(FORTUNE Magazine) – The reference in our headline is to political correctness (not personal computers), and the principal PC mugger in this case turns out to be Bernadine P. Healy, director of the National Institutes of Health. Until August, NIH was planning to fund what sounded like a fascinating conference at the University of Maryland on genetic factors in criminal behavior. But the PC watchdogs started baying, and Ms. Healy withdrew all support for the conference. The record resoundingly supports a statement by research scholar David Wasserman, a principal organizer of the event, who told the Chronicle of Higher Education: ''It has become very clear that ((she)) is just viewing this as a political liability, and that she just wants it to go away.'' Why would it be a liability? Scholars in recent years have studied links between genes and many other behavioral traits, including alcoholism and sexual preference, and most of this research has been supportively reported as enlarging our understanding of society. The difference here is, obviously, the racial angle: American blacks are disproportionately represented in the data on crime, and now -- according to the Justice Department's published crime statistics -- account for the majority of all violent crime in American cities. So the Congressional Black Caucus instantly turned on the conference when some PC academics alerted the solons to it. NIH then suddenly discovered that the conference was flawed because its title (''Genetic Factors in Crime: Findings, Uses, and Implications'') seemed to imply that there surely were such factors. John W. Diggs, deputy director for outside research at NIH, was quoted in the New York Times as decrying the project because of its ''unjustified leap to the conclusion that there is a genetic predisposition to crime . . . There is no basis for that.'' What is bizarre about all this is that many different studies -- most of them involving white populations, incidentally -- persuasively suggest that there is a link between crime and genes. It is well documented, for example, that identical twins are substantially more likely to be ''concordant'' for crime -- i.e., for both to be either criminal or noncriminal -- than are fraternal twins. In both groups, the sets of twins share the same environment. The difference between them is that the identical twins are genetically indistinguishable, while the fraternal twins, like siblings generally, have only about half of the same genes in common. Bottom line: The more similar you are genetically, the more likely you are to have similar propensities toward crime. Is it really impossible for the U.S. government to confront such data? The answer may be yes. In the PC age, the emerging rule for bureaucrats is that it is safe to be officially ignorant of politically incorrect data -- and very dangerous to suggest the data might actually enlarge understanding.