The plagiarist, the fraudster and the cheat may be rare in academe, but that still leaves scope for some dubious practices, says Harold Hillman.
Scientific fraud has been going on since the Greeks initiated the scientific method. In addition to the fraud committed by the great pioneers, most contemporary academics and research workers know of examples of smaller or greater fraud by their colleagues. They would not be wise, and it would not further their careers, to expose such practices, which by their nature are difficult to prove and document.
However, in historical times the intellectual climate was quite different. It was considered proper to report only the data that supported one's own beliefs. The concept of a "control" experiment was not known until the l9th century.
Much of belief was based on the authority of the person who said it, probably because science was intimately bound up with religion. For the same reason, students were discouraged from asking fundamental questions. Theirs was not to reason why; theirs was but to do or die. In all branches of thinking, such questioners were regarded as heretics.
Nowadays, all academics agree about the concept of fraud. For example, one cannot claim to have done a certain number of experiments or observations if one has done fewer or none; one must not discard some values in experiments, which do not fit in with one's belief; one cannot plagiarise other people's data or ideas; one must not misquote other people's publications, etc. The whole academic and research community condemns such practices, and sometimes punishes the miscreants.
But in addition to fraud, there is a wealth of publications containing mistakes, made usually in good faith. These include mistakes in measurements, readings, calculations, calibrations, typing, etc. Only occasionally, corrections are published subsequently, and they are unlikely to be seen and taken into account by interested readers. One of the biggest problems besetting all modern academics is the plethora of publications, so that most honest research workers simply do not have the time to be aware of and to evaluate critically the whole literature crucial to their projects.
Although no one knows the extent of fraud, my judgement would be that its total impact, at any time on biological science at least, is relatively small. And I think most scientists would agree.
However, another huge but overlapping area of academic research and thinking is regarded as normal practice, and is not talked or written about. One may call this "para-fraud". It may be defined as "unwillingness to enter into proper dialogue with people with whom one disagrees, or who have challenged one's findings, assumptions, theories, or published opinions". It probably arises from the understanding that knowledge is power, and questioning or debunking someone else's firm beliefs threatens their power and self-esteem. I believe that para-fraud is real and widespread, although it should be totally alien to academic practice.
Examples include:
* speakers who do not answer challenging questions crucial to their beliefs at learned society meetings or in correspondence; * research workers who do not report their own experiments or observations that are incompatible with their beliefs; * academics who do not quote publications whose conclusions they do not like;
* scientists who do not carry out the relevant control experiments, either by omission, or refusal to do so, when attention has been drawn to them. An experiment or observation is only as good as its controls. Elsewhere, I have published evidence that in several widely practised areas of basic medical and biological research, such as subcellular fractionation, electron microscopy, immunocytochemistry and histology, crucial control experiments have never been published. These procedures are used in trying to understand the biochemistry of cancer, multiple sclerosis, muscular dystrophy, schizophrenia etc. The failure to carry out crucial controls means that the experiments are incomplete, and it puts at risk all findings, theories and treatments based on these experiments;
* the referees of manuscripts for publication in learned journals and for applications for grants for research are usually anonymous, and do not have to enter into dialogue with applicants or give detailed reasons when they reject them. Failure to publish or refusal of funds discourages research into alternative hypotheses. The referees may be competitors with the authors or applicants, and the value of their own work may be threatened by the publication of the manuscript or the expected results of the project. Thus, the referees often have divided loyalties; * the major speakers at national and international meetings of influential learned societies are usually decided by confidential meetings of the executives of the societies. Applicants for substantive presentations of say 30-60 minutes have no right to know why their particular contribution has not been chosen. Those with whom the organisers do not agree are usually allowed to present their ideas on metre-square posters, of which there may be several hundred at a meeting: interested parties who speak to the exhibitors of posters block their access to many others, so that the authors can only address a very small audience. However, the exhibition of a poster allows the organisers and the authors to pretend that all participants at a meeting have had a reasonable and fair exposure, when they have not;
* academics challenging fundamental aspects of a consensus are denied access to the popular media, including the British Association for the Advancement of Science, New Scientist and Scientific American. Unfortunately, the popular media are extremely influential on students at schools, who may be intending to enter the particular research field, as well as on interested lay persons, and on collectors for research charities;
* some supervisors expect to share authorship of research work, in which they have made little or no intellectual or physical contribution. In the expanding research groups which are proliferating at present, senior authors may not see the jobbing worker frequently or even at all, so may not know about the techniques, errors, calculations or literature in relation to a project, in which authorship implies intellectual responsibility. Sometimes when fraud has been discovered subsequently, senior authors have then written to journals to abrogate responsibility.
Colleagues to whom I have spoken admit that examples of para-fraud are very widespread. I would maintain that they are extremely influential in creating the consensus of belief, which is basic to medical, biological and technological research. Some colleagues say that one does not expect academics to behave in a more moral way than the public at large (I am afraid I do). Others have said that "the truth will out in the end". Does this permit them to continue to make dogmatic statements in areas of controversy without considering the evidence currently available?
Other answers have been that in many professions and vocations such as in theology, cosmology, archaeology and medicine, the fundamental science is sufficiently distant at present from the practical applications, that the truth of the basic beliefs at this time is not so important practically. Many teachers feel that they themselves are not knowledgeable enough to assess the evidence for conflicting viewpoints, and so they teach the consensus views found in the textbooks. Many academics admit in private that their desire to keep their jobs or obtain promotion discourages them from being involved in controversies. They have families and mortgages to support.
One has an uneasy feeling that one has heard of para-fraud before. Not answering questions, being economical with the actuality, avoiding making crucial observations, emotive dismissal of evidence against their own view, and monopolising the media, are normal tactics of politicians, and potentates.
Are we not entitled to expect academics and research workers to be more intellectually honest than this?
Harold Hillman is reader in physiology at the University of Surrey, and director of the Unity Laboratory of Applied Neurobiology.