Writing a Paper With Co-Authors Who Write Differently

A paper assembled from sections written by different people reads as assembled. The vocabulary shifts, the level of detail varies, the same concept is named three ways, and the seams are visible to any reader.

The fix is not better editing at the end. It is deciding, at the start, how the writing will actually happen. For a concrete operational example of how a workplace tool approaches a related coordination problem, see read more.

Decide the process before writing

Who drafts what, and whether anyone drafts at all.

Three workable models:

One drafter, others comment. Produces the most coherent text. Requires the drafter to have the time and the others to accept the voice. Usually the best choice, and it should be paired with clear authorship agreement so that drafting is not confused with contribution. See authorship.

Sections divided, then one person harmonises. Faster to a complete draft, and the harmonising pass is real work that must be assigned to someone and scheduled.

Sequential. One person writes, passes to the next, and so on. Slow, and it works for short papers.

What does not work: everyone editing simultaneously in a shared document from the start. It produces conflicting revisions, lost text and a document nobody feels responsible for.

Agree the vocabulary first

The cheapest fix and it is almost never done.

A short list of terms and how they will be used. What the outcome is called, what the treatment is called, whether it is participants or subjects or respondents.

Ten minutes, and it removes a category of confusion that otherwise persists into the published version and confuses readers.

Agree the tense and voice conventions too, briefly, since fields differ and people carry habits between them.

Structure before prose

Write the outline together, in enough detail that each section has a stated purpose and a stated claim.

Then the numbers and figures, before the prose. If the empirical content is settled, the writing is describing something fixed rather than a moving target.

Then draft.

Rewriting a fully written section because the structure changed is the most common waste in collaborative writing, and it happens whenever people start drafting before the outline is agreed.

Whose voice wins

Somebody's has to.

Name the person who owns the final pass, at the start. Their job is consistency: one vocabulary, one level of detail, one register.

Everyone else agrees in advance not to relitigate style in that pass. Substance, yes. Whether a sentence should have been phrased differently, no.

This is easier to agree at the start than to introduce at the end, when the person doing it looks like they are overriding colleagues.

Feedback that does not stall the project

Deadlines for comments. "Comments by the 14th" gets comments; "let me know what you think" produces silence for six weeks.

Say what kind of comment you want. Structure and argument at the outline stage; prose only once the structure is settled. Line edits on a draft that will be restructured are wasted work and they discourage the restructuring.

One round at a time. Circulating a new version while people are still commenting on the last one produces chaos.

Use tracked changes and comments, not silent rewriting. A co-author who cannot see what changed cannot review it.

And respond to comments visibly, even briefly. A comment addressed silently leaves the commenter unsure whether it was seen.

When people write at very different levels

Common in collaborations spanning career stages or first languages.

Do not silently rewrite a junior co-author's section. They learn nothing and they will notice. Comment first, let them revise, then edit.

Distinguish clarity from style. A section that is unclear needs work; a section written in a different but clear style may not.

For co-authors writing in an additional language, the useful help is specific: this sentence is ambiguous, this term is not standard. General remarks about the writing are not actionable and land badly.

And be aware that the final pass will erase their voice by design. Say so in advance so it is not experienced as a judgement.

Practical mechanics

One current version, one location, always. See collaborating across institutions.

Version control or a clear naming convention.

Say who is editing what, so two people are not revising the same section in parallel.

A shared reference library, so citations are consistent and nothing is added twice.

Agree who submits, and who is corresponding author — a separate role from authorship position.

Before submission

One person reads it end to end, aloud or on paper. Seams are audible.

Check the vocabulary is consistent against the list you agreed.

Check the numbers in the text match the tables. In multi-author papers these diverge during revision more often than in single-author ones, because no one person tracks both.

Everyone reads the final version. All authors are accountable for it, which means all of them should have read what is being submitted. This is regularly skipped and it is the point of authorship.

The short version

Decide the process before writing. One drafter produces the most coherent paper; simultaneous editing produces the least.

Agree the vocabulary in ten minutes and save a category of confusion.

Outline and empirics before prose, or sections get rewritten when the structure moves.

Name who owns the final consistency pass, at the start, and agree not to relitigate style in it.

And everyone reads the final version — that is what authorship means.

For an independent authoritative reference, consult the EQUATOR Network for reporting guidance for research writing.