LETTER TO MR. ROBERT W. LUCKY FROM JAMES L. CALKINS

• Document Type: CREST
• Collection: STARGATE
• Document Number (FOIA) /ESDN (CREST): CIA-RDP96-00787R000200080043-7
• Release Decision: RIFPUB
• Original Classification: U
• Document Page Count: 10
• Document Creation Date: November 4, 2016
• Document Release Date: November 5, 1998
• Sequence Number: 43
• Publication Date: April 26, 1976
• Content Type: LETTER

\:,-. tppro ev d 'lFor~ fea a X1/03/26: CIA-RD~bt`-& dYft ~'bC~ ~ '3=~1j` "nwyl :.tt-:.t 1
WAY
(.(illc~~,t~ 01 I IU11),111it ieS
& Soci .iI Sciences
I )t'ImItIllent- of I'tiv( hology
.. tilrrinllt);y
(2I ) ti')~i-1+1x3
JEEE - EDF1ORiAL,
Mr. Robert W. Lucky, Editor
Proceedings of the IEEE
335 Enat 47th Street
New York, New York 10017
Dear Mr. Lucky:
April 26, 1976
I have just read, with considerable interest, an article in the March issue of
the Proceedings of t?he 71:1111; by It. E. Puthoff and R. Targ entitled "A Perceptual
Channel for Information Transfer Over Kilometer Distances: Historical Perspective
and -Recent Research." These authors have undertaken) a series of potentially
significant studies into a topic of obvious interest to an increasingly large number
of individuals, doing so utilizing what appears to be appropriate techniques tnkoll
from,' L scicntist'c arsenal of objective_ :netl;odo!ogiwy, many or. t;e h i . < - , 1 a the i1 a t ? s rrs
helve previously distinguished themselves using in research in tho natural r)cienccn,
it is certainly an understatement to any that. such a controversial topic as ES1' re-
ytil1I 5 Oxt)-aordi.naarily precise and careful methods, especially well suited to excluftc+
the basis for all those nagging criticisms often irreverently hurled at the para-
psychologi st-'-fraud, inadequate controls, inlprecir:c and incomplete reporting of dlytrtl la,
iulproper statistics, proper atatistics improperly used, and, especially, the prc)Ienct!
of numerous confounding variables which vary with the independent variabl:cn In the r
experimental designs in such a way that theso other uncontrolled variables remain 1.{)
potentially account for the given results. Indeed, the authors t:herttsielves apparently
set out to achieve this goal, what they refer to as their "principal responsibility--
to resolve under unambiguous conditions the basic issue of whether, or not thin clasts
of paranormal perception phenomenon exists" (pps. 334-335) . Unfortunately, the model
for their study follows the traditional atratagcros of the I)araapsycholol;istra iii the
United States and Europe, rather than the method they no doubt otherwise use in their
non-behaavioial research in the natural sciences (rand I nnight add that is largely
followed in experimental psychology). As we shall indicate, the consequence of this
is that they must necessarily fall far short in fulfilling their stated "principal
responsibility."
It is the essence of the c periniontal method--in contrast to naturalistic,
observation, the survey technique, correlational procedures, field studies, and the
"Theoretical model" of the parapsychologists (Girden, 1962, p. 360)".+-to in fact, create
Approved For Release 2001/03/26 ~ [ A RD.P96-OO787ROOO2OOO8OO43-7
(;r,Illlmll'S1lit 111IIll' I1111.11`'~III'Ill1'
~~;;5),~'~r?
01111'Ill'IIIIII ', Iit IAI IIV Ill r',lt,IIIrif I
t {{I11'IU+111 t llll!III'1'lll!II ? l \1'11111111 1+111'!11'
J3
t u11lq+,I, Itl 111H11,1111111'4 i11111'dil 11tt'.I a I,.l a
 Mr. lioIAPpmvP F9r IJ?8se 241/03/26: CIA-RDP96-00787ROQ 200080043Qi 2(, 19 /n
1'a;occeding;s of the IEEE Page 2
the conditions necessary foa.' "unambiguous" ret::olution of fundamental quest ions since
on1 this method permits manipulation and control of potentially confounding; vnriablCtt
by the eminently sensible method of vrnryin the critical, factors under study, and
systematien1ly observing; their effects upon other selected and measurable variablora
while holding these potentially confounding vsr.ialien under tight control. vin such
techniques as randomization, constancy, counterbalancing, matching, et:c. The mani-
pulnted variables are called independent' variables (IVs), and the variables sensitivo
to the effects of the IV is called, the dependent var.iabl.es (DVn). Thin has been
expressed quite eloquently by l;bbinghaus (1964, p. 7; original: 1885) in his
incorporation of this expcrimental method into his researches into human memory,
as successful effort which went far in illustrating in 1885 the col '?ac:r of this method
in scientifically understanding human behavior--including al e.rce.ptiYon:
"We all know of what this method consists: an attempt is
made to keep constant the mash of conditions which have proven
themselves causally connected with a certain results one of
these conditions is isolated from the rest And varied in a way
that can be numerically described; then the accompanying change
on the side of the effect is Ascertained by measurement or
computation."
The simplest experiments, therefore, are those employing but a single IV and
it single DV, and, in the fundamental situation in which an effort is being made to
demonstrate the sheer- existence of a phenomenon (as in the present study, without:
inquiring fcrtbeu into composition and Contingencies), the two basic values or
levels or variates of the IV may be simply designated the "experimental" (i.e. , the
factor appears--operationally defined--in some amount) and "control"" condition
(i. e. , the factor appears as a zero amount) .
A helpful example might be a drug, study in which we only wanted to know if the
drug; makes running a mare difficult--or not; In this situation, we wou;.d experi.w
manta l l.y compare the drug condition's effects to a condition otherwi. SC i (lent: ice l in
which, however, the drug, is absent; This would constitute one of tlae most i:und11111wntal
cr p1.r.ical control conditions used in experimental science in general, and experim"mtal
psychology in particular, Although of course other empirical control conditic"nn are
po;,ni.hle, depending on what one is Controlling for We speak, by the way, of "control
Condition", and not "controlled" condition, since we nuvuise all coaaciitici-arr in the
experiment are controlled in some waay. This is apparently a source of much confunlon
in the pariapsycholog;ical literature, where constant reference is mode to "controlled
laboratory conditions," "scientifically Control.lc:d conditions," And no on, thereby
creating the illusion that basically well conceived control conditions nr.e being used.
In the proselat example, if one is concerned that the .ce?Iure of injection causes
errors to be made on the maze, then we need a j .ncebo control in which S is trent.ed
identically an in the drug condition, but saline rather than the drug is in.ycctc!d.
In this case, we "control for" the injection procedure 'n effects on the DV by holding
thin potentially confounding variable constant and can thereby evaluate it and prevent
it from creating the erroneous impression that the drug, per se, produced the errors S
made on the maze, when in fact the drug injection procedure itself may have produced
that effect on the DV.
As we know, in the present study, the basic procedure described was carried out
with all the subjects (six in number for section III studies through subsection I'D",
Approved For Release 2001/03/26 : CIA-RDP96-00787R000200080043-7
 Mr, RAppromecLFoa'yFeJe 1/03/26 : CIA-RDP96-00787RO 0008004 ri 26, 1.976
Proceedings of the IEEE Page 3
two more in section "E", and five more in section "F"; we limit our Critique to the
more detailed accounts given in sectionsA-1) rather than the very sketchy rnrate:rial
in sections E and F, although all these Studies used essentially the same procedure
and varies mainly on the basis of subject charractcri?rtics,). That is, all the
mull ects were administered, as it wera, tlac sWane bnsi.c t:rontment con~lition, and were
thus all part of the same "clairvoyant" or ''remote viewing" group, consisting
essentially in the Ss making an effort to somehow envision a remote target, Ac tua i ly,
all we know of the instructions to the S, so critical in determining tile'
he operational
definition of this condition, is Clint the "remote-viewing subject was naked to
describe his impressions of the target site into a tape recorder and to snake any
drawings he thought appropriate (p.335)," since the authors (10 not give tin the actutai
instructions, In any case, for each S, there (lid indeed exist a "remote target"
designated by a "target team" or "demarcation team" situated at some geographical
location nearby the SRI laboratory, although the theoretical rationale for this team.
existence and its role in the procedure is never really logically presented by the
author- save only for an historical Precedent established working informally with
Mr. Ingo Swann (p. 334) and perhaps the anecdotal "pilot experiment" (p.330) where
one of the authors functioned in this capacity as it "dumnrention team'" anal, we infer,
grew quite excited aaad impressed with a subject's closer iptions of some sites he
visited, where the subject supposedly had no particular prior information about the
site in question. This situation, we shall detail later, results in considerable
ambiguity in the designation of the "target" (is it. the "target.-perceived-by-the-tears",
or is it the physical stimulus of the target, and no on?).
The whole procedure, carried out under what the nut:hor(a describe as "rigidly
controlled scientific conditions (p.334)", and including the subsequent judging
procedure we shall discuss in some detail later, could have been called an "experi-
mental'' condition comparable to the "drug present'" condition in the example above i is
there had been a control condition with which to compare it (comparable to the "placebo,
or "no\drug" conditions in our example:), that is, to condition in which all of the
preceding procedures were exactly followed but ill the absence of what was previously
LL,rationnlly and ab ectively defined as the "remote-viewing" condition (note the
importance of oh ecrive specification of the conditions under which the "remo.o-vl.c'win);'
is to occur, because without that there can be no objective controlled vnri.ntt(,n of
the condition since one would never know quite when the experimental conaliti.on exi rated
in the first place;). As the maLt~ r stands, since only one condition wan run in thin
study, we really have neither experimental nor control conditions (because the ternw
are defined relative to each other); and since that is the case, we really have no IV;
and since we have no IV, we actually do not even have nee experiment (let alone the 50
or so claimed by the authors in this single publication--p. 330).
The DV,,of course, also deserves careful scrutiny, since hypothesized IV effects
are evaluated in terms of changes in J)V measures. In this study, the actual "number"
obtained an the basic raw datum was the individual ranking or "match" number of a
subject's tape recorded description (supposedly of to ''target"), with some aspect
(perhaps) of a nearby geographical location. The number could assure any value from
1"1"' to "9", where "1" referred to the judge's estimate of a "bent" match, through "9"
which was assigned to a match if it was a "worst" match for a given target, . Nine tar-
gets and nine (lescriptionaswere obtained for each S. This was apparently an ordinal
scale of measurement. We must assume, in the absence (also) of a specific report of
the instructions to the judge, that the judge knew exactly what his task was in deter-
mining the degree of correspondence between the Ss' descriptions and the judge's own
unspecified perceptions when physically present. at the so-called target,
Approved For Release 2001/03/26 : CIA-RDP96-00787R000200080043-7
 MI , ltobA ipIovedcFpr Release 1 /03/26 : CIA-RDP96-00787R0 200080'0 2 , 0 / 0
Proceedings of the IEEE rag o 4
1,1 r., should, therefore, consider the precise meaning of a "hit" under these
circumstances. Evidently, a "hit" for the 1,'s occurred when the geographical locus
they called the "target" is judged to correspond to the S's description obtained
while the "demarcation team" was visiting a particular site. This was also, therefore,
a situation where the S's corresponding desci.ption was assigned a rank of "1". Since
9 targets were "experimented" with a given S, we have 9 separate judgments of rank
carried out by a single given judge. The. sarm of ranks for the given S of all the
descriptions for the theoretically associated targets was then compared to n probability
distribution for that statistic, a situation we claim is really improperly serving as
a substitution for a test based on some statistic (which could be the Sum of ranks)
obtained from the comparison of the matches for this hypothetical "remote viewialp,"
data to the watches obtained under similar judging conditions where there was no
remote viewing possible (other control conditions will be described below).
'However, there are numerous peculiarities about this judging process possibly
unresolvable on the basis of the authors' scanty and ambiguous description of this
critically important aspect of their study. Let us consider a few of these problems.
1. What Was the uclS,Ids . Ili,? The' judge for a given S's performance for a
given target (what the authors refer to incredibly as an "experiment") was successively
driven to each geographical location prcvioussly vi sited by the peripatetic En. Since,
as we previously noted, we do not know precisely what aspects of the geographical
location constituted a "target" in the original "experiment" when the demarcation
team, was present, and since it is even more ambiguous now what the judge was viewing,,
as well as what he was su j)(2sed to be looking at while he reviewed the S's packages of
9 descriptions, we seem in this procedure, therefore, to actually be dealing with two
(and perhaps three) recognizably distinct categories of "targets"; one in constituted
by the perceptions of the demarcation team; a second by the perceptions by the
and a third by direct physical aspects of some geograpllictal location (photographs are
used in the report and labelled "target" to further complicate target: delineation-
e.g., Figure 4). It is difficult to evaluate how potentially dissimilar these various
"targets" were in the absence of clarifying and detailed accounts of the specific
instructions to the teams and to the judge. After all, the judge or demarcation terns
may have fixated the horizon, focused on ptassing vehicles, noticed n sign, and no on,
Consequently, the reporting of the target simply as a "target location", or "remote
location" (e.g. , "Marina,) Redwood City") an the authors do, belicrs the fact tnnt we
are dealing here with both their clni.ryoynnco belief and in the perceptual !henry
that the way things appear to a viewer and the way they tire physically constituted
may not be at all identical (e.g., the. sensation "reel" of a perceived stop sign, and
the atomic structure of the paint pigments on its metal surCace). The possible cony
sequence of this lack of precise. definition of the target in that the Judge Hale .
considerable latitude in fitting S's description to the scene as he. perceives it,
possibly even loollicable, instructions so the reader could eva ii,ncc
the degree to which we may have given our S any information about: the type, condition
or location of the target, all of which information may help diminish the population
of total targets S thinks might be under consideration, thus significantly increasing
the probability of making a hit--a circumstance that may help account: for such out-
standing "hits" in the authors' report' as that of White Plaza).` The differential
. Approved For Release 2001/03/26 : CIA-RDP96-00787R000200080043-7
 Approved For Rel.`Oase X1/03/26 : CIA-RDP96-00787R0Qg200080043-7
Approved For Release 2001//281; CIk'7R000200080043-7
 IZ boast W. ck ~d "a. tor a a
l,rQee lea ,r v~~ R ; 'se 1/03/26 : CIA-RDP96-00787R092000800 "-7"
- r !
Page 7
presence of such information, varying with the "inntrue ti-oil" l:V, would be an
excellent example of a confounding Variable in tin experilnent of this type. in tiny
case, the use of these control conditions (no target, no remote-viewilni; demand in
various combinations designated by cells A, C and 1)) permits an empirical deter-
mination of the extent to which such "hits" in matching may occur simply by virtue
of the similarities judges may read into this matching material due to its inherent
unstructured subjective quality, much as human observers of Rornchach,ink blots nee
a map of Ireland or a bust of Caeser, an organization imposed on perceptual materials
that perceptual theorists refer to as "Gestalts," often reflecting unconscious needs
of the observer (such as, "I have psychic powers I must demonstrate!").
As for the sequence to be followed in actually running S through these different
treatment combinations, a simple randomized arrangement would suffice, that is,
assuming 20 targets, with numbers 1-5 in cell A, 6-10 in cell 11 and so on, we simply
run the treatment combination according to the next number in the random series
if 13 is the first random number, we run S according to the conditions
obtaining for cell C). Later the judge can in random sequence consider one cell 1alnc~
of descriptions and targets after the other, until all, four cells have been "matched,"
where naturally the judge is not informed of the treatment combination condition. lilt
rates are determined, for example, and then appropriate statistical tents for simple
effects, main effects, and possible interactions determined. The "remote -viewing"
hypothesis would be confirmed if cell "lt" had significantly more hits than Coll "A"
(no target),cell "C" (target, no R-V instruction), and certainly than coil "D" (whore
both target and appropriate instructions are absent). The present reported study,
by-the way, consists only in cell "B'", making appropriate comparison impossible. In
any rase, lack of such significant differences will merely dis:confix'rn this para-
psychological hypothesis (not "disprove"), thereby providing no one with any cupiricti1
reason for affirming belief in the existence in this unusual phenomenons existenc(J--
at least, as evaluated in this suggested experimental design.
Furthermore, the underlying logic should be clear, that the clairvoyant hypo-
thesis cannot be evaluated when matches or hits (cell "A") are compared to some
hypotletica1 "chance" level (no matter how .accurate the statistical tent be according
to individual statisticians, pronouncements of statistical associations-, or as found
in statistics texts) that fails to represent the normal, or nveraLe number of hits
that might be really exjcted under these identical. conclitions whhen. clairvoyance sloes
not exist (as cells "u" especially, and also "All and "C" establishes, ac- explained
above).
The,enaential experimental. invalidity of this parapsychological tactic of
comparing; their so-called "experimental data" to some hypothetical "chance" level
especially in the absence of experimental control conditions, in perhaps most clearly
illustrated in the psychokinetic ("mind over matter") literature. In those studies,
classically championed` by Rhine some years ago (e.g., 1947) , a tumbler might tons
out onto a table some 600 dice; the S being evaluated for his alleged "psychic" skills}
makes some effort to mentally or "psychically" influence each die and have, nay, an
many 5s turn up as he can. A tabulation is made, and it is found that, nay, 235
5 s have indeed turned up--a startling result, indeed, especially when it isa considered
that only 100 should have so turned up by "chance." The problem of this i-iterpretaatio
is the same as that in the present "remote -viewing"sLudy. This "chance level" is a
the"oret.ical, mathematical abstract model of the behavior of 600 ideal dice, not
Approved For Release 2001/03/26 : CIA-RDP96-00787R000200080043-7
 itr. 1~ be -t cnlcr v+ I dit or A aL'a,t Z", 19/o
111:0ceARPE1 It jRase X1/03/26 : CIA-RD P96-OO787R0,QP2OOO8OO43-7 Pago ii
necessarily related to the die actually being used in the study at all, not to
mention to the procedures and physical conditions operating in actually Using and
"tossing" them. An cm)i( ticsl control condition is necessary to obtain this "litnn-
dard" for comparison, one representing; the actual concrete di.e used in the study,
and accounting for such potentially confounding variables influencing the throw such
its non-horizontal tables, loaded (lice, etc. (all as discussed above). Thin control
condition would exactly duplicate the experimental condition, save only for the
omission of soiree critical factor under' study in the experiment, such as (in thin
cane) psychokinesis itself! In other words, we run a non-PK condition, in which
all is done as in the experimental condition, but S does not when the die in
thrown.
