In response to:

Funny Coincidence from the May 26, 1966 issue

To the Editors:

Among the critics of parapsychology, a small number have always approached the subject as if ESP just has to mean “error some place.” The review by Martin Gardner of the book by C. E. M. Hansel in your May 26 issue seems to indicate that the reviewer belongs along with the author in this minority group. They both write as if they are rendering to mankind the great service of helping to rid science of the fallacy of ESP and that any means is worthy in the pursuit of this end. Certainly your readers could easily see that many of the criticisms in Mr. Gardner’s review go beyond proper scientific and ethical bounds. In other instances, however, the facts will be known only to an insider in parapsychology. May I give a few illustrations?

Mr. Gardner says: “I once tried to get a few easily remembered details out of Pearce by correspondence…” Easily remembered—after more than 20 years? Psychologists have long known that details regarding even the most vivid experiences are not accurately recalled over a period of time. What could Pearce have said beyond expressing confidence in the experimenter and his report, and what valid purpose could it have served if he had tried to do more?

The reviewer continues. “…but he flatly refused to discuss the incident.” I must thank Mr. Gardner for informing me about his 1951-56 correspondence with Pearce. Hubert answered one letter saying that he did not wish to write about his ESP work at Duke. When the second letter, the one referred to in the review, was received Pearce apparently did not reply. But surely there are many innocent reasons why he may not have done so, including the fact that his earlier letter had already given his answer. Under the circumstances his failure to write again was hardly a flat refusal to discuss the matter. Yet in the review the omission is treated—could anyone fail to get the implication?—as a sign of a guilty conscience.

The reviewer accepts without question Hansel’s statement that Hubert could have peeked to see the cards by hiding in a room across the corridor from the experimenter’s room in the Pearce-Pratt Series. Hansel shows a floor plan of the rooms and corridor that he admits is “not to scale.” But how is the reader to know that in his plan the positions of the illumination windows have been changed and that a true floor plan would show that the supposed peeking was impossible?

I strongly protest against the tone of the whole review, which is one of scorn and ridicule for an on-going field of research that the reviewer has not accepted. Mr. Gardner has every right, of course, to his opinions. But am I alone in feeling that the type of critical attack exemplified by his review is foreign to the spirit of scientific inquiry? The methods he (and Professor Hansel) use for guarding the present frontiers of science against ESP could keep out any new discoveries that would require revolutionary changes in our views of the nature of the universe.

J. G. Pratt

Assistant Professor of Psychiatry

University of Virginia


Martin Gardner replies:

Mr. Brier’s first point is a linguistic one. I meant to call attention to the well-known fact that the overwhelming majority of professional psychologists are not only skeptical of ESP, but have performed scores of ESP tests, regularly reported in the journals, that have failed to support the findings of parapsychology. Those psychologists who are also parapsychologists have, naturally, made tests that support parapsychology, but their results are published almost entirely in parapsychology journals. The gap between this small group, and the large group of psychologists who are not parapsychologists, not only exists, but is so large that it is the object of constant complaints by the parapsychologists themselves.

Mr. Brier’s second point is conceded. There is no reason why a person who received his Ph.D. in botany, as Rhine did, could not become a competent experimental psychologist.

And he is correct in saying that if one goes over a table of, say, random numbers, and looks deep enough for statistical anomolies, he is likely to find them. This is a point that I stress in the chapter on Rhine, in my book Fads and Fallacies, where I point out (pages 34-35) that many of the positive results reported by Rhine may actually be no more than statistical anomolies of just this sort. The question of how strong such a pattern must be in order to be taken seriously as indicating a bias in an experimental situation is a question that touches on deep aspects of inductive logic. There are no simple answers. Suppose that someone has recorded a thousand throws of a die and found that it came up an ace more often than it should. The hypothesis that the die is loaded has a probability, or what Carnap calls a “degree of confirmation,” that depends on how many times the die showed an ace. If it rolled 1,000 aces, the degree would be close to certainty that the die is not a fair one. Suppose, however, that the bias is small. Then you discover, after checking the records more carefully, that the bias is confined entirely to every tenth roll of the die, and that records are on a sheet that divides the scores into groups of ten. This anomoly would be harder to explain than the overall bias. Since there is no connection between the symmetries in the experimental situation and the number 10, you would be led toward hypotheses that concerned the score sheets rather than the die.

This is similar to the situation presented by Soal’s score sheets. Soal himself, in published replies to Hansel’s criticism, has not regarded the anomoly as merely accidental. He considers it a “segmented salience” effect of unknown origin. Hansel reports in his book on a test he made with a group of students, asking them to place dots at random on the spaces of a form similar to Soal’s score sheets. The patterns they produced showed salience effects similar to those on Soal’s records of tests with Shackleton. Of course this proves nothing. But it does suggest the hypothesis that the records were doctored by someone who did not make the doctoring sufficiently random: a suggestion which gains credence by Soal’s statement that he lost the original records, and by the charges of Mrs. Gretel Albert, who was one of three assistants in these tests. (The other two were Soal’s wife and his barber.)

I was incorrect in saying that Mrs. Albert told Hansel personally that she had seen Soal doctoring the records. She reported this to Mrs. K. M. Goldney, Soal’s collaborator and a council member of the Society for Psychical Research. The society did not publish Mrs. Albert’s charge until about eighteen years later, when it appeared in their journal (vol. 40, p. 378, 1960) along with Soal’s reply. (It was said at the time that Soal was being harrassed by two enemies, Hansel and Gretel.) Nor was I quite correct in saying that Hansel had been requested by Rhine to terminate his visit at Duke University. Hansel has since informed me that he left a week before he was supposed to because Rhine refused to continue to let him have original data sheets unless he signed a form saying that he would publish nothing about them without Rhine’s permission. This Hansel refused to do. Since there was no longer any point in staying, he left.

Professor Pratt excuses Pearce from replying to my letter of September 7, 1953, on the grounds that it would not be easy for Pearce to recall details of that historic occasion on which he, alone with Rhine, correctly named 25 ESP cards without seeing their faces. Since this was by all odds the most sensational ESP demonstration ever witnessed by Rhine, so unequivocally unexplainable by known natural laws, then surely it is a scandal that neither Rhine nor Pearce immediately recorded every detail so that the event could later be described with great accuracy. In his infuriatingly vague description. Rhine says that “each card was returned to the pack and a cut made.” I asked Pearce if this meant that, each time he guessed a card correctly, Rhine was holding that card and looking at its face. It is hard to believe that Pearce would not remember this aspect of the procedure. Pratt writes that Pearce “apparently” did not answer my letter. From later correspondence, which I photocopied and sent to Pratt, it is quite clear that Pearce did not answer that letter.

In his first letter, Pearce had said that he would be delighted to do whatever he could to assist me, but he added that his replies would have to be confidential, because the leaders of his church (Pearce is a Methodist minister) did not look with favor on his connection with ESP work. I would not have written him a second time unless I thought that he was willing to answer a few questions off the record. I am responsible for Pratt’s misunderstanding on this point, since I did not include my first exchange of letters with Pearce in the group of subsequent letters that I photocopied.

As for the floor plans of Pratt’s old office, Hansel says in his book that when at Duke he had asked for the details of the structural alterations. “These details were to be forwarded to me,” writes Hansel, “but I never received them. I wrote again requesting them, but had no reply.” If Pratt possesses the “true floor plan,” it would have been helpful if he had made it available to Hansel. Indeed, ruling out all possibilities of peeking into the room was so absolutely essential to Pratt’s test with Pearce that his failure to print the room’s floor plan, when he first published his results, remains a serious blot on that report. However, the exact floor plan is now seen to be amusingly irrelevant. Even if it had not been possible for Pearce to stand across the corridor, in another office, and peek through a transom into Pratt’s room, there was nothing in Pratt’s clumsy experimental design to prevent Pearce or a collaborator, had he wished to do so, from standing on a chair in the corridor and peeking directly through Pratt’s own transom.

This Issue

July 28, 1966