In legalese, the words that distinguish good science from bad are "mainstream" or "generally accepted." This is about as far as the law can go. This standard places a serious responsibility on the scientific community to ensure that mainstream science is indeed good science.
Institutions are scrambling to develop means to monitor and deal with departures from proper scientific practices. Science reported1 that in Europe "a rising tide of retracted papers and some high-profile fraud cases are finally stirring research groups into action ... the very structure of research institutions might be creating an environment that encourages research misconduct." The American Association for the Advancement of Science (AAAS) has published a quarterly Professional Ethics Report since 1988, with scores of reports in each issue on events relevant to scientific ethics. AAAS has also announced a Demonstration Project on Providing Independent Experts to the Federal Courts and is cosponsoring, with the U.S. Office of Research Integrity (ORI), a conference in Washington, D.C., April 10-11, 2000, on "The Role and Activities of Scientific Societies in Promoting Research Integrity." ORI has decided to "downgrade the role of ORI in policing scientific misconduct" and issued a "government-wide research misconduct policy" statement2 that defines misconduct, limiting it to fabrication, falsification, and plagiarism as described therein, and will "assist institutions unwilling or unable to do their own investigations [and] continue to review the results of university investigations and propose sanctions."
Illustration: A. Canamucio
Scientists Will Be Held Accountable
Baylor College of Medicine found its molecular biologist Kimon Angelides guilty of fabricating data, stripped him of tenure, and evicted him from his lab.3 He sued the university and 14 of its officers and scientists, but he withdrew the action after the federal appeals board backed Baylor's findings that Angelides had "committed scientific misconduct." The appeals board concluded that the evidence showed "not honest error, not preliminary results that later proved overly optimistic, not even carelessness, but rather intentional and conscious fraud."
Lawrence Berkeley National Laboratory (LBNL) took similar action against its staff biochemist Robert P. Liburdy, finding that Liburdy "deliberately reported selected data, omitted contradictory data, and falsified results."4 As a result, Liburdy was forced to retire, withdraw his findings, and be barred from all federal research for three years. Congress is pressing the laboratory to return at least some of the money that funded Liburdy's work, which may dampen the enthusiasm for any laboratory to find scientific misconduct in its midst.
Strikingly similar kinds of scientific misconduct seem to be happening in the radiation area. There are research reports widely cited as supporting current policy that, on closer scrutiny, can be seen to have conclusions that are contradicted by their own data. This is because the data have been improperly manipulated in the same way as the cases of scientific misconduct cited above, to bring about the desired but unwarranted conclusion. Such cases, if allowed to stand, tarnish the whole scientific enterprise. They can, and they will, be challenged.
An important fact to face here is that there are cases where mainstream scientific opinion is bad science--where scientific positions are maintained long after their invalidity becomes clear to any who will unblinkingly examine the refutory evidence. When reputations and incomes have been made and maintained by the "established wisdom" and the interests of funding agencies, there are strong personal and institutional incentives to hold off any changes as long as possible, pleading that we still do not know enough for sure, and a little more research money may make the situation clearer.
This situation can damage the integrity and credibility of the entire scientific enterprise if the mainstream opinion is later shown to be invalid and was under serious challenge by other credible scientists at the time in question. If opposing views are openly and honestly debated, there is no problem. The problem arises when the challengers are numerous and credible, and the challenges are not adequately considered and responded to by the mainstream science establishment.
Radiation Protection Standards
Current policy presumes that any dose of ionizing radiation, no matter how small, can lead to cancer and must be considered hazardous. This policy is contradicted by a large body of highly credible scientific evidence, but it provides support and professional benefits for the mainstream of the radiation protection field. But its continuance creates unwarranted public fear and hundreds of billions of dollars spent for no discernible public benefit. People are being scared away from life-saving medical treatments that involve radiation. People are dying from foodborne pathogens that could be killed by radiation. Scientists are forced to use less effective analytical techniques to avoid burdensome regulations associated with radioactivity. And fertile topsoil less radioactive than good farmland in Colorado or Norway is being dug up and hauled away for treatment because it is "radioactive."
Credible challengers to the policy have become so numerous and so vocal for so long that mainstreamers now describe the situation as controversial. But this has not led to the needed examination of the refutory data and open debate as to its significance. Instead there have been three-year studies by mainstream committees such as the International Commission on Radiological Protection, the U.S. National Council on Radiation Protection and Measurements, and the committees on Biological Effects of Radiation. These committees are staffed by prominent scientists from prestigious institutions, but they self-select their membership and their successors, and they carefully exclude all of the many respected scientists whose work contradicts existing policy and provides replicable, statistically significant evidence that low-dose radiation is not harmful and is often beneficial. This evidence is not adequately dealt with in the reports that these committees produce, which create and support policy.
The normal procedures for challenging bad science are thwarted in this situation. Relevant research is reported in peer-reviewed journals and is formally presented to the committees and directly to the appropriate government regulatory agencies. But the evidence they present is ignored or summarily dismissed. The associated professional societies, such as the American Nuclear Society and the Health Physics Society, occasionally issue mild position statements and then back away from any follow-up action. Conferences of scientific and policy experts are called, but the meetings are not open and challengers are not invited. Sometimes a representative of an antinuclear organization such as Greenpeace is invited, but this is a political move, and their presence adds nothing to resolving the science issues.
Proposed radiation regulations are subject to public input. But we find that although other aspects of a proposed regulation may be discussed, the question of radiation health effects is considered defined by the review bodies and is specifically excluded from public challenge.
So the problem is routing out bad science when the bad science underlies a mainstream scientific position. Here hope lies in two facts. First, rules of good science are widely agreed upon, and the bad radiation policy persists, not because it is well defended as good science but only because it has never been forced out into the open. Second, although the policy is defended by the mainstream of the radiation protection community (which has much to gain personally and professionally from it), the wider scientific community is excluded. It is reasonable to expect that the mainstreamers would object, if given a chance. Although the funding of contradictory work is constrained, there are respected biologists, medical researchers, statisticians, epidemiologists, and other scientists outside the radiation protection community who have no conflict of interest in this matter, who could be expected to judge the issue fairly. They must be brought into the arena. If this problem is left for too long without a constructive response from the science community, the public reaction will be to question the integrity of all of science.
Several recent actions are accelerating this process. U.S. Senator Pete V. Domenici (R-N.M.) has asked the General Accounting Office to investigate and report back by next June. Physicist-lawyer Richard Meserve, who played a leading constructive role in many of the recent actions to bring good science into policymaking, has been named chairman of the Nuclear Regulatory Commission. From the other side, Energy Secretary William Richardson announced that radiation is causing cancer and premature deaths to workers in nuclear plants, implying that scientists have known this for decades but have covered it up.5 The Energy Department then stated,6 "We estimate that over the next 30 years, there will be between 250 and 700 radiation-induced cancers, of which about 60 percent will result in death." This contradicts the position of virtually all responsible scientists in this area for over 50 years. The time has come for the scientific community to speak up on this issue.
1. M. Hagmann, "Scientific misconduct: Europe stresses prevention rather than cure," Science, 286:2258-9, Dec. 17, 1999.
3. J. Kaiser, "Scientific misconduct: Baylor saga comes to an end," Science, 283:1091, 1999.
4. "Report of the Formal Investigating Committee Inquiring into Alleged Scientific Misconduct of Dr. Robert P. Liburdy," July 7, 1995.
5. M.L. Wald, "U.S. acknowledges radiation killed weapons workers," New York Times, Jan. 29, 2000.
6. DOE News Release, July 15, 1999.