A friend of mine sent the following:
Harry, I'd be interested in a skeptical take on the following: I stand face toface with someone, an arm's length apart. He puts his arm straight out from the shoulder, palm up, resting his wrist on my shoulder (his right wrist on my left shoulder). I tell him I am going to try to bend his straightened arm, and he is to resist as much as he can. I cup my hands from both sides of his arm over the inside of his elbow and pull down hard, trying to get his elbow to bend downward. Depending on the person's size and strength, I have to use more or less force, but I can always get the elbow bent at least 60 degrees, usually more.

We then reset in exactly the same position with the same task as before. This time, however, he is instructed to visualize 'energy' flowing up from the earth, through the soles of his feet, up his trunk, into his arm and out through his fingertips and beyond, like a five-nozzled fire hose shooting out water under high pressure. He is told not to focus on resisting, but simply on maintaining the visualization. Once he indicates he is ready, I try again to bend his elbow, and I can't, no matter how much pressure I put on it.

I have done this dozens of times, with all kinds of people, with myself and/or others as subjects. Aside from whatever psychobabble and metaphors are used to describe what happens, the result is the same every time.The only "objective" difference is that the person is apparently thinking differently from one trial to the next; how does that make them (apparently) stronger?

This is a great example of the kind of incomplete thinking for which scientfic rigor was invented. Imagine that you wanted to test the hypothesis that some kind of "energy" could be used by some sort of"mental control" as described above. The null hypothesis is that there is no such effect apart from that produced bythe belief itself. To test this and eliminate experimenter and subject bias, you would want to have a double blind experiment. This would be one in which neither the subject nor the "tester" would know what exactly was happening.

If the effect was only produced when both parties to the experiment knew which experimental condition was in place, you would know that the effect was the result, not of the experimental condition, but of the belief/credulity of the parties in the experiment. That is, of course, exactly how the experiment described by my friend works. What you observe,therefore, is NOT some mystical and unexplained force occurring through some sort of equally ill-described mental energy, but the "placebo" effect itself. The amount ofenergy difference in the two conditions described is exactly the definition of the amount of energy produced by a belief system itself.

My friend has come upon a perfect demonstration of the placebo effect. What you believemakes a difference, not in the way the universe operates, but in how you operate. Occam's Razor states that when there are two hypotheses that explain the same data, you should be biased in favor of the simplest. It is usually stated in the form "Do not multiply complexities". In practice I think this means that a higher standard of experimental proof is required for an hypothesis that requires a radical change in the way the world is viewed. Sometimes that needs to happen, and our world-view is wrong and needs to change. But the level of proof needed is still a high level of proof. In the example given by my friend, the standard of proof that is offered is just the willingness of the participants to believe that "something" is happening. That standard of proof gets us belief in the supernatural, the phlogiston theory, humours, astrology, and the endless pantheon of silly and ill-informed beliefs that seem to constitute proof for much of the world.

Rationality may not seem to be enough, but it's way ahead of whatever is in second place. Just because a ready explanation isn't available doesn't mean we should leap to an irrational one.