The fact of plagiarism itself is easy to establish in most cases and the willingness of scientists to come on record to say so would show that the science enterprise in India is robust and well.
This is the fourth in a series of notes in connection with the workshop on “Science, Journalism, Media: Communicating Science in a Changing India”, held at the Institute of Mathematical Sciences, Chennai. The first of these outlines the aims of this series. It is available here.
Prasad Ravindranath (The Hindu) raised an important issue for a science journalist covering instances of scientific plagiarism. He said that he often found it difficult to get scientists to place their views on public record in these matters. In some cases, these scientists were from the same institutions where the alleged plagiarist occupied a position of some power and may have thus felt constrained. But even scientists elsewhere, when contacted, often declined to go on record in similar cases.
The decision of an individual scientist to go on record or not depends on multiple factors. Some scientists might not want to be associated with an issue that could be thought of as controversial, even if they agreed with the substance of the allegations. Others might have other reasons, for example professional associations with the alleged plagiarist. Some might fear retribution, while others could be motivated simply by a desire to wait for any official investigation to play out.
My own view? Plagiarism is the easiest thing to prove. Unlike cases of harassment, there is no “He said, she said”. The evidence is documented. Plagiarised documents can be compared to their originals. There is sometimes supporting evidence of unethical conduct from lab-members or colleagues of the scientist involved.
In some other cases, for example checking whether a Western blot has been duplicated in a publication at two separate points or simply copied across publications, more analytic methods are required. These are of course cases of actual scientific fraud and misrepresentation.
In one famous case, my colleague Rahul Siddharthan (IMSc, Chennai) made an animated gif file illustrating the similarity between two sets of Western blots. These were supposed to be different according to the corresponding author of the papers in question, but the issue was easily answered at a visual level by simply toggling between them. The Society for Scientific Values (SSV) had flagged the similarities in images earlier but Rahul’s demonstration was particularly graphic.
But in many other cases, involving the cutting-and-pasting of text wholesale, the otherwise untrained eye can see what an expert can see too. (The Wikipedia page on “Scientific plagiarism in India” is a useful resource of past cases.)
So scientists can and should be able to comment in public if their attention is drawn to such cases. The ethical responsibility of being a scientist is too central to let issues like this slide. Also, students must be alerted to the consequences of professional misconduct. This is no substitute, of course, for an investigative procedure that helps to establish what happened. Was it bad oversight, did a student send off the paper without showing it to an advisor, did a collaborator provide text for incorporation that was inadequately verified and so on?
But the fact of plagiarism itself is, as I said, easy to establish in most cases and the willingness of scientists to come on record to say so would indicate that the science enterprise in India is robust and well.
All that said, let me point out some shades of grey. In my mind, there are a hierarchy of offences against the practice of science. The worst of these, if we exclude cases of harassment, corruption etc, all of which have a social component, is the deliberate falsification and misrepresentation of data. Copying a Western blot from one paper to another ranks with these, because it misrepresents data. Producing a fake graph, or representing data measured in one case as being measured in another are all examples. Making up the results of large cohort studies in biomedicine, as in a recent Japanese case, is another example.
These attack the foundation of science as a body of trusted work that is further built on by the efforts of many. We think of the next step because we trust the work that has led us to the present step. A foundation of lies is impermanent.
But not all is necessarily entirely black-and-white in the issue of plagiarism: To deliberately pass off large swathes of another’s text, whole chapters and entire sections as one’s own is the worst form of plagiarism and one that deserves to be called out loudly and clearly. Stealing someone else’s idea and passing it off as one’s own, equally so.
Copying a line or two from another text? Well, reprehensible but perhaps less so and perhaps just the result of carelessness. Much would depend on whether there is a pattern of similar behaviour. Of course, the probability of this being an accident decreases exponentially with the length of text that is identical between articles.
