1959 00 00 American Psychologist - Vol 14 No 9 - Crumbaugh

Magazine Overview

This issue of THE AMERICAN PSYCHOLOGIST, dated 1958, features a significant debate on the scientific validity of Extra-Sensory Perception (ESP) and the concept of 'flying saucers'. The primary focus is on two articles: "ESP AND FLYING SAUCERS: A CHALLENGE TO PARAPSYCHOLOGISTS" by James C. Crumbaugh and a response titled "HOW DOES ONE DECIDE ABOUT ESP?" by J. B. Rhine.

ESP and the Challenge of Repeatability (Crumbaugh)

James C. Crumbaugh begins by addressing the claim that the reason for the lack of a "repeatable" ESP experiment—one yielding favorable results in a majority of repetitions by skeptical experimenters—is due to subtle, uncontrollable variables of personality and attitude in both experimenters and subjects. He notes that while this criterion of repeatability is fundamental to scientific proof, some have blandly denied its necessity.

Crumbaugh references J. B. Rhine's argument that positive experiments have hit upon the correct combination of variables, thus constituting proof, even without specifying repeatable conditions. Crumbaugh finds this tantamount to denying the necessity of repeatability for scientific proof, citing E. G. Boring's caution against sponsoring evaluative judgments based solely on published reports. He quotes R. A. Fisher emphasizing that "Very long odds are much less relevant to the establishment of the facts of nature than would be a demonstration of the reliable reproducibility of the phenomena."

Crumbaugh states that the explanation of "failures" in terms of skepticism and subject handling has caused many reputable experimentalists to lose interest in ESP. To test the hypothesis that his own results were a function of his "experimental personality," he obtained grants from the Parapsychology Foundation. In collaboration with the Duke Parapsychology Laboratory staff, he designed an experiment to fractionate experimenters and subjects based on self-confidence versus insecurity and belief versus disbelief in ESP.

His first experiment (completed in 1955) yielded marginally significant differences from chance expectation (5% level of confidence), indicating results favorable to ESP. However, a repetition of the same basic experiment in 1956, and the combined results of both experiments (totaling 1,920 ESP "runs" by 24 selected experimenters and 24 selected subjects), were negative, showing results expected by chance.

Crumbaugh argues that for experimental proof of ESP, experimenters and subjects should be equivalent to many times the number of unselected cases. He asserts that both flying saucers and ESP may exist, but few trained experimentalists will accept them as proved until they meet all criteria of controlled experimentation, including repeatability. He challenges parapsychologists to accept the responsibility of specifying conditions for consistent observations, noting that the burden of proof rests on positive claimants.

To provide the necessary repeatability for experimental proof of ESP, Crumbaugh proposes a specific procedure:

1. Parapsychologists, represented by the Duke Laboratory, should select one specific experimental design (or the Anderson-White experiment) and submit it to interested experimenters via the Journal of Parapsychology.
2. A committee appointed by a scientific organization should act as moderator and call for experimenters to examine the design, including qualifications for experimenters.
3. Time should be allowed for criticisms of the design to be returned to Duke.
4. The Duke Laboratory should synthesize an exact procedure acceptable to the majority of experimenters, based on the criticisms, and resubmit it.
5. A committee should make a final selection of experimenters willing to repeat the final form exactly. This group should include three equal, odd-numbered groups: (a) those who previously obtained positive ESP results, (b) those who previously obtained negative ESP results, and (c) those who have done no previous ESP experimentation. A minimum of five experimenters per category is suggested, with several months for completion.
6. Upon completion, each experiment's report should be submitted to the committee, who would announce all findings.

Crumbaugh outlines considerations for evaluating the over-all results, emphasizing that proof would require a majority of experimenters in each of the three categories to obtain significant positive results. He stresses that proof should not be concluded from the significance of pooled data alone, as spurious results could skew the outcome. If results are consistent across categories, it would strongly indicate the "error" lies with a specific type of experimenter. Finally, a committee augmented by neutral experimental scientists should render a final report.

Rhine's Response: Clarifying ESP and Scientific Method

J. B. Rhine acknowledges Crumbaugh's contribution as a title maker and expresses sympathy for his efforts in mastering interpersonal relations in ESP testing, but views Crumbaugh's challenge as impatient and based on confusion.

Rhine corrects two minor factual errors. First, he disputes Crumbaugh's claim that "most reputable experimentalists" have lost interest in ESP, stating that there has never been wholesale interest to lose, but that existing interest among scientific professions is growing. Second, he denies rejecting observations of experiments with skeptical observers present, noting that S. G. Soal and associates have specialized in this, but personally prefers good experimental design and independent confirmation, as supernumerary observers can be a hazard to the subject's effectiveness.

Rhine then addresses the two main issues: Crumbaugh's proposed "majority vote" method and his conception of parapsychology's current standing.

Regarding the "majority vote" method, Rhine questions its scientific validity, stating it's the first time such a standard of evidence has been proposed for a scientific question. He finds it difficult to see how a quantitative result like a majority of positive researches contributes to the finality or acceptability of the ESP hypothesis, viewing it as a purely personal requirement.

On the second point, Rhine believes Crumbaugh misunderstands what parapsychologists claim to have established. He argues that Crumbaugh's proposal is not a scientific procedure and that his conception of parapsychology's status is confused.

Rhine reiterates that the problems of ESP are complex and cannot be solved easily. He emphasizes that the results of repetitions might leave questions to be answered by further study but would clarify issues and point to next steps. He proposes the following:

1. Parapsychologists should settle on the Anderson-White experiment or any one specific design and submit it to all interested experimenters through the Journal of Parapsychology.
2. A committee appointed by a scientific organization should act as moderator and call for experimenters to examine this design, including qualifications for experimenters.
3. Time should be allowed for criticisms of the design to be returned from prospective experimenters and forwarded to Duke.
4. From the criticisms, the Duke Laboratory should synthesize an exact procedure acceptable to the majority of experimenters, which would then be resubmitted.
5. The committee should make a final selection of experimenters willing to repeat this final form exactly. This group must include three equal, odd-numbered groups: (a) those who have previously obtained positive ESP results, (b) those who have previously obtained negative ESP results, and (c) those who have done no previous ESP experimentation. Several months would be needed for completion.
6. When completed, a report of each experiment should be submitted to the committee, who would announce the findings of all experiments.

Rhine then outlines considerations for evaluating the over-all results, similar to Crumbaugh's points, emphasizing the need for a majority of experimenters in each category to obtain significant positive results. He highlights that if the majority of experimenters who previously obtained positive results now obtained positive results, while those who previously obtained negative results now obtained negative results, it would indicate the results are a function of the experimenters. In such a case, the third category (those with no prior ESP experimentation) would be crucial for determining the final answer.

Conclusion on the Debate

Both Crumbaugh and Rhine agree on the need for a structured, collaborative approach to establishing the validity of ESP. Crumbaugh's challenge is to initiate a large-scale, controlled study to overcome the problem of repeatability, while Rhine's response clarifies the scientific standards and the current state of parapsychology, emphasizing the importance of rigorous methodology and independent confirmation.

Recurring Themes and Editorial Stance

The recurring themes in this section are the scientific methodology required for parapsychology, the challenge of repeatability in ESP experiments, and the role of personality and attitude in experimental outcomes. The articles highlight the ongoing debate within the field about what constitutes sufficient proof for phenomena like ESP. The editorial stance of The American Psychologist appears to be one of facilitating open discussion and critical examination of scientific claims, providing a platform for prominent researchers like Crumbaugh and Rhine to present their views and engage in scientific discourse.