Strict Internal Review May Curb Research Fraud

William Brinkman's Opinion piece "Integrity in Industrial Research" (Physics Today, March 2003, page 56) is commendable for setting the record straight. There are many misconceptions about pressure in the industrial environment, but my own experience was very different. The years I spent at Bell Labs (1975-77) were the happiest and most important period of my professional life. And I was under much less pressure during that time than after I became a tenure-track assistant professor.

However, the issue raised by Jan Hendrik Schön's case is not industrial versus nonindustrial research. The real problem is that the company allowed so many grossly fraudulent papers to be submitted for publication under its name. While working at Bell Labs, I published more than 20 articles in refereed journals. I remember the internal review process as an unmitigated nightmare, much tougher than the journal refereeing process.

The pressure, in fact, was on being right rather than on achieving at all cost. One grossly fraudulent paper would have been stopped; a series of flashy results would have triggered merciless scrutiny. Why is this no longer true?

Brinkman says, "Most of the management was paying more attention to applied research because we really were trying to deliver technical value to the company." That focus, however, does not justify the failure of the internal screening process, which, instead of involving upper-level managers, should be handled by peers and the direct supervisor.

The screening that took place in the Bell Labs of 2000 could not match that of the Bell Labs of my youth. The deterioration in the screening process is part of the problem--and solutions should be sought there.