Terence Hines - Pseudoscience and the Paranormal

Parapsychologists often complain that their critics require higher standards of proof for claims of psi phenomena than for other types of claims. They argue that critics reject as proof of the reality of psi evidence that they would accept as sufficient to prove more mundane claims. As was shown in chapter 1, this is both true and reasonable—because extraordinary claims demand extraordinary proof.

The issue of cheating and fraud also divides parapsychologists and their critics. There can be no doubt that, historically, cheating and fraud have been much more common in parapsychology than in other areas of scientific investigation. The reason is probably quite straightforward—the dearth of positive results pushes more investigators in parapsychology to cheat, compared to other research areas in which positive results are obtained much more commonly. This is not to say, of course, that cheating does not go on in other areas of scientific investigation. It clearly does. It remains, however, less common than in parapsychology.

The relative frequency of dishonesty in parapsychology, by either investigators or their subjects, is no excuse for critics to fall into what Hyman (1980–81, 1985a) has called the false dichotomy of accounting for all reports of experiments supporting the existence of psi by assuming that either the report is true or fraud was involved. There are a host of other explanations for positive reports that involve neither the reality of psi nor fraud. These explanations turn on what are often very subtle methodological flaws in experiments that can produce results that seem to support the reality of psi. As Hyman (1985a) has pointed out, it is the job of the responsible critic of parapsychology to work with parapsychologists toward the goal of eliminating such methodological errors and problems in a joint attempt to produce the best possible studies of psi. Only in this way can the critic really contribute to productive investigation of the existence of psi.

Historically, the first type of evidence used to argue for the reality of ESP and related phenomena was drawn from spiritualism. As was noted in chapter 2, such evidence was inadequate because of widespread fraud and the fact that witnesses who reported impressive spiritualistic phenomena were untrained in magic and sleight of hand.

In the 1930s a new, more laboratory-oriented approach to the study of ESP gained popularity among parapsychologists. Subjects were to identify long series of hidden objects; their “hit rate” would be compared to that expected by chance. Usually, the objects to be guessed were cards or numbers. The most famous practitioners of this type of experimental parapsychology were J. B. Rhine of Duke University in Durham, North Carolina, and S. C. Soal of England. Rhine’s experiments are well-known in this country, and his procedures for studying ESP have come to form the popular view of a typical ESP experiment. Rhine actually coined the term extrasensory perception in 1934 and popularized the use of the now-famous Zener ESP cards. These cards were named for a Duke University psychology professor, Carl Zener, who developed them when working with Rhine. Zener cards consist of a deck of twenty-five cards, five each with a different design printed on the face: square, circle, star, cross, and three wavy lines.

In the standard ESP experiment using Zener cards, the cards are arranged in random sequence, then shown—face down and one at a time—to the subject, who must identify which of the five designs is on the card. Since there are five possible designs, the rate of correct responses (the “hit rate”) expected by chance is 20 percent. Hit rates significantly above chance or significantly below chance (the latter being termed psi missing) are taken as evidence of ESP in one form or another. We will see that numerous factors other than ESP can produce hit rates above chance.

The results of Rhine’s experiments can be fairly easily summarized. For the most part, subjects scored at chance levels. That is, the hit rate when guessing which design was on the card was the 20 percent expected by chance. Occasionally, however, a few subjects were found to score significantly above chance for at least some period of time. This effect typically declined as testing of these “gifted” subjects continued.

The basic experiment can be varied in numerous ways. For example, the cards to be guessed can be shown to another person, who then attempts to send this information to the subject doing the actual guessing. Clairvoyance can be studied by having the subject guess the sequence of designs before the cards are randomized.

Many people, having heard about ESP and related phenomena in the popular media, tend to think there is a considerable body of scientific evidence for such phenomena. This is simply not so. The major problem in parapsychology is the lack of any repeatable paranormal phenomenon. Occasionally, seemingly impressive results pop up, but they have the curious property of not being repeatable in other laboratories—at least not when procedural, statistical, or other flaws in the construction of the experiment are corrected. Several examples of this will be discussed below. As in the case of N rays and polywater, lack of repeatability is a sign that the alleged phenomenon is merely an artifact, the result of some experimental flaw, and not a real effect. It is this lack of repeatability of paranormal phenomena that has convinced the great majority of the scientific community that psi phenomena are nonexistent.

The ubiquity of negative results has been troubling to parapsychologists as well as to their critics. Dommeyer (1975) has commented that

The reader inexperienced in parapsychology is likely to believe… that psi phenomena are relatively commonplace. The scientific investigator knows that this is not so…. The present reviewer, after spending the greater part of two summers in the 1960s at the Parapsychology Laboratory at Duke University, was unable to observe over those months a single identifiable instance of ESP or PK. (p. ii)

Crumbaugh (1966) made a similar point, but his experience covers a much longer period of time:

At the time [1938] of performing the experiments involved I fully expected that they would yield easily all the final answers. I did not imagine that after 28 years I would still be in as much doubt as when I had begun. I repeated a number of the then current Duke techniques, but the results of 3,024 runs [one run consists of twenty-five guesses] of the ESP cards as much work as Rhine reported in his first book-were all negative. In 1940 I utilized further methods with high school students, again with negative results. (p. 524)

Beloff (1973) made the same point:

I recently completed a seven-year programme of parapsychological research with the help of one full time research assistant. No one would have been more delighted to obtain positive results than we, but for all the success we achieved ESP might just as well not have existed…. I have not found on comparing notes with otherparapsychologists… that my experience is in any way out of the ordinary.

These, and additional testimony making the same point can be found in Alcock (1981).

The most common rationale offered by parapsychologists to explain the lack of a repeatable demonstration of ESP or other psi phenomena is to say that ESP in particular and psi phenomena in general are elusive or jealous phenomena . This means the phenomena go away when a skeptic is present or when skeptical “vibrations” are present. This argument seems nicely to explain away some of the major problems facing parapsychology until it is realized that it is nothing more than a classic nonfalsifiable hypothesis, like those discussed in chapter 1.