WHAT will 2002 be remembered for? Publication of the mouse genome or the emergence of RNA interference, with all their potential for treating human disease? The solution to the solar neutrino problem? Or as the year that physics lost its innocence?
Ask physicists and they will tell you that their subject is intrinsically trustworthy because it leaves little room for subjective thinking. No matter what question they ask, eventually they hit nature鈥檚 bedrock, where the true shape of reality reveals itself.
Not everyone agrees. Some thinkers argue that scientific knowledge depends on social and cultural conditions and is not a version of a universal truth. This group so incensed the American physicist Alan Sokal that in 1996 he sent a spoof article loaded with sociological jargon to the journal Social Text. When the paper was published, Sokal crowed that the editors had not had the wit to realise they had been duped. Their field, he said, was replete with 鈥渘onsense and sloppy thinking鈥�.
Advertisement
But this year, the tables were turned. Rumours circulated that physics journals had unknowingly published four spoof papers on the origins of inertia and space-time by the French twins Igor and Grichka Bogdanov. Or had they? The brothers, who recently received their PhDs, insist their papers are serious. The irony runs thick here. After six weeks of discussion on the Net, some of the world鈥檚 leading mathematicians and physicists cannot decide whether or not they are spoofs.
How can such a mess happen? Mathematician Peter Woit of Columbia University, New York, put it best by decrying 鈥渢he low standard of coherence in the whole field鈥�. This sounds awfully like 鈥渘onsense and sloppy thinking鈥�. Nobody would argue that science can operate without theories. Even theories that will never be tested have a role in challenging the way people think or opening new avenues for enquiry. But there is a fine line between deep wells and murky shallows, which PhD supervisors, journal editors and referees should be aware of.
Theory is one thing, but in experimental physics, the bedrock of reality ensures that any sloppy or suspect results never stand for long. At least, that used to be the case. This year two apparently brilliant physicists were accused of making up results over several years. Victor Ninov was one of 16 reseachers who, in 1999, announced they had created elements 116 and 118. But it turns out that Ninov fabricated the crucial evidence on which the announcement hung. In May, he was sacked from Lawrence Berkeley National Laboratory in California.
But the biggest shock came in October when Hendrik Sch枚n was sacked from the prestigious Bell Labs in New Jersey after being found guilty of fabricating a host of results. Sch枚n had claimed significant advances in nanotechnology and superconductivity, but investigators found doubtful results in 17 papers published over three years.
For most honest, hard-working physicists, these incidents may seem irrelevant. But they are not. The fraud genie is out of the bottle and it will not go back in. The integrity of research results is paramount within physics not least because pay and promotion depend on numbers and quality of papers published. Funding from outside depends on the trust that the public and politicians place in physics. These high-profile cases have damaged that trust.
To its credit, the American Physical Society acted swiftly, recommending that all institutions adopt the US government鈥檚 guidelines on research misconduct. It also urges training on research ethics for students. Yet, sadly, it chose to dilute the responsibilities of co-authors. Ninov and Sch枚n had dozens of co-authors who basked in the reflected glory of the stars鈥� work but never questioned it. Surely all co-authors have responsibility for checking the integrity of the results they put their names to. But the APS advice reads like a get-out-of-jail-free card for anyone working alongside a fraudster. The best guidance comes from Germany, where Ninov and Sch枚n studied. 鈥淎ll authors of a scientific publication share the responsibility for its contents,鈥� states the German Physical Society. 鈥淓xceptions should be clearly indicated.鈥�
One excuse used widely after the Sch枚n affair by co-authors, journal editors and referees was that they didn鈥檛 look for fraud because it happens so rarely. That excuse simply won鈥檛 wash any more. To misquote a famous US President, the price of trust is eternal vigilance.