I personally see the question of self-plagiarism as less important than plagiarism of another scientist’s work. At least the original version was that of the author and, especially if these are a small number of lines of boilerplate text from the introduction, usually part of a general description of prior work and background. So, journalists should call attention to such instances by all means, but should take it less seriously. This would be my own view. (I should point out that this particular view is not shared by all.)
There is also be the question of pattern. A one-off case would really be ignorable. A pattern of papers where lines of central text are repeated, much less so. This would be even more reprehensible if these lines are from a methods, data or conclusions part of the text, rather than from an introductory paragraph, in the case of self-plagiarism. Copying at a larger scale from other authors who are not acknowledged, credited or referenced appropriately is a serious matter indeed.
As was pointed out in the aftermath of the Watergate scandal: “It’s not the crime, it’s the cover-up”. This is certainly true. In some cases, efforts have been made to shift the blame to the student author or authors on the paper. This won’t wash. The responsibility for a paper rests on all authors, but more particularly so on the so-called “corresponding author”, the one who takes overall responsibility for the paper, who corresponds with editors, coordinates the response to the referees and so on. So journalists should be alert to any move to dodge responsibility by the senior author.
Equally, science journalists should be positive in their coverage of situations where the authors seem to have made a genuine effort to set things right, accepting responsibility, withdrawing the paper if required or certainly publishing a correction, as well as responding honestly to press queries.
We all make mistakes. We might remember a nicely written line from a paper that we’ve read but when it comes to writing our own, what enters our mind unbidden might be traces of that memory. We trust a student to write their own text so that they get experience in writing – in fact we encourage it – but we might miss the line stolen from Wikipedia, especially if we’re in a hurry. We trust the integrity of collaborators who contribute text or sections to our paper. We hope that our collaborators will not send a paper off with our name on it without telling us and before we have had a chance to review our contributions. And journals are often reluctant to amend the published record.
All of this also points to the complexity of doing science in the modern world, where numbers of publications and impact factors of journals matter in terms of professional recognition, grants and public profile. There are certainly temptations to take the short cut, perhaps more in some fields than in others.
The culture of the field matters too in this respect. A culture of writing a large number of inconsequential papers is more likely to encourage ethical lapses. Misconduct seems to correlate to more hierarchical organizations, not just in India, where willingness to call out unethical behaviour can be tempered by a feeling of general powerlessness.
What is irreducible, however, is scientific honesty, coupled to the willingness to set things right. And here, there’s no question that we should all be on the same page.
Gautam Menon is a Professor at The Institute of Mathematical Sciences, Chennai. He can be reached at firstname.lastname@example.org.
Since this article is based on what I had said during the workshop, I would like to place on record certain other things that I did not tell then. In the case involving image duplication (mostly Western blots) in 18 papers by two faculty members at Bose Institute, Kolkata, I did get in touch with two other scientists from different institutes to tell me if the charge of image duplication (as highlighted by people on Pubpeer) was indeed correct. I wanted this confirmation even before I contacted the concerned authors at Bose Institute. And both scientists were very prompt in their response. Since the two Bose Institute authors admitted that there were issues with the images, there was no need for me to get a scientist to go on record confirming image duplication. But I am not sure if I would have been lucky in finding scientists willing to be quoted saying that the images have been duplication if the authors had contested.
Unlike plagiarism, which is quite easy to confirm by science journalists, especially when the copied portions are highlighted by a plagiarism detection software, the same cannot be said about image duplication/manipulation, particularly Western blot. (But even in plagiarism we need independent scientists to confirm it so Editors are more likely to publish the articles.) Even though many science journalists have a science background, not everyone can look at a Western blot and be sure if it has been duplicated/manipulated. What is troubling is that, I have been seeing more and more instances of papers by researchers from Indian lab/institutes with such troubling image duplication/manipulation. And there is a pattern to this -- the authors of such papers tend to be serial offenders. Will scientists familiar with Western blot be willing to go on record to confirm image duplication/manipulation?