While the investigation into the work of Bell Labs physicist Hendrik Schön is not a criminal inquiry, it is as serious as science gets.
According to federal guidelines, scientific misconduct is not a matter of proposing a controversial theory, submitting sloppy data or retracting a paper. Nor is it a matter of publishing results that other scientists have difficulty duplicating.
As Bell Labs spokesman Saswato Das pointed out, it took seven years of experiments to confirm the presence of particles known as neutrinos predicted by physics theory. While that’s true, the Schön case isn’t a question of unproven theory, but a question about experimental results.
Scientific misconduct is by definition a matter of falsifying or fabricating results, or plagiarizing someone else’s work.
As the principal investigator on the papers under review, Schön would have been responsible for the accuracy of the data.
Schön has been a prolific scientist. Moreover, much of his work has been high-profile — “hot papers” as one physicist described them — that garnered attention and, say other scientists working on molecular electronics, some jealousy.
“This work was pretty spectacular,” said Paul Grant, a fellow at the Electric Power Research Institute and a former IBM scientist who supervised a group working on exotic superconductivity at the company’s Almaden Research Center in Silicon Valley.
“For the last year, the barroom buzz whenever physicists got together was, ‘How did Schön do it? How did he achieve such high fields given the dielectrics of the materials involved?’ ” said Grant, whose article on the investigation — “Is a bell tolling for Bell Labs?” appeared in the June 20 issue of Nature.
Grant said that in examining several of the Schön papers, the data in different graphs appeared to be very close, but not exactly the same. “It doesn’t look good,” but doesn’t appear to prove anything. “In Scottish law there can be three verdicts: guilty, innocent or not proven. I think this falls into that last category,” he said.
As Grant explains, scientific papers are typically reviewed by two or more anonymous “referees” who funnel their comments to the journal editor that serves as the ultimate arbiter between the author and the referees.
Grant said that he had not served as a referee on any of the Schön papers, but was contacted by one editor “as a safety check.”
While at IBM’s Almaden Labs, Grant considered the quality of researchers’ work as journal referees part of their job performance. However, he noted that not all journal article referees take their responsibility as seriously as they should. Some may be distracted by the demands of their own work; others may not be up to speed on leading edge research they may be asked to review.
With respect to the Schön work he’s looked at, Grant said that the results were provocative, but there was no reason to disbelieve them. “They did not appear to be breaking any laws of physics in the way that, say, cold fusion seemed to have,” he said.
For the last three years, Schön had been producing such a steady stream of papers on electric fields in crystals and molecular transistors that some people believed he might be working his way toward a Nobel. The majority of the papers appeared in Science, but others were also regularly showing up in Nature, Physical Review Letters and Applied Physics Letters.
Over this productive period, some scientific peers began noticing that Schön’s papers were brilliant but somewhat sloppy, with multiple errata filed in subsequent journal issues. Moreover, no one had been able to reproduce the results of the superconducting buckyball work, including other Bell Labs scientists.
Scientific papers don’t generally include an exact recipe for how a result may have been accomplished. There’s also an increasing tendency to leave out critical information to protect a patent application, especially if the patent is initially declined and needs to be refiled.
But as Grant wrote in Nature about the Schön flap, “if you’re playing the sport of experimental physics, someone else had better reproduce your results or you’re out of the ball game. If you’ve got an exciting result that may send you to Stockholm, the next thing to do, after you’ve established publication and patent priority, is to get your worst competitor to reproduce it, helping where necessary. Your management shouldn’t have to set up a special commission to pore over your notebooks and files.”
While a lone allegation of scientific impropriety in nanotechnology may not may seem like a heavy blow, the Bell Labs flap does comes at an inopportune time. The field is only beginning to attract broader investor and public attention, and is already contending with concerns of dot-com-tinged overpromise.
Several industry watchers note that the investigation also comes amid corporate fiascoes — from Enron to Tyco to Arthur Anderson to WorldCom — that have sapped the public’s trust in large corporations and raised market apprehension.
If the panel does find clear evidence of research misconduct — the most serious accusation that can be made against a scientist — what impact would it have on a small tech industry striving to find its feet?
While Mark Modzelewski, executive director of the NanoBusiness Alliance, believes Schön must be given every benefit of the doubt, he said that the investigation alone was a sharp blow. “This really pulls the rug out from under many of the businesspeople who were just beginning to look at investment opportunities.”
Modzelewski is also concerned that the Bell Labs investigation will erode the public’s and market’s trust that small tech research is making progress and producing important breakthroughs.
“This is such a new industry, and it was already facing falling confidence and concerns about hype,” he said. It would be easy, he fears, for the media to jump to conclusions that nanotechnology is not the Next Big Thing, but the Next Cold Fusion.
Also in this report
Did Bell Labs’ rising nanotech star commit the cardinal sin of science?
Controversy proves that science is battlefield of conflicting ideas