India has seen a rise in research paper retractions in recent years. In the newly released 2025 National Institutional Ranking Framework (NIRF) rankings, the framework now deducts points for research papers retracted in the past three years, signalling a stronger stance on integrity, with penalties expected to tighten over time.
At first glance, that sounds sensible. Why should institutions benefit from work that turns out to be incorrect and is retracted from the published literature? Reputational pressure can improve behaviour,
India has seen a rise in research paper retractions in recent years. In the newly released 2025 National Institutional Ranking Framework (NIRF) rankings, the framework now deducts points for research papers retracted in the past three years, signalling a stronger stance on integrity, with penalties expected to tighten over time.
At first glance, that sounds sensible. Why should institutions benefit from work that turns out to be incorrect and is retracted from the published literature? Reputational pressure can improve behaviour, so sending a signal that integrity matters is overdue.
However, in practice, this idea may have unintended consequences. A blunt penalty may push cases into the shadows. Institutions may cooperate less with journals that publish their authors’ research. Second, not all retractions are due to fraud, and failing to distinguish between causes may prevent the disclosure of honest errors.
A journal can screen a manuscript for plagiarism. It can run basic image checks. Editors can spot statistical red flags. What a journal cannot do is walk into a lab and pull the notebooks off a shelf. It cannot examine raw instrument files on a protected server or interview lab members. That is the institution’s job.
Most serious ethical cases start with a concern from a reviewer, a reader, a whistleblower, or a post-publication forum. The journal writes to the authors and to the university. An integrity officer asks for lab records, reagent logs, original files, and the history of analyses. Sometimes the issue is clear-cut: duplicated images, fabricated data, or a ghostwritten paper with fake peer review. Sometimes it is not: a messy analysis, a reagent gone bad, or a bold claim that does not hold up when others try to repeat it. Only the institution can get to the bottom of these cases. Journals do not have that power and must rely on cooperation.
Also read: Scientifically Speaking: Fake scientific research on the rise, says study
That is why a blunt penalty can backfire. If every retraction automatically hurts a university’s rank, administrators may decide that cooperation is risky. They may stall requests for data. They may fight over notice wording for months.
There is another basic point the policy must acknowledge: retraction does not equal fraud. Retractions happen for two broad reasons. The first is misconduct, including plagiarism, image manipulation, invented data, and fake peer review. Penalise these cases heavily. The second is honest error or non-reproducibility, where the authors made a mistake or subsequent work showed the claim does not stand. This is science doing its job.
Consider the recent retraction of the 2010 “arsenic life” paper in Science. It made a provocative claim about a bacterium that could use arsenic in place of phosphorus. Many scientists were skeptical, and over the years, the evidence against the claim grew. Science retracted the paper this July without alleging misconduct. We can disagree with the timing, but the message is clear: sometimes a retraction is a course correction, not a confession of wrongdoing. If NIRF treats both categories the same, it will punish transparency and discourage self-correction.
The record itself must be clear on this matter. Retraction notices should use a standard set of reasons so that rankings can classify them accurately. Vague language helps no one. Most reputed journals follow guidance that distinguishes between retractions for ethical violations and non-ethical reasons such as honest error or non-reproducibility.
Also read: Scientifically Speaking: How a fake journal is hijacking science
Journals need better tools to detect manipulated images and text. Institutions need trained research integrity officers, clear investigative procedures, and whistleblower protections. The cost of putting these safeguards in place is far smaller than the cost of letting flawed science accumulate.
Finally, the process must be transparent. NIRF should publish exactly how it classifies and tallies retractions and give institutions a way to appeal. Whether the data come from journal notices or other sources, the methodology should be public. Sunlight protects everyone, especially honest researchers.
I support docking points for fraud. It shows that integrity matters. But without separating fraud from honest correction and addressing the broader ecosystem of research quality, the policy risks driving problems underground.
(Anirban Mahapatra is a scientist and author, most recently of the popular science book, When The Drugs Don’t Work: The Hidden Pandemic That Could End Medicine. The views expressed are personal.)
One Subscription.
Get 360° coverage—from daily headlines
to 100 year archives.
Archives
HT App & Website