The Anomalist

Site Archive: Commentaries

Extraordinary Claim? Move the Goal Posts!

A Commentary by Patrick Huyghe
Reprinted from The Anomalist 3

If you've heard it once, no doubt you've heard it a million times.  "Extraordinary claims require extraordinary proof." That old saw has become  the skeptic's number one attack against claims that threaten to overturn their cherished applecarts. And it's a good one, for one simple reason: they're right.

But behind this squabbling over the would-be extraordinary, some  rather startling back-room maneuvering may take place. As the arguments fly over what exactly constitutes the necessary proof, there's often some hasty rewriting of the rules of the game. For the would-be extraordinary, for the unorthodox claim  on the verge of scientific success, the ground rules are gratefully changed. This  practice, often referred to as "Moving the Goal Posts," is an extraordinary  phenomenon in itself and deserves recognition.

The phrase evokes an image of the goal posts in, say the endzone  of a football game, that are slowly moved to the back of the endzone, or beyond,  as one team threatens to score. The other team resorts to cheating by changing the rules of the game in an all out effort to prevent a loss.

Well, it's not likely to happen in football, but here's the way  it happens in science. I'll illustrate this "moving the goal posts" phenomenon  with two examples, one from the field of geophysics, the other from linguistics,  but the same phenomenon can be found to occur in a host of less orthodox disciplines such as parapsychology, for example. I'll begin with the geophysics example because  I am intimately familiar with the details of the controversy in question. It just  so happens that I helped the scientist involved write a book on the subject.

The book, called The Big Splash (Birch Lane Press, 1990;  Avon, 1991), involves Louis A. Frank. Frank is a physicist at the University of Iowa  and a highly respected member of the space science community. In 1986 he found evidence  in satellite images that the Earth was being bombarded by about twenty house-sized  comets per minute. These ice comets are so small, he said, that they break up and  turn to water in the upper atmosphere. And over the age of the Earth, Frank reasoned,  these incoming small comets would be responsible for all the water in our oceans  and then some.

The astronomers' response to Frank's discovery was not unexpected.  "If these things exist," they said, "we would have seen them."  Of course, astronomers really had never considered that comets could be so small,  as they normally measure comets in kilometers. Nor had they ever conducted a search of near-Earth objects that might have revealed the existence of such small, dark incoming objects. But nevermind, astronomers had no interest in searching for these  objects because they knew the outcome in advance.

One physicist, however, decided to prove Frank wrong the old fashioned  way--by conducting a telescopic search. The physicist's name was Clayne Yeates. In the late 1980s he worked as the project manager of the Galileo mission for the Jet  Propulsion Laboratory (JPL) in California. Yeates, who has since passed away, obtained  funding from JPL and rented the Spacewatch Telescope at Kitt Peak run by the University  of Arizona. A search conducted in January of 1988 produced some stunning results--actual  images of the small comets.

When the images were presented to scientists at a meeting of the  American Geophysical Union a few months later, however, many were unconvinced. They  thought the so-called small comet streaks in the images were merely noise--fluctuations  in the data due to chance. The standard of proof in astronomy is to have two images  of the same object. When Yeates wrote up a paper announcing the results of his search,  the editor of Geophysical Research Letters informed him that "for your  paper to be accepted for publication, the referees must be convinced that you have  seen the same object in two consecutive exposures."

As it turned out, Yeates had already conducted such a search and  had obtained just that--two consecutive images of the same object. In fact, he had  six such pairs of images. Yeates then provided the editor of Geophysical Research Letters with a pair of successive exposures that showed the same object. But when the referees of Yeates' paper saw the double images, they must have been taken  aback, for they decided to change the rules of astronomy just for him.

Despite having meet the editor's requirements of proof, Yeates'  paper was rejected. One of the referees said that three consecutive images of the  same object were need for him to believe the streaks were not noise. Yeates was angry  and rightfully so. It seems as if suddenly astronomers had decided to change the  standard rules of confirmation. Rather than having two images of the same object,  astronomers now randomly decided that three were necessary. But if Yeates  had then produced three, surely astronomers would have asked for four. And if he  had had four, they would have wanted five.

This was my first encounter with a blatant example of "moving  the goal posts." I've witnessed many other examples since then, most recently  in a bitter controversy taking place in linguistics--Can chimps really learn to use language?

A decade and a half ago, the claims of animal language researchers  were discredited as exaggerated self-delusions. The critics insisted that such claims were merely exercises in wishful thinking. You can train animals to do all kinds  of amazing things, they said, like teaching bears to ride motorcycles. They said  that the chimps had learned nothing more sophisticated than how to press the right  buttons or make the right utterances in order to get humans to cough up those much-loved  bananas and M&Ms. There is no evidence, the critics concluded before slamming  the door shut on the subject in the early '80s, that the chimp utterances even remotely  resembled the linguistic abilities of a young child.

But recent research by Sue Savage-Rumbaugh and other scientists  at the Language Research Center at Georgia State University in Atlanta appears to refute that view. Her pigmy chimps, which some scientists believe are more intelligent  than the common chimpanzees studied in the earlier "flawed" language experiments,  appear to have learned to understand complex sentences and seem to use symbolic language  to communicate spontaneously. Her chimpanzees demonstrate the rudimentary comprehension  skills of two-and-a-half-year-old children.

The critics will have none of this, of course. And all the  claimants can do is shake their heads in frustration. Stuart Shanker, a philosopher  at York University in Toronto and a co-author with Savage-Rumbaugh on a new book, insists that linguists are applying a double standard to this new work. The critics  are dismissing skills like putting together and noun and a verb to form a two-word  sentence which they would consider nascent linguistic ability if seen in a young  child. "The linguists kept upping their demands and Sue kept meeting the demands,"  Shanker told George Johnson of the New York Times in a story that appeared  on June 6, 1995. "But the linguists keep moving the goal posts."

Ah, yes. Shanker is obviously quite familiar, not to mention frustrated,  by this "moving of the goal posts" business. Extraordinary proof often  seems to mean a change of the basic rules of the game, a change in the standards  of proof. While claimants consider this unfair, and I can easily see why they would  think so, such an action might be acceptable if at least the rules were changed in advance. But unfortunately, it often seems as if the rules are changed as the game is being played. All of which gives a truly extraordinary meaning to  the phrase "extraordinary proof."

Copyright 1995 by The Anomalist