QUESTION: When a free afternoon has been provided in the middle of a four-day
biological convention to give some relief from listening to technical papers,
what would be better than a picnic or a sight-seeing tour?
Answer: Another scientific meeting to hear two more technical papers - by two
parapsychologists invited especially for the occasion!
This, at least, was the answer the Canadian Physiological Society gave to the
question when it met in Winnipeg in June 1960. Those responsible for planning
the activities for the free Thursday afternoon boldly decided to schedule more
of the serious business of science while the other organizations were
interrupting their conventions to relax and play. The physiologists asked Dr.
Gardner Murphy of the Menninger Foundation to talk on the investigation of ESP,
and they invited me to speak on the laboratory research on PK (psychokinesis:
the direct influence of mind upon matter). The members of the arrangements
committee warned us before the meeting that we might be speaking largely to
empty seats. To their surprise and delight, the large lecture hall was filled,
and the meeting time was extended for a second hour until the chairman had to
call a halt because of other scheduled activities. Those who were too interested
to heed the closing gavel took Dr. Murphy and me to dinner, and about thirty of
them asked us to continue the discussion, which we did until about ten o'clock
in the evening.
What did we say that so captured and held their attention? Dr. Murphy's topic
dealt, of course, with matters such as those already covered in Chapters 2 and 3
of this book*. My own paper dealt with the research developments that will be
described in the next several pages.
* ISS note: Chapter 2 - "Puzzling Experiences";
Chapter 3 - "The Breakthrough in ESP: Telepathy and Clairvoyance
Can wishing for something help to make it happen? I do not mean can it help
anyone work harder to get what he wants, but can the mind somehow act to
influence directly, even if to only the slightest degree, the course of physical
events? This idea is not a stranger to any of us, for we all went through the
period of childhood when the heroes of fairy tales and mythology were our daily
companions. We admired them when they triumphed through their good wishes, and
most of us have thought of what we could do if only we were granted our
Yes, you may say, but surely we left this all behind us, even long before we
But do we really outgrow this fanciful way of thinking? Must we accept as final
truth the teachings of modem scientists who would sentence the mind to life
imprisonment in the physical brain and who would deny it not only any wishing
power but all other powers as well? If they are right, yes, of course we must.
But, if they are not right in wanting to banish the power of thought from
the universe, what could be more important than using the methods of science to
keep man from literally losing his mind?
The discovery that we have abilities which allow us to know about things
happening at a distance, beyond the reach of our senses - that is, the discovery
of even the simplest forms of ESP - may by itself be enough to prove that mind
exists. At least, no one has yet developed a theory that succeeds in explaining
telepathy and clairvoyance in purely physical terms. But scientists should not
be satisfied simply because they have demonstrated experimentally some of
the unique actions of mind. If there are ways other than ESP by which its
existence may be known and its nature investigated, we must explore and attempt
to explain these other psi mysteries as well.
In daily life, the use of the senses is closely linked with things that we do. I
see the open box of candy and I reach for a piece and plop
it into my mouth. You hear your name, and you turn your head to
see who is calling you. In fact, almost every instance of sensory perception
leads to some action or muscular response.
Does not this close linkage of sensing and doing in ordinary
experience justify our wondering whether the same is not true in the realm of
psi phenomena? Does not ESP, in which the mind is directly influenced by events
in the outside world, lead us to expect that PK should also occur? If you think,
for example, of clairvoyance as the influence of matter upon mind, should we not
find that the same general psi capacity works in the opposite direction - an
effect of mind over matter?
