Take, for example, that a student is nearing the end of their thesis project.

The supervisor has an idea for a new method that can help to verify the results of part of the project. He does not think much of the significance of the idea at the time, [writes up a document outlining the procedure] and passes it to the student to implement.

A week later, a problem arises. The supervisor questions whether the idea was in fact more significant than he originally thought - i.e. could form a paper on its own, and has realised that answering the students questions on the specifics of how to implement the said method would take up more time/effort then just doing it themselves.

So, how would the supervisor go about retracting the new project from the student? From the student's point of view, they would have a strong motivation to try and implement the idea that has been brought to their attention since it is directly relevant to their thesis work. Yet the supervisor does not want to be hassled by the student on learning the specifics (that would form the content of the paper), and does not want the student to claim partial ownership of the idea. The student has not yet produced any results using the method that has been passed to them.

The issues that I see are as follows:

  • If the supervisor tells the student to relax and think about something else instead, they probably would not listen
  • If the supervisor works independently from the student to produce the paper (figures, text, observing trends), then the student could still claim that they were involved (?) since they would have probably started working on it too
  • 43
    and does not want the student to claim partial ownership of the idea. This to me sounds extremely unethical.
    – AndrejaKo
    Commented Apr 2, 2014 at 17:00
  • 23
    The only "issue" here (as everyone is pointing out) is the unethical alleged behavior of the supervisor.
    – Suresh
    Commented Apr 2, 2014 at 18:00
  • 13
    If "writing procedure into computer code and creating figures/observing features" was not an original contribution, all of us would be out of work.
    – Alexandros
    Commented Apr 2, 