A high-flying young physicist, who rapidly published a series of remarkable discoveries, has been sacked by Bell Labs, New Jersey, for 鈥渟cientific misconduct鈥�.
Hendrik Sch枚n鈥檚 dismissal follows an independent investigation at the labs that found 鈥渃ompelling evidence鈥� of data manipulation and misrepresentation in 16 of 24 papers examined.
The investigation 鈥渃ompletely cleared鈥� 20 co-workers of scientific misconduct, but raised questions about the extent to which co-authors of papers and lab heads should be responsible for published material.
Advertisement
Sch枚n鈥檚 work was remarkable, and some had tipped him as a future Nobel prize winner. It included apparent demonstrations of molecular-scale transistors and superconducting buckyballs.
No witnesses
But, with hindsight, there were warning signs. Sch枚n, 32, nearly always conducted his experiments and processed the data alone, often in a laboratory in his native Germany. The report says: 鈥淣one of the most significant physical results was witnessed by any co-author or other colleague.鈥�
Sch枚n鈥檚 productivity was astonishing too. In 2001 he is reported to have averaged a new paper every eight days including four in each of the leading journals Nature and Science. Most scientists would be delighted with four papers a year in any journal.
In an appendix to the report, Sch枚n says he disagrees with several of the report鈥檚 conclusions but admits to have made 鈥渕istakes鈥�, which he 鈥渄eeply regrets鈥�. He adds that he remains convinced that the physical effects he reported are real, and will be reproduced in future.
Identical graphs
Bell Labs was alerted to Sch枚n鈥檚 falsification and fabrication of data in May after other researchers studying published papers noticed identical graphs supposedly representing data from different experiments. It then formed a five-strong committee to conduct an investigation, the results of which were released on Tuesday evening.
The report says Sch枚n kept no proper lab records, had deleted nearly all the primary data files, 鈥渞eportedly because the old computer available to him lacked sufficient memory鈥�. Equipment that could have been used to repeat his key experiments has all been damaged or discarded, it says, meaning the validity of the claimed results cannot be assessed.
However, some scientists have already stated that such large-scale misconduct discredits the whole body of work. Bell Labs is working with the co-authors to determine which papers should be withdrawn.
Karl Ziemelis, physical sciences editor at Nature, which published five now-discredited papers, says the report is 鈥渄amning and depressing鈥�. Ideally, the papers would be withdrawn by the authors, says Ziemelis. But, speaking in a personal capacity, he says if this does not happen within weeks 鈥渢hey should be withdrawn come what may鈥�. It would be unprecedented for a journal itself to withdraw a paper, he notes.
Science鈥榮 editor-in-chief, Donald Kennedy, said that if no action was taken by the Bell Labs researchers, the journal would have to make its position clear. But both editors believe that the events have not exposed any major flaws in the reviewing process, which is not an investigative process, they say.
Bell Labs says it is reinforcing publication procedures and the process of internal peer review. Jeff Jaffe, president of research, said: 鈥淎n experience such as this heightens everyone鈥檚 awareness of the importance of ensuring scientific research integrity. Integrity is at the foundation of scientific research.鈥