If you feel that this line of reasoning is too far removed from daily life to be
taken seriously, pay a visit to the nearest bowling alley. There you will see
husky youths, hard-headed men from the workaday world of industry and finance,
and practical housewives all demonstrating that they do not stop trying to
influence the ball after they have released it. The bowler sways and strains to
guide the ball as it moves toward the pins at the far end of the alley. True, if
you ask him whether his efforts are doing any good, he is likely to say no. But
the fact remains that he acts, under the stress of the moment, as if he
is guiding the ball after it has been sent on its way. If the very idea of mind
over matter is so foreign to our vaunted scientific culture, why should we all
so openly fall uncritically into superstition under stress, when we forget about
May the truth not be that we really know, in spite of all the efforts to
educate us to the contrary, that the mind has a force of its own that may at
times be called upon to influence a physical situation, such as a faulty bowling
As long as we remember that this question will not answer itself, but that it
must be decided by careful, objective, experimental research, what do we have to
lose by raising the issue? For science and for mankind generally, the
irreparable loss would come from mistakenly closing our eyes to the possibility
that there may be something to the old and familiar idea of mind over matter,
something just waiting for the right scientific method and the right moment in
the advancing tide of research to be discovered.
Thomas Huxley said: "It is the customary fate of new truths to begin as heresies
and to end as superstitions." But the old adage of mind over matter has long
been so very prominently labeled as superstition without ever having been
credited as truth that it is doubtful if it could be brought into the arena of
scientific investigation through purely logical considerations alone.
Fortunately, Nature herself occasionally gives us some encouragement to think
that there is something to PK. Mysterious physical effects in the realm of
experience, spontaneous unexplained objective events that appear to have some
personal significance, do take place. These PK experiences are not so frequent
as those in the area of ESP, but they are sufficiently numerous to raise the
question for the psi research worker. Some of these cases from daily life were
cited in Chapter 2 - the stopping of a person's clock or watch or the
unexplained falling of his picture from the wall at the time of his death - and
they also include the examples of persistent household disturbances, the
so-called poltergeist cases, with which we will be concerned in the next
Interest in PK is not new in parapsychology. Not only has the spontaneous PK
occurrence long been the object of serious study, but there were decades both
before and during the first fifty years when investigators seriously attempted
to test the claims of individuals who were reputed to possess remarkable powers
of mind over matter. Some of these studies, such as those Sir
carried out with the famous D. D. Home*, were reported in the scientific
literature with startling findings, and they remain even today as a challenge to
the open-minded explorer. But others of these earlier efforts at PK
investigation became hopelessly lost in the darkness of the mediumistic séance
room. Clearly, the study of PK was stalled unless and until someone could find a
suitable method of tackling the problem, a method similar to the one by which
ESP had been successfully brought into the laboratory.
* ISS note: See
Investigation of a New Force.
After the first publication on the ESP research at Duke in 1934, Dr.
[J. B.] Rhine
thought that the time had come for making a new attack upon the PK problem. But
how was this to be done? The answer came from a casual visitor to the Duke
laboratory. He claimed that he could successfully will dice to fall so that they
would give the numbers he needed in order to win. His confident boast reminded
Dr. Rhine that many experienced players of the game of craps believe that they
are able to control the dice.
The visitor went on his way, but he left his beautifully simple idea behind. Dr.
Rhine had his method for testing PK. Both he and Mrs.
[Louisa] Rhine, using themselves
and a few friends and students as subjects, set out to see if this was the
answer. The first tests were made by throwing two dice for high dice (a total of
8 or more on both faces), low dice (6 or fewer), or sevens. The first work
naturally was exploratory, and the dice were thrown under a variety of
circumstances. Some of the trials were made as in ordinary dice games: the cubes
were shaken in cupped hands and rolled upon a blanketed surface or bounced off a
"wall" before they stopped rolling. For other tests, the dice were shaken in a
cup and immediately thrown blindly from it. In still others, the dice were
rolled down an inclined chute onto a table or the floor.
These first PK tests yielded results that were consistently beyond the average
expected by chance coincidence. The subjects maintained their success over a
large number of throws and the odds against chance were very great. The
experimenters, of course, had confidence in their own results, but they wanted
confirmation by others before publishing such startling findings. Dr. Rhine
therefore quietly informed a few people about the work and encouraged them to
undertake PK tests. Several of them did so.
This stage of quiet research on the PK hypothesis lasted from 1934 through 1942.
Very early during this period, variations in testing procedure were made. For
example, some subjects preferred to try for a designated target face instead of
the sum of the faces on a pair of dice. This kind of test was more adaptable
since it allowed the subject to throw any number of dice at the same time. Tests
were made with from 1 to 96 or more dice per throw. As the experiments advanced
beyond the exploratory stage, adequate safeguards were introduced to control
against physical bias of the dice by throwing equally for every face, and
machines were made to throw the dice for trial after trial. Once the dice were
put into the machine, no one touched them again until the experiment was
Most of these series, like the first tests made at Duke by the Rhines, gave
highly significant results. As each experiment was finished, the results were
noted in terms of the over-all scores, and the records were sent to the
Parapsychology Laboratory to be added to the PK file. This process continued
over the eight-year period before the Duke experimenters felt that the degree of
confirmation justified publication.
Then, in the course of re-examining the data before reporting the findings, the
investigators discovered a new type of internal evidence. This discovery
provided a clear-cut basis for excluding the counter-hypotheses to PK even in the
more exploratory series. The evidence was the observation of the fact that
highly significant rhythmic changes had occurred in the level of the subject's
success in relation to his progress through the PK test. By chance, of course,
there is no reason to expect that one section of the record page should
consistently do better than another. As the subjects worked their way through
the trials that the procedure required to complete a particular unit of the
test, their success had gone up and down in a remarkably regular and lawful way.
No one had anticipated or noticed this variation in performance while the tests
were being conducted. Consequently no one gave any thought to such position
effects. Therefore neither the experimenters nor the subjects were
consciously motivated in one part of the record page more than in another.
This discovery made it possible to consider as evidence for PK the results of
exploratory experiments in which subjects threw for the face of their choice
instead of trying equally for all faces to guard against any effect of physical
bias in the dice. If the 6-face, say, was used as target throughout an
experiment, a high total score might be due to the fact that the larger number
of spots (holes) on that face made it lighter and this caused it to turn up more
often. But a statistical test based upon the difference between the
numbers of 6-faces found in two different sections of the record page could not
be interpreted as something that happened because of physical bias in the dice.
The 6-face would not be favored because it is lighter during the throws
made at one particular moment and the 6-face on the same dice then be
disfavored because it is heavier in the throws made at the next moment.
What were the unexpected position effects found in the PK records? When the
first series was being reanalyzed, the investigators observed that there was a
general decline of success between the top half and the bottom half of the
record page. Similarly, they noticed that there was a decline of success between
the data recorded in the left-hand and right-hand columns. As a means of
getting, a standard and optimal test of these two trends, they divided the page
into four equal quarters by means of horizontal and vertical lines. Then they
made a statistical evaluation of the difference in the hits scored in the upper
left and the lower right quarters of the page, a test they named the "quarter
distribution" or QD analysis.
Having discovered a decline effect in the first records examined, the
experimenters went on to apply the QD analysis to all of the other records
already in the files. A total of eighteen separate experimental series were
found to be suitable for evaluation by this method.
In twelve of these series the subjects had been throwing for a designated
face of the die as target. In all but one of these the QD analysis showed a
higher rate of scoring in the upper left quarter than in the lower right quarter
of the page. Statistically, a difference as large as that observed for all
twelve series between the total score of the upper left and the lower right
quarters would be expected by chance one time in more than 30,000,000 such sets
The other six series were experiments in which the subjects threw a pair of dice
for a target sum of the two faces. The QD results once again showed that
in all but one series the upper left quarter had given a higher score than the
lower right. Here the difference between the total scores would be expected by
chance one time in more than 150 such sets of records. Even this result has
better than 1 in 100 chance odds, and so it is statistically significant.
The QD results for the PK record page were discovered and reported by Dr. Rhine
and Miss Betty Humphrey of the Duke Parapsychology Laboratory. Seeing this new
evidence as solid proof of the PK hypothesis, they issued an invitation for any
qualified scientists to come and recheck the results for themselves. When no one
else accepted the invitation, I took the opportunity. Although I was a member of
the staff, I was away from the laboratory when their analyses were made, and
thus I was qualified to make a completely separate recheck. I rechecked the ten
out of the eighteen series which contributed most of the PK evidence from the QD
analysis. A few minor errors of tabulation were found, but these did not make
any real difference in the results. Thus my own findings verified the discovery
of a lawful grouping of hits in the data that could not be written off as chance
and for which there was no explanation except that of mind working with
spontaneous bursts of force at the start of a new page.
The QD analysis was carried through two further stages to see if the decline
effect was present in units of the records smaller than the page. The final
stage was one that I suggested as a check that was completely independent
statistically of the page QD that had already been found. The results of this
analysis carried as much weight as they would have if a totally different set of
records were being analyzed. This was a study involving the quartering of record
units lying wholly within the four quarters of the page, a sort of
quarter distribution analysis of the original page quarters.
I had originally urged that this study should be made before I had seen any of
the PK data. Thus there could be no question regarding whether I had simply
looked over the old records until I found something that seemed good and then
had planned an analysis to take advantage of a lucky hindsight. This would not
have done at all, of course. It would be like betting on a horse race after it
has been run and the results were known! We were never so foolish in any of the
QD analyses, but it was nice to have the results of this third study to prove
the findings to the hilt.
Out of the original eighteen series, only eight were suitable for this third QD
study. These showed the predicted decline effect with the odds for chance
occurrence of 1 in 200,000 This study by itself was strong evidence of PK,
especially for the reason that it confirmed the results previously found in the
study of the QD of the page.
This is how the case for PK stood in 1945, eleven years after the first dice
tests were started and three years after the publication of the first results.
The case was conclusive - as conclusive, at least, as it could be on the basis
of work largely centered in or directed from one laboratory. From that time, the
hypothesis had scientific status, commanding the close attention of research
workers in parapsychology. The question was: Could the evidence be confirmed by
investigators in other research centers?
Since 1945 there have been many confirmatory PK experiments, but I will present
the basic details regarding only three of them here and a fourth one involving a
new type of test a bit later.
In 1946, Mrs. Laura A. Dale reported a PK experiment which she had conducted at
the American Society for Psychical Research in New York. The experiment was
planned from the outset to guard against every imaginable way in which non-PK
factors might influence the results. Fifty-four college students were used as
subjects, each one taking part in a single session. Two dice at a time were
shaken in a cup and poured into the top end of a chute. They tumbled about three
feet down an incline and came to rest in an enclosed area at the bottom. Both
the experimenter and the subject silently recorded the faces of the dice after
each throw. The score of the entire experiment of more than 31,000 die-throws
was above chance to the degree represented by odds of 1 in 200.
Another effort to repeat the PK work was made at the University of Pittsburgh.
This experiment by Dr. R. A. McConnell and two associates used 393 subjects and
involved about 170,000 die releases. A special feature of this experiment was
the use during two thirds of the trials of an automatically operating electrical
apparatus for throwing the dice. The dice, as they lay, were photographed by a
camera attached to the machine and they were also recorded by the experimenter.
The photographic record made it possible to recheck the scores from the films
and thus to find any errors in the experimenters' records. The total score of
the experiment was not significantly different from chance, but the decline
effect on the record page showed a drop in scoring with chance odds of 1 in 500.
A third effort to confirm the PK work done at Duke was an experiment conducted
by Mr. G. W. Fisk and Dr. D. J. West in England with a selected high-scoring PK
subject. An unusual feature about this work was the fact that the subject was
not informed which face of the die was the target for any given set of trials.
Thus the subject had to use ESP to get the target and PK to make the dice fall
to match it. The subject worked in her own home where she simply threw the dice
to match the unknown targets as they were set up a number of miles away. The
first series of 10,000 trials gave results above chance with odds of 1 in 625.
In three further series this subject also obtained consistently positive scores
under the conditions that required the use of both ESP and PK. These further
results had chance odds of 1 in 5880.
