ESP and Psychokinesis
ESP and Psychokinesis
�LllDPllDIL
INllillPll Temple University
Philadelphia, Pa. 1 9 1 22
A Philosophical Examination
Stephen E. Braude
� Philosophical Monographs
Braude, Stephen E. 1 94 5 -
E S P a n d psy chokinesis.
(P hilosophical monographs)
Bibliography: p.
Includes indexes.
1 . Ex trasensory perception. 2. P sychokinesis.
I. Title. II. Serie s : P hilosophical m on ographs
(Philadelphia, 1 9 7 8- ).
B F13 21. B 7 2 1 33.8 79-27 1 6 1
ISB N 0 - 8 7 7 2 2-163-4
IX
.
Preface
PART I
A. Conceptual Foundations 3
1. Preliminary Termino logical Remarks 3
2. Telepathy 7
3. Clairvoyance 15
4 . Precognition 19
5 . Psychokinesis 26
6 . Experimental Ambiguity and Purity 30
7 . Experimental Replicability 41
B. The Data 72
1 . Introduction 72
2. PK 74
a. History o f Research 74
b. PK and RNG s 76
c. Retroactive PK 81
d. Further PK Research 94
e. Theoretical Issues 99
3. Clairvoyance 1 24
a. The Pearce-Pratt Tests 1 24
b. Clairvoyance and RN Gs 127
c. Pre-Conscious Clairvoyant Interaction 1 28
d. Free-Response GESP Tests 131
e. The Hypothesis of Trans-Temporal
Inhibition 1 40
Vll
4. Telepathy 1 45
a. Experiments in C ard-Guessing 145
b . Pre-Conscious Telepathic Interaction 146
c. Hypnogenic Telepathic Interaction 149
d. Theoretical Issues:
Telepathy as Energy Transfer 152
PART II
IX
x / Preface
to this pro bl em, and trust only that readers will be tolerant
o f passages aimed at readers at some other level of philoso
phical or para psychological proficiency .
My decision t o think seriously about the data o f parapsy
chology is relatively recent , and was m ad e in the summer
of 1 9 7 5 . I decided then to become part of the community
of parapsychologists, in order to learn as much as possible
about the latest ex perimental and conceptual d evelopments
in the fiel d . I must thank the m any scien tists working in
parapsychology who have aided m e in my attempt to mas
ter the data and who have graciously and enthusiastically
welcomed me int o their fold. I have the utmost respect for
the intellectual honesty and flexibility they have shown in
encouraging m e to express my philosophical perplexities and
misgivings about their own theoretical work. Despite my
continued efforts to comply with these invitations, I fe ar I
have rendered them less of a service than they have rendered
me. Among the m any parapsychologists to whom I am
indebted, I w ish especially to thank the team at SRI (Harold
Puthoff, Russell Targ, Edwin May ) , Charles Honorton , John
Palmer, and Charles Tart ; also, John Beloff and my colleague
Bruce Goldberg, whose suggestions and criticisms have been
invaluable in writing the theoretical portio ns of this book;
also J an Ludwig, for his commen t s and criticisms o f an early
draft of the m anuscript . For their helpful criticisms and
suggestions , and for their willingness to discuss issues with
me, I wish also to thank Gerald Barnes , Jule Eisen bud,
Michael Hooker, and my UMBC colleagues Thomas B en son,
Audrey McKinney, and Alan Tormey. I wish also to acknow
ledge several very distinguished philosophers, who unlike
many other of my colleagues ( I 'm sorry to say) have dis
played the intellectual courage and open-mindedness to
support my interest in investigating the subject o f parapsy
chology. In particular I wish to thank Bruce Aune, Roderick
Chisolm, Wilfrid Sellars, Hector-Neri Castaned a (who invited
me to submit my paper 'Telepathy' t o No us [ v. 1 2 ( 1 9 7 8 ) :
2 6 7-30 1 ] , which I have revised and woven into the text), and
Joseph Margolis, who invited me to prepare this book for
Preface I xv
3
4 / Conceptual Foundations
2. Telepathy
* See, e.g. , the case of Rider Haggard and the dog Bob, Proc. S.P.R. 3 3
( 1 9 2 2): 2 1 9ff.
Conceptual Foundations I 9
*The re ader may no tice that this point is similar t o one of Broad's
basic limiting p rinciples in · [ 24 ] : 3. See II.C. f or a discussion o f the se.
1 4 I Conceptual Found ations
3. Clairvoyance
4. Precognition
Perhaps the first thing we must note is that the loose and
popular ways o f talking about precognition are extremely
misleading. First of all, many events tentatively classified as
precognitive are not cognitions at all. This should come as no
surprise, having already noted that telepathy and clairvoy
ance seem to be non-cognitive. It may even be that precogni
t ive events are simply a subset of the set of telepathic and
clairvoyant events. W e can, therefore, rej ect at once one cus
tom ary way of defining 'precognition'-namely, as 'the non
inf erential knowledge [or , even less plausibly, as the percep
tio n ] of some future state of affairs'. *
5. Psychokinesis (PK)
tress the e xperimental evidence for psi even i f w e are not sure
who is responsible for the good scores. It is, however, also
possible that the experimenter (say, through an unconscious
wish· to fail or to discredit a colleague ' s work, or perhaps
b e cause of certain personality traits) may somehow influence
scores negatively, by 'cancelling out ' a subj e c t 's psi e ffort
so that scores hover around chance levels. Such an effe c t �
*See the next section for futher remarks a b o u t sheep/goat results and
the concept o f negative psi.
