Most authorship disputes are not caused by dishonesty. They are caused by assumptions: one collaborator assumes that collecting data earns authorship, another assumes it does not; one supervisor expects to be listed, one student expects to be first author, and nobody discussed it until the manuscript was ready. By then, every option feels like a slight to someone.
This guide sets out who deserves to be an author, who belongs in the acknowledgements instead, how author order is decided, and how a short written agreement at the start of a project prevents most disputes. It connects to our guides on research ethics and consent and writing acknowledgements, and to our guide on responding to reviewer comments, because authorship problems often surface during revision.
Authorship disputes usually trace back to three causes. The first is that contributions were never discussed: people joined a project at different times, did different work, and formed different expectations without ever comparing them. The second is that contributions changed during the project, so an early agreement no longer matches what actually happened, but nobody updated it. The third is that power is uneven: a junior researcher may feel unable to challenge a senior colleague's expectation, and a senior researcher may not realise a decision feels unfair.
Disciplines also differ in their conventions, which causes confusion in collaborative and interdisciplinary work. In some fields, the first author did most of the work and the last author is the senior investigator; in others, authors are listed alphabetically, or the principal investigator comes first. A collaborator from another field may read an author list by their own field's rules and misunderstand it completely. Stating the convention being followed, in the project and where the journal allows it, prevents that confusion.
The cost of a dispute is high. It delays submission, damages working relationships, and, if it reaches the journal, can delay or derail publication. Journals take authorship disputes seriously and will usually pause consideration until all parties agree, which means a dispute that could have been settled in an hour at the start of a project can cost months at the end. Prevention is not bureaucracy; it is the cheapest insurance a project has.
The most widely used standard, from the International Committee of Medical Journal Editors, requires all of the following: a substantial contribution to the conception or design of the work, or to acquiring, analysing, or interpreting data; drafting the work or revising it critically for important intellectual content; final approval of the version to be published; and agreement to be accountable for all aspects of the work. Other fields use similar principles even where they do not cite this standard directly: real intellectual contribution, involvement in the writing, approval, and accountability.
Notice what this standard excludes. Providing funding alone, supplying routine technical help, collecting data to order, or holding a senior position does not by itself qualify someone as an author, however valuable that contribution was. Such contributions deserve recognition, usually in the acknowledgements, but authorship claims a share of responsibility for the work's intellectual content, and that responsibility cannot be delegated or honorary.
Accountability is the part most often overlooked. Every author should be able to explain and defend the work, or at least to identify which co-author can answer for each part, and should be willing to cooperate if questions arise after publication. Anyone who would not want their name attached to the paper if its conclusions were challenged should think carefully about whether authorship is appropriate, and anyone listed should be asked that question directly before submission.
Acknowledgements exist precisely for valuable contributions that do not meet authorship criteria. Typical examples include technical assistance, data collection carried out to instruction, language editing, statistical advice on a specific point, access to facilities or samples, and helpful comments on a draft. Listing such contributors by name, with their permission, gives them visible credit without misrepresenting their role.
Permission matters. A person named in the acknowledgements is publicly associated with the work, and they may not wish to be, especially if they disagree with its conclusions. Always ask before naming someone, and describe their contribution accurately. Vague acknowledgements, such as thanking a whole department without saying why, give little credit and can imply endorsement that was never given.
Funding sources, institutions that provided facilities, and ethics committees are also acknowledged, usually in a form the funder or institution requires. Check those requirements early: some funders mandate specific wording, and adding it during revision is easy, but missing it can cause problems after publication. Our guide to writing the acknowledgements section covers wording and order in detail.
Author order should reflect relative contribution, under the convention of the field. In many fields, the first author made the largest contribution, the last author is the senior or supervising investigator, and middle authors are ordered by contribution. In some fields, alphabetical order is used, and in large collaborations, group authorship or a consortium name may be used with individual members listed elsewhere. Whatever the convention, it should be stated and agreed within the team, not assumed.
Order should be discussed early and revisited when contributions change. A student who began as a data collector but ended up designing the analysis and writing the first draft has earned a different position from the one first assumed, and the team should recognise that openly. Conversely, a planned lead author who could not complete the work should not retain first authorship out of politeness. Both adjustments are easier when the team has agreed that order follows contribution and will be reviewed at milestones.
Equal contribution deserves explicit handling. Some journals allow a note that two or more authors contributed equally, and where that matters to the authors, it should be requested at submission, not added after acceptance. If equal contribution is claimed, the team should be able to explain what each person did that justifies it. Shared first authorship is a real recognition, not a way to avoid a difficult conversation about order.
The corresponding author is the team's representative to the journal, and the role carries real responsibility. They submit the manuscript, communicate with the editor and reviewers, circulate drafts and revisions, and confirm that every author has seen and approved the submitted version and each revision. They also handle post-publication correspondence, including corrections and queries about the work.
Because the corresponding author controls communication, they must be organised and transparent: keep co-authors informed at each stage, share reviewer reports and response drafts, and obtain explicit approval before resubmitting. A corresponding author who revises without telling co-authors, or who submits a version nobody else has seen, creates exactly the kind of dispute that journals must then resolve.
The corresponding author is not necessarily the first or senior author, and the choice should be practical: whoever will be reachable throughout review and revision, and for a reasonable period after publication, is the right choice. If the corresponding author changes, for example because someone moves institution, the journal should be told promptly, and co-authors should agree the change.
Contribution statements, often using the CRediT taxonomy, record who did what in standard roles such as conceptualisation, methodology, formal analysis, investigation, writing the original draft, and supervision. Many journals now require such a statement, and even where they do not, writing one is excellent practice: it forces the team to describe contributions precisely, and it makes authorship discussions concrete instead of impressionistic.
A good contribution statement is specific. Rather than saying everyone contributed to everything, it says who designed the study, who collected which data, who ran which analysis, and who wrote which sections. Specificity protects junior researchers, whose work is most easily overlooked in a vague statement, and it protects the team if contributions are ever questioned.
Contribution statements should be written while the work is fresh, and updated as roles change. Reconstructing contributions months later, from memory, produces exactly the disagreements a statement is meant to prevent. Keeping a running record, even a simple shared document updated at project milestones, turns a difficult end-of-project negotiation into a routine review of facts.
Two practices deserve explicit warning. Gift authorship is listing someone who did not contribute enough to qualify, often a senior figure added out of deference or in the hope of lending prestige. It misleads readers about who did the work and exposes the listed person to responsibility for work they did not do. Ghost authorship is the opposite: omitting someone who did qualify, often a junior researcher or a professional writer, which denies them credit and hides a potential conflict of interest.
Both practices are misconduct under most publication ethics guidance, and both can lead to correction or retraction if discovered. Red flags include author lists that change substantially between submission and revision without explanation, authors who cannot describe the work when asked, and acknowledgements that thank someone for work that looks like authorship. Reviewers and editors do notice these signs.
If you are offered gift authorship, decline it politely and explain why: authorship should follow contribution, and an unearned listing helps neither the paper nor the person. If you believe you have been omitted despite qualifying, raise it first within the team, with a clear account of your contribution, and, if that fails, through the institution's or journal's process. Documenting contributions from the start makes either conversation far easier.
The single most effective step is a short written agreement at the start of the project, revisited at milestones. It need not be a legal document: a page that lists the team, the expected contributions, the intended author order or the rule for deciding it, the corresponding author, and how changes will be handled is enough. Every team member should see it and be able to raise concerns about it.
The agreement should also cover what happens if someone leaves the project, if contributions change, and if the team disagrees. Agreeing a process for disagreement, such as discussion followed by mediation through a supervisor or department head, is far easier before a disagreement exists than during one. Journals and institutions also have processes for authorship disputes, but they are slow and should be a last resort, not the plan.
Before submission, run a final check: every listed author meets the criteria, has seen and approved the manuscript, and accepts accountability; every significant contributor not listed is acknowledged with permission; the order reflects contribution under the field's convention; and the contribution statement is accurate. That check takes an hour and prevents the disputes that cost months. Authorship handled well is almost invisible; handled badly, it overshadows the research itself.
| Contribution | Usually author? | Notes |
|---|---|---|
| Designed the study or analysis | Yes | If combined with writing, approval and accountability |
| Collected data following instructions | No | Acknowledge, unless the role grew into design or analysis |
| Analysed data and interpreted results | Yes | With drafting or critical revision and approval |
| Wrote or critically revised the draft | Yes | Required alongside a substantial contribution |
| Provided funding or facilities only | No | Acknowledge the funder or institution |
| Gave technical or language help | No | Acknowledge with permission |
| Supervised without intellectual input | No | Supervision alone does not qualify; real input does |
| Commented helpfully on a draft | No | Acknowledge, unless comments reshaped the work |
These authoritative external resources explain the standards and guidance mentioned in this article:
Does providing funding make someone an author?
No, not by itself. Funding is acknowledged, but authorship requires a substantial intellectual contribution, involvement in writing or critical revision, approval of the manuscript, and accountability for the work.
Should a supervisor always be an author?
Only if they meet the authorship criteria through real intellectual contribution, writing or critical revision, approval, and accountability. Supervision alone, or a senior position, does not qualify someone.
What if contributions change during the project?
Revisit the authorship agreement at milestones. If someone's role grew into authorship, or shrank below it, update the list openly and record the change. Adjusting order to match actual contribution is fair; hiding the change is not.
How is author order decided?
Usually by relative contribution, under the field's convention — often first author for the largest contribution and last author for the senior investigator. Agree the convention and the order early, and review both when contributions change.
What is gift authorship, and why is it wrong?
Gift authorship is listing someone who did not contribute enough to qualify, often out of deference or to add prestige. It misleads readers and exposes the listed person to responsibility for work they did not do.
What should I do if I am omitted despite qualifying?
Raise it within the team first, with a clear account of your contribution against the criteria. If that fails, use the institution's or journal's authorship dispute process. A written record of contributions from the start makes the case much easier.