Table-Turning, Parapsychology and Fraud

Abstract

According to sceptics, all of the experiments of parapsychology may, so far, be divided into two classes. The first class consists of experiments which are, prima facie, poorly controlled and thus inconclusive. Those of the second class are properly controlled according to the standards of 'normal' psychological research, but contain insufficient safeguards against fraud on the part of both experimenters and subjects. The possibility of fraud needs to be taken seriously both because the field has a long history of deception and because, unless deception is ruled out, we are not entitled to suppose that there exist any data which cannot be accounted for by 'normal' means. If the sceptic's claims are correct, parapsychology is, so far, a science without a subject matter. Closely connected with the possibility of fraud is the sceptical charge that parapsychological experiments are not replicable by sceptical or neutral scientists who employ strict controls. The replicability question is particularly important to parapsychology for two reasons. First, it is often maintained, the possibility of replicating experiments and observations provides a line of demarcation between science and pseudo-science. Second, if parapsychology's experirnents were to prove generally replicable, the charge of fraud would become irrelevant to the acceptability of parapsychology as a science, whereas if experiments were not appropriately replicable, fraud would be an attractive way of accounting for the positive results which some experimenters have obtained. In an ingenious argument, Trevor Pinch' has attempted to turn the tables on this line of attack. The very critics who have taken the position that fraud is a scientific and parapsychology an unscientific hypothesis have also insisted that replication provides a criterion for demarcating science from nonor pseudo-science. And yet, says Pinch, if we apply this criterion to the fraud hypothesis itself, we find that it has not, so far at least, generated any more replicable experiments than have parapsychological theories. Furthermore, it, like parapsychology (in the eyes of its critics), seems to be metaphysically biased, theoretically inadequate, and incapable of falsification. It follows that all of these purported reasons for rejecting parapsychology should have served also to reject the fraud hypothesis. Pinch concludes