In my last column I promised to talk about junk science. I define this as writings or theories that use the vocabulary of science and often sound like science, but are actually some greater or lesser form of gibberish.

This article will be difficult to write, not because there is so little junk to choose from, but because there is so much. Sometimes I feel that junk science is like junk mail, which in my mailbox so swamps the real thing that I worry about accidentally throwing away the latter along with the former. Robert Park wrote a whole book, "Voodoo Science" (his name for what I call "junk science") without by any means exhausting the subject. I am doubtful that I can cover all the ground in a single column.

Let's begin by putting the matter in context. Much important science remains to be discovered, and no genuine scientist will ever pretend that we know all there is to be known. Since this is the case, how do we distinguish between new science that may prove one day to be a good description of the way that the universe operates, and non-science that makes the same claim?

We have to careful, because there is much truth in the dictum that all great scientific advances begin as heresy. However, there are certain markers that I have learned to associate with junk science.

If we are dealing with observational evidence, be wary of the experiments by one person that others are unable to reproduce. One good recent example is cold fusion. In 1989, Pons and Fleischmann at the University of Utah reported on an experiment that showed the presence of excess energy. That energy, they asserted, must derive from nuclear fusion. Unfortunately, no one else could duplicate their results, nor could any physicist suggest a plausible mechanism for the process.

We have to be rather careful here because occasionally someone reports a finding that no one else can repeat, but the discoverer is ultimately proved right. A famous example in astronomy is provided by Ariel and Umbriel, two of the moons of Uranus. William Lassell insisted that he had seen them, in 1851 and 1852, but for more than twenty years no one else did. Lassell was simply a superb observer (he also found Triton, the large moon of Neptune, only ten days after the planet itself was discovered). However, if only one person can perform an experiment or make an observation, we should be at least suspicious.

We should be suspicious, too, of events which never happen where and when logic suggests they should. Aliens never choose to land on the White House lawn in broad daylight. Aliens, presumably for good reasons of their own, do not abduct scientists. Politicians are not taken for rides in alien spaceships, nor are women leaders in our society ever impregnated by aliens. Yet there remains an abiding belief by many people that aliens have landed on Earth, only to have their existence concealed by the U. S. Government.

Be wary, too, of phenomena that refuse to appear when skeptics are present. This is particularly true for claims of paranormal powers. Experiments in telepathy, psychokinesis, spoon-bending, clairvoyance, teleportation, and levitation have a very high (I am tempted to say total) failure rate when experienced professional magicians are in the audience.

This does not mean that the people claiming to find evidence for these powers are necessarily frauds, trying to trick the public. J.B. Rhine for many years conducted experiments on telepathy at Duke University. He sincerely believed that he had discovered certain individuals whose hit rate at describing the symbols on cards they had not seen was far beyond anything one could ascribe to chance. Now it seems that Rhine was fooled by his own desire to believe, by poor experimental design, and by simple statistics.

Suppose, for example, that we have a large pool of people (say, ten thousand) available to take part in our experiments; and suppose that one person in ten scores significantly higher than chance would predict. One person in ten will also score lower than chance, but we are unlikely to find them of interest. We take the thousand who did better than chance (the "talented" ones) and test them again. One in ten of them will again, for purely random reasons, score better than chance predicts. Take those amazing hundred, test them again. Repeat the process two more times. Finally we will be left with one individual who has never failed, whose ability to predict what is on the cards seems beyond all explanation other than telepathy or clairvoyance. It is, however, explained perfectly well by the laws of probability. (I know a way to use these same laws of probability to make a lot of money, and I am not at all sure that it would be illegal. Maybe I'll describe it in a future column.)

I believe that Dr. Rhine to the end of his life was convinced that he had found undeniable evidence for the reality of telepathy and clairvoyance. More than that, surveys show that to this day far more people believe in telepathy than are skeptical of it. The popularity of psychics, faith healing, the interpretation of dreams, dowsing, palmistry, astrology, and reincarnation remains as great as ever. One of the oddities of the latter is that celebrities who believe in reincarnation were always someone famous in previous lives. Apparently celebrity passes down the generations, as a kind of psychic DNA.

The popularity of astrology in many ways represents junk science's most baffling success. Horoscopes appear every day in newspapers that rarely offer an article about real science. However, the astrology used to cast horoscopes is based on the solar system as it was known before 1781, when William Herschel discovered Uranus. Why are astrologers ignoring that planet, and Neptune, and Pluto, and all the members of the asteroid Belt, and all the planetoids that are now known to exist beyond Neptune and Pluto?

The answer, I suspect, is that astrologers are doing just fine without all those extra celestial bodies. Adding Uranus to the mix would not increase or decrease the number of believers by one individual.

Examining what I have written, I am forced to conclude that junk science is doing just fine. Many of you may believe in one or more of the subjects for which I have recommended skepticism, since junk science is probably better accepted by the public than genuine science. The latter is often regarded as the potential source of all kinds of unpleasantness, everything from nuclear war to genetic meddling.

However, I am not quite ready to leave junk science. Almost all of what I have said so far has been based on observation and experiment rather than theory. Next week I propose to consider "deep" junk science, where unconventional notions are used to define a whole new universe.

"Borderlands of Science" is syndicated by: