survival of bodily death. Longmans, Green & Co., London, 1903.

17. . Gurney E, Myers FWH, Podmore F. Phantasms of the living. Trubner & Co, London, 1886.

18. . Brian D. The enchanted voyager: The life of J.B. Rhine. Prentice Hall, Englewood Cliffs, NJ, 1982.

19. . Koukkou M, Lehmann D, Angst J (eds). Functional states of the brain: Their determinants. Elsevier, Amsterdam, 1980.