Lawrence Krauss

The Citizen-Scientist's Obligation to Stand Up for Standards

April 22, 2003
By LAWRENCE M. KRAUSS


On April 3, I appeared at a symposium for students and teachers sponsored by the Illinois Math and Science
Academy, a remarkably successful high school founded by Dr. Leon M. Lederman, a Nobel laureate in physics, to foster young people's interest in science.

The symposium, called "Science, Technology and Society: Ethical Awareness for Tomorrow's Leaders," was convened to discuss the way ethical issues might be explicitly raised for young scientists.

I was somewhat hesitant to appear on a panel on ethics because, like almost all scientists I know, I have no
formal training in this subject. Indeed, like many of my colleagues, I have been reluctant to include formal courses on ethics in the physics curriculum, and I have tended to suppose that students should learn the ethos of science "by example."

Presumably, in laboratory courses and in research projects with faculty, students can learn the values of honesty, creativity and full disclosure that are the hallmarks of good science. Also, in spite of the implicit hierarchy associated with education, students should get a sense of the anti-authoritarianism of science: that there are, or should be, no scientific authorities whose views are not subject to question.

Indeed, proving one's colleagues (and oneself) wrong is one of the great pleasures of scientific progress.

Scientific ethics have been mightily tested of late. In my own field of physics in the past several years, two
important examples of scientific fraud were uncovered in subfields as diverse as molecular electronics and nuclear physics. In each case the fraudulent results were brought to light relatively quickly, but not before they were published in articles involving numerous co-authors who should have been more skeptical.

This lack of internal critical review has prompted much hand-wringing. It has also raised an issue of ethical
responsibility: do scientists who take credit as co-authors of papers need to verify all of the results cited in those papers?

The problem is that by nature science does not deal well with fraud. Scientists assume some basic level of honesty in the scientific enterprise, and while we expect mistakes to occur, we do not anticipate deliberate obfuscation of the facts.

Moreover, scientists tend to expect that ultimately the truth will win out without explicit and immediate action on their part. Future experiments that do not reproduce earlier results will expose fraudulent experimentalists,
while theoretical nonsense will be exposed when it leads to nonsensical predictions.

Nevertheless, confronting misconceptions, deliberate or not, our own or others', is probably the single most
important factor driving progress in science, and in a broader sense society. Scientists must not allow nonsense to remain unconfronted, regardless of whose sensibilities we offend. Once we allow empirical truth to be blurred with impunity in one important area of human activity, we jeopardize the very basis of a healthy democracy.

Only when we are willing to accept the universe for what it is, without myth or fear or prejudice, can we hope to build a truly just society. So I found myself in Chicago in early April proposing a possibly unpopular thesis: scientists have a special ethical responsibility at this particular time to question our government's actions. It appears that this administration is marginalizing the recommendations of major scientific organizations on the one hand, while defending artificial research to support political goals, or, worse still, manufacturing it.

Empirical constraints that may otherwise guide sensible policy making seem to be evaporating.

When a Bell Labs scientist was shown to have based some of his results on fraudulent data, his other scientific
results, no matter how exciting, lost credence. We should be prepared to apply the same skepticism to the political arena.

Last month, the National Academy of Sciences presented the reports of an expert panel that assessed current plans for examining the effects of global warming. The scientists concluded that the research program proposed by the administration lacked the most basic elements of a strategic research plan.

In particular, the panel said it lacked "a guiding vision, executable goals, clear timetables and criteria for
measuring progress, an assessment of whether existing programs are capable of meeting these goals, explicit
prioritization and a management plan."

In short, it lacks the characteristics on which empirical science is based.

A year ago, the American Physical Society passed a resolution calling on the government to delay deployment of a missile defense system until it was demonstrated to be workable against realistic threats.

Yet the administration scrapped a longstanding international treaty, committing billions of dollars to the
deployment of a missile defense system that even under the most liberal interpretation of the data has a success rate of 40 percent.

We would not accept such innumerate policies in the private sector. What if Detroit put on the assembly line a new breed of S.U.V.'s that toppled over when executing curves at greater than 30 miles an hour 60 percent of the time, or if the makers of nuclear power reactors demonstrated that prototypes catastrophically failed 40 percent of the time?

Dr. Shirley Tilghman and Dr. David Baltimore, internationally known biologists, and the presidents
respectively of Princeton and Caltech, wrote recently in The Wall Street Journal that human reproductive cloning and therapeutic cloning to produce stem cells that might be used for research were completely different biological investigations.

Further, they said a wholesale ban on cloning designed to stop efforts to produce the former would have dire
consequences for important biological research on the latter. Yet the White House has supported a wholesale ban on cloning, driven it seems by inappropriate fears of science.

Equally worrisome is what apparently is the distortion of the results of medical studies in government Web sites, like the National Cancer Institute's. It used to state that the best studies showed "no association between abortion and breast cancer," but was altered to say that the evidence was inconclusive until a scientific review panel insisted the original language, which correctly reflects current research, be reinstated.

Or consider the Web page of the Centers for Disease Control and Prevention, which used to point to studies showing that education on condom use did not lead to earlier or increased sexual activity; now, it omits this discussion.

A democracy, like science, functions best only when all actions are open to question, and when we require the
highest levels of accountability. If there is a risk that politics is being placed above empirical truth on issues of
vital national importance, inaction by scientists may be unethical.

http://www.nytimes.com/2003/04/22/science/22ESSA.html?ex=1052025529&ei=1&en=40650a42d736f5f6
Copyright 2003 The New York Times Company