Should scholarly publishing have a three-strikes rule?
Imagine a researcher who has had three papers retracted because of proven research misconduct. They submit another manuscript, perhaps to a different journal or publisher. Should that submission be considered in the normal way, or should the researcher face restrictions on their ability to publish?
My starting proposition is deliberately provocative: three retractions and you’re out.
Perhaps that means being prohibited from publishing for three years, five years, or even permanently. The precise sanction is something we could debate, but the underlying principle deserves consideration. Should repeated research misconduct eventually lead to consequences that extend beyond the individual papers involved?
At present, journals can retract papers, institutions can investigate misconduct, and publishers can take action against authors. However, there is no universal, internationally recognised system that automatically prevents someone with a history of serious misconduct from submitting to another publisher.
Is that an acceptable situation?
A retraction is not necessarily misconduct
Before considering any form of three-strikes policy, we need to recognise an important distinction. A retraction does not necessarily mean that an author has committed misconduct.
The Committee on Publication Ethics (COPE) makes this distinction in its retraction guidelines, updated in 2025. Retractions are intended to correct the scholarly record rather than punish authors. Papers can be retracted because of honest mistakes, unreliable findings, fabricated data, manipulated peer review or other serious problems.
An author who discovers a significant mistake and voluntarily requests a retraction should not be treated in the same way as someone who knowingly fabricates research results.
Indeed, we should encourage researchers to acknowledge and correct their mistakes. A policy that penalises every retraction equally could discourage precisely the behaviour that research integrity requires.
I would not advocate counting every retraction as a strike. The proposed system would need to distinguish between genuine errors and cases where there is sufficient evidence of misconduct.
This introduces an immediate complication. Who makes that judgement, and what standard of evidence should be required?
Who receives the strike?
Consider a paper with ten authors that is retracted because one author fabricated part of the dataset.
Should all ten authors receive a strike, or only the person responsible? What if the corresponding author was unaware of the fabrication but had responsibility for checking the data?
There is also the question of multiple retractions arising from a single incident. Suppose a researcher used the same unreliable dataset in five papers, all of which were subsequently retracted.
Should that count as five strikes, or one case of misconduct affecting five publications?
Conversely, imagine a researcher who deliberately fabricated results in three separate projects over ten years. Should that be treated differently from someone whose five retractions all resulted from one investigation?
These distinctions matter because the number of retractions is not necessarily a reliable measure of the seriousness, frequency or responsibility associated with research misconduct.
If we are to introduce a system of sanctions, we need to consider the circumstances rather than simply count retracted papers.
Would three strikes be the right threshold?
The number three has an obvious appeal. It is easy to understand, memorable and suggests that researchers have been given opportunities to change their behaviour.
But why three?
Someone who commits one particularly serious act of deliberate fabrication might warrant stronger action than someone involved in three comparatively minor cases of misconduct.
Similarly, an author with three retractions over a thirty-year career might warrant a different response from someone who accumulates three within eighteen months.
Perhaps a more appropriate system would have several stages. An initial finding of serious misconduct could trigger additional scrutiny of future submissions. Repeated misconduct could lead to a temporary restriction on publishing, with longer restrictions for subsequent offences.
A permanent exclusion might be justified in exceptional circumstances, but it would need especially strong safeguards.
There is also the question of whether restrictions should apply to all research publications or only to journals participating in an agreed scheme. A publisher can decide not to accept submissions from a particular author, but a global prohibition would require much more coordination and authority.
Protecting researchers from victimisation
Any proposal of this kind must address a serious concern: could the system itself be used to victimise researchers?
Academic disagreements can become personal. Researchers may be involved in disputes over findings, authorship, intellectual property or institutional decisions. Allegations of misconduct can sometimes be malicious, mistaken or motivated by personal interests.
We should not create a system in which accusations alone can damage someone’s right to publish.
At a minimum, any formal sanctions system would need to ensure that:
- A strike follows a properly established finding of misconduct, rather than an allegation or retraction alone.
- Responsibility is attributed to individuals, not automatically to every co-author.
- Researchers have the right to respond to allegations and challenge findings.
- Decisions are made or reviewed independently, with appropriate protection against conflicts of interest.
- Sanctions are proportionate, time-limited where appropriate, and subject to appeal.
- Findings can be reconsidered when new evidence emerges or a retraction is overturned.
Even these safeguards would not eliminate every risk. Universities and publishers do not always have identical procedures or standards, and an internationally recognised system would require considerable confidence in the fairness of those making decisions.
However, the difficulty of designing a fair system does not necessarily mean that we should dismiss the idea. Approaches that strengthen the integrity of the scholarly archive while protecting individual rights are worth considering.
Who would enforce the rules?
This may be the greatest practical challenge.
If a researcher is prohibited from publishing by one publisher, what stops them from submitting to another? What happens if they change institutions, use a different email address or publish under a variation of their name?
One possible solution would involve cooperation between publishers, supported by reliable author identification and an agreed framework for sharing verified findings of misconduct.
There are already developments in this direction. For example, the STM Integrity Hub supports publishers in identifying potentially problematic submissions and sharing relevant research-integrity intelligence.
However, detecting suspicious manuscripts is quite different from maintaining a system of sanctions against individual researchers.
Any such system would have to address confidentiality, data protection, accountability and the possibility of incorrect identification. It would also need to decide who could impose sanctions and how participating publishers would recognise them.
COPE provides guidance on publication ethics, but it is not an international disciplinary tribunal. Universities can investigate researchers, but their authority does not necessarily extend beyond their own employees and students.
Perhaps what is needed is not a global blacklist, but an independent mechanism through which substantiated findings of serious misconduct can be considered when assessing future submissions.
That would still be a substantial undertaking, but it might be more defensible than automatically excluding researchers based on a numerical threshold.
Should repeated misconduct have wider consequences?
The purpose of retracting a paper is to correct the scholarly record. I agree with that principle, but it leaves another question unanswered.
What should happen when the same researcher repeatedly publishes work that later requires retraction because of established misconduct?
Should editors and publishers simply evaluate each new manuscript without considering the researcher’s history? Or should repeated serious breaches of research integrity affect the level of trust placed in future submissions?
There are reasonable arguments on both sides. Research should be judged on its merits, and even someone who has previously committed misconduct may subsequently produce legitimate, valuable work.
On the other hand, journals depend on authors providing accurate information, following ethical standards and responding truthfully to editorial enquiries. Repeated misconduct can undermine that trust.
There is also a wider responsibility to protect the scholarly record, reviewers, editors, readers and researchers who may build on published findings.
A carefully designed system could seek to balance these competing interests. It might impose temporary restrictions in serious cases, require enhanced verification of subsequent manuscripts, or allow researchers to demonstrate that they have addressed earlier concerns.
Such measures would require evidence, transparency and procedural fairness. They should not be used simply to punish people for having made mistakes.
Three strikes and you’re out?
I started with a deliberately simple proposition: three retractions and you’re out.
Having considered the complications, I am not convinced that an automatic three-retraction rule would be fair. There are too many differences between honest errors, isolated misconduct and repeated deliberate manipulation of the publication process.
However, I remain attracted to the underlying idea that repeated, proven research misconduct should have consequences beyond the retraction of individual papers.
Perhaps three substantiated findings of misconduct should trigger an independent review rather than an automatic ban. A temporary restriction might be possible when a pattern of serious misconduct has been established, with a right of appeal.
Whatever the mechanism, I think scholarly publishing needs a more coherent way of addressing researchers who repeatedly abuse the trust placed in them.
Would you support a system that temporarily prevents researchers with repeated findings of serious misconduct from publishing, provided there are independent investigations, safeguards and a right of appeal?
Or should every new manuscript be judged independently, regardless of the author’s publication history?
I publish two free newsletters each week
Governance & Research in HE
A weekly brief for higher education leaders and researchers examining governance, leadership and research in practice.
https://buff.ly/GhsZqVl
Publishing with Integrity
Examining scholarly publishing through an ethical lens, challenging assumptions and supporting academic career development.
https://buff.ly/WiG1yFb
If you subscribe, you will be alerted whenever I publish a new edition.
About the author
Professor Graham Kendall is an Emeritus Professor at the University of Nottingham and a former Vice-Chancellor, with extensive experience in university leadership, research and scholarly publishing. He has published approximately 300 peer-reviewed papers and has served on several international journal editorial boards, including as Editor-in-Chief of IEEE Transactions on Computational Intelligence and AI in Games. He writes regularly about research integrity, higher education governance and academic publishing.