Would not such a comparison change our interpretation of this whole situation
if it was found that a n about 230 dice came up "5:3"? We would then obviously
look for those other alternative, non-parapsychological explanations for the high
5'a count (such as the possibility of loaded (lice). An excellent review. of the
PK literature through 1962, and general critique of this "theoretical model", is
that of Girden (1963).
It may be true, as the parapsychologists claim, that ESP is real and represents
agreat latent power of the human mind, one day Co emerge in full recognition by
science as another momentous step in the evolution of man and his mind; but it-Is'
t.r'uth remains to be demonstrated through use of the f.x rerimental. method, and Until
it. is, in the same way as Ohm's law or Pavlov's conditioned reflex paradigm ht, r)
been so demonstrated, parapsychologists ought not consider- psychologists and other
scientists and engineers calcified conservatives blindly refusing to sec the obvious
"fact" that ESP, etc., exists, because the parapsychologists themselves seem to
virtually intentionally avoid using the only techniques which in the lone; run will
prove persuasive to the scientific community, and those are ob cct?ive experimental
procedures.
Perseverance of belief in ESP .and related phenomena built upon such shabby
and preposterous evidence as is usually offered as "scientific proof" on behalf
of the ESP proponents (and the present study is Co some extent an example) itself
requires some explanation, since such perserverance is often cited an itself somehow
evidence for the existence Of these "paranormal" phenomena (important people
wouldn't believe it if it wasn't true.'). It is not hard to find such an explanation
in the literature of the social psychology of social movements and cults. since
therein it is well understood how organized groups of individuals may band toy;etiter
with their own ideologies, their own clubs, their own in-group publications and
sacred works, their own symbols, passwords and slogans, in the interest often of
providing Some sense to life, some direction, some, compensation for a sense of
personal loss, insignificance, or inferiority, which in provided by becoming; a "true
believer," as lloffer (1951) put it, in some special and unique movement. Especially
is this true of ideologies that persist despite virtually universal rejection on
some rational grounds, suggesting, in contrast, that the belief in question answers
to needs other than those that are rational. The particular difficulty the porn-
psychologists seem to encounter in that (most fortunately for humanity) the scienci.Nc
method represents one of the few really pristine exemplars of the r.ationril uric of the.
human mind, thereby guaranteeing a clash with their own vested irrational ideological
systems. Add to this the consideration that we are presently witnessing; n strong
resurgence of popular interest in-the occult and supernatural, with a proportionate
increase in the volume of superstitutious behavior prevalent, we can see that the
Approved For Release 2001/03/26 : CIA-RDP96-00787ROO0200080043-7
 f7 Approved Fgr , NP 1, Rejgase,9 01/03/26 : CIA-RDP9.6-00787R0Q#20008004~r71 26, 1976
Proceedings of the 7Lk,13 `..0'"
present study decidedly fits into c:lic "Zcitgcist" of contemporary demtintin "T'his is
especially unfortunate for psychology, since this also means considerable talent
arad potential expertise that could have pushed bacl( frontiers of new understanding
of the normal complex realities of human perception and its relation to the nervous
system, and physical reality, should instead to diverted into inquiry into the
bizarre circuitous vagaries of the so-called 11 lie
REFERENCES
1. Ebbinghaus, R. Memory la Contributioaa tea i?xea imrntal. Psychology. New York:
Dover Publications, 1964 (original: 1885)
2. Girden, E. A Review of Psychokinesis (PK). Ps.ycholo icrl 13u11etin, 1962, 59,
353-388.3, Iloffer, H. The True Izelievcr. New York: New American Library, 1964 (original
1951).
4. Puthoff, H.E., and Targ, R. A Perceptual Channel for Information Transfer, over
Kilometer Distances: Historical Perspective and Recent Research. Prod .eclin"
of the IEEE, 1976, 64, 329-354.
59Rhine, J. B. The Reach of the Mind. New York: W. Sloane Associates, 1966
Sincerely yours,
9wvvJ ;
James L. Calkins, 11h, D,
Associate Professor
JLC:rm
. Approved For Release 2001/03/26 : CIA-RDP96-00787R000200080043-7