Upload
quirin
View
215
Download
1
Embed Size (px)
Citation preview
David Cyranoski, TokyoWhere does the border lie between simpleerror and scientific misconduct? That is thequestion thrust in the face of the Japaneseresearch community by the unsavourymeans of an anonymous letter.
The letter was sent to funding agenciesand publishers throughout Japan, Europeand the United States in August 2001. It concerned papers by Takuji Tanaka, a cancerresearcher at Kanazawa Medical University.The anonymous author claimed that therewere some unusual similarities between fig-ures in an article on foods that may help toprevent cancer published in September 1997in the Japanese Journal of Cancer Research (T.Tanaka et al. Jpn J. Cancer Res. 88, 821–830;1997) and those in a 1999 paper in Carcino-genesis (T. Tanaka et al. Carcinogenesis 20,1477–1484; 1999). The letter says that bothpapers include two identical figures forpolyamine compounds in rats, even thoughthe two experiments used different numbersof rats for different lengths of time. The letterwent on to allege that this was just one of atleast 30 discrepancies in articles by Tanaka.
In January 2002, the Japanese Journal ofCancer Research ran a correction to the 1997paper (Jpn J. Cancer Res. 93, 109; 2002). “Itwas a clear mistake,” admits Tanaka, “but Ihave never fabricated data.” He also deniesany misconduct in the other cases.
Tanaka’s supporters say that he is the target of an unjustified personal attack, and
they reject the anonymous author’s claimthat he or she is acting on behalf of anunnamed watchdog group dedicated to protecting scientific integrity. The author’sallegations are limited to Tanaka. “He or shejust wants Tanaka to lose his job,” says HidekiMori, head of the medical department atGifu University and a co-author on the twopapers mentioned in the letter. “Tanaka’sreputation has suffered unfairly.”
Mori suggests that Tanaka should take the case to a lawyer or the police. “The authorof the letter is like a stalker,” notes Mori, who claims to have a very good idea of theauthor’s identity.
But it would be healthy for the scientificcommunity to have a mechanism in place toinvestigate such charges properly, Moripoints out. Japan lacks anything like the US health department’s Office of ResearchIntegrity, and officials at both the ministry ofeducation and ministry of health say thatthere is no system in place to look at cases of misconduct.
Masaaki Terada, editor of the JapaneseJournal of Cancer Research and a formerdirector of the National Cancer Center inTokyo, says that he did wonder whether theallegations should be raised with anyoneother than his journal’s executive editors. Heagrees that Japan needs a system to investi-gate misconduct accusations, and says thathe plans to raise the question with the healthministry. “There should be a government-
level body to deal with this,” says Terada.The problem of misconduct is especially
difficult in Japan because the society doesnot look kindly on insiders who blow thewhistle, says Tokai University’s KiyoshiKurokawa. “Even if you want to report aproblem to your professor, he may be afraidof his dean or hospital director, who will beafraid of the education ministry,” saysKurokawa. “Such reporting may backfire.”
Some critics say that the reluctance to criticize others, especially those in establishedpositions, was to blame for one of Japan’s most notorious misconduct case — the 25years of research in which an archaeologistpushed back Japan’s stone age some 600,000years with his ‘discoveries’ of planted relics.After the fraud came to light, manyresearchers claimed that they had doubted thefindings all along.
More broadly, a spate of scandals inindustry and government circles hasbrought the problem of encouraging andprotecting whistleblowers to centre stage.The cabinet office is currently considering asystem to protect informants.
Kurokawa, who is vice-president of theScience Council of Japan, an organization that represents thousands of scientists, hasestablished a committee to come up withguidelines on the subject of scientific miscon-duct. Such a system could also offer protectionfrom false accusations, Mori points out. ■
Additional reporting by Oliver Schmidt.
news
348 NATURE | VOL 420 | 28 NOVEMBER 2002 | www.nature.com/nature
Japan ponders steps to probe data errors
Maths adds up for Berlin centreQuirin Schiermeier, Munich A centre that will apply mathematics to problems in physics, biology,telecommunications and transportation has opened its doors in Berlin. Advocatessay that it could catapult the German capital to the top of the league inmathematics research.
Over the next four years, the DFGResearch Centre for Mathematics for KeyTechnologies will receive 5 million euros(US$5 million) per year from the DFG,Germany’s main research funding agency,and a further 3 million euros per year fromBerlin’s state government.
“This is big money for mathematics,”says Martin Grötschel, vice-president of the Konrad Zuse Centre for InformationTechnology in Berlin and spokesman for the new centre.
“The nature of our discipline allows us to invest the bulk of the money in people,rather than in equipment,” says Grötschel.Six new full professors, seven young
principal investigators and 60 otherscientific staff will be hired at the centre in the coming months.
Günter Ziegler, a mathematician at theTechnical University of Berlin, who last year won the DFG’s prestigious Leibnizprize, is enthusiastic about the newinvestment. He says that the funding boostshould help to block the brain drain ofmathematicians from German academicpositions to industry and to the UnitedStates, where mathematics is winningincreased support from the National Science Foundation (see Nature 414,676; 2001).
Any concern that the new DFG researchcentre’s emphasis on applications will be at odds with fundamental research isgroundless, says Grötschel. “Modellingcommon problems in industry, banking andscience will require new algorithms,” hesays. “This will trigger a loop of researchthat is equally beneficial” to both theoreticaland practical mathematics. ■
Summing up: computer modelling will help toboost biology at Berlin’s maths institute.
ZIB
© 2002 Nature Publishing Group