(Inted in the United StatesAppmued For Release 2000/08108 CIA-RDP96-00789ROO2200120001-4 Parapsychology: Science of the anomalous or search for the soul? James E. Alcock Depam"ent of Psychology, Glendon College, York URIverSIty, Toronto, ontado,Canads WN 3M6 Abstract: Although there has been over a century of formal empirical inquiry, parapsychologists have clearly failed to produce a I " " ' " h omena. Although many parapsychological research projects have been single reliable demonstration of "paranorma, or psi, p en carried out under what have been described as well-controlled conditions, this does not by itselfmake a science, for unless and until it can be demonstrated that paranormal phenomena really exist, there is no subject matter around which a science can develop. Indeed, parapsychologists have not even succeeded in developing a reasonable definition of paranormal phenomena that does not involve, or iniply, some aspect of mind-body dualism. Moreover, parapsychology has developed several principles (such as the experimenter effect) which can be used to explain away failures, and the use of these principles contributes to making the psi-hypothesis unfalsiflable, The "anything goes" attitude in parapsychology, which seems to lend credence to virtually any "paranormal" claim, serves to weaken the credibility of parapsychological endeavors in the eyes of critics. This general willingness to suspend doubt is another inAilcation that parapsychology is more than the quest to explain anomalous experiences, as is claimed. it is argued in this paper that parapsychological inquiry reflects the attempt to establish the reality of a nonmaterial aspect of human existence, rather than a search for explanations for anomalous phenomena. Keywords: anomaly; causality; dualism-, ESP; experimental method-, explanation; methodology; parapsychology, philosophy of science; replication 0. Introduction 1973; Irwin 1985a; McConnell 1977; Sheils & Berg 1977) have found personal experience to be the major reason It is curious that, precedented given by respondents for their belief in this age of un literacy in paranormal phe- and unceasing scientifichnological nomena. This is not surprising: Given and tec progress, their often power- many people are prepared ful emotional impact, combined with to accept that spoons a lack of under- can be bent by the power of standing about the myriad "normal" the mind alone, that ways in which these disease can be cured by the laying experiences can come about (e.g., on of hands, that see Alcock 1981; Marks water can be located by means of a forked & Kammann 1980; Neher 1980; Reed willow stick, or that 1972; Zusne & the mind can influence the decay Jones 1982), it is easy to ascribe of radioactive substances. paranormal explanations It is even more curious when such to odd experiences that one cannot claims are put forth readily explain and de- fended by people trained otherwise. in the ways of science. Most BBS readers, I "Parapsychology" is defined as the would imagine, have scientific study of little diffi- culty dismissing popular paranormal phenomena (Thalbourne occult beliefs in 1982). The study of astrology, palmistry, the tarot, the paranormal was historically associated or biorhythms. However, with the so- those same readers may not called occult sciences such as astrology be nearly so cavalier and numerology; a about dis- regarding such supposed more direct progenitor was the spiritualism "paranormal" (also craze of the syn- onymously referred late nineteenth and early twentieth to as "parapsychological" centuries. However, or "psi") phenomena as extrasensory parapsychology stands well apart perception ("ESP") from these belief sys- or psy- chokinesis ("PK"). tems in a number of ways: ESP refers to the supposed ability to obtain knowledge of a target object or of another per- son's mental activity 0.1. Scientific orientation. For in the absence of over a century, there has sensory contact, and PK is the putative been careful and deliberate investigation ability of the mind of psi phe- to influence matter directly. Belief nomena by people trained in the methods in such phenomena of science. In is actually very widespread, not the past 50 years, much of this research only among members has been labora- of the gener- al public but also tory-based and carried out in university among university students settings. Cur- (e.g., Al- cock 1981; Gray 1984; rently, parapsychological research Otis & Alcock 1982). is being conducted at Such belief is no doubt such prestigious academic institutions tied, at least in as the University part, to the fact that many people, perhaps of Edinburgh and Princeton University. even most, have from time to time had direct personal Throughout the last century and continuing experiences that seemed to the to be 11 telepathic" or "precognitive" present, a number of very prominent or "psychokinetic." natural and social In- C094WT a numbe*--s ayp~F.ke,AT scientists have been proponents and pprM S~Ja&&vans supporters of para- 08: C IA-RDP96-00789ROO2 200120001-4 0 1987 Cambridge University 553 Press 0140-525X187 $5.00+ 00 Alcock: Parapsyk ,4W efte ve r Release 2000/08/08 : CIA-RDP96-00789ROO2200120001,4 -psy6hological nomena research (Otis (see & Hyman Alcock 1995a; 1982); Rogo there 1986), was no clear dif- including ference physicists between Sir representatives William of Crookes, the Lord sciences Rayleigh and the (Nobel humanities. Prize, This 1904), is Wolfgang consistent Pauli with (Nobel the Prize, results 1945), of a smaller Brian survey Josephson conducted (Nobel at Prize, two 1973), other and Canadian David universities Bohm; naturalist (Alcock Alfred 1981). Russell Yet, Wallace; Wagner chemist and Robert Monnet Hare; (1979), in a much physiologist larger Charles survey Richet of (Nobel professors Prize, at 1913); 120 psychol- colleges and universities ogists in William the James, United William States McDougall, (response Carl rate Jung, 49.5%), Sir found that Cyril 73% Burt, of and the Hans respondents Eysenck; from anthropologist the Margaret humanities, arts, and Mead; education mathematician indicated John they Taylor believed (who ESP became to con- be either an vinced established of fact the or reality a of likely psi possibility, phenomena whereas on only the 55% basis of Uri Geller's of purported the feats respondents (Taylor from 1975], the only natural subsequently sciences and 34% of to the repudiate psychologists his did belief likewise. in Whether such the phenomena differences [Taylor & Balanovski between 19791); the and results Robert of Jahn the of two the surveys Engineering reflect differences Department in at the Princeton questions University. asked or differences in the groups There sampled has (the also former been study a was history limited of to professional respondents interaction between from conventional two science large and and parapsychology prestigious at universities) sci- is not clear. entific (It conferences, should through be symposia noted on in the any paranormal case that such surveys are and always invited subject addresses to by a parapsychologists response (e.g., bias, Ameri- in that there is likely to can be Association a for differential the response Advancement rate of as Science, a 1975, function of attitude 1978, toward 1984; the American subject Physical matter Society, being 1979, addressed.) American Psychological Although Association, all 1966, of 1967, this 1975, might 1984, suggest 1985), that parapsychology although is admittedly a such serious opportunities and for professional para- research discipline that is psychologists viewed to with present respect their within ideas university and settings, evidence at have best been parapsychology limited. struggles to maintain a toe-hold at the fringes of academia; mainstream science continues vir- 0.2. tually Organization. to As ignore a its research subject discipline, matter parapsycholo- or even to reject and gy ridicule is it. organized One very finds much no the mention way of various psi disciplines phenomena of in mainstream textbooks science of are. physics There or are chemistry professional or bodies biology. Lecturers do that not emphasize address empirical the enquiry paranormal using in scientific undergraduate meth- or graduate odology science and programs. that Psychology encourage students high are research rarely standards. taught (One anything of about these, the the subject. Parapsychological Parapsychological Association, research over half pa pers are only very infrequently of published in the jour- whose 300 or so members hold doctorates in science, nals of "normal" science, and parapsychologists engineering, have or medicine [McConnell 19831, has been criticized affiliated leading with scientific the publications American such Association as for Science, the Advancement The of American Science Journal since of 1969.) Physics, Annual and research American Psychol- conferences ogist are for held. suppressing Research the grants dissemination are of awarded. para- There psychological is research substantial findings empirical (Honorton literature 1978a; in Mc- the field - including Connell several 1983). research Funds journals for and parapsychological many research books, are some usually of generated which within have parapsychology been itself published or by come leading scientific publishers (e.g., the Handbook of from parapsychology private donors; the agencies that fund normal sci- [Wolman 1977a], the Foundations ence of parapsychology turn a blind, or even hostile, eye toward para- (Edge psychological et research al. proposals. 19861, The and United the States series gov- Advances in para- psychological ernment, research however, [Krippner has 1977; provided 1978a; multi-million-dollar 1982a; 1984]). support for psi research into remote viewing at SRI International in California (Targ 6r Harary 19841.) 0.3. What Academic accounts Involvement. for Courses the in disparity parapsychology between what would are seem offered to for be academic a credit substantial at degree about of 50 professionalism colleges in and universities parapsychology in on the the United one States hand, (McConnell and 1983); the a continuing few even relegation grant of degrees parapsychology in to the the subject fringes (see of Stanford science 1978). on Ph. the D.'s other? have For been one awarded thing, for parapsychology parapsychological continually re- search encounters at opposition Cambridge from University, mainstream the psychology; University psy- of Edin- burgh,.Surrey chologists University, appear Purdue to University, constitute the the Uni- most skeptical group versity concerning of whether the psi Witwatersrand, is and likely the to City exist University (Alcock of 1981; New Wagner York, & among Mormet others. 1979). The Second, University people of who Edinburgh may serve has as recently the established "gatekeepers" the of Koestler science, Chair in in that Parapsy- they are very chology, influential which in is determining endowed what by is a and bequest is from not the the late proper Arthur subject Koestler, matter a of long-time science, supporter are of skeptical parapsy- about psi. A chology. recent survey of "elite" scientists (Council members and How selected do section members committee of members academia of view the claims American about psi? in Association one for survey the of Advancement humanities of and Science) science revealed professors the at two large highest universities level (University of of skepticism Michigan regarding and ESP University of any group of surveyed Toronto; in response the rate last 53%), 20 only years about (McClenon one-third 1982): of Fewer the than respondents 4% indicated of believing the in 339 paranormal respondents phe- (the response rate was Approved For Release 2000/08/08 : CIA-RDP96-00789ROO2200120001-4 554 BEHAVIORAL AND BRAIN SCIENCES (1987) 10 4 Approved For Release 2000/08/08 CIA-RDP96-00789ROO020012400".-.-b., .,i~wea ESP as scientifically established. (However, :her 25% considered it to be a likely possibility, cating about the same proportion of favourableness as )rted by Otis and Alcock (19821, cited above.) Fifty ,ent considered ESP to be impossible or a remote AilitY. McClenon's (1982) view, this negativity is based on threat that paranormal phenomena, were they to it, would pose to the prevailing scientific worldview. A ,er different viewpoint, which is part of the thesis of ; paper, is that parapsychology, over its century or so of stence as an empirical research endeavor, has simply ed to produce evidence worthy of scientific status. Of irse, both these views could be correct. ro facilitate the discussion of this issue, I shall proceed posing a number of questions I consider to be impor- it concerning psi and parapsychology: i .What is psi; how is it defined? 2, is psi "possible"? 3. if psi exists, how can it be detected? 4. What is the evidence that psi exists? 5, Does parapsychology follow the rules of science? 6. Are the critics fair? 7. Is rapprochement possible between psychology and trapsychology? et us consider each of these questions in turn. What Is psi? Although only a few parapsychologists appear to share his conservatism, Palmer (I.Ma; 1986a) argues that until parapsychologists have produced a positive theory of psi which describes the properties that must be present in order to claim that psi has occurred, all they can claim to have demonstraied is the occurrence of a number of anomalies which themselves constitute the subject mat- ter of psi. Seemingly paranormal events might be explica- ble in terms of conventional science or science as it will be understood in the future, he says, or, indeed, such events might be due to errors in interpretation or measurement or statistical analysis. He recommends that the term 11 paranormal phenomena" be supplanted by a much less committed term such as "ostensible psychic events." Palmer's circumspection is commendable and would find favour with most critics of parapsychology. However, it is rare to find parapsychological research reports or other kinds of literature treating apparent anomalies in such a noncommittal fashion. Most, in fact, treat psi not as a description of an anomaly but as a causative agent. There is a second and more important sense in which psi is negatively defined, albeit implicitly, and that is in terms of its incompatibility with the prevailing scientific worldview (Boring 1966; Flew 1980; Mackenzie & Mac- kenzie 1980): In some way or another, psi phenomena, to be considered as such, are impossible if the current worldview is correct. There are two different camps within modern parapsychology regarding this incom- patibility (Beloff 1977): Ithough it may at first seem straightforward to define or atalogue paranormal phenomena, it turns out to be a ifficult task indeed, for there is a considerable spectrum f opinion even within parapsychology as to which osten- ible phenomena are likely to be genuinely paranormal -nd which are probably based on error and self-delusion. ~or example, although many parapsychologists might coff at such claims, some believe that "psychic healers," hrough the laying on of hands, can speed the healing of vounds and slow the growth of fungi (Krippner 1982b); ,thers believe that some gifted persons can project im- ges onto photographic film (Eisenbud 1977), that water ources, or even lost treasure, can be located by "dows- rig" with a willow stick (Bird 1977; Schmeidler 1977), that eincarnation warrants serious investigation (Child 1984; )Itevenson 1977), that one's personality can leave and return to the body at will and may even be able to travel through outer space (Targ & Puthoff 1977), and that deathbed visions may be indicative of survival after death (Otis & Haraidsson 1978). Because there is no general agreement on what psi is, or at least how it manifests itself, parapsychologists have found it easier to define it in terms of what it is not. The term "psi" itself was introduced by Thouless (1942) as a neutral label in order to avoid the many associations that terms such as "psychic phenomena" and "extrasensory Perception" have developed over the years, and psi is defined simply as "interactions between organisms and their environment (including other organisms) which are not mediated by recognized sensorimotor functions" (Krippner 1977, P. 2; my italics). Psi 'phenomena, then, are explicitly defined in a negative manner: To demon- strate that psi has occurred, one must first eliminate all notwwl sensorimotor explanations. 1.1. Incompleteness of current science. just as the scien- tific worldview changed to accept the extraterrestrial source of meteorites and the constancy of the speed of light, so too, according to this viewpoint, it must ulti- mately accommodate psi. Thus, "paranormal" phe- nomena are part of the natural order, but a part of that order which is not yet understood; as soon as scientific knowledge advances to the point that the paranormal is comprehensible, then the latter will become part of an expanded normal science (Truzzi 1982). This process has been manifested already in several instances: Bat navigation was taken to involve psi until the echo-sounding apparatus of bats was discovered, at which time it became part of the normal scientific domain and of no further interest to parapsychologists (Boring 1966). Bird navigation (Pratt 1953; 1956) and hypnosis (see McConnell 1983; Spanos 1986) are other examples of phenomena that have passed from the realm of the paranormal to the normal. 1.2. A nonphysical dimension of existence. According to this perspective, paranormal phenomena mark the outer limits of the scientific worldview, and beyond those limits "lies the domain of mind liberated from its dependence on the brain. On this view, parapsychology, using the methods of science, becomes a vindication of the essen- tially spiritual nature of man which must forever defy strict scientific analysis" (Beloff 1977, p. 21). of these two perspectives, the incompleteness ap- proach would no doubt be more acceptable to most scientists. Yet, it does not really capture theflavour of the paranormal. Whereas anomaly is, it would seem, a neces- sary condition for paranormality, it is not a sufficient one. Approved For Release 2000/08/08 CIA-RDa6AQQZRSMOA26(klt2GOW(l4B7) 10A 555 Alcock: p,&ppf""&L%AJW"se 2000/08/08 : CIA-RDP96-00789ROO2200120001-4 Were without some it reference sufficient, to the independence then of the mind all anomalies throughout the history from the materialistic of realm (Rhine, science L. E. 1967). would have to have been considered .. paranormal, " whereas it is clear that they have not been considered 2. Is pal as "possible"? such (Braude 1978). Braude (1978) suggests that a definition of the paranor- mal Psi phenomena must are defined go implicitly beyond in terms anomaly of their to include the notion that it 1. thwarts incompatibility our with the familiar contemporary expectations scientific about world- what sorts of things view. Although can many parapsychologists happen (e.g., Rao to 1983) the sorts of objects involved" (p. 241). believe that Yet, only a major as revolution Mabbett in scientific (1982) thought points out in response to Braude, could lead experimental to the accommodation parapsychological of psi, there studies have been that are taken attempts to to reconcile demonstrate such phenomena the with modern reality of psi typically produce scoring science. For rates example, that although are it would only seem that slightly psi above chance; these hardly cannot occur thwart without violating peoples' well-tested expectations, laws of physics and "even the thoughtful to regard - such as layman such the law ofconservation would of matter be and energy unwilling and results k without the inverse as, a little square law evidence of energy of propagation anything (signal but luc assurance strength is or proportional instruction to the inverse from square of the the expert statistician" (p. 340). distance) - or violating the logical principle that an effect On the cannot precede other its cause, hand, ad hoc explanations the of how psi bizarre and paradoxical proper- ties might occur of without such light, violation as have been described proposed by relativity theory, would no doubt (Collins & have Pinch 1982). been As an example, unexpected with regard by to the laymen as well as by scientists presumed impossibility prior of seeing to into the Einstein, future, one Mabbett says, yet most peo-- ple could posit would that what not appears to have be precognition regarded is really these properties as paranor- mal. psychokinesis: Mabbett The individual argues uses PK to that bring about paranormal the phenomena are psychological events he in believes the have been sense foreseen that precognitively. they In involve mind or consciousness, a similar whatever fashion, these one may be may able to construct be, other ad and that they reflect hoc explanations a relationship to overcome between all the various the incom- mental and physical worlds patibilities that that appear is to exist radically between physical different science from that conceived of by science. and parapsychology, although such contrived mecha- What nisms are y to satisfy most scientists. is not likel being struggled with here by Braude and Mabbett A more direct is attempt to that, render psi more compatible than with being simple anomalies, paranormal contemporary phenomena science has have been made a special through efforts and to particular relationship show that to such phenomena the are not inconsistent human with mind. Indeed, as I have dis- cussed quantum mechanics. in In recent greater years, there detail has been elsewhere (Alcock 1985), it is hard to escape considerable the discussion conclusion in parapsychology, that led by para- the concept of paranor- mality physicists implicitly (parapsychologically involves oriented mind-body physicists) dualism and (see Wol philosophers, man about some 1977b), of the paradoxes the of quantum idea that mental processes cannot be reduced mechanics to and about physical how it is processes possible and to suggest that solu- the mind, or part of it, tions to these is paradoxes nonphysical that imply in the direct nature. influence of The the mind on late matter, allowing Gardner for - or Murphy even demanding (1961), - once president of the American psi (e.g., Psychological Oteri 1975; Association Schmidt 1975; and Walker 1974; one 1975). of parapsy- chology's This has generated most negative erudite reaction and even within persuasive para- proponents, ar- gued psychology that (e.g., Braude even 1979a), with if some para- the paranormal were to be defined only in terms physicists of such as Phillips anomaly, (1979; 1984) this arguing that would the still lead to a dualism of some orthodox view sort of quantum because mechanics of does not its lead to independence from considera- tions paradoxes of that necessitate time the introduction and of mental space. Indeed, parapsychologists have at times influences. insisted Phillips that describes psi the difficulty phenomena and the ar- are distinguished from the bitrariness other of interpreting phenomena the mathematical of picture psychology by virtue of the fact that served up they by quantum are theory: "Me of predictions a nonphysical of quan- nature (e.g., Rhine, J. B. & Pratt tum mechanics 1957). have been Although verified, the and there boldness is little of such a declaration might doubt that well the mathematical raise formalism the is correct. hackles Con- of some modern para- psychologists, structing most a physical of picture to them correspond do to the mathe- seem to accept such dualism matics is (Thalbourne much more 1984). difficult, The and authors influence dffer in of what dualistic thinking they find creates intuitively a deep appealing schism and philosophically between satis- parapsychology and factory" (1984, modern p. 298). science. In summary, Even if quantum then, mechanics although did allow some for psi - modern a notion pam- psychologists few mainstream prefer scientists to would be speak likely to only accept at of anomalies, these anomalies, present - if that would they not in itself are make the to reality of be psi of continuing interest to parapsychology, any more likely. must Flying cows ultimately are not inconsistent involve with some radically different quantum mechanical relationship notions, between but as far consciousness as we know, and they the physical do not exist. world What is missing than in such discussions that of psi is held to be possible by contempo- rary the phenomenon science. itself. Until Some there is parapsychologists clear evidence might that deny being mind-body psi exists, dualists, it is surely but premature they to try to would bend quantum do well to consider just *~6&0* tRe-416% how ~~Mkffl-g :"Ae-PDPAe*Affl:8&MUV0 14 Alcock: Parapsychology critique Approved For Release 2000/08/08 CIA-RDP96-00789ROO2200120001-4 I.-I pal exists, how can It be detected? psychological concepts. If there were unobserved weak nesses in the controls, if some unknownprocess were code based of -iere are three major sources of evidence for psi: (a) iecdotes of spontaneous personal experiences, (b) dem- ,,trations by "gifted" psychics, and (c) laboratory ex- ~riments- The early studies of psi examined anecdotal sports in great detail, but gradually the realization grew iat such evidence is just too unreliable to serve as data ,r science (H6velmann & Krippner 1986; Rhine, L. E. )77, Bush 1986a), "Gifted" psychics have provided the most spectacular si claims, both in the early days of psi research and more ,cently (Rush 1986b). For some parapsychologists (e.g., ;e1off 1985), such demonstrations still stand as strong ?stimony to the reality of the paranormal. Yet, once gain, this evidence is unsatisfactory in the extreme, iecause of both the history of fraud involving reputedly -ifted psychics (e.g., see Girden 1978) and, more impor- ant, the fact that such psychics have as yet been unable to wrform their feats under controlled conditions for neu- ral or skeptical investigators. For example, Uri Geller vas taken by a number of parapsychologists (e.g., Beloff ,975; Cox 1976; Eisenbud 1976; Hasted 1976; Moss 1976; ?uthoff & Targ 1974) to have genuine paranormal powers intil a conjurer's investigations (Randi 1975) showed to :i,vat people's satisfaction that Geller was using trickery. Some parapsychologists (e.g., Schmeidler 1984) insist that the fact that a psychic is caught cheating does not weaken the evidential value of those demonstrations during which the same psychic was not caught cheating. Given the rarity of such supposedly gifted individuals, it is not surprising that investigators are loath to terminate their research with an individual just because fraud has been detected on some occasions. However, it is no easy task to guard against fraud if a subject is determined to cheat, and what better indication is there of such deter- mination than the subject's being caught at it? It was because of dissatisfaction with both anecdotal evidence and uncontrolled demonstrations that Joseph Banks Rhine, in the 1930s, set up an experimental labora- tory for the study of psi. The hope was that through rigorous application of the methodology of science, psi would soon be put on a solid empirical footing. Rather than simply relying on the ability of self-proclaimed psychics to demonstrate their skills, Rhine began the systematic study of both gifted and ordinary individuals in a number of "guessing" tasks in which probabilities of success could be calculated. If one makes a prediction, based on a probability model, as to how well a subject should score in a guessing task, or if one predicts the distribution of events whose occurrence depends on a random process (in Rhine's day, dice-throwing; nowa- days, subatomic particle emission) which the subject attempts to influence mentally, then if all known normal forces have been ruled out, statistically significant depar- tures from the prediction are taken to indicate the in- volvement of a psi process. Thus, experimental parapsy- chology, just as conventional psychology had done before it, took on a pronounced statistical flavour. If one could reliably demonstrate departures from some statistical model, this would call out for explica- tions. There would be no justification, however, for. beginning with an explanation based on para- on silent some involved (e.g., the use counting, or the use of "silent" dog whistles that children, but not adults, can hear [Scott & Goldney 19601), if there were equipment problems or biases in the random gener- ator, if the statistical model were inappropriate, or if errors were made in the recording or analysis of the data, the paranormal explanation would be erroneous. just as important, in the absence of a positive theory of psi, even if an observed effect is not due to artifact, one is left only with an anomaly. The availability of the psi hypothesis can distract the researcher from other, normal, explanations and thus impede the development of the understanding of anomalies (Blackmore 1983a). What would constitute "solid" evidence of psi? Ob- viously, no evidence is ever 100% solid, because we can never be sure how new discoveries will change our understanding of processes that we currently think we understand. Furthermore, evidence that seems uncon- vincing or unimportant in the light of one theoretical worldview may be viewed as much more important if the prevailing theory changes. An extraordinary degree of evidence is often de- manded in support of extraordinary claims. We are gen- erally less demanding of evidence in the case of claims that "fit" with existing theory or knowledge. When one is weighing evidence in law, the distinction is made be- tween "beyond all reasonable doubt" and "on the balance of probabilities." The former, applied in criminal cases, demands virtual certainty of guilt; the latter, used in civil litigation, refers to the notion that the defendant is more likely than not to be guilty. Because psi is a concept that would probably revolutionize science (Rao IM), most skeptics implicitly use the criterion of beyond all reason- able doubt, while accepting conclusions made on the balance of probabilities where only "normal" and 'non- controversial phenomena are involved. However, al- though the controversial nature of psi may justify the use of tougher criteria, this view has been attacked as being another tactic for denying legitimacy to controversial claims (McClenon 1984). Before we accept that psi (even in the simplest sense as an anomaly) has been demonstrated in the laboratory, three important factors must be considered: 3.1. Intemal validity. Psychologists use the term "internal validity" to refer to the degree to which experiments are free of the influence of extraneous variables that might introduce alternative explanations for the observed re- suits (Berkowitz 1986). Most criticisms of experimental studies of psi concern internal validity. Randomization maybe inadequate, sensory leakage (i.e., communication of information by normal sensory means) may have oc- curred, and so forth. McClenon (1984) argues that such methodological crit- icisms of psi experiments are often unfair. By refusing to accept the shared assumptions that are implicit in any experiment, he says, the critic will sooner or later "ask for information that is no longer available, or for a degree of experimental control and exactitude that is desirable in principle but impossible in practice" (p. 89). Thus, the 1. perfect" ESP experiment is an impossibility, McClenon Approved For Release 2000/08/08 CIA-RDP96-00789ROO2200120001-4 BEHAVIORAL AND BRAIN SCIENCES (1987) 10:4 557 Alcock- PAW6V&10~%,ic1*W§se 2000108/08: CIA-RDP96-00789ROO2200120001-4 contends, for one has no clear rationality, can always suggest its only chance that the experiment- of demanding er was incompetent scientific attention or that trickery is replication" was involved (see (p. 4), also Honorton 198 1). Despite On this basis, repeatability McClenon's concerns, is, in general, there is a less imPor. considerable difference tant in psychology between making unsubstantiated than in parapsychology. Even so, charges of incompetence psychologists pay or trickery and pointing far too little to attention to the importance methodological flaws. of repeatability If the flaws are (Epstein 1980; there, para- Fishman & Neigher 1982; psychologists should Furchtgott 1984; run the experiments Heskin 1984; Sommer again - without & Sommer 1983; the flaws - rather 1984); replication than argue about studies account the motivation of for a very small the per- person who pointed centage (3% or less) them out. in leading psychology journals Rather than rerunningperiments (Bozarth & Roberts the ex correctly, 1972; Sterling a 1959). This has led on more usual response he critic. occasion to the is to attack t For widespread dissemination example, of information critics have been that is subsequently chastised for pointing found to be unreplicable to flaws without (see, for demonstrating that example, Marshall these flaws are capable & Zimbardo 1979; of generating Maslach 1979; the observed departures Schachter & Singer from chance (Honorton 1979). 1975; 1979; Palmer 1986a). Even when replication This criticism does is attempted, its not stand up, for importance two reasons. First, often depends on critics are usually who conducts it. not advocating the We are not likely to acceptance of an alternate accept a wild claim hypothesis but asking supported by the only that research of only one claims of psi be suspended person, whether until properly controlled that research has been replicated by that studies are carried person or not (Hyman out (Akers 1984; 1977a). Similarly, Hyman 1981). Sec- a failure to ond, such flaws need replicate by a student not be the cause in a high-school of the statistical science class will deviations, but they carry little or are symptomatic of no weight, whereas lax research a failure to replicate by a standards (Hyman 1985b). well-respected scientist One should hardly will be much more have confi- seriously dence in the experimental viewed (Collins controls if one is 1976). It is also faced with difficult to know just what evidence of violations constitutes a replicatione & Morris 1986); of proper procedure. (Edg there are Akers (1984) uses the "dirty test-tube" in fact several s of replication analogy: A chemist different kind that one can would have little confidence provide (Alcock in a colleague's 1981; Lykken 1968). findings if it were Beloff (1984) differ- ob- served that a test entiates between tube used in the "weak" and "strong" experiment was replicability, contaminated. where the former term refers to a situation in which an It is not so difficult experiment or phenomenon to design and execute has been independently an experiment con- that is methodologically firmed by at least and statistically one other investigator, sound. Psycho- and the latter logical experiments refers to a situation published in the in which any competent better psychology researcher, journals stand in following the prescribed evidence of this. procedure, can obtain the re- ported effect. Although parapsychologists have present- 3.2. Consistency. ed, as evidence Before accepting for psi, studies the reality of a that have been phe- replicated nomenon, one generally by other parapsychologists, looks for signs that there has never there is a been a psi consistent pattern demonstration that of results across is replicable in experiments. The the strong sense lack (Be- of any consistent loff 1973; 1984; pattern in the research Palmer 1985b). findings is one of Indeed, para- the most serious weaknesses psychologist/psychologist in the evidence offered Susan Blackmore for (1983a) re- psi (Blackmore 1983a). cently referred Unfortunately, it to unrepeatability is standard prac- as parapsychology's tice in parapsychology only finding. to take one pattern of data as evidence for psi in Of course, even periment is replicated, one experiment, then if a psi ex that to disregard its absence and take some by itself does not ffect has a paranormal other pattern as mean the e cause. evidence for psi in another experiment. On the other hand, the inability to repeat an experiment or a demonstration cannot by itself rule out the truth of 3.3. Repeatability. the psi claim. Poor Not only should there repeatability could be consistency conceivably stem in the pattern of from factors other data across experiments, than the nonexistence but individual of psi (Palmer experiments should 1986b). It is possible be repeatable by that certain conditions others. Repeat- are necessary ability is an important for the production safeguatd, albeit of psi, and given only a partial one, that no one knows just against error or fraud . what these conditions (Sommer & Sommer are, it could be 1984). 0b- that an essential viously, however, element is missing replication by itself when an experiment is not enough. if fails to replicate. someone is dishonest of It has also been in the actual reporting the suggested that re- psi could turn out to be search, reports of hor inherently unlawful replication by the will (Palmer 1986b; same aut not Rao 1982), although eliminate the dishonesty this position is (Casrud 1984). difficult to defend (Hbvelmann & Kripp- Yet, as Rao (1985) ner 1986). From points out, repeatability this viewpoint, is not a it has been argued that matter of primary the quest for repeatability concern in normal should be abandoned science. Only if (Pratt some important and 1974). controversial finding is made is replication likely Despite the arguments to be attempted, about the relative and this will often unimpor- be undertaken by others tance of repeatability, who have competing the history of theories that science demonstrates would not accommodate that unrepeatable the finding. When experiments or observations demonWations should are consistent with be viewed with a theory, replication very cautious eye, is less important. Most para- However, as Murphy psychologists probably (1971) commented: would not dispute "If the event is this point. unclassifiable, then Indeed, the claim it is doubly important is made that the that it have a level of repeatability rational interpretation aarapsychology exceeds that is, one ~A ftit Anet6q RV p ~ - th thought pattANONRi e &C i Ica social sciences; the l al rep Alcock: Parapsychology critique n-rav ongest.claim in thPregard ~ct, for which replicability is said to be in the area of % (Honorton 1976; 1978b). This research is discussed the next section. in summary, then, although one cannot set precise ,ndards that evidence of psi must meet, judgment ould be suspended until there is at least some con- itency among research findings from a body of meth- jologically irreproachable experiments, at least some of ,lich are repeatable in Beloffs (1984) strong sense. were seriously flawed, and even the 8 that were conducted wit4 reason- able care were not methodologically ideal. The problems fell into several categories, including randomization failures, sensory leakage, inadequate safeguards against subject cheating, the possibility of errors in the recording of the data,, errors in statistical analysis, and failures to report important procedural details. Akers concluded that these 54 experiments taken together were too weak to establish the existence of a paranormal phenomenon. is there any substantive evidence that psi exists? 'ithin parapsychology itself, there are arguments about e strength of the evidence adduced for psi. Some argue .at no substantive evidence has yet been found (e.9., irker 1978), whereas others consider the laboratory ,,idence for psi convincing (e.g., Schmeidler 1984); still hers believe that psi can even now be harnessed - for (ample, to guide stock market investments (Targ & .arary 1984). On the whole, it would appear that most arapsychologists believe that psi has already been dem- astrated. Schmeidler (1971) reported that almost 90% of er small sample of members of the Parapsychological s-ciation indicated they believed that ESP had been stablished so firmly that any further research aimed only t demonstrating its existence would be uninteresting. ubsequently, in a survey of all 241 members and associ- tes of the Parapsychological Association (which yielded a esponse rate of 84%), 68% indicated complete belief in he reality of psi (McConnell & Clark 1980). The average. trength of belief across all respondents was 93%. Many studies have been carried out and published that )Urport to provide statistical evidence for paranormal )rocesses. However, even if we were willing to treat -ertain statistical deviations as evidence of psi, such widence has been unsatisfactory: A number of recent nalyses have demonstrated a serious problem with the uality of the methodology used in parapsychological ,search. For example, Akers (1984) cites a survey of 214 K experiments (May et al. 1980), in which the authors )ncluded that none had been properly designed and ,ported. In order to explore in more detail the state of the vidence in parapsychology, five major areas of contem- )orary parapsychological research will be discussed -)elow. 4-1. Out-of-body experiences. Blackmore (1982; 1984), ifter carefully studying both the anecdotal and research 'literature on out-of-body experiences (experiences in which the individual believes that the physical body has been left behind and that travel through physical space is therefore unencumbered by limitations imposed by the aesh) and after conducting her own research, came to the conclusion that normal psychological theories are capable of accounting for such experiences and that nothing paranormal is likely to be going on. 4.3. The psi ganzfeld effect, As mentioned earlier, studies of ESP in a ganzfeld (a condition of reduced sensory stimulation typically produced by covering a subject's eyes with halved Ping-Pong balls and shining a white light onto them while playing white noise into the subject's earphones) have been very promising in that they have appeared to demonstrate a replication rate of 50% or higher (Blackmore 1980; Honorton 1978b). Hyman (1985b) has completed an exhaustive analysis of virtually all psi ganzfeld research, using a data base of 42 studies conducted between 1974 and 1981. Hyman's analysis leads him to conclude that the replication rate exhibited in this collection of studies is probably very close to what would be expected by chance. Several flaws of procedure - including less than adequate randomiza- tion, the possibility of sensory leakage, and erroneous statistical analysis - plagued these studies; not a single study was flawless, he reported. He suspects that most of these studies were not well planned, and he concludes that this data base is too weak to support any assertions about the existence of psi. However, Honorton (1985) disputes Hyman's conclusions, arguing that his assign- ment of flaws is itself seriously flawed, and he maintains that these studies do indeed indicate a significant psi ganzfeld effect. Hymanand Honorton (1986) prepared a joint paper as a follow-up to the two papers discussed above. With refer- ence to the data base discussed earlier, they agree that the experiments as a group departed from ideal stan- dards on aspects such as multiple testing, randomiza- tion of targets, controlling for sensory leakage, applica- tion of statistical tests, and documentation. Although we probably still differ about the extent and serious- ness of these departures, we agree that future psi ganzfeld experiments should be conducted in accor- dance with these ideals. (p. 353) They go on to say that whereas we continue to differ over the degree to which the current ganzfeld data base contributes evidence for psi, we agree that the final verdict awaits the outcome of future psi ganzfeld experiments - ones conducted by a broader range of investigators and according to more stringent standards. (pp. 352-53) Thus, although the ganzfeld studies have been offered as the strongest evidence for a repeatable psi effect, any conclusion about a psi ganzfeld effect must await future research carried out more carefully than these studies were. 4.2. Personality/attitudinal variables and psi. Akers (1984) 4.4. Remote-viewing studies. In 1974 Nature carried an evaluated 54 experiments that studied the influence of article by two physicists (Targ & Puthoff 1974) in which altered states and of personality/attitudinal variables on they described their successful demonstrations of "re psi and that had been cited as significant confirmations of mote viewing," a talent by means of which subjects are Approved For Release 2000/08/08 CIA-RDP96-00789ROO2200120001-4 13EHAVICRAL AND BRAIN SCIENCES (1987) 10:4 559 Alcock: ParA"rbQie0yFZwM0ease 2000/08/08 : CIA-RDP96-00789ROO2200120001-4 able to describe geographical locations being visited by other people without having any normal form of commu- nication with them. This putative skill is said to be within everyone's capability (Targ & Puthoff 1977). For a period of time, this research seemed to promise a breakthrough in the search for a demonstrable psi effect. However, Marks and Kammann (1978; 1980), unable to replicate the remote-viewing effect themselves, discovered serious flaws in the remote-viewing procedure - flaws that they argued accounted for the observed effects. The principal flaw concerned the judging procedure: judges were asked to match up a series of responses against a set of targets. Marks and Kanimann argue that because the transcripts of the subjects' reports were not edited to remove cues that would assist the judges in identifying the targets, the judging procedure itself - and not any psi effect - produced above-chance matching of transcripts with targets. Tart et al. (1980) responded to this criticism by first having the transcripts edited to remove any possible extraneous cues, and then having them rejudged. They reported that this did not eliminate the remote-viewing effect. Hbwever, Marks and Scott (1986), after obtaining access to the relevant findings (they had until recently been denied access to the raw data), report that the editing of the transcripts had failed to eliminate all the extraneous cues and that enough cues remained to account for the above-chance scoring rate. There have been other criticisms of the remote- viewing studies as well, including concerns about statis- tical problems that could give rise to above-chance scor- ing rates (Hyman 1977b), and about the lack of adequate controls and control groups (Caulkins 1980). A number of replications and extensions have been reported (e.g., Bisaha &- Dunne 1979; Dunne & Bisaha 1979; Schlitz & Gruber 1981; Schlitz & Haight 1984); only the Schlitz and Haight (1984) study appears to avoid the weaknesses of the Targ-Puthoff series, but even here, there was no control condition to allow proper assessment of the back- ground "coincidental" scoring rate. Thus, the Targ-Puthoff series is too flawed to be of evidential value, and none of the subsequent published studies have been carefully enough controlled to bear testimony about psi. 4.5. Schmidt's random-event generator (REG) studies. For almost 20 years, Helmut Schmidt has been conduct- ing research into the ability of subjects to predict or influence the radioactive emission of subatomic particles. His research enjoys generally high regard from other parapsychologists: Beloff (1980), for example, views some of Schmidt's research as being among the most evidential in all of parapsychology, despite his own inability to replicate Schmidt's findings. Schmidt has published a considerable number of stud ies. Unfortunately, this investigator typically completes a study and then - rather than focusing on a given research question, or refining his measurements, or examining the effects of various parameters in that particular situation, or working with one type of generator over a period of time so that he and others can come to appreciate its idiosyncrasies - he moves on to a totally different situa tion altogether (Hansel 1980)i changing the design and components of his genexator-as. he mati r AW91- To oq V 1981). This" reading his research reports to learn the limitations of his generator or his procedures. Little of Schmidt's research is free from serious meth- odological shortcomings (Hansel 1980; 1981; Hyman 1981). Consider, for example one of his initial studies (Schmidt 1969b), which has iteen favorably cited many times in the parapsychological literature. The situation was as follows: A subject was seated before a panel Of four lights and four corresponding buttons. On each trial, the subject would press one of the buttons to predict which light would next illuminate, something that would be determined by particle emission from a strontium-go source. The light would then illuminate, giving immedi- ate feedback. if the light corresponded to the depressed button, it was a "hit." In the first experiment in this report, Schmidt com- bined the results from his three subjects and obtained a hit rate significantly higher than would be expected by chance: 0. 261 as compared to 0. 250 (p < 2 X 10 - 9). In the second experiment, subjects were allowed to choose to try to make a high or a low number of hits. Here, the combined scoring rate of three subjects was 27%, again significantly higher than chance expectation (p < 10- 10). Both experiments suffered from less than optimal ex- perimental control; as in most of Schmidt's studies, sub- jects were usually unsupervised, and there was a general lack of rigour in the control of experimental conditions. Hansel (1980) objected to the fact that the exact numbers and types of trials undertaken by each subject were not specified in advance, and also to the fact that the equip- ment, although partially automated, did not rule out cheating during data classification. There is a more fundamental concern about these experiments: the target series (Hyman 1981). Schmidt compared the subjects' hit rates to chance expectation, but this assumed that the target series was random. (Particle emission is presumably random; the output of his generator was not necessarily so.) Schmidt's ran- domization checks were carried out on target strings much longer than those used in the experiments, and therefore did not allow the detection of possible short- term biases in the generator which could give rise to nonrandom target strings. Because immediate-feedback VVUb PIUVLUCU U11%JUY,11%JUL UIV 'ZILIJUILUIVUL, iulu UCLAWbu subjects were free to "play" with the equipment and to decide when to start and stop a given session, any un- detected short-term bias in the generator might give the subject the impression of being "hot" and therefore lead him to initiate a session, which he would probably end once he seemed to turn "cold." This, of course, could produce above-chance scoring rates. It would therefore be important and appropriate to analyze the actual target sequence in terms of how well it conformed to what would be expected by chance. How- ever, were one to find that the target sequence was nonrandom, this could, after the fact, be taken as evi- dence of PK. Indeed, Schmidt reported that after the testing one subject said he had tried to affect the outcome rather than just predict it; be had tried to produce more illuminations of lamp no. 4, he said. It was found for this subject that there was indeed an excess of 4s in his target series. No indication is given in the report as to whether out for other subjects, gra) CORY, Schmidt subsequent1v Approved For ReWse 20.00 F/08d0q!P6"SPft_ffi~g 0 &Q99SQkW"1-0ropertyofpsi. d this same piece of-apparatus in a expenme Irnidt & Pantas For example, the 1972) in which decline effect the only task in one experiment was to try was ,nfluence the interpreted as machine to a "sign of psi" produce an that was taken excess of to strengthen 4sl )ve-chance rates were reported the claim of a scoring in that in- genuine psi effect (Bierman & Weiner lee as well, ed Schmidt again to 1980). which l conclude that was operating. 3. In a related The skeptic vein, Schmeidler is left wondering (1984) reports whether that PK t apparatus effects are often simply produces strongest just an excess after a session of 4s from has termi- time :iMe. Certainly, natted or during nothing can a subject's rest be concluded period. Rather from such than ,orts until ignoring data more is known accumulated after about the the session is target series over, this is duced by the taken to reflect generator. another psi phenomenon, and has been rhus, a study given two names that seems - the "linger at first to effect" and the offer considerable "release of dence of an effort effect. anomalous " if this is process is to be taken seriously, found to be then all badly ,ved. it would researchers should make sense report not only for Schmidt the presence to redo the of such .dy, taking an effect, but steps to make its absence as these criticisms well: were this unnecessary. done, the -iierally lacking frequency of the in Schmidt's effect may well studies is turn out to be a proper control within the (idition: One bounds of normal should generate statistical expectation. pairs of runs, with one f, designated, 4. Some parapsychologists seem consistently on the basis to obtain of some random procedure ch as the toss the results they hereas others are of a coin, desire w unable to find as the experimental and the her as the significant departures control for from chance (Palmer each trial 1985b). The (Hansel 1981). The problems failure of one in this study researcher to recur over obtain significant and over in results using hmidt's research the same procedure (Hansel 1980; that yielded 1981; Hyman significant results 1981). for a1v one of another researcher, his studies rather than being appears well taken as a failure designed (Schmidt to et 086). However, replicate or as we must wait a hint that extraneous to see whether variables may the psi be Feet apparently producing artifactual obtained in results, is often this very interpreted in recent study terms stands ) to replication. of the experimenter There have effect. This been many effect is so psi studies common in psi (e.g., irg & Puthoff research that 1974) in the it has even been past that described by at first appeared one para- ~vond reproach, psychologist as only to be parapsychology's found later one and only to be seriously finding -twed. (Parker 1978)1 To describe the fact that two researchers In summary, obtained different these various results by calling areas of research it an experimenter are lagued by methodological effect is quite and statistical appropriate. flaws of one After all, the sort experimenter r another. effect as such Until research is by no means is undertaken unique to parapsychology, that is meth- Jologically and a great deal well planned has been written and well executed on the subject - as Hyman with regard to research ,id Honorton in psychology (1986) recommend and other domains with regard (see to the anzfeld - there Rosenthal & Rubin is little 1978). However, point in debating in psi research whether or the not ie existing term is all too evidence establishes often used more a case for as an explanation psi. than as a description, and that is because it is considered that the Does parapsychology effect may result follow the not only from rules of experimenter error (in science? that one experimenter may be more successful in obtain- ing psi effects than another because he unwittingly allows ,f course, more artifacts by using the to contaminate term "rules his procedure), of science," or from one could -)en up all differences in manner of personalities dispute because (in that some of the difficulty experimenters iat exists may put their in listing subjects into those rules a more comfortable or in demarcating and psi- science om pseudoscience conductive frame (e.g., see of mind than Bunge 1984; others), but Edge & Mor- also from the s 1986). Rather psi influence than tackle of the experimenter that conundrum, himself (Krippner it is more rofitable to 1978a; Palmer examine several 1985b; 1986b). aspects of If psi exists, parapsychological of course, it ndeavor that would only make appear to sense that the run counter experimenter, to the spirit who natu- of ;eientific rally wants his inquiry; each experiment to is discussed succeed, might below: un- knowingly bring his psi influence to bear, whereas a 5.1. Unfalsiflability. skeptical or neutral There are experimenter a number of might not use principles psi at all, in parapsychology or might use it that can be to prevent the used to explain appearance of away failures a subject psi to find empirical effect. This whole support for problem leads a hypothesis, Palmer (1985b) thus creating to de- a situation of scribe the experimenter unfalsifiability: effect as the most important 1. Perhaps challenge facing the subject parapsychology did significantly today. It is worse than hard to imag- expected by ine scientific chance. If inquiry of any so, this may sort if the results be taken as of the evidence of psi, because investigation it seems to are determined be psi-missing, by the psychic something influence of which occurs the investigator so often that (Alcock 1985; it is now see also Krippner taken to be 1978b). a man- ifestation The experimenter of psi (e.g., effect (or the Crandall & experimenter Hite 1983). psi ver- 2. If outstanding sion of it) provides subjects subsequently a powerful method lose their for undermining psi ability, or failures to replicate, if subjects and is sometimes do more poorly resorted to forjust toward the end of a session or that purpose. of a series For example, of trials, when Blackmore this is labeled (1985), a the decline effect (e.g., devoted parapsychologist see Beloff for many years, 1982). Rather found herself than being taken as a Possible consequence becoming increasingly of either skeptical about statistical psi as a conse- regression or the tightening quence of her up of controls inability to (when that produce experimental has occurred), evidence the decline for it, she noted effect qften that "many parapsychologists takes op the suggested ower of an explana- v C CIA-RDP96-00789ROO2200120001-4 Approved or elease 2000/08/08 BEHAVIORAL AND GRAIN SCIENCES (1987) 10:4 561 Alcock:. ParaPAJT~PAWPMRV& lease 2000/08/08 : CIA-RDP96-00789ROO2200120001-4 that the reason I didn't get results was quite simple - tne. Perhaps I did not sufficiently believe in the possibility of psi" (p. 428). In summary, it is the way such "effects" are used - and not, in principle, the research procedures - that vitiates the scientific respectability of parapsychology, for they make the psi hypothesis unfalsiflable by providing ways to explain away null results and nonreplications. These descriptive terms have mistakenly come to be taken as properties of psi, which leads to the circularity of explain- ing an observation by means of the label given to it. Moreover, as important properties of psi, their nonap- pearance in a psi experiment should weigh against any conclusion that psi has occurred; this never happens in the parapsychological literature. 5.2. All things are possible. Another aspect of parapsy- chology that makes critics uncomfortable is what seems to be almost an "anything goes" attitude, with no specula- tion seeming too wild. For example, so-called observa- tional theory based on paraphysical interpretations of quantum mechanics, predicts that random events can be affected simply by being observed, even if the observa- tion occurs at some time in the future (see Bierman & Weiner 1980). In line with this notion, studies have been done which claim to show that subjects can exercise an -influence backwards in time ("retroactive PK") so as to affect the choice of stimulus materials preselected for the study in which they are participating (e.g., Schmidt 1976). This also means, of course, that the present is possibly being influenced by future events (Martin 1983). A "checker effect" has also been postulated, in which E S P scores may be retroactively and psychokinetically influ- enced by the individual who checks or analyzes the data (Palmer 1978; Weiner & Zingrone 1986). Schmidt (1970c) reported that cockroaches were able to influence a random-event generator in such a way as to cause them to be shocked more often than would be expected by chance. He suggested that perhaps his own psi, fueled by his dislike of cockroaches, accounted for the increase, rather than a decrease, in shocks. Not only can psi apparently transcend temporal bound- aries, it also seems that no effort, no training, and no particular knowledge are required to use it. Indeed, modern PK studies appear to indicate that psi is an unconscious process, but a goal-oriented one in that it helps the individual attain desired objectives: Success in a PK experiment does not require knowing anything about the target, or even knowing that one is in a PK study (Stanford 1977). Thus, psi appears to operate very much like wishful thinking. For example, going back to the Schmidt (1969b) study, all that was needed, it seems, was for that one subject to wish for a particular light to come on and it would light up statistically more frequently than the others. (Of course, when subjects do score above chance, neither they nor anyone else can say which hits were brought about by psi and which were the conse- quence of chance.) As I have argued earlier (Alcock 1984), the fact that no physical variable has ever been shown to influence the scoring rate in psi experiments (Rush 1986c), combined with the apparent total lack of constraints on the condi- miles, between humans and objects, between humans and animals, or even between animals and objects) serves to weaken the a priori likelihood that psi, as ' any sort of force or ability, exists. After all, most Psi experi' ments are very similar, in that all that is typically done is to examine two sets of numbers, representing targets and responses in an ESP experiment or outcomes and aims in a PK experiment, for evidence of a nonchance associa. tion. It may simply be that the enterprise of parapsychol. ogy generates, from time to time, significant statistical deviations - be they the result of artifact, selective reporting, or whatever - which are then independent of the research hypothesis, so that no matter what the researcher is examining - the effects of healing on fungus, PK with cockroaches, ESP across a continent, or retroac- tive psi effects - the likelihood of obtaining significant deviations remains the same. (For example, if an REG produces an excess of 4s on a short-term basis, and if the procedure allows subjects to tap into this, then it should make no difference in principle whether the targets are generated on-line or were recorded a week earlier: If the subject aims for more 4s, he wfll obtain them.) Difficulty in replication by other researchers using their own equip- ment or slightly different procedures would, of course, follow from such a state of affairs, as would the experi- menter effect. This psi-as-artifact notion is not offered as an em- pirically testable hypothesis. I only mean to show that the lack of constraints on the appearance of psi undermines rather than strengthens its credibility. It would be hard enough to accept that a philosopher's stone can turn base metals into gold, as alchemists believed. It would be harder still to believe that it can turn anything into gold and that anyone can use it without any training. 5.3. Lack of rapport With other areas of science. Parapsy- chology, despite its efforts to find common areas of interest with other research fields (see the Handbook of Parapsychology (Wolman 1977a]), has failed to establish any genuine overlap with other disciplines, because, so far at least, other disciplines do not seem to need psi. if 1. normal" explanations for strange physical or psychologi- cal phenomena were exhausted, and/or if the influence of the researcher's consciousness appeared to have an effect on the way matter behaved in "normal" experiments, then a much greater number of scientists might be more open to the possibility of psi. Indeed, if parapsychologists are right about psi, then the well-tested theories of physicists and neurologists are wrong (Hebb 1978). It is perhaps noteworthy that the claims that psi can influence radioactive decay do not come from particle physicists in the course of their everyday work. 6. Are the critics fair? Some parapsychological proponents, such as Child (1985), argue that few in "normal" science bother to immerse themselves in the details of parapsychology, and instead gain a false or misleading impression from the accounts given by their colleagues who serve as critics of the field. Such critics are accused of unfair tactics, such as (a) arguing that unless fraud can be ruled out, it is the re~ Pa en Pa ha J, tir St. ca al: re bi SU fa re C1 tf n( W (I e, el d ai S( C( Si ef V fi F 0 t( S, n s tions under which psi can be manifested (whether for- most parsimonious explanation of psi claims; (b) setting ward in timeAp&4;~Wg V1pqtiWp~&%s?0W"M :OPIA-FRRROPS&7§PBCMgpi%WA%lC)$.caI research than for Alcock: Parapsychology critique Approved For Release 2000/08/08 : CIA-RDP96-0078-9ROO220012000.1-4 ,rch in the realm This process, of normal sciences although sometimes and (c) simply seemingly cruel and ting the possibility dogmatic, is of psi out-of-hQ perhaps necessary (see Collins & to allow scientists to focus h 1979). on claims that appear most promising, rather than being iarles Tart (1982; distracted 1984), a former by others president of the that appear to have little to recom. psychological Association,gests that mend them. sug there is an Sooner or later in science, it seems, the truth tional basis for lingness to will out, and critics' unwil welcome error falls by the wayside. Even acu- psychology into the puncture, long scientific fold, regarded as an argument that being nothing short of super- been repeated by stition, is Schmeidler (1985) now regarded and Irwin as capable of producing limited 5b), among others. pain relief Tart posits that (Zusne & Jones a widespread and 1982). jinscious fear of If the insulin-binding psi has developed antibody, either because biological prepared- ng psi ability would ness, and acupuncture disrupt social functioning analgesia (be- won accommodation in ;e we would have science, it access to one another's is because true feelings the evidence for them became so thoughts) or because strong that of what he calls they had to "primal conflict be accommodated. A century of -ession": A mother parapsychological often feels angry research has toward her child gone by, and the evi- keeps her cool and dence for psi speaks to the child is no more in a positive, convincing now than it was a portive way. The century ago. child, if psi is already operating, is ~d with a frightening ,It seems accordingly conflict of messages that parapsychologists and learns to who attack ress psi altogether scientists so as to avoid the and critics information for their refusal to recognize the nnel creating such importance conflict. Targ and of psi and Harary (1984), on of psi research are attacking the other hand, argue messengers that skeptics base because they their opposition cannot accept the messages on rationality but they bear. on religious conviction. Suppose that instead of psi, parapsychologists ;uggestions about were promoting fear and religious a cure for conviction are too baldness, but that the amount A and ad hoc to require of hair produced rebuttal. Collins by the treatment & Pinch's was tiny and detectable 79) concerns, on only by some the other hand, researchers, are important. How- sometimes. If the effect is ~r, they could be unreliable equally relevant and unrepeatable, to any controversial if it also contradicts all that im, and thus nothing is known about abnormal seems to hair growth be going on in and alopecia, and if there is no ~ .ritical reactions theoretical to parapsychology. mechanism The scientific put forth for the putative effect, ,na is a tough one; then one would many ideas march hardly expect in to do battle; the scientific community to ne survive, but just cheer the end as many perish. of baldness. Numerous other Science will never take para- itroversial claims psychologists have faced hostility simply at and even deri- their word; they must offer a in from scientists; clear, replicable some of these have demonstration won out (e.g., of a basic phenomenon in ntinental drift - order to gain see Hallam 1975); acceptance others (e.g., poly- in science. kter - see Franks Moreover, one 1981) have not. can seriously Psychologists were challenge at the claim that st unwilling to believe practitioners in the notion of of normal biological pre- science do not give, or have not redness with regard given, parapsychology to learning (i.e., its day in the idea that court. As was mentioned ganisms, including at the outset, humans, are biologically a number of prepared professional scientific organiza- learn certain kinds tions have of aversions more invited parapsychologists rapidly than to address them or hers), and the leading have set up journals refused symposia on to publish re- the subject. True, para- irch reports on the psychological subject, reports ideas have that are now hardly been embraced with ,wed as being among open arms, the most important but that does in their field not mean that scientists are ,ligman & Hager 1972). motivated by This concept is fear or blind now part of prejudice or ignorance or ,instream psychology. distorted interpretations Many psychologists purveyed by also refused unreliable believe in biological skeptics. constraints on intelligence (or, at st, racially determined Indeed, when gy ones); and as a parapsycholo began result of such to take shape as a gged refusal to believe, serious research number the fundamental field, a good of studies in this psychologists ea - reported by and others Sir Cyril Burt - immediately were eventually took up the challenge of investi- (posed as fraudulent gating claims (Kamin 1974). When, of spiritualistic in the late communication, telepathy, 460s, Neal Miller clairvoyance, announced that he and so on. and L. Dicara had All that was lacking to make emonstrated operant parapsychology conditioning of part of mainstream heart rate in rats psychology was evi- A iller & Dicara dence that 1967), many experimental there was psychologists a phenomenon to investigate. At the efused to believe Fourth International it, despite Miller's Congress of high reputation Psychology, as held in n experimental psychologist. Paris in 1900, Ultimately, Miller an entire him- section was devoted to psychical elf, when subsequently research and unable to replicate spiritualism, his own and the president, Ribot, an- tudies, publicly nounced the withdrew his claims founding of (e.g., Miller 1978). a psychical research institute in cience refused to Paris (L'Institut publish, on the G6n6ral Psychique) grounds that it (McGuire 1984). was rroneous, the initial Membership research of Solomon in this institute Berson and included a number of prom- tosalyn Yalow (Yalow inent psychologists subsequently won such as Janet, the Nobel Prize) Richet, James, and in the insulin-binding Tarde. In 1895 antibody, research Binet published that was funda- some case studies of nental to the development telepathy. of the radioimmunoassay However, as McGuire (1984) points out, psy- echnique (Garfield chologists 1986; Yalow 1978). were already Albert Einstein becoming very uneasy about the bsolutely refused growing link to believe that between psychical "God plays dice," research and spir- lespite the implications itualism; this of quantum mechanics; mistrust began he chose to show itself at the Fourth believe the theory Congress, and to be in error due subsequently to incomplete- many French psychologists hical research. ess. Science is full 8 ~mfil-ftm-ouUPT~B~g6b~2RI2' 0001-4 4PI'dVe0ftleRelease 2000/08/0 BEHAVIORAL AND BRAIN SCIENCES (1987) 10:4 563 For R lease 2000/08/08 : CIA-RDP96-00789ROO2200120001-4 Alcock: RaPaprosycNology cn% que Psychologists Pieron, PsychologicalAbstracts, Janet, and Dumas conducted which is published by a the Amer- number of seances in which ican Psychological Association they reexamined mediums (McConnell 1983). who had produced positive What more should parapsychologists outcomes in earlier studies expect, given the at the Institut M6tapsychique. track record they- have One medium was caught produced? I am of the strong flagrantly cheating, and opinion that rejection these psychologists concluded of, or dissatisfaction with, paranor- that no psychical phenomena mal claims is not based had been observed under on narrow, dogmatic prejudice the careftilly controlled but on the fact that after conditions. LeBon offered a century of research, a large there is stili reward to anyone who could nothing substantive to produce the mediumistic showl effects in his laboratory, but once informed of the strin- gent controls, no one ever underwent the test (McGuire 1984). 7. Is rapprochement between The American Society for psychology and Psychical Research was parapsychology possible? set up in 1885 to examine apparent psychical phenomena (Moore 1977). Its officers In 1982 psychologists Zusne included prominent psychol- and Jones's Anontalistic ogists such as Prince, Psychology was published. Hall, Jastrow (later This book constituted to become an a outspoken critic), and milestone in the course James. When they failed of interaction between to find any psychol- evidential basis for mediumistic ogy and parapsychology claims, most members by virtue of its attempts to lost interest; the group establish a framework for was disbanded, and its the psychological study remnants of the merged with the British phenomena taken by parapsychologists Society for Psychical to be paranormal. Research (SPR). James continued Blackmore (1983a), coming to support and believe from the parapsychological in psychi- cal research, and later side, and just as she was became president of the renouncing her belief SPR. in the psi In the 1930s, parapsychology hypothesis, also called had another opportunity for the study - within psychology to persuade mainstream - of the experiences that science about the importance appear to people to be of paranor- psi research. A poll conducted mal. Palmer (1986a) calls in 1938 found that 89% for a collective focus of by skeptics psychologists felt the and parapsychologists on study of ESP was a legitimate finding explanations for anoma- scientific enterprise lous experiences and phenomena, and 79% felt such research whether the explana- was a proper subject for psychologists tions prove to be mundane (Moore 1977). In that or not. These actions may same year, a round-table parapsychologyreflect what Truzzi (1985) discussion of views as a movement toward was sponsored by the Americanogical Associa-rapprochement between psychology Psychol and parapsy- tion. Parapsychologists chology. did not succeed in their attempts to gain the psychologists' Unfortunately, I doubt support for the study that such a rapprochement of psi. will The 1970s provided another ever occur, for I believe period when mainstream that those in parapsychology who science seemed ready to move closer to the skeptical give parapsychology a side will fail to draw chance. the rest As mentioned at the beginning of parapsychology along of this article, the with them. That is not Para- to say that psychology Association there will not be cooperation had gained affiliation between psychologists with the and American Association for parapsychologists in the the Advancement of Science study ofanomalistic experiences, in 1969. In 1974 one of the something which should world's leading scientific be strongly encouraged; jour- nor is it nals, Nature, published to deny whatever movement an article by paraphysicists there has been toward Targ and Puthoff in which they better mutual understanding detailed their claims and respect. about scientific evidence for However, finding explanations the paranormal, based for ostensible anoma- largely on research with Uri Geller lies is not what parapsychology (Targ & Puthoff 1974); is really about for most true, the journal did precede the parapsychologists. If it article with an editorial were, much more effort dis- would be claimer, but the research made to try to find psychological nonetheless appeared. and neuropsychological Al- though some parapsychologists explanations for such experiences were irked by the edi- before even con- torial "inoculation" Nature templating the radical provided for its readers, psi hypothesis. (Indeed, such one must a disclaimer proved to wonder why parapsychologists have been prudent, because, seem not to concern as discussed earlier, Uri themselves with the actual Geller was subsequently experience, or with how exposed such as a fi-aud (e.g., Randi experiences are generated, 1975). or with how the supposed Although mainstream psychological phenomena work (Scott 1985]. journals continue Why, for example, do to be reluctant to publish they not set out to try parapsychological research, to produce in subjects the subjec- that is not to say that tive impression of telepathy, these journals are totally instead of merely conclud- closed to parapsychologists; occasionally ing that subjects in a articles do appear (e.g., guessing task must have experienced Layton & Turnbuil 1975). telepathy on some of the Atnerican Psychologist trials? Studying guess re- rates is not cently published an article the study of the telepathic (Child 1985) that presented, experience.) along with his criticisms If parapsychology is not of skeptics' interpretations primarily motivated to of explore parapsychological research, anomalies in an open-minded the results of a meta-analysis fashion, what is its moti- of the classic Maimonides vation? Why does parapsychology dream studies. Child persist after a century con- eluded that something of failing to produce compelling important is going on, evidence of psi? Why although, in my view, his analysis does the psi hypothesis is unlikely to impress survive? To be fair, of many psychol- course, ogists. Parapsychology normal science does not was discussed in an open-minded reject working hypotheses just fashion, albeit very briefly, because they fail to be in a recent issue of confirmed empirically the Annual - although Review ofPsychology (Tyler they rarely, if ever, showy. For 1981). Since 1950, more such longevit example, than 1,500 pai%ppmW*tg**IRVWabevkdomWWZLMA4RDPgSaOCYAMOOWAbt2MG6i&ion blood depended of the Apprgqd Forfelegse ~ 00/013/08 th.~'existejice of capi aries, an stic capiRaries 1,ot be, observed with the naked eye; but the fAure er"t! them did not lead to rejection of the theory. .igators continued to seek them until, with the aid of ,,-opes hey were at last discovered (Gregory 1'et, ihe're is a difference between, on the one ,jot giving up a preferred hypothesis when that seems to promise more explanatory power ,,istilig theories about a range of observations and, t. other hand, the discounting of failure to find J,~d statistical deviations in a psi experiment. In the ,base, one is trying to establish the existence of a ,)[liellon that is not required by the existing body of I jilc data, nor is it predicted by theory, nor would it ljl~, or clear up current anomalies in physics or kology or biology. ,j. dispute about psi reflects the clash of two funda different views of reality. The first of these is the ,riajistic, monistic view that the human mind is some oI'cmergent manifestation of brain processes, where ,t. ccond is the dualistic position that maintains that li,iman mind/personality is something beyond the (datoms and molecules. Parapsychology grew out of sccolld of these; it developed directly from attempts, I it) Europe and the United States, to put the post W111 survival of the human personality on a sound ,,t;1',c footing (Cerullo 1982; Mauskopf & McVaugh 0. Nloore 1977). It is the search for the soul - not the il,LS it is described by various religions, and perhaps even the secularized soul sought by the psychical (IM'diers of the late nineteenth century during the daN, of spiritualism (Cerullo 1982), but a soul all the tc. Because, if the mind can operate separately from phN-sical brain, as the psi hypothesis would suggest, ,it it possesses much of what has been ascribed to the & ,%lost religions teach that the Soul survives death in it(- form. The question of survival of the parapsycho- sts soul" or "mind" or "personality" after death is, t,ii inany leading parapsychologists agree, an important 4-stion tor parapsychology to consider (e.g., Krippner S3~ Palmer 1983; Roll 1982). Blackmore (1983b) sus- cts that just as it was thefundamental question to many the early psychical researchers, it is still so for many of r f'ellow researchers today. it is important in any debate about parapsycholo- to inake clearjust what is being debated. Is the debate ,mit whether or not there exist "natural" phenomena lat science has so far failed to recognize, or is the debate ,mit whether or not dualism, as opposed to materialistic ,(Mism, is the correct view of nature and of mankind's lace in nature? Or, is the first question very often the ,rface issue, while the hidden agenda is the question of lialism? Conclusion Either parapsychology is a harvest offalse illusion, or the ,eat and fibre of biology, the focus of psychology, and ven the material conception of physics on which all dence stands" (Walker 1984, p. 9). These words by a laraphysicist should remind us that the existence of psi is : CIA-RDP96-QQ789ROO2209120001-4 no trivial matter. X et, to accept the reality ofpsi, we must accept that soxne force or process exists which cannot at this time be described in terms of positive properties, but only in terms of what it is not; a force which is capable of allowing for direct communication between two brains, regardless of the distance between them; and which allows the mind directly and often unconsciously to influ- ence matter in such a way as to gain some desired goal, again without any effect of distance, physical barriers, or even time. To accept the reality of psi, we must discount a hundred years of failure to find substantive evidence; there is not a single demonstration that is repeatable in Beloffs "strong sense." We must also accept that there are fundamental problems with well-tested physical and neurophysiological theories, We must accept all this in the face of the inability of parapsychologists to sort out whether, in a given experiment, a statistical deviation is due to PK or to ESP, whether it is due to the subject or to the experimenter, and whether the source of psi is acting in the past, the present, or the future. Furthermore, we must overlook the fact that even the best research pro- grams in parapsychology are seriously beset by meth- odological weaknesses. We must ignore history as well, for as Hyman (1981) points out, each generation of para- psychologists has put forth its current candidates as pro- viders of proof of psi - experiments that supposedly should have convinced any rational person were he to examine the evidence fairly. Yet, these candidates keep changing, and if prior history is a reliable guide, today's most promising research programs in parapsychology may well be pass6 in a generation or two. If parapsychologists really are dedicated to the study of anomalous experience, then it should make more sense to follow Blackmore's (1983a) lead and focus on the anoma- lies while putting the concept of psi aside until, if ever, it is needed. This is unlikely to happen, however. Psi has been postulated not because normal psychology is incapa- ble of accounting for people's apparently psychic experi- ences, nor because of inexplicable findings in physics or chemistry; nor is it the logical outgrowth of some compel- ling scientific theory. Rather, the search for psi is now, as it has been since the formal beginning of empirical para- psychology over a century ago, the quest to establish the reality of a nonmaterial aspect of human existence - some form of secularized soul. All that is needed to turn the attitude of the scientific establishment from doubt to serious interest with regard to psi is to produce some clear, substantive evidence of a psychic phenomenon. Without it, parapsychology can never become a science. ACKNOWLEDGMENTS I sincerely wish to thank both Professor Graham Reed and the BBS reviewers for their careful reading of the first version of this manuscript and for their thoughtfid comments and suggestions, most of which have been heeded in this final version of the paper. Approved For Release 2000/08/08 : CIA-F~Mi§.~ft7'ggftM_"ItM~*oqZ40:4 565