38 / Concep tual Foundations
holding the photos had already been led to expect the ratings
to be of one sort rather than another. But in the first repli
cation the photos were mounted on cardboard and labelled,
so that the subj ect could evaluate the photographs without
looking at the experimenter and then just call out the rating.
The reason for doing this was to reduce the likelihood that
the subject was receiving subtle sensory cues from the experi
m enter-for example, from his hand as he held the photos,
or from his face, which in the first exp eriment would have
been in the subj ec t 's field of vision during the rating process.
It was by m e ans of such ex tremely subtle sensory cues that
the horse Clever Hans had been able to perform his appar
ently astounding feats of intelligence (see Pfungst [ 1 0 6 ] ) .
I n spite o f these additional precautions, however, this p arti
cular experiment showed the largest rating biases of the
three. Although this result does not lead directly to cosmic
conclusions about the nature of the exp erimenter influence
in such experiments (much less their p aranormal nature),
it nevertheless strongly suggests that such influence is quite
unlike that which Pfungst discerned in the case of Clever
Hans. From the first replication, it appears that experim enter
expectancy led to biases in the ratings, even though experi
m enters did not handle the stimulus m at erial and they were
not always in the subj ec t ' s field of vision. Clever Hans, o n
t h e other hand, lost his 'abilities' when his interrogator,
or the experim enter, was located outside the horse ' s field
of vision. I am not suggesting that the next natural conclu
sion is that experimenter psi ( e . g . , PK) accounts for the biases
in the ratings. Still, this remains a live optio n, depending on
the strength of the evidence for various sorts of PK, or per
hap s telepathy (as in the Soviet studies in telepathic sugges
tibility, de scribed in I . B .4. c . ) .
O n e final complication; i f the so-called o bservational
theories of psi (see I . B . 2 . e . ) are correct in maintaining that all
psi influence reduces to a form of retroactive PK, then
we should expect to be able to find evidence of retroactive
exp erimenter effects. Recent studies by Broughton [ 2 5 ] and
Conceptual Foundations / 4 1
7. Experimental Replicability .
But this hop e that such general criteria (or recipes) o f repeat
ability are specifiable, I will now argue, is nothing m ore than
wishful thinking. Not only does it ignore certain interesting
facts about the ways scientists actually operate, as well as
most of what is interesting about exp erimentation in parapsy
chology, but it also ignores some extremely important
philosophical truths.
Perhaps the best way to approach this topic is t o consider
various sorts o f reasonable possible d i fferences between an
exp eriment and its replication attempts, and then see whether
46 / Conceptual Found ations
sion that the results of E 1 are spurious. After all, the replicat
ing scientist might have had a bad day . He may have been
feeling negative or irascible ; perhaps he d i d not 'hit it o ff
with the subj e c t ; or perhaps he was harboring an unconscious
d esire t o discredit a colleagu e ' s work. In fact for these
reasons, even repeated failures by the replicating scientist
need not conclusively discredit E 1 .
Thus, when we consider the possibility that the state o f
m i n d o f t h e exp erimenter m ay b e causally-and i n p articular,
psychokinetically-related to the exp erimental outcome, and
when we realize that the experim enter's influence or psi
functioning m ay be long-term or short-term , and m ay be
unknown to the experimenter and to all others, we cannot
help but appreciate the difficulty in determining, in p arti
cular cases, whether the experimenter's attitude or interests
render him unable to serve as a replicating scientist. For
example, there is no way to be sure whether his presence
would tend to suppress or cancel out the subj ect's psi activity.
I .
Such considerations frustrate all our attempts to specify ' .
from the subj ect's using some psi ability (presumably subcon
sciously) in the service of producing poor scores, perhaps t o
discredit parapsychology, or for s o m e other reason . * In such
cases the subj e c t might use ESP to identify the correct target
in order to be sure to 'guess' the wrong one. Otherwise, it
seem s , there would be no way of ensuring the continued
below-chance scoring, and scores would hover around chance
levels.
Rhine, however (in [ 1 2 1 ] ), has argued that psi-missing
may not be a function simply of the subj ect's mo tiva tio n to
produce low scores. He observes that psi-missing often occurs
under conditions of s tress . For example, in studies conducted
in India ( Ra o , Kanthamani and Sailaj a [ 1 1 7 ] ) , subj e cts tested
b e fore an important interview showed significant psi-missing,
although positive scores were o b t ained in p ost-interview ses
sions. Rhine suggests that, while under stress subj ects may
inadvertently alter whatever subconscious 'strategy' or pro
cedure they employ when using E SP ( e . g . , instead of re
sponding to the first idea that comes to m in d , they may wait
for the m ost vivid one, or they may m od ify guesses in order
to avoid the appearance of patterns in their calls). Thus, a
'change of method' (as he calls it) m ay account for negative
deviations. But whether these 'methods' do or do not involve
the use of any psi ability requires further probing.
First of all, the idea that psi subj e ct s employ subconscious
-and often maladaptive- calling strategies has been explored
in another context by Tart (see I . B . 3 . e . ) . This idea strikes me
as at least reasonable, and perhaps even plausible. Certainly,
there is nothing obj ectionable in supposing that a person's be
havior can be guided by goals and strategies of which h e is not
consciously aware. Consider, for example, people who
tend t o behave in a manipulative or controlling way , or
who tend to draw attention to themselves. Such p eople are
1 . Introduction
*See Rhine [ 1 2 2 ] , [ 1 2 3 ] As the case of the late Cyril Burt reminds us,
.
