that we do not run red lights, nor is it the primary incentive that keeps us honest in our scientific research.

But how much has honesty in science got to do with such mundane matters as traffic lights? Are we comparing apples and oranges? Since the hospital president used the traffic light analogy in a discussion of integrity in science, he probably takes the view that the principles guiding a scientist in research are not significantly different from those affecting behavior in other facets of life. That is the position taken in this booklet, but it is not a universally-held view. For example, some would argue that science requires higher standards of ethical behavior than can be expected in the world at large. Others prefer to believe that the nature of science is such that ethical questions are less important than in the rest of life: how we deal with traffic lights, or with our friends and enemies, involves moral judgments and ethical standards, but the structure of DNA and the origin of submarine canyons are not affected by the character of the scientists who study them.

The latter view misses the point. Scientific problems such as the structure of DNA or the origin of submarine canyons are investigated by scientists, who may be all-too-human in their capacity to make mistakes, to miss or misinterpret critical pieces of evidence, and, on occasion, deliberately to fake research results. Science may be morally neutral, but so is a traffic light; car drivers and scientists are not.

This does not mean that mistakes and omissions are frequent in science, still less that fraud and dishonesty are commonplace. Most of us follow the rules most of the time, in our daily lives as in our scientific activities. We make occasional scientific mistakes, and on deserted streets at four in the morning we may occasionally be tempted to run a red light. But accuracy and responsible behavior are much more common than their opposites.

There are, nevertheless, many scientists who believe that to stress the fact that scientists are fallible human beings does imply that mistakes, omissions, and unethical behavior are common in science. They feel that this is not merely bad for the image of science but is simply not true. Few of them, probably, believe that research scientists can somehow avoid the temptations and frailties that affect humanity in general, but they would argue that the scientific method has, over the centuries, come to incorporate so many checks and balances that the mistakes and misinterpretations which do occur are inevitably detected and corrected. Scientists may be fallible, but science is self-correcting.

Such contrasting attitudes are evident in the responses of different scientists to the instances of scientific fraud that have been exposed from time to time. To many people, such spectacular cases are probably the visible tip of an iceberg of unknown but substantial dimensions. How



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement