I am about to start a research project with two other people. I was wondering if there are any countermeasures if someone stops working halfway through but doesn't officially withdraw. Do I still have to give them credit? Or what would happen?

  • How, exactly, do you know that they have "stopped working"? Did they say so? Oct 30, 2023 at 18:35
  • 3
    If you're spending time wondering about what to do if your collaborators back out before you even begin work, perhaps you don't have the right collaborators. Oct 30, 2023 at 22:44
  • 5
    1. If people participate in building half a house, you need to give them credit for participating in building the house. 2. Don't be miserly about giving credit.
    – einpoklum
    Oct 31, 2023 at 7:32
  • 4
    This question doesn't meet site quality standards in my opinion because you are asking an open-ended, hypothetical question: “What if ______ happened?”. Do you have a "practical, answerable questions based on an actual problem that you face"? Oct 31, 2023 at 21:01

6 Answers 6


While this is a frustrating part of collaboration (aka grown up group projects) it isn't something you can prevent. If someone contributes early on and then drops off the grid you still have to acknowledge their contribution - they must be given credit for their work. That may mean they still retain some middle authorship or it may mean only a thank you in the acknowledgements. This would depend on the specific circumstances.

The best way to mitigate the chances of this happening are to set expectations for contribution early and assign achievable, discrete goals for every member of the team.

Past that it's just luck of the draw, at least until you can really control who you work with.


This is quite discipline specific; here's a mathematician's perspective.

The social contract and the rule of thumb that I've often heard cited and that I abide by is that if a person contributed 10% of the work on the paper, they deserve authorship. For a two-author paper, this means one person might end up doing 90% of the work, and I personally find it helpful to make peace with this possibility before entering a collaboration (your mileage may vary, of course). I've not been in a project that's nearly as skewed as 90:10, but I've been in situations where it was 70:30 or 30:70 (in my subjective judgement). In the end, it tends to even out over a sufficiently long period of time.

It's unfortunate when one author stops contributing halfway through, but assuming that they made significant contributions until this point they should be given credit (as a naïve approximation, half the contribution for half of the time is one quarter of the total work, which is fair for coauthorship). This situation may or may not be their fault, and it's best to give people the benefit of the doubt whenever possible. In either case, maintaining good relationships in the community tends to be more important than upgrading co-authorship to solo authorship on a specific paper (the same point applies even more strongly to papers with more than two coauthors).

The best countermeasure is to avoid collaboration with colleagues that you've had negative experiences with. It's also helpful to talk about expectations ahead of time, especially when there is significant uncertainty involved and if the publication is particularly important for one of the participants. (For instance, if me and a colleague come up with an idea for a paper and we're fairly sure it will work out, he's uncertain whether he'll have the time to work on it in near future and I need to get the paper on arXiv before April to include it in my CV when applying for a job, we might tentatively agree to work on it together with the contingency that if he doesn't make significant contribution by February I'll do the remainder of the work and publish it alone.) Finally, you don't have to rely just on your own experience, you can also ask others about their history of working with a specific person. Coauthors who ended up doing more than their fair share of work tend to be at least mildly annoyed and happy to talk. Just make sure these conversations don't turn hostile!


While you need not over worry about it, nonetheless, what a splendid research approach; carrying out a risk analysis and ensuring risk mitigation. The threat factor, you've identified, is valid and 'common'.
Having the 'right' collaborators in the first instance is golden. Ensure your circles are of the right type.

Ordinarily, being part of a research project ought to attract co-authorship, however, this is not (shouldn't be) cast in stone. There're other expectations of writing up and especially reading (or signing off) the final manuscript.

Beyond these, as you rightly identified, co-authors might not pull their weight and put the research projects at risk.


