Authorship: Who Goes on the Paper, and in What Order

Authorship disputes are among the most damaging things that happen in research groups, and nearly all of them are caused by the same thing: the conversation was not had until the paper was nearly finished.

By that point positions have hardened, effort has been invested under different assumptions, and the disagreement is about fairness rather than about a decision. For a concrete operational comparison, Monitask explains accountability and responsibility at work in practical terms.

Conventions differ enormously by field. What follows describes the common principles and the main variations; check what your discipline and institution expect.

The criteria that are generally recognised

Most formal guidance converges on a version of this: an author has made a substantial contribution to the work, has been involved in drafting or revising it, has approved the final version, and accepts responsibility for it.

The last one is the part people forget. Authorship is accountability, not credit alone. If the paper is challenged, every author is answerable for it.

What does not by itself justify authorship: providing funding, providing data or materials, holding a supervisory position, or general departmental support. These are acknowledgements.

And what is frequently underweighted: substantial intellectual contribution to the design, even by someone who wrote none of the text.

How order varies by field

This is where cross-disciplinary confusion causes real harm, because people apply their own field's convention to a collaboration.

Alphabetical. Common in economics and parts of mathematics. Order carries no information about contribution, and everyone in the field knows this.

First author most, last author senior. Common in biomedical and life sciences. First author did most of the work; last author is typically the principal investigator or lab head; middle positions are ordered by contribution with the second position significant.

Descending contribution. Common in psychology and much of social science.

Corresponding author is a separate role from position and does not always signal seniority.

The consequence: a scientist reading an alphabetical economics paper may assume the first-listed contributed most. An economist reading a biology paper may not know the last position is the senior one. In interdisciplinary work, state the convention explicitly in the paper, and increasingly journals require a contribution statement, which resolves most of it.

Have the conversation early

At the point the project becomes a project, not when the draft exists.

What to settle:

• Who is expected to be an author, on what basis
• What order, and by what convention
• Who is corresponding author
• What happens if someone's contribution changes substantially
• What happens if someone leaves

Write it down. An email after the meeting is enough. This is not distrust; it is the same reason you write down anything else that several people will remember differently in eighteen months.

Revisit it when the project changes. Contributions shift. A conversation at the midpoint is much easier than a dispute at the end.

The conversation that is difficult

Two situations account for most of the pain.

Someone contributed less than expected

Address it when it becomes apparent, not at submission. "The scope of your part has shifted; can we talk about what that means for the paper" is an uncomfortable conversation in month four and a serious dispute in month twelve.

Distinguish contribution from effort. Someone can work hard on something that did not end up in the paper. That deserves acknowledgement and an honest conversation, not automatic authorship.

There is a power difference

The common case: a student or postdoc, and a supervisor or head of group.

The junior person usually cannot raise it safely, which is precisely why the senior person should raise it first, early, and unprompted. If you are the senior person, this is your responsibility and not theirs.

If you are the junior person and it was not raised: ask early, in writing, framed as clarification rather than as a challenge. "Just so I plan — how are we thinking about authorship on this?" is a normal question in month one and a loaded one in month ten.

If it goes wrong, most institutions have a research integrity office and most journals have authorship policies. These exist for this. They are unpleasant to use and they are better than the alternative.

Things that are not acceptable

Worth stating plainly, because they still happen.

Gift authorship — adding someone who did not contribute, usually senior, usually for political reasons. It dilutes the credit of everyone who did contribute and it is a form of misrepresentation.

Ghost authorship — omitting someone who did contribute, whether to hide an interest or because they left.

Requiring authorship as a condition of access to data, equipment or funding.

Adding or removing authors without telling everyone. All authors should see the final list before submission.

Contribution statements

Increasingly required, and useful beyond compliance.

They make the actual division visible, which reduces the load on positional convention.

Filling one in is a good forcing device. If someone's contribution cannot be described in a sentence, that is information.

Do it during the project, not at submission, so it reflects what happened rather than what people recall.

When someone leaves

A departing collaborator retains authorship for work already contributed. Leaving does not remove a contribution.

Agree in advance what happens if someone becomes uncontactable — most journals have a procedure, and knowing it beforehand prevents a submission stalling.

Keep the record. Who did what, dated. See a research record that survives you leaving.

For supervisors specifically

Raise it first. The student cannot.

Explain the convention in your field, because they may not know it, and may be too uncertain to ask.

Be conservative about your own inclusion. Supervision alone is not authorship under most formal criteria, however common the practice is in some places.

And say what you expect from them in return for the position — being an author means approving and defending the paper, and a student should know they are entitled to read it properly before that.

See supervision: what a student actually needs.

The short version

Authorship is contribution plus accountability, not credit alone.

Order conventions differ fundamentally by field. In interdisciplinary work, state which one you are using.

Have the conversation when the project starts, write it down, and revisit it when scope changes.

If there is a power difference, the senior person raises it — the junior person usually cannot.

For an independent authoritative reference, consult the Open Science Framework for infrastructure for transparent research practice.