Edward Ginzton, one of the founders of Varian Associates, once remarked concerning his colleague Russell Varian that "he had many modes of thought, of which logic was only a special case." So it is, most likely, with all great inventors and so it is, I believe, with all truly productive scientists. In this essay, I will argue that scientists need at least three modes of thought that I call "curious," "creative" and "critical."

These requirements, though they may be quite general in their applicability, come sharply into focus when one deals with anomalies within mainstream science or with anomalous phenomena that seem to reside outside of science as we know it.

Let us take just one example from within mainstream science. It has been claimed for some years by Halton Arp of The Max-Planck Institute for Astrophysics in Garching, Germany and by William Tifft of the Steward Observatory in Tucson, Arizona, that there is evidence indicating that our interpretation of the redshifts of astronomical objects is incomplete. Their results, if taken at face value, contradict the usual assumption that the redshift of distant objects (such as distant galaxies and quasars) is due almost entirely to the expansion of the universe. Arp and Tifft have been curious in examining strange patterns that arose in their early observations; they have also been creative in trying to seek an interpretation of their results; and they have been critical of their own work by seeking new observations and encouraging others to make their independent observations.

Then what is the problem? It is that the astronomical community has, by and large, applied only critical thinking to the same problem. There have been some attempts to reproduce Tifft's results, with mixed success, but the general attitude has been "It cannot happen, therefore it does not happen," just the opposite of good advice once given by the great physicist Robert Leighton of the California Institute of Technology, "If it does happen, it can happen." As a result of this attitude, Arp and Tifft have come to be regarded somewhat as heretics. Indeed, Arp lost his observing privileges at the Mount Wilson and Palomar Observatories, forcing him to leave the United States to go to his present home in Germany. An even more disturbing and challenging situation arises if a scientist takes an interest in a topic that is outside of mainstream science and is believed by the scientific community to represent "pseudoscience," the "paranormal," or "pathological science."

Some of the best known examples that are regarded in this light are "parapsychology," "ufology" (the study of UFO reports) and "cryptozoology" (the search for zoological anomalies, including "Big Foot" and the so-called "Loch Ness Monster" ). Even the mention of such terms will send a shudder through the frame of almost any self-respecting scientist. Why is this so?

Typical responses to this question are in fact indicated by the terms I have just used. If I assert that a subject is "pseudoscience," I am stating that the activity is not truly scientific but merely pretends to be scientific. However, such an assertion is indefensible. A "subject" is neither scientific nor nonscientific. It is only the study carried out by a particular person or group of persons that can be so described. One may be able to make a legitimate case that this person who studies parapsychology is being pseudo-scientific, but that does not mean that it is impossible for some other person to carry out a study in the same field that meets the highest standards of the scientific enterprise.

There is a similar problem connected with the term "paranormal." If I assert that a subject is "paranormal," I am implying that I know what is "normal." I am further implying that any subject that is not "normal," according to my definition of the term, does not accord with scientific knowledge and must be rejected as bogus. This would be a huge responsibility to take on. If pressed on this issue, most scientists would agree that science is incomplete. They would agree with Isaac Newton who stated that he felt like a boy "finding a smoother pebble or a prettier shell than ordinary, whilst the great ocean of truth lay all undiscovered before me." If we do not know all there is to know about the universe (including human beings and everything else in the universe), then clearly we cannot claim to know what is "normal," and it is therefore foolish to use the term "paranormal."

On the other hand, the term "pathological science" is somewhat more useful. It refers to poor, slipshod and misleading research that yields results that turn out to be false. However, the term was originally coined by the Nobel Laureate Irving Langmuir to imply that certain fields are made up only of bad science. If this were the case (an assertion that is in itself debatable), it would prove only that better work needs to be done, not that such fields should be placed off-limits to scientific research.

In thinking about such questions over the years, I have come to the conclusion that the problem with such topics is not a purely intellectual difficulty in trying to understand the nature of the phenomenon or to assess the quality and conclusions of the research. I have concluded that the key to the puzzle is to be found in non-scientific and non-intellectual considerations. Although the prototype of a scientist is that of a lone researcher following the truth according to his own light, with little heed to the world around him, such is not the scientist that we know today, and perhaps the image was never more than a myth. Science today is a collective enterprise. Much of the work is done in groups. Even an individual scientist is dependent upon the good will and support of his fellow scientists for the wherewithall to continue his work.