Honesty in Science – A Thing of the Past?

By Helmut Schwarz

It is painful to have to admit that, ever more frequently, the research community is being forced to address a problem which is not, however, particularly new: the number of publications being withdrawn is on the increase, scientific scandals hit the headlines too often, the reproducibility of data leaves a great deal to be desired in far too many studies – and this is just the start of the list of shortcomings.

Helmut Schwarz  (Photo: Humboldt Foundation / David Ausserhofer)
Helmut Schwarz  (Photo: Humboldt Foundation / David Ausserhofer)

If these findings are indicative of an honesty problem, then the entire science system is affected, research funders included, as the recent, controversial case at the Humboldt Foundation proves. A researcher who had been chosen to receive the Humboldt Professorship and its five million euros in funding was accused of scientific misconduct. The person involved, who denies the charges, relinquished the award before the Humboldt Foundation could hold deliberations on whether to grant sponsorship which it had temporarily suspended following the announcement of the allegations.

How would we have come to a decision if the researcher had not stepped down? This question uncovers a general dilemma. A funding organisation like the Humboldt Foundation is not an institution for examining scientific misconduct; neither does it conduct reproducibility studies or contract others to do so. It does not have a team of experienced statisticians who could scan publications for methodological errors and it is not even entitled to revise the judgements of committees or ombudspeople at home and abroad.

„Fraud allegations turn into an instrument of combat.“

The Foundation has to have trust, must be able to have trust – be it in the honesty and integrity of academics and research institutions, the critical, fair appraisal and healthy scepticism of reviewers or, finally, that the selfcleansing powers of science actually work. In the vast majority of cases, this really does seem to happen. But we have to recognise that the review system has long since reached its limits given the sheer volume it is faced with today. Concern that something might get overlooked is simply increasing. And how do we deal with allegations? Are there errors (“venial sins”) that are forgivable? Are researchers totally innocent until proved indubitably guilty? Or when the first serious doubts are raised, is it a case of “if there is doubt, there is no doubt” to quote the famous dictum of Massachusetts Institute of Technology?

We should not and shall not look for easy answers. Matters have to be weighed up carefully because the consequences for the career and life of the person involved may be irreversible. Moreover, it does not require an overactive imagination to envisage an unholy alliance between the honesty issues outlined here and the ever fiercer competitive spiral which could generate yet more allegations of fraud. The accusation of falsification, invention or plagiarism could develop into a crippling instrument of combat, used by every competitor to disqualify others from the race for funding or positions. I fear that we will not be spared this horror scenario in times to come.

But we are not defenceless, and the hope that science knows how to defend itself is not built on sand. At the annual meeting of the influential US National Academy of Sciences, Ralph Cicerone, its President, held a remarkable speech which not only included a selfcritical appraisal of the problems but also numerous suggestions for improving the situation: from replicating important studies, which is already being funded by the Laura and John Arnold Foundation, particularly in medical and experimental psychological research, via alternative, open publication paths like eLife, to voluntary commitments by the leading scientific journals and research institutions.

In order to solve science’s honesty problem, however, more than this is required. A clear stand should be taken against the trend to classify “cheating” – whether at school or later on in research – as a trivial offence.

No: honesty and integrity are not negotiable.

published in Humboldt Kosmos 104/2015

Professor Dr. Helmut Schwarz is the President of the Alexander von Humboldt Foundation. A much-decorated, internationally eminent chemist, he teaches at Technische Universität Berlin.