It’s not a new virus. Various types of frauds have been there throughout the history of science. The fake fossil human skull famous by the name Piltdown man created its own history. Some experimental results by great personalities such as Gregor Mendel and Jacques Monod are said to be “too good to be true”, i.e. so well fitting their hypothesis that such a result is statistically highly unlikely given the inherent biological variability. It is possible therefore that they themselves, or someone obliging them manipulated results or cherry picked only the favorable ones. There can be many more examples that never got exposed. But in spite of such examples the core spirit of science has largely survived and maintained its reputation.
However, the rate at which misconduct is mounting today seems to be unprecedented. There is reproducibility crisis, doubts raised about designs and conduct of clinical trials, statistical analysis twisted to support the favored hypothesis, manipulated images and cooked up data. It is difficult to decide whether the frequency of scientific misconduct has increased or only the rate of getting exposed has. The rate at which image manipulations are getting detected is due to the efforts of groups like ‘pubpeer’ (https://undark.org/2020/07/23/cracking-down-on-research-fraud/, https://blog.pubpeer.com/). Image manipulation is only one type of fraud. Currently most of the flags raised are based on detecting image manipulation. Many other types of frauds may have just escaped because we don’t have tools to detect them. If this is true, the actual frequency of misconduct must be substantially higher. ‘Retraction Watch’ (https://retractionwatch.com/) makes data on over 30,000 retractions and the reasons behind them accessible to people. Retractions on getting the misconduct exposed include papers by reputed laboratories and mentors. Occasionally the responsible people have to pay the cost of the deed through their nose but many others appear to escape more or less unhurt. The culprits may not pay the cost but science may be paying a big cost in terms of its reputation. Certainly I find it hard to prevent myself from losing faith in published science and I am not the only one. Richard Smith, the former editor of British Medical Journal, remarks that it is time to assume that all health research is fraudulent until proven otherwise!! (https://blogs.bmj.com/bmj/2021/07/05/time-to-assume-that-health-research-is-fraudulent-until-proved-otherwise/).
There have been many discussions on how to prevent the scientific frauds of various kinds. People have been debating the pros and cons of better ethics education, institutional mechanisms of vigil , strict action and punishment as deterrent and so on. But I am troubled more by another innocent but unpleasant question that keeps on peeping in my mind in spite of my repeated attempts to suppress.
The high frequency of retractions makes me wonder about how science works today. I thought science works by building on prior knowledge and working the way ahead. In my mental model of science, every piece of evidence, every bit of data, any new concept, model, analysis is crucial for progress. One published piece of work lays the foundation of further work and so on. If this was true, any paper retracted would have made some concept collapse, some paradigms failed, some lines of work given up, some technologies defunct and so on. But I am surprised that following over 3000 retractions per year there is hardly any collapse seen, no fundamental change in the direction of work, no reconsideration of the existing paradigm, no rethinking of any prevalent theory. If retraction of thousands of papers has no major effect on science, it only means that these papers never had any relevance to science. Their being in a state of ‘published’ or ‘retracted’ makes no difference to the field. Perhaps they are being published only for the benefit of the authors in building their CVs and getting better positions. If they really made a difference to the field of science, their retraction should have affected the field quite badly, but that is not the impression one gets after looking at the work of pub-peers and retraction watch. Thousands of papers have been retracted but no scientific theory has collapsed or no technology has been withdrawn.
If papers are irrelevant, so must be the career of a researcher. If in judging the career of a researcher, we go by the number of such publications and weigh by impact factors and so on, we are counting the irrelevant. This way we are building more and more irrelevant science in our institutions. No wonder if the institutions themselves become irrelevant soon. They already have to a large extent, as the apathy of common man, government and media shows.
I still hope and believe that interesting, fundamental, relevant and important science is being pursued somewhere. But it may not be with the big people working in the prime institutions of science and publishing in big journals. It may be happening in some obscure lab somewhere, some teacher with a handful of undergraduate students, some farmer, some uneducated, illiterate, humble, unnoticed individuals in some corner of a third world addressing a basic question. May be that kind of science will count ultimately, some day, some time. Let us wait till them.
What else can we do? Tell me if you know.