ONE of the more elusive discoveries scientists are chasing is an effective answer to the hoary question: how to get professionals to live up to the ethics of their profession. In other words, how to cope with scientific misconduct and research fraud. This has nagged the scientific professions for years without resolution. Now, parts of the United States government are beginning to force the issue.
The Department of Health and Human Services has proposed rules and procedures to investigate and punish ``serious deviation'' from accepted research practice. That means plagiarism and data faking as well as outright legally culpable fraud. Also, spurred by a couple of dramatic hearings last spring, there is continuing discussion in Congress of legislation that would criminalize scientific misconduct.
This is a clear signal to the United States scientific community - especially the academic element - that its own efforts to deal with the issue seem halfhearted and self-serving.
That community, generally, has taken the attitude that misconduct is rare. Daniel E. Koshland Jr., editor of the journal Science, has maintained that ``99.9999 percent'' of papers published are accurate. Yet in 11 percent of its random audits of research over a 10-year period, the Food and Drug Administration found misconduct.
The National Institute of Mental Health brought one of the more spectacular cases to light when it reported phony work by psychopharmacologist Stephen E. Breuning. He had claimed that using Ritalin or Dexedrine improved mentally retarded children. His indictment last spring for submitting fraudulent reports in applying for grants is one of the first criminal charges of research fakery.
Many research institutions - especially universities - have taken some steps to deal with scientific misconduct. These include formal investigative processes. The staffs of the two House subcommittees that held hearings, however, had no trouble finding witnesses who testified that these processes could sometimes seem a means of avoiding the issue. This led some representatives to question the propriety of having institutions, in effect, investigate themselves.
The new action by Health and Human Services would partly meet that objection by providing oversight of a grantee institution's misconduct probes. Meanwhile, various scientific bodies such as the National Academy of Sciences and the American Association of Universities are holding workshops and coordinating studies of how to handle misconduct.
There is much at stake. Once uncovered, cases of fraud can be dealt with under existing law. But much scientific misconduct falls short of that. It ranges from careless data taking and claiming credit for another's work to research fakery that does not involve legally defined fraud.
Trying to police this range of malfeasance could bring an unwelcome McCarthyism that would sour the research enterprise. It would be better to find ways to encourage an individual sense of ethical responsibility.
Harvard University made an effort at this last summer in publishing guidelines that deal with some of the root causes, such as pressure to ``publish or perish'' and lack of supervision. The guidelines suggest using only a limited number of papers in promotion decisions. No more than 10 should be considered for full professor, seven for associate professor, and as few as five for assistant professor. Supervisors should take greater responsibility for juniors in their laboratories. And any co-signer of a research paper must take responsibility for its content along with the main authors.
There's nothing novel in the Harvard guidelines, except that they show a new earnestness to pursue the kinds of reforms many academics have talked about. Such reform should be pursued even more vigorously across the whole scientific community. Otherwise, scientists may find government defining and enforcing their research ethics for them.
A Tuesday column