Sunday, 18 December 2016

Honesty and Science

Honesty is a virtue. The strange thing about honesty is that we do not seem to see even the simplest aspect, telling the truth when it is owed, as a duty. People who would be horrified at hurting anyone will trim, twist, exaggerate and lie at the drop of a hat, especially when it advances their ideological agenda, and will not feel they have done much wrong at all. Worse, they will anathematise those who utter inconvenient truth and feel highly righteous in doing so.

Science can be pure enquiry but the questions it seeks to answer are often those with significant practical import. The reason we subsidise science is because of the benefits it promises. Getting the benefits depends on scientists finding out and telling us the truth. From this point of view, then, honesty is a prime virtue of science and to be honest is a stringent duty owed to us all by scientists. It is unclear to what extent scientists feel properly bound by this duty.

First of all, it has come to light that for many years many disciplines have promulgated fraudulent research without noticing it. There have been shocking high profile cases such as Marc Hauser, but anyone who has attended to the Retraction Watch blog will see that there is a steady stream of such blatant dishonesty in science. Recently we have Stapel promoting his left wing prejudices by making up stuff  to prove that failure can make you happier than success and untidiness promotes stereotyping and discrimination. Secondly, there have been equally shocking ideologically based witch-hunts, such as that perpetrated by the Marxist Leon Kamin and those we see pursued today over environmental issues. Thirdly, there are the ostensibly milder cases of trimming and twisting. This category may be where we should be most concerned since it exploits the rules by conforming whilst misleading. Consider all those pharmaceutical results where the positive trials get published and the negative trials left on the shelf. Consider also those experiments where researchers keep gathering data until they get a statistically significant result and then stop.

To some degree the success of these dishonesties has been the result of the high level of trust among scientists leading to a reluctance to believe the worst of anyone. This is interesting and important in its own right. To some degree science has worked precisely because scientists have been able to rely on one another’s honesty. One might say the same of society in general. It is an achievement of civilization that we can generally trust one another. But the very fact of widespread warranted trust increases the returns to cheating.

Obviously one mode of dealing with this problem is to have mechanisms for detecting cheats, and I am very glad to see that scientists are implementing such mechanisms. Most recently, and the efficient cause of this blog, the deliberate cherry picking of data by Smeesters was exposed by ‘Clever statistical sleuthing by an anonymous fraud hunter’. This case is interesting from out point of view because
Smeesters said this type of massaging was nothing out of the ordinary. He "repeatedly indicates that the culture in his field and his department is such that he does not feel personally responsible, and is convinced that in the area of marketing and (to a lesser extent) social psychology, many consciously leave out data to reach significance without saying so."
This seems to me demonstrate a failure of another mode of dealing with this problem that what we might call moralisation. A different example from some time ago manifests the same failing. Stephen Schneider said scientists should, ‘offer up scary scenarios, make simplified, dramatic statements, and make little mention of any doubts we might have…. Each of us has to decide what the right balance is between being effective and being honest.’

What is wrong in both these cases is that the scientists concerned do not seem to be clear on the nature of their duty. They are not free to massage results just because some of their colleagues do, nor are they free to balance honesty against the political agenda they wish to promote. When they take a job as a scientist they promise to tell us the truth. They may not  promote a personal interest or political prejudice instead.

Originally at

No comments:

Post a Comment