The above three examples of independent confirmations of the early Duke PK
results are offered here both because of the experimental conditions and because
of the results. I could have chosen a number of other series with comparably
well-controlled conditions that did not yield as striking results and others
that did not measure up to these in safeguards but gave better scores. The PK
research has not proved to be any royal road to discovery for the investigators,
for in several instances they have met with success in their first experiments
only to find that efforts to repeat their work were fruitless. These
unsuccessful tests have been cited by some critics as grounds for discounting or
rejecting the positive cases. But the real lesson they teach us is that the
essential psychological conditions for getting PK to work in a test situation
are difficult to provide and even harder to keep. The case for PK needs to be
considered on the basis of the total evidence, not on that of one worker or one
group or one period. But it is interesting to observe the extent to which the
results of the second period of the research confirmed the findings of the first
phase as done in or directed from the Duke laboratory. This is why I have called
attention to the outstanding experiments of the period after the first
announcement of the PK evidence and when the problem had been taken up in other
centers. On the basis of both the safeguards and the results, the
three studies described above should provide the answer to the question of any
open-minded person who might wonder whether the case for PK depends entirely on
work done in the Parapsychology Laboratory.
In all the work described so far the target was the face of a die or a
combination of two faces. We come now to PK work that introduced a radically new
departure in method - efforts to influence the placement of falling objects by
In 1951, Mr. W. E. Cox reported an experiment in which subjects attempted to use
PK on a tumbling die in two ways at the same time: one, to get target faces; and
the other, to make the dice stop on those squares on a checkerboard surface
which were marked with the target-face number. Both the target-face data and the
placement data were statistically significant.
This beginning on the study of PK placement was followed up by other
investigators, chiefly by Mr. Haakon Forwald, an engineer in Sweden.
Since 1951, Mr. Forwald has been the most active investigator of PK placement.
By the end of 1957 he had published eight reports on his research. In this work
the investigator, serving as his own subject, mechanically released cubes to
roll down an inclined plane and spread out on a horizontal throwing surface. A
center line divided the horizontal surface into equal right- and left-hand areas.
The subject's aim was to influence the falling objects to stop on the side
chosen as the target. The two sides were used as target the same number of
times, and thus any physical bias in the apparatus was controlled.
During the first series, the results were scored only in terms of the division
of objects between the target and non-target sides. Nevertheless, the number of
objects falling within the target area was highly significant.
Mr. Forwald then introduced a simple change in his apparatus with the purpose of
achieving a more sensitive measure of the PK placement effect. He drew lines
parallel to the center line of the table at one-centimeter intervals and
numbered them to provide a "scale." Thereafter the cubes were scored on each
throw to show the actual degree of displacement.
The experimenter released six cubes on each trial and ten releases were recorded
together as a set, the first five for the right-hand or A-side of the table as
target and then five for the left-hand or B-side. The data were evaluated in
terms of the difference between the cube distribution when A was target and that
for B as target.
As Mr. Forwald continued his research, two general facts became clear. One was
that he was getting, with almost marvelous regularity, an effect upon the
placement of the dice in the direction of his wishes. The second was that this
effect was not uniformly distributed throughout the set, but was more
concentrated in the first throw of the set for each target side. As evidence
that something more than chance was operating, his data were conclusive. Because
the placement effect was found in the first throw of the set, it became standard
procedure in the evaluation of his data to select these results for separate
Parapsychologists have long been subjected to criticisms of a kind that
scientists in other fields do not need to worry about. When any psi investigator
working alone has obtained significant results, it has been commonplace to hear
that he could have made errors of observation and recording and thus have
deceived himself. To forestall such criticism, it was necessary for Mr. Forwald
to repeat his tests in the presence of a witness and independent recorder.
This need was met in the fall of 1957 during a visit by Mr. Forwald to the Duke
Parapsychology Laboratory. The purpose of the visit was tacitly understood but
not overtly stressed. We recognized that any PK subject would be placed at a
psychological disadvantage if he felt "put on the spot." Consequently we were
interested in approaching the crucial stage of his visit by slow degrees. Dr.
Rhine asked me to supervise Mr. Forwald's work at Duke, and in the end I joined
Mr. Forwald in reporting the research.
As a starting point Mr. Forwald worked through two series entirely alone as he
had done in Sweden. These series give significant results for the first throw of
the set (the basis of the statistical test, selected in advance on the strength
of his previous work), with odds of 1 in 140 and 1 in 58, respectively.
He then tried repeating this success when a member of the laboratory staff was
assigned as an independent observer and recorder. Two series done under this
condition were totally without statistical significance.
In the next stage, three different members of the laboratory staff in separate
series participated as co-subject as well as independent observer and
recorder. With one of these laboratory members, Mr. Forwald's results were
significant, with odds of 1 in 166. So far the results were merely exploratory
as far as the real purpose of the visit was concerned.
Finally plans were made for a confirmatory test set up on the basis of the work
done up to that point. In this test Mr. Forwald had as his co-subject Mrs. Peggy
Murphy, the member of the laboratory with whom he had previously worked most
successfully in exploratory series. This member of the laboratory also made an
independent record of the cubes. The two observers compared their records on the
spot and reached agreement before the cubes were disturbed. The results showed a
high level of significance, with odds of 1 in 5000. Thus, in spite of the
psychological difficulties which had to be overcome, the Duke series confirmed
this subject's abilities under conditions excluding subjective errors of
observation and recording.
But Mr. Forwald's objective in his work as a whole has not been limited to
piling up more and more evidence for a PK placement effect. Almost from the
beginning of his research he has been trying to gain some insight into the
dynamics of the PK process. He has, for example, compared cubes of different
kinds of materials, weights, roughness, and surface coatings. At the same time
he has worked out mathematical formulas for converting the effect obtained
(taking into account the relevant aspects both of the cube movements and of the
cubes themselves) into the physical energy equivalents for bringing about the
result. It is too soon to attempt to make any general scientific evaluation of
these research efforts, since they are still at an early stage of development. I
mention them only as an indication of the fact that the investigators of the PK
effect are aware of the need to find ways of relating this function to more
familiar scientific principles. Parapsychologists not only want to know whether
PK occurs; as soon as this conclusion is reached, they turn to the much more
difficult task of trying to learn something about the nature of the effect.
Other investigators have also continued to contribute to the study of the PK
placement effect during the past decade. There has been further research from
Mr. Cox, who was the originator of this type of test, as well as by others. But
thus far it is the work of Mr. Forwald which stands out in this area, and it is
work which presents some very challenging questions for future investigators of
The emphasis in this chapter has been put largely on the question of evidence
for the occurrence of PK. This is appropriate, because the establishment of a
direct, extra-muscular influence of thought upon physical systems has even
greater revolutionary implications for psychology and biology than does the
discovery of ESP. The general acceptance of the PK effect would require a
reorientation of scientific thinking regarding the nature of the living organism
- a reorientation in which the influence of mental factors would be recognized
in fact and not in name only. But I need not dwell on the importance of
the discovery that thought processes have real force!
But something should be said regarding the secondary problems - questions about
the nature of PK - which have received some attention in the research. We have
made a start toward finding out whether the PK effect is limited by the space
and mass aspects of the physical situation. Thus far we have found no
differences in relation to the number, mass, shape, and distance of the objects
the subject was attempting to influence. Also, subjects have tried to influence
a range of types of objects with apparent success, including dice and
cubes, coins and other discs, spheres, roulette, a spinning pointer, the
swimming of paramecia, and the growth of plants and molds. On these points the
present results are inadequate for final conclusions, but the findings encourage
further explorations in search of the scope and limits of PK in relation to the
Investigations bearing upon the psychology of PK have also shown some progress.
Motivation appears to be a factor of paramount importance. Subjects who
succeeded in their first series of tests have often failed in later efforts to
demonstrate their PK abilities: the excitement and eager curiosity which marked
their first experience could not be recaptured. Thus investigators have learned
that success in a test for PK, as with ESP, is not to be taken for granted. In
general, exact duplication of the results of experiments has been as difficult
for PK as for ESP.
The findings of parapsychology form but a beachhead on a new continent of the
world of science. The facts I have been presenting show that PK, like ESP, is a
part of that beachhead. There are uncertainties in the situation but they are
not concerned with the evidence for the occurrence of the phenomena. They
relate, rather, to what we will find as we move on beyond the perimeter of our
small beachhead of unshakable evidence and extend our lines of exploration and
discovery. The uncertainties are merely the challenge to further research.
The article above was taken from J G Pratt's "Parapsychology: An Insider's
View of ESP" (1964, Doubleday & Co, Inc).