A paper is submitted with four authors.
During revision, a fifth author appears. Before acceptance, an author disappears. Then, after the paper has been accepted, two more authors are added.
Should the journal simply update the author list?
I would hope not.
There are legitimate reasons why authorship may need to change after a manuscript has been submitted. But changing the names attached to a paper is fundamentally different from correcting a spelling mistake, updating an affiliation or changing a corresponding author’s email address. Authorship determines who receives credit for the research and, importantly, who accepts responsibility for that paper.
In this era of paper mills and authorship-for-sale, perhaps we should regard a request to change authorship not simply as an administrative amendment, but as a potential research-integrity concern.
There can be legitimate reasons
It is certainly not the case that every authorship change indicates misconduct.
Reviewers might request additional experiments. A researcher who was not involved in the original study might undertake that work and make a contribution which justifies authorship. Deeper analysis might be required, which requires statistical expertise that the current authors do not possess. Perhaps somebody who clearly qualified as an author was inadvertently omitted at submission.
These things happen and we have to provide procedures to deal with them.
Indeed, publishers recognise that authorship sometimes needs to change. Springer Nature provides specific training for editors [1] on assessing requests to add, remove or reorder authors and distinguishing acceptable from problematic requests. Wiley requires authorship changes to be formally requested [2], with agreement from all co-authors, including those being added or removed, and it records the reasons for the proposed change.
But the existence of legitimate reasons, or procedures, to change authors does not mean that authorship changes should be routine.
Paper mills change the equation
There is another reason to take authorship changes seriously.
Paper mills do not necessarily sell complete papers with a fixed author list. Some sell authorship positions while manuscripts are moving through the publication process. A study in The BMJ [3] said:
“In some cases, they sell the authorship before publication, they then handle the submission and the peer review process. Other organisations sell the authorship after the manuscript has been accepted for publication in a legitimate scientific journal. When this scenario occurs, the organisation includes the author or authors who bought the authorship on the list of named authors, which amounts to a (sometimes total) change in authorship.”
Think about what that means. The researchers whose names appeared on the manuscript submitted for peer review might not be the researchers whose names eventually appear on the published article.
That should change how we view authorship amendments.
What should make an editor suspicious?
No single authorship change necessarily demonstrates wrongdoing. But some cases deserve attention.
- Adding one researcher during a major revision may have a straightforward explanation. Adding four authors from previously uninvolved institutions shortly before publication deserves rather more scrutiny.
- Removing an author should generate questions. If somebody satisfied the journal’s authorship criteria when the manuscript was submitted, why do they no longer satisfy those criteria? Alternatively, if they never satisfied them, why were they listed as an author in the first place?
- One change might be understandable. Several additions and removals during successive rounds of review begin to suggest that something unusual is happening.
- A change in corresponding author alongside substantial changes to the rest of the author list is a strong signal that deserves investigation and clarification.
- The appearance of authors with no obvious intellectual, institutional or collaborative connection to the work could be another red flag.
What about after acceptance?
For me, adding authors after acceptance is particularly difficult to justify. At that point, the research has been conducted, the manuscript has been written, revisions have been completed, peer review is finished and the journal has decided that the paper is suitable for publication.
So what justification can there be for adding, or removing, an author at this stage?
There may be exceptional circumstances in which there is a convincing answer. But the burden of explaining that contribution should be substantial and would be exceptional, rather than routine. I have edited journals and authored many papers and I have never seen an author added, or removed, after the paper has been accepted.
Some publishers already take a very firm position. Elsevier states that authors should provide the definitive author list when a manuscript is originally submitted, that subsequent changes should occur only in exceptional circumstances, and that authorship changes are not permitted after acceptance [4]. Other publishers allow greater flexibility. Wiley, for example, provides mechanisms for correcting authorship even after acceptance or publication where there is a valid reason and the required approvals are obtained [2].
The differences between publishers raise an interesting question. Should they not agree on how to deal with authorship changes so that every journal operates in the same way? What justification is there that different journals/publishers operate in different ways in this important aspect of scholarly publishing?
Should we simply prohibit changes?
There is an attractive simplicity to say that the author list is fixed when the manuscript is submitted.
If the authors subsequently decide that the paper has the wrong authors, they withdraw it, resolve the issue and submit it again with the correct author list. That would certainly make authorship trading more difficult, though not impossible.
But it may also be unnecessarily punitive when circumstances genuinely change during peer review. A substantial revision can involve new experiments, new analysis and sometimes new expertise. Academic publishing is already slow enough without forcing legitimate manuscripts back to the beginning of the process.
A compromise might therefore be more sensible. Perhaps authorship changes should be allowed before acceptance, but the process should be deliberately difficult.
Make authors explain themselves
Perhaps any request to add or remove an author should require more than an email saying, “Please add Professor X.“
The journal could require the corresponding author to explain why the change is necessary and, where somebody is being added, exactly what that person contributed. They could also be asked why that contribution was not anticipated or recognised when the paper was originally submitted.
Where an author is being removed, the journal should ask why that person qualified for authorship previously but apparently no longer does.
Every existing author, every newly added author and every removed author could be required to confirm the change independently, using their own email address. Contribution statements and declarations of competing interests should be updated. More unusual changes could be referred to an editor responsible for research integrity before the manuscript proceeds.
And once the manuscript has been accepted, perhaps the default answer should be no, unless genuinely exceptional circumstances can be demonstrated.
None of these requirements would make improper authorship changes impossible. But they would make them harder.
Perhaps inconvenience is the point
We often try to make publishing systems easier for authors. Usually that is entirely sensible. But perhaps changing authorship is one area where friction is desirable.
A journal might spend months examining a manuscript. Reviewers scrutinise the methodology, results and conclusions. Editors assess revisions. Software checks references, plagiarism and sometimes image manipulation. Yet the names of the people claiming credit and accepting responsibility for that research can, in some circumstances, change during the process.
Perhaps the question is not whether authorship changes should ever be allowed. It is:
How difficult should it be to change the authors, what evidence should be required, what changes should trigger an integrity review, and at what point should the answer simply be no.
What would you do?
Would you allow authorship changes throughout peer review? Freeze the author list at acceptance? Or go further and require authors to withdraw and resubmit whenever the author list changes?
I am genuinely interested in where you would draw the line.
About the Author
Graham Kendall is Vice-Chancellor of GlobalNxt University, Malaysia, and an Emeritus Professor at the University of Nottingham. He has published more than 300 peer-reviewed papers and has held senior editorial roles, including Editor-in-Chief of the IEEE Transactions on Computational Intelligence and AI in Games and Associate Editor of the Journal of the Operational Research Society.
Publishing with Integrity explores the challenges facing scholarly publishing, research integrity and academic practice.
[1] https://www.springernature.com/in/editors/editor-courses/authorship-changes, accessed 14 Sep 2026
[2] https://authors.wiley.com/ethics-guidelines/index.html, accessed 14 Sep 2026
[3] https://doi.org/10.1136/bmj-2022-071517, accessed 14 Sep 2026
[4] https://www.elsevier.com/about/policies-and-standards/publishing-ethics, accessed 15 Sep 2026