  • engage and have clear expectations from the onset; including authorship, communication strategy, conflict management...
  • address delays, incommunicado, below par contributions collegiately yet firmly and outcome oriented
  • do not shy away from quick phone calls, quick chat/SMS, friendly reminder/nudging
  • ditto: initiate/call for project risk/crisis meeting
  • where and when need be, set clear specific deadlines
  • ...
  • do what is academically necessary to get the research project going/completed

Couples of previous questions to pay attention to and factor into your risk analysis and treatment

  1. What can I do if my coauthor takes a long-time/unreliable to finalize/submit a paper?
  2. Former advisor/coauthor not answering emails
  3. How to politely and diplomatically ask one "half-senior“ coauthor to contribute more?
  4. How to handle a coauthor who is very late in finalizing a research paper?
  5. Coauthor always postpones working on article
  6. How should I confront an under-performing coauthor?
  7. How to treat an under-performing coauthor?
  8. Coauthor wrote a small part of the paper then ghosted us
  9. Unresponsive coauthor
  10. Non-responsive co-author
  11. What can I do if one coauthor tries to prevent a journal paper from being published?
  12. How to submit errata when a coauthor disagrees

Not losing sight of How to politely ask a co-author to remove his/her name from my manuscript?
Coauthor used large part from common paper almost verbatim in another paper

All the best and may you have like-minded positive collaborators.


Academics is a repeated game.

Having a non-performing co-author, if there are already three authors, comes at almost no cost to anyone.

Just put their name as co-author and never work with them again.

Prior to submission, all co-authors should approve the manuscript to be submitted.

If an author is being unresponsive, is fine to send an email a week or so before submission saying, "Please review this manuscript and make sure you agree with all statements made. Please alert me to any typos or other errors. If you do not respond by dd.mm.yyyy, we will assume you are no longer interested in co-authorship and you will be thanked in the Acknowledgements." Send a follow-up message three days later.

(By the way, our Departmental musical group is called "The Non-Contributing Co-Authors". More than once I have knocked people off of short lists or declined writing them an evaluation letter because they were non-contributing co-authors. Justice is often served in the long game, and not in individual papers.)


What do you mean by "countermeasure"? I know you don't mean to threaten any potentially lazy coauthors :) Truth is you can't make someone do the work you want them to do, but if they provided even an iota of intellectual propriety they should be considered according to the journal's authorship criteria - which are very detailed.

Since the submission process goes between named authors and the journal, I'm sure lots of collaborators "fall off" either due to malice or otherwise - but that gets hairy if the offended non-author discovers the paper and takes issue with the editor. A bad reputation is just the beginning in that case.

I do stats. I've had several uncomfortable discussions about not gaining authorship upon doing a final analysis because the prior statistician did several times more work than I, although it was all for nought because the team had no well defined hypothesis or understanding of the data. But Lancet is Lancet, JSON is JSON, etc. etc. There are many papers where one author contributes heavily at the front end and very little to the final manuscript. In another example, I stressed the importance of predetermining the entire content of the paper, and I wrote my section, performed all analyses and provided very little additional work even when it was requested. I did this to enforce a bit of data integrity toward the design. Many content experts naturally posit more and more hypotheses as they see more and more results, but we know it turns into a fishing expedition and the result is confusing and not meaningful. My strategy was good in this case, the paper was exceptionally focused. The team however despised my approach and felt probably similar to your concerned scenario. By giving less, I was confident I gave them more.

So at a minimum you can a. Share the authorship guidelines from the target journal and b. solicit some acknowledgement of the work expectations for authorship, but it's far easier and more political to give credit where it's not due than it is to withdraw credit from where it might be due.


A good collaboration should result in a whole that is worth more than the sum of its parts, such as when experiment and computation combine into a coherent description of a phenomenon.

Such a collaboration creates a powerful disincentive for your collaborator to "defect", in that if they leave the collaboration, they lose a very nice piece of work that would otherwise help their career. Additionally, most people simply enjoy the feeling of having helped someone else. You should know these are powerful motivators -- after all, I'm sure those are the reasons you want to collaborate with them!

So the most efficient thing to do is to let these motivators have their way, and to smoothen the collaboration with clear, consistent communication. Be sure that you both know what to do, and that you both know that you both know. (An indisputable paper trail happens to be very important for claiming priority should a collaboration go wrong.)

Of course, some collaborations will nonetheless fail because of unfortunate circumstances (pandemics and the like), but there's nothing either you or your collaborators can do about those besides being resilient and planning well.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .