Flim-Flam! Psychics, ESP, Unicorns, and Other Delusions - James Randi [161]
We must understand the status of both Davies and Evans in this matter. David Davies, editor of the science journal Nature, brought Evans with him as a referee. The year before, Crussard had submitted to Nature a report that told of 116 bending feats performed by Girard. It was admitted that "in some cases the experiment was confused and a trick may have been possible. But we never saw any tricks." An experiment involving a stainless-steel bar was performed in a closed—but not sealed—tube, said Crussard. The repertoire of feats read more like a theatrical bill than a series of scientific experiments, but Crussard's reputation as a metallurgist made Davies take him more seriously than he might have otherwise. I was reminded of his dilemma with parapsychologists Russell Targ and Harold Puthoff and their vague "scientific" paper, which Nature did publish.
Evans, in the company of Davies, had already met Girard and had seen some bending done. He had claimed no expertise and merely reported that he had witnessed something he could not explain but that he thought should be seen by a conjurer. All of us agreed that the conditions imposed by Crussard had been inadequate, to say the least. The metal samples in particular had not been well controlled, and Girard had even been invited to take bars home with him to practice with, thus making it possible to substitute an already bent bar on his return. The taped and filmed tests we had seen always showed a great number of bars strewn about the table rather than just the one in use at the time. But this, remember, was the way Girard liked to work. Again, we found that the mouse was running the experiment.
Davies and Evans, in their first contact with Girard, had made no attempt to control anything. The method was to allow things to happen comme d'habitude—without interference. This way, the defects, if any, would show up. They showed up. And our efforts that afternoon had assured us that the new conditions Girard would face were adequate to prevent any deception, but in no way would they inhibit any genuine powers from operating. Davies and I wrote out in detail the simple rules, and Evans was put in charge of the box containing the samples to be used. We were prepared.
What followed was a comic opera. Girard stroked himself into a froth trying to produce a bend. At one point a tray of glasses was upset in the kitchen with much breakage. Many eyes turned toward the noise, but three pairs never left the test. Ranky, peerless observer that he was, succumbed to good wine and food and several times snored so loudly that we had to wake him up. Sanlaville never took his eyes off the experiment—except to cross the room to refill his glass, to light up another in an endless chain of noxious cigarettes, to chat with Ranky when he was awake between naps, and to doze off serenely a few times himself. If these were the standards they had previously used to put Girard to the test, it was little wonder they were bamboozled. As for Davies, Evans, and me, our eyes were falling out with exhaustion. Girard was as fresh as when he had started, and as he ebulliently stroked away we noted several times that his enthusiasm included a move we knew was quite sufficient to put the proper kind of pressure on the sample. We would interrupt the process at that point to replace the sample with a fresh one.
That afternoon Chris Evans had inadvertently ruined one sample. It was a bar marked "999," and we discovered that it was 99.9 percent pure aluminum. Tiny amounts of other metals in an aluminum mix produce astonishingly different qualities, but the almost-pure metal is also surprising. Chris had put a very small amount of pressure on the thick bar and it had bent so easily that he was amazed. We thus learned that almost-pure aluminum is very easily bent with very little pressure, and we wondered how many times Girard had been successful