October 1, 2002
Who do you trust? ... Let's take five with Moira Gunn. This is "Five Minutes".
It was a whole lot like an episode of Law and Order, only there were no cops and no lawyers. This time, there were only scientists.
It all began with what looked like the inadvertent insertion of the wrong graph in a prominent journal, and the highly-regarded and incredibly-prolific young author simply submitted a replacement. But before long, scientists throughout the entire research community began looking more closely at the 80 papers he had published within a two-year period.
Detail by detail, an alarming picture emerged: Identical results when testing completely different materials. His data was simply "too perfect." Still, his research was hailed as a huge breakthrough in physics, giving a whole new platform for the computer chips of tomorrow.
No wonder it was widely believed he was on the fast track to a Nobel Prize ... and no wonder 20 scientists leapt on the bandwagon and served as co-authors on this mountain of publications. To be published with a Nobel Prize winner - and to have added your own research twist. It might guarantee much needed funding in future. And who knows? Perhaps it might get you one step closer to the coveted prize itself.
--
Alas, all that glittered was not gold, and before long it became obvious that an official inquiry had to be undertaken. Lucent Technologies, the parent company of Bell Labs, where the scientist was employed, assembled an independent blue ribbon panel to investigate. And that's when the questions really got tough.
Five months later, the committee confirmed scientific misconduct, and the scientist was dismissed. You can read their 127-page report online, and it's frankly, very painful.
For one thing, the young scientist asks us to believe that he deleted his raw scientific data because he ran out of disk space - if you know scientists, they guard their raw data with their lives. Repeatability is the key to proof in science. Furthermore, you might see something else when you go back to it later.
--
While I have faintly suggested that his scientific collaborators might perhaps be guilty of at least weakened character, please duly note that they have been completely exonerated. You see, the young scientist represented his data as trustworthy.
I remember working at NASA. I led a team of 6 programmers building a complex computer application. When it was shown to work, we made an elaborate presentation to management for which we received many kudos.
The next day I took my team to lunch, and while we were waiting for our pizzas, everyone started to look very anxious. One of the programmers turned to me, and said, "I need to tell you something." He said that when he and his wife came across the border from Mexico into California, he had stashed marijuana in the hubcap of their car, but he didn't tell her about it. He said that if she knew, she would never have been convincing at the border.
I asked him, "Why are you telling me this?" He looked me straight in the eye, and said, "The program doesn't work."
I've got an inkling as to how those scientists must feel. It will take a while before they believe in themselves again.
I'm Moira Gunn. This is Five Minutes.