2. PK
a . History of Research
In order to place recent experimentation against an appro
priate background, I shall first briefly summ arize the history
of laboratory research in PK. In the mid 1 9 3 0 s , J. B. Rhine
began conducting quantitative experiments in PK (see Rhine
& Rhine [ 1 2 5 ] , and Rhine [ 1 1 9 ] ). The principal method
The Data I 7 5
b. P K and RNGs
Accordingly, recent PK research has charted a new direc
tion . One of nature's most elementary random processes i�
radioactive decay. For example, atomic nuclei in the radio
active element Strontium-90 decay spo n taneously after an
average life of 30 years. It is not that the atoms age, or that
anything occurs which causes them to decay. The way most
physicists understand the process is that radioactive decay is
a genuine random occurrence typical of events at the atomic
level. Moreover, this randomness is of an especially deep
kind, since on the received (Copenhagen) view, there is
literally nothing which is the cause of the decay of an atom
of Sr-90. We can, of course, predict with considerable accu
racy how many such atoms will decay in a given period o f
t im e . B u t there i s no basis i n principle for predicting when
particular decays will occur. Now, i f all this is true, then
atomic decay could provide PK researchers with a paradigm
natural random process to use in their exp eriments, one
whose statistical prop erties are well known and not subj e ct t o
The Data I 77
30 / sec 40,000 5 1 .6 % 7 x 10 9 to 1
300 / sec 400,000 5 0 .3 7 % 4 x 10 5 to 1
c. Retroactive PK.
By now, the reader should have s o m e sense o f the sophisti
cation of Schmidt's PK experim ents. We may turn, therefore,
to some of his most peculiar results, particularly his astonish
ing prima facie evidence for retroactive PK. Let us b egin by
briefly considering an experiment designed by Schmidt which
suggests that PK success is independent of the complexity
of the PK task (see [ 1 3 9 ] ). In this experiment, subj e cts were
tested on RNGs having different degrees of internal complex
ity. The simple RNG was one o f Schmidt's coin-flippers. The
com plex RNG was also a binary RNG, but it selected a + 1 or
a - 1 by first generating 1 0 0 binary random events (at the rate
of 30/sec) and then taking the 'maj ority vote'. In the case of
a tie, no de cision was made and the test went on to the next
82 I The Data
( [ 1 44] : 5 3 8).
were recorded, and for each test session, only one mem ber of
each pair was used, the other serving as a control sequence.
The decision as t o which member of each pair was to b e the
test sequence was made after the target generation but before
inspection of the content of the tape. Test and control runs
were thus generated under identical conditions, and before
a decision had been made as to which runs would be which.
The results of this portion o f the experiment match those
of the conventional portion. 30 subj ects again participated ;
and for the t o t al of 600 runs, the average run length was
5 6 . 3 7 , again significant at the . 0 0 1 level. The control runs,
however, evaluated after the test sessions, showed n o signifi
cant deviation fro m chance, the average run length being
63.08.
The results o f t h e t w o experimental arrangements thus
produced very similar ostensible PK effects. Apparently, the
production of such effects did not depend on whether targets
were generated during or before the test. Since the only appa
rently relevant difference in the conventional and u n conven
tional portions of the experiment was that, in the latter,
targets were generated before the test, and since the conven
tional portion suggested that PK activity occurs during the
t e s t , it seems that, in tests w i t h pre-recorded targets, the ac
tivity of the subj e c t during the test (or, perhaps, the activity
of the experimenter) is a causal condition of the earlier non
random behavior of the RNG. The only relevant difference
between test and control tapes was that the former were
played back to a subj ect m aking a kind of PK e ffort.
Those sympathetic t o the existence of psi phenomena,
but wary of backward causation, may wish t o propose
another explanation. The least outrageous possibility is that
the experimenter (or perhaps the subject) unconsciously used
precognition before t arget generation t o determine (a) when
exactly, the 60-run sequences would be generat e d , and ( b )
which of these sequences would be chosen a s t e s t sequences,
and then (c), again unconsciously , the experimenter (or sub
j ect) used PK during target generation t o produce nonrandom
behavior in the RNG j ust for the test sequences. A variant
The Data I 8 7
and does affect his ext ernal environment with the same
degree o f refinement and persistence used to affe ct the b o dy,
as in cases of biofeedback, hypnotic suggestion (for e xample
the raising of welts on the skin), the placebo effect, and other
continual psychosomatic processes. Perhaps we always take
an active role in literally shaping or causing the events around
us, though not by the familiar forms of agency; and perhaps
this explains why some people are lucky or unlucky, or why
some people experience m ore coincidences than others. Per
haps, just as we m ay be continually controlling our body's
health through subtle mental influence, and j ust a s this might
explain why some people are perpetually healthy or sick, w e
might also b e continually exerting a n e ffe c t on o u r physical I
d. Further PK Research
Many studies in PK, as well as in E SP, suggest that subjects
are more likely to obtain above-chance results when they do
not actively try to affect the experim ental outcome. That is,
a state of passive a ttention or expectation o ften seems more
conducive t o successful scores than a state in which the sub
j ect makes a conscious effort to succee d . For this reason, sub
j ects in psi e x periments are often asked merely to w ish for
good results at the beginning of an experiment rather than at
tem·pt consciously to make som ething happen.
This apparent finding is consistent with what we know
about the operation of biofeedback and the efficacy of
placebos. Biofeedback research has established that ordinary
persons can learn to control, voluntarily, numerous physio
logical processes customarily regarded as involuntary-for
e xample, skin temperature, brain wave frequency, and even
the activity of a single cell (see B asmajian [ 4 ] , [ 5 ] , and
Green and Green [ 6 0 ] : 3 1 - 3 2 ) , simply by observing feed
back displays which monitor these processes. Research has
shown that these self-regulatory activities are most effective
when the subj e ct simply desires the intended result and al
lows it to occur, rather than striving or exerting a conscious
e ffort . Two leading biofeedback researchers. Elmer and
Alyce Green of the Menninger Foundation, call this passive
volitio n . The m echanisms of successful self-regu lation through
biofeedback are not at all well-understoo d . But enough is
known about the surface aspects of its operation to make
The Data I 9 5
e. Theoretical Issues.
Sum ming things up thus far, there is a growing (if not al
ways coercive) body of evidence that success in PK experi
ments does not depend , or depends very little, on subjects'
knowing (at least by normal means) such apparently relevant
facts as the nature, mechanics, or existence of the PK t arget
system, or even whether they are being tested for PK. There
is evidence, also, as I ob served above, that subj ects tend to
perform best when they do not actively try to affect the
experimental outcome. And when we add these to the results
of Schmidt's experiments, we begin to get a rather surprising
picture of PK. In fact, to some, it begins to look as though
success in PK tasks might be accomplished without any form
of computation or information-processing by subj e cts. In
other words, the evidence may point away from what
Stanford calls the cybernetic model of PK.
Lately, the idea that PK violates a cybernetic model has
received considerable attention in parapsychology. There are
two m ain reasons for this. The first concerns the recent emer
gence and development of various speculations lumped to
gether under the heading of observa tional th eories. These
o ffer an analysis of psi functioning quite unlike that sug
gested by conventional models-an analysis that seems consis
ten t , in fac t , with a good deal of otherwise puzzling evidence.
The second con cerns the influence of a certain line of think
ing championed most conspicuously by Stanford [ 1 5 6 ] ,
1 00 I The Data
learns that he correctly guessed , say, the 6th card down, this
permits him to influence by means of retroactive PK his guess
at the previous time.
As parsim onious as all this sounds, I confess I can m ake
little sense of it. Consider, first, the case of the reduction of
E SP to retroactive PK. The observational theories ask us t o
concede that
(a) S u bj ect S's observation o f feedback of a hit
is causally necessary for
the other han d , since human beings also o ften exhibit the
decline e ffect in psi tests, the decline during the fish experi
m ents m ight equally be evidence for experimenter psi.
In any case, no matter what our final assessment of the
work conducted with animals thus far, Stanford's position
against the conventional models of PK and ESP is hardly
outlandish. But my own view is that, though it is fo rceful,
Stanford's verdict m ay be a bit hasty. To see why, notice
first , that Stanford m akes two claims, not clearly separated,
one of which is stronger than the other. The m ore m odest
o f the two is (i) that PK violates a cybernetic m o d el accord
ing to which an organism acts on a system and guides and
m onitors its own PK activities by means of (presumably para
normal) feedback loops. But Stanford feels that the data also
support the m ore sweeping claim (ii) that no psychobiolog
ical model of PK is satisfactory-that is, that PK is not at all
like ordinary motor actions, and that any m o d el of PK b ased
on familiar psychological or biological activities must be
system atically misleading. In m aking this wider claim , Stan
ford argues that PK seems t o b e accomplished without any
form o f computation or information processing by the
organism involved . I believe, however, that the evidence
probably does not support either of St anford 's claim s-par
ticularly the stronger.
Let us look first at the weaker claim , that PK cannot b e
m o d eled cybernetically. This amounts t o t h e denial that,
during PK, the organism guides and m onitors its actions by
m eans of a continuing supply o f feedback. But the evidence
seems able to support only the still weaker claim that if PK
can be modeled cybernetically, the fee d b ac k loops involved
must be somewhat unusual-not quite like those involved ,
say, in steering a car (which Stanford treats a s a paradigm
case of an activity that may be construed cybernetically ) .
B u t t h e d a t a m ay not support even this strong a claim.
Even the results o f Schmidt's tests for PK with RN Gs seem
compatible with the view that the subj ect monitors his PK
activities b y m eans of ESP, and that, through ESP, the
subject keeps tabs on such things as moment-to-m oment
1 1 6 I The Data
(a) and (b) suggest further that PK does not rely on m oni
toring of mom ent-to-moment changes in the targe t system ,
since the conditions which seem to contribute causally to
PK success occur after such changes have taken place.
The Data / 1 1 7
3 . Clairvoyance
A 12 +5 9 P < 1 0- 14
B 44 +75 P < l o-6
c 12 +28 P < 1 0-4
D 6 +26 p < 1 0-6
Combined 74 + 1 88 P < 1 0- 22
The Data / 1 2 7
cially interesting was the fact that signi ficant missing tended
to occur only for small temporal displacements. That is, there
tended to be fewer correlations between calls at ti and targets
at li + 1 and ti - 1 (and often t i - 2 ) than between calls at l i
and targets further removed from the ( ti ) th targ e t . Tart also
fo und that the degree of m issing on immediately past and
1 42 I The Data
4. Telepathy
a. Experiments in Card-Guessing
As we saw , some of the experiments described in the sec
tion on clairvoyance are best described as GESP experiments,
since positive results may also b e attributed t o telepathy. In
fac t , as already m entioned , the M aimonides Hospital experi
ments were regarded as experiments in telepathy. Although,
in general, the ambiguity between telepathy and clairvoyance
is not circumvented in telepathy experiments, it tends to be
overshadowed by another-the ambiguity between telepathy
and PK (see I . A . 6 . ) ; although the latter is also present in
clairvoyance experiments, perhaps it does not loom as large.
The evidence for telepathy takes various form s. The phe
nomenon does not lend itself to quantitative analysis as easily
as PK or certain forms of clairvoy ance ; in fac t , parapsycholo
gists have had to muster considerable ingenuity in order t o
design quantitative m easures o f its prima facie occurrence.
Still, despite this difficulty, the m ost dramatic and conclusive
evidence favoring telepathy was, until recently, the card
guessing experiments conducted in England by S. G. Soal
with two star subjects, M r . Shackleton and Mrs. Stewart .
Shackleton's results were especially interesting, since they
appeared to furnish evidence o f a kind of precognitive tele-
1 46 I The Data
cent to that of the subj ect viewed cards with names written
o n them . Some names were of persons known only to the
subj e c t , others had names of persons known only to the
agent , and still others had names selected fro m the telephone
directory. For an additional control measure, there were
some periods in the experiment where the agent viewed blank
cards. The measurements o f the plethysmograph were d one
on a double-blind basis, and the results showed the subject
names to be correlated with larger vasoconstrictions than
were the agent- and neutral-names and the blank cards (the
magnitude of the associated vasoconstrictions of all these
being about the sam e ) .
In a fallow-up series of ex periments conducted by Dean
and Nash, which in general supported the results of the first
series, the exp erimenters confronted a problem which led to
some interesting observations. For each of a protracted series
of trial runs, the subj ect was asked to submit five new names,
and it proved to b e not long b e fore he ran out of names to
use. Initially, t h e experimenters dealt with this problem in
subsequent trials by randomly selecting names from the pool
already given by the subj ec t . B u t the results of those trials
did not confirm Dean's earlier results. The experimenters
then reviewed the successful names used , and found that they
were the names of people with whom the subj ect had been in
contact recently, or with or about whom the subj e ct had
strong feelings or associations.
Accordingly, Dean and Nash altered the selection process
so that the subject-names had at least one feature of strong
contact or association , and this time , they found significant
vasoconstrictions when the agent viewed names of persons
known to the subj ect . Moreover, they found that the subj e ct
showed significant vasoconstrictions when the names viewed
were of persons recently contacted by the age n t , and nonsig
nificant vasoconstrictions when the agent-names were of
people n o t recently contacted .
It m ay also be significant that one of the names given b y
the subj e c t w a s o f a person h e saw almost every d ay , a n d that
out of the 43 times the name was viewed by the agent, the
1 48 I The Data
age time for the onset of sleep was 6 . 8 minutes (with an aver
age error of ± . 5 4 m in . ) . It thus took almost three times as
long for subj ects to fall asleep in the absence of m ental sug
gestion.
In a subsequent and more elaborate series of e x p eriments,
the onset o f sleep in the absence of mental suggestion was
compared with the onset o f sleep (with m ental suggestion)
under conditions both of lead screening of the agent and
without screening of the age n t . In these, agent and subject
were in d ifferent rooms separated by several closed chambers;
the recording apparatus was in another room adjacent to the
age n t ' s . Various precautions were taken against cheating and
against the transmission of subtle cues to the subject (see
Vasiliev [ 1 84 ] : 1 40-4 7 ) . In 29 experiments, 1 0 were con
ducted without mental suggestion, and the average time for
the onset of sleep was 7"40 min. In 1 0 experiments with
mental suggestion under conditions of screening, the average
tim e was 4. 7 1 m in . ; and in 7 exp eriments without screening,
it was 4. 2 1 min.
In order to test the sensitivity o f telepathy to distance
and to rule out completely the possibility of hyperaesthesia
of hearing frequently observed in hypnotized subjects, som e
exp eriments were conducted from Sebastopol to Leningrad
(approx. 1 , 700 k m ) . Altho�gh only two experiments were
conducted at this distance (several others were conducted at
distances from 2 5 meters to 7, 700 m eters), the results were
very interesting. Days and times for the experiments were
arranged in advance; also agent and the observer of the sub
j e ct checked their watches, by radio, with M oscow tim e . The
agent, incidentally, also had an observer, but he was ignorant
of the purpose of the experimen t . On the first day of the
experiment planne d , the agent became ill and , unknown to
his colleagues back at the Institute for Brain Research in
1 5 2 I The Data
1. Anomalous Mo nism
logical kind term may not m ark off a genuine kind ; hence, n o
predicate of <I> n e e d s t o correspond to this predicate o f '11 . B u t
if so , then we have n o reason t o suppose that we shall ever
form an expression in the l anguage of the physical sciences
having the same referent as 'm . If the kinds of phenomena
'
serts, then, is that there must be some brain activity for a per
son to have m ental states-perhaps simply that we must be
alive to have thoughts. Now it is true that my earlier objec
tions do not touch this position. Although so constru e d , ( a )
is not altogether trivial- for it seems incompatible with cer
tain dualistic and especially spiritualistic views according to
which mental processes are autonomous-it is no longer an
explanatory thesis. It cannot explain , in physiological terms,
why m occurred rather than som e other mental state. Nor
would it help explain why m is j u st the kind of mental state
i t is. To do this in physiological terms, we must revert to a
type-type correlation view-which is unsatisfactory. In any
case, if it does not matter what the physiological features o f
brain state-token b are, then ( a ) promises n o phy siological
u nderstanding of why m occurred (rather than a token of a
d ifferent type), or of wha t , phy siologically speaking, it means
to have a thought of type M. If Smith is thinking that Jones
is a coward , partisans of (a ) can say no m ore than that Smith
has this thought because he is in some brain state or other.
They cannot tell u s why, physiologically , Smith has this
p articular thought rather than another though t , or what it is,
physiologically speaking, to h a v e ( s a y ) a thought that J o n e s is
a coward (or even w h a t it is t o h a v e a n y thought about
cowards).
Nor can (a ) provide the foundation for a theory having
any predictive utility. In fac t , if b can be any brain state at
all, then ( a ) provides no ph y siological basis for the science of
psy chology . We would have no way of describing or pre
dicting, in physiological terms, what mental state a person
was or would be in. Understood in this exceedingly weak
way, (a ) is actually contrary to the original spirit o f material
ism , which was initially conceived as a program for ex
plaining why a person was in one mental state rather than
another or what it is (in physiological terms) to be in certain
types of mental states, or how on the basis of one's physio
logical states, we might predict what a person's mental states
would be. But these explanatory and predictive features o f a
materialist theory may be secured only if there are type-type
correlations between the m ental and the physical.
1 8 4 I Psi and the Philosophy of Mind
tion, he find s that , for the first tim e , he can indeed conjugate
new ar-verbs, and has thus acquired an understanding he did
not possess previously. Or, imagine that we are trying to
t e a c h a child the color red by teaching him t o distinguish r e d
things from non-red things. A n d suppose t h a t i n this process,
suddenly , the child clearly understands what to do. The
assumption we tend to make in such cases is that something
has happened, within the person, between the time he did
n o t understand and the time he did. In other words, we tend
to explain the newly-acquired understanding in terms o f
s o m e associated struc tural modifica tion o f t h e person-most
would say it was a change in his brain.
Or imagine that Jones, who for many years had been a
callous and frivolous person, suddenly has a religious experi
ence that completely transforms his character, so that he
becomes a loving, serious, and God-fearing person. We tend
to assume that the change in Jones' character m ust be due
to some corresponding physiological change within Jones
himself (specifically , a change in his brain) produced by the
religious experience. We assume that something relevant
about the way Jones is put together is different from what it
was b e fore. (I say 'relevant' because, after all, Jones must be
changing physiologically all the tim e , whether or not he
undergoes a religious conversion . )
O r im agine that one day I see a person whom I recognize
as my old friend A , whom I have not seen in 1 0 years. We
ordinarily assume that I could not remember A were it not
for something "'ithin m e (a m emory trace) caused by my
previous acquaintance with A , and that I could not recognize
A after 1 0 years were it not for some specific physiological
process occurring within me (the process of recognition ) .
These assumptions, I believe, are entirely false a n d seri
ously misleading. Of course I cannot hope to d emonstrate in
a detailed way how very nearly every aspect of even just
cognitive psychology (to take one general field) is based on
a mistake, by exam ining, why these assumptions or their
variants are unj ustified in memory theory, recognition theory,
learning theory, and so on. The general mistake appears in
Psi and the Philosophy of Mind / 1 8 7
much greater detail (and, espe cially in Malcolm's case, from a somewhat
different point of vie w) the sorts o f criticisms I offer here. If these
books were t o have the effect they deserve on the intellectual commu
nity , many neurophy siologists and neuropsych ologists might find
themselves out o f work. Bursen is more blunt o n this p oint than Mal
colm, and charges (correctly) that trace the ories are scientific the ories
only on the surface and that, at bottom , they require that memory is
magical.
1 9 0 / Psi and the Philosophy of Mind
*How the 'right ' trace gets activated is a problem that m e m ory theo
rists tend not to see clearly. They tend to suppose that this can be ex
plained b y reference to structural isomorphism between the inner state
(the trace) and some other e xternal thing (like a picture of A or A
himself). B u t , as we saw in the case of the E T theory, this idea rests on
the untenable claim that isomorphism between two things is intrinsic to
t h o s e things, and t h a t rests on the unacceptable i d e a that s o m e struc
tures are functionally unambiguous, that their topologies determine
their own representational properties or rules o f proj e ction. (How fan
tastic this idea is becomes clear, if we recall that a brain state does n o t
represent o r resemble a person A i n t h e w a y t h a t even a photograph
of A represents or resembles A . )
In the next chapter and in the fallowing section, we shall examine
yet another aspect of this mistake-namely, the unacceptable idea
that things or events have a preferred parsing.
1 9 2 / Psi and the Philosophy of Mind
* T h e type of pos sibility here i s empirical rather than logical. That is,
the trace theorist is committed t o saying that as a matter of empirical
fact (i.e. , given the laws of this world) the trace cannot have more than
one associated effect, although there may be p ossible worlds in which
some trace t has an effect it doesn ' t have in this world.
Psi and the Philosophy of Mind / 1 9 5
the organism for which certain kinds o f 'why? ' questions are
no longer appropriate. If so, then to explain how I can re
m em ber A , we need appeal only to certain events in my past.
Similarly, when w e consider changes in disposition ( e . g . ,
J ones' conversion from being frivolous t o being serious), w e
need s a y n o m ore than t h a t t h e change m a k e s sense relative
to a certain episode in Jones' life (his religious experience,
say ) . Here then, w e may abandon the search for a further
explanation in terms of som ething happening within Jones
himself ( e . g . , within his brain ) . No thing of that sort could
e xplain the change .
Many readers will no doubt blanch at my suggestion that
we stop seeking explanations of abilities, dispositions, etc . , in
terms of underlying physiological structures or m echanisms.
Some m ay protest, for example, that , since we obviously do
not remember e very event in our past, we must do m ore than
explain present remem berings in terms of past experiences; in
particular, we m us t appeal t o some persisting physiological
m o d ification produced by the relevant past event . I concede
that more can b e said than that certain events in my past
enable me to remember A now. But granting the foregoing
obj e ctions, I would argue that we should not appeal to inter
nal mechanisms. What we can say is that certain events in my
past, involving A , were particularly intense or important at
the time and that I tend to remember intense or important
events. ( In fact, even on m em ory trace theory we must ex
plain why some past events and not others produce traces . )
O r perhaps I a m simply a person able t o d o certain things
rather than others. I might remember A 's phone number b u t
not his nam e , because I h a v e an ability to remember numbers
and not names. Here, we explain my remem bering A 's phone
number in terms o f other regularities o f m in e , like my ability
to remember numbers generally. But that general ability is a
fact about me for which no structural explanation (say, in
terms of mem ory traces) exists or can exist.
Another objection , bound to b e raised sooner or later, is
that studies in cerebral lesions and the like show that m ech
anisms for at least som e abilities and dispositions do exis t .
Psi and the Philosophy of Mind I 2 0 1
1. Introduction
The past few y e ars have seen the re-emergence (or resur
rection, depending on your point of view) of an idea-it
hardly d eserves its usual title of 'theory' -associated mainly
with C. G. Jung (see [ 7 5 ] ). Roughl y , this is the idea that
some events occur together, n o t due to any causal connec
tions between t he m , or between the group of events and
some common causal ancestor, but simply in virtue of simi
larities in the content or m eaning of the events. The attention
now being lavished by parapsychologists on the idea of acaus
al but meaningful connections-synchronistic connections
between events stems, as far as I can see , from two main
sources. First, many ( following Jung) have turned to the
notion of an acausal connecting principle (as Jung put it)
in the hope o f finding a way of explaining the exp erimental
data of parapsychology, one more promising than m echanis
tic accounts of psi functioning, especially those positing the
existence of physical carriers of psi information. The idea
here is that if correlations between (say) a subj e c t ' s calls and
targets on E SP tests represent acausal connections, then we
needn't bother looking for the kind o f causal chains that
have so long eluded parapsychologists. The other-and prob
ably the main-reason for the interest in synchronicity is
that many people feel that things happen to them in the
course of living that have a kind of significance or numinos-
217
2 1 8 / The Theory of Synchronicity
cupy ' an atomic 'orbit ' ) , even though this order need not be
m anifest in causal chains. Koestler then asks why, if acausal
principles m ay be scientifically legitimate, we should not also
regard as legitimate something like Jung 's principle of syn
chronicity (according to which nature is governed by rela
tionships of meaningfulness or significance between events).
Koestler never answers this question, as though its rhetorical
nature suffices to legitimize the notion of synchronicity. But
the question does have an answer, and the answer has two
parts. First, synchronistic or m e aningful connections are not,
as J ung and Koestler allege, clearly acausal. And secon d , even
if they w ere acausal, not every putative principle of organiza
tion in nature rests on the same presuppositions, and in some
cases the presuppositions involved are false, or at least suspi
cious.
I shall address b o t h parts of this response to Koestler's
question. I shall argue that synchronistic connections count
as acausal only on a certain kind of pre-Humean conception
of causality, and that even on this view of causality the idea
of synchronicity m ay presuppose a causal cosmology. And
with respect to the idea that nature is organized around
m eaningful relations between events, I shall argue that this
idea is a causal view in disguise (and not the radical thesis it
is claimed to be), and m oreover that it rests on an unintel
ligible notion of what it is for an event to have meaning, or
what it is for a description of an event to be a preferred
description . These two lines of thought run together, and I
shall not try to keep them entirely separate.
*Of course this is simply a version of the p oint made against the E T
theory in I . B . 4 . d . and later in I I . A . 3 . , except we were there concerned
specifi c ally with a slice of mental history.
The Theory of Syn chronicity I 2 3 1
5. Conclusion
1 . Introduction
2. Preliminaries
242
The M eaning of 'Paranormal' I 243
unusual is (to use Scriven ' s terms) one o f 'ex ceptional idio
sy ncrasy ' or 'generic di fferen c e ' . These expressions, however,
as well as Scriven ' s term 'order of existence', are really too
vague to be satisfactory. Moreover, the term 'order of exis
tence ' is presumably tailored to fit Scriven's discussion
of the class of supernatural phenomena, phenomena that fall
(as Scriven puts it) outside the natural order. But I think we
can state with somewhat more precision what we need in
clause (b) for an account of the paranormal, whether or not
it is what Scriven was trying to convey.
To say that P is especially unusual from the point of view
of science is at least to say that scientific explanations of P
demand new descriptive categories or new concepts. This
hardly suffices, of course, since we must sometimes develop
new concepts in order to explain phenomena that , h o w ever
novel, are not 'especially unusual' in the sense here required
(or intended by Scriven ) . This was recently the case, for
example, when, in developing the theory of plate tectonics
to explain such things as the intense geological activity along
certain contentintal coastlines, scientists introduced such new
concepts as plate subductio n . So for a phenomenon P t o be
especially unusual in t h e required sense, it c a n n o t m erely be
the case that scientific explanations o f P d e m a n d the use of
new descriptive categories or new concepts. It m u s t , in addi
tion, be true that the employment of these new categories
causes m aj or ripples elsewhere in the conceptual pool. In
order to explain P scientifically, we must transform or amend,
in some m ajor way, science as we know it. So perhaps condi
tion ( b ) m ay be recast as fallows.
(b' ) P cannot b e explained scientifically without m ajor re
visions elsewhere in scientific theory.
Condition (c) of ( D S ) is perhaps the most interesting com
ponent of that definition. Its inclusion in ( D S ) acknowledges
that extreme strangeness is a necessary but not sufficient con
dition of a phenomenon's paranormality . Since manifesting
consciousness is another necessary condition, ( D S ) l e aves
open the possibility that some phenomena may be extremely
2 5 8 / The Meaning of 'Paranormal'
265
2 6 6 I Bibliography
303.
[15] Bender, H . and Vandrey, R . , 'Psychokinetische Experimente
,
1 97 5 ) .
[55] Forwald, H . , Mind, Matter and Gravitation (New York: Para-
p sychology Foundation, 1 969).
[56] Frank, J., Persuasion and Healing (Baltimore : Johns Hopkins
Press, 1 9 6 1 ) .
[57] Gatlin, L . L. , 'Meaningful Information Creation: A n Alterna-
tive Interpretation of the Psi Phenomenon', JA SPR 7 1
( 1 977) : 1 - 1 8 .
Bibliography I 2 6 9
(58] Gauld, A . , ' ESP and Attempts t o Explain I t ' , i n Thakur
[ 1 75 ] : 1 7-45 .
(59] Goldber g, B . , ' A Probl em with Anomal ous Monism ' Philo- '
'
and Ullman, M . , 'Telepathy and Dreams: A Controlled
Experiment . . . ' , J. of Nervous and Men tal Disease 1 5 l
( 1 9 7 0) : 3 9 4-403 .
[88] - - -, Ullman, M . , and Honorton, C . , 'A Precognitive Dream
Study with a Single Subj e c t ' , JASPR 65 ( 1 9 7 1 ) : 1 92-203.
[89] Ludwig, J. (ed.), Philosophy and Parapsychology (Buffalo :
Prometheus, 1 9 78).
[90] Lurie, Y . , The Correspondence Thesis, Doctoral Dissertation,
Cornell University, 1 9 73 (Ann Arbor: University Micro
films , 1 9 77).
[9 1 ] Malcolm, N., Memory and Mind (Ithaca: Cornell University
Press, 1 977).
[92] Martin, M., 'The Philosophical Importance of the Rosenthal
Effe c t ' , J. Theory of Social Behavior 7 ( 1 977): 8 1 -9 7 .
[93 ] May, E . C . , Targ, R . , and Puthoff, H . E . , 'Possible EEG Corre
lates to Remote Stimuli Under Conditions of Sensory
Shielding', paper presented at IEEE Electro/77 Special
Bibliography / 2 7 1
[ 1 44] ___
'
'Evidence for Direct Interaction Between the Human
M ind and External Quantum Processes', paper presented
at Int. Conf. on Cybernetics and Society, Washington, D.C . ,
Sept. 1 9-2 1 , 1 9 77.
[ 1 45 ] _ _ _
'
' Ca n an E ffect Precede its C a u s e ? A Model o f a Non-
C ausal World', Foundations of Physics 8 ( 1 978) : 463-480.
[ 1 46] -- - , 'Use of Stroboscopic Light as Rewarding Feedback in
a PK Test with Pre-Recorded and Momentarily Generated
Random Events', paper presented at 2 1 st Annual Conven
tion of the Parapsychological Association, St. Louis, Aug.
8 - 1 2 , 1 9 78.
[ 1 47 ] - - - , and Pantas, L., 'Psi Tests with Internally Different
M achines' , JP 36 ( 1 972): 222-2 3 2 .
[ 1 48 ] Scott, C . , and Haskell, P . , 'Fresh Light o n the Shackleton
Experiments?' PSPR 56 ( 1 974): 43-72.
[ 1 49 ] Scriven, M. , 'Explanations of the Supernatural', in Thakur
[ 1 7 5 ] : 1 8 1 - 1 94.
[ 1 5 0] Shapiro, A K . 'A Contribution to a History o f the Placebo
. ,
'
and Vaughan, A . , Dream Telepathy (New York: Mac-
millan, 1 9 7 3 ) .
[ 1 83 ] Vallee, J . , Hastings, A ., a n d Askevold, G . , 'Remote Viewing
Experiments through Computer Conferencing', Proc. of
the In stitute of Electrical and Electronics Engin eers 64
( 1 976): 1 ,5 5 1 - 1 ,5 5 2 .
[ 1 84] Vasiliev, L . L . , Experiments in Distant Influence (New York :
Dutton, 1 976).
[ 1 85 ] v o n Bekesy, G., Sensory Inhibition (Princeton: Princeton
Univ. Press, 1 967).
[ 1 86 ] Wade, N. , 'IQ and Heredity: Suspicion of Fraud Beclouds
Classic Experiment ' , Science 1 9 4 ( 1 9 7 6) : 9 1 6-9 1 9.
[ 1 87] Walker, E . H . , 'Foundations of Paraphysical and Parapsycho
logical Phenomena', in L. Oteri (ed.), Quantum Physics
and Parapsycho logy (New York: Parapsychology Founda
tion, 1 975) , 1 -5 3 .
[ 1 88 ] ___ , 'Consciousness and Quantum Theory ' , i n E . D . Mitchell,
e t al. (ed . by J. White), Psychic Exploration (New York:
Putnam , 1 9 74) , 544-5 6 8 .
[ 1 89] Wal ti, B . . 'Die Silvio-ProtokolJe 1 9 76 - 1 9 7 7 ' , Zeitschrift
fu"r Parapsycho logie und Grenzgebiete der Psychologie
20 ( 1 97 8 ) : 1 -4 6.
[ 1 90] Warner, L . , 'The Role of Luck in ESP D a t a ' , JP 1 ( 1 937):
84-92.
[191] Watkins, G . K . , and Watkins, A . , 'Apparent Psychokinesis on
Static Objects by a "Gifted" Subject: A Laboratory Demon
stration ' , RP ( 1 9 7 3 ) : 1 3 2 - 1 34.
Bibliography I 2 7 7
1 82, 1 87 Haskell, P . , l 4 6n
E quivalence hypothesis, 8 2-8 3 , H astings, A . , 1 3 2 , 1 3 9
89, 1 1 1 H olographic model, 2 1 0-2 1 2
Experimental a m biguit y , 1 0 , 1 9 , H onorton, C . , 3 5 -3 6 , 6 4 , 1 2 3 ,
Index I 2 8 1