The root of the problem, it seems to me, lies in the premise—ironically shared (at least implicitly) by both the CTs and many parapsychologists—that the burden of proof falls exclusively on the claimant and that the only claimants are the parapsychologists. The parapsychologist must prove that psi exists; the CT does not need to prove that psi does not exist.
On the surface this position seems plausible, for several reasons. It is difficult if not impossible to prove a negative proposition, especially in the case of OPEs where the potential domain of their occurrence is so vast that the poor CT would have to debunk every case that comes down the pike to prove the nonexistence of something just because someone suggests it as a possibility? We don't ask them to prove that fairies do not exist, so why psi?
There is another way to deal with the burden-of-proof question, however, and that is to propose that the question "Does psi exist?" is the wrong question, or at least a wrong phrasing of the question. It is certainly odd, if not downright inappropriate, to apply the verb to exist to a hypothetical construct such as psi, and this reification is implicit in the line of argument summarized in the preceding paragraph. But I think I can attack the traditional question most effectively by proposing and defending what I consider to be a better question.
Before I do that, however, I would like to digress a moment and direct a little fire at my own field of parapsychology. It turns out that many of the formal definitions of our field are
patently invalid. For instance, in Michael Thalbourne's glossary , parapsychology is defined as the scientific study
of paranormal phenomena
(p. 51). Taken literally, this definition implies that we have concluded that a given
phenomenon is paranormal before studying it. In fact, in most cases the investigation is undertaken to
determine if the phenomenon is in fact paranormal. Thus, the definition should read, "Parapsychology is the scientific study of certain ostensible or potential
paranormal phenomena."
I doubt that many parapsychologists would defend the traditional definition if the above point were brought to their attention. The problem is that this definitional inaccuracy has gone so long unnoticed. I think this unawareness is symptomatic of a deeper and more serious problem, the ramification of which I can only touch upon in this talk. The problem is that we make no clear distinction in parapsychology between the phenomena under investigation and the principles proposed to account for them. We use the same terms (e.g., psi, ESP, PK) for both.
We may also have confused some CTs owing to this use of language. Some of them seem to feel that it follows from the conclusion that psi (i.e. paranormality) has not been established that there is no subject matter in parapsychology. But a moment's reflection can reveal that such an inference is incorrect. Surveys have shown that ostensible psychic events have been reported by over half the American population . Parapsychologists have carefully documented thousands of these spontaneous cases. Moreover, hundreds of documented reports testify that such events occur in the controlled setting of the laboratory. Even if one exercises the behaviorist option and equates the phenomena themselves with the reports of the phenomena, those reports still constitute a subject matter, a database in need of explanation.
By appreciating this point, one can see that the better alternative to the question "Does psi exist?" is "How can OPEs be best explained?" Although I cannot offer a neat deductive argument, I think it is our failure to appreciate the distinction between psi as an anomaly (or a report or an anomaly) and psi as a paranormal process that has prevented us from formulating our fundamental research question in this more constructive and scientifically typical manner.
The new question invites us to seek a real understanding of the subject matter in a way that the old question—"Does psi exist?"—simply does not. However, its most important implication for present purposes is that it places the burden of proof on anyone who proposes to explain the anomalous reports. Strictly speaking, it is improper to speak of just one explanation, as inevitably no single explanation will account for the entire database. The range of possible explanations varies from pure fabrication at one extreme to some paranormally mediated process at the other. The point is that any of these explanations must be backed up by sound empirical evidence if it is to be considered acceptable. The situation is no different that it is for any other topic in psychology, and all the problems we have in psychology about how to define good evidence, how far we can generalize research findings, and so forth, tag along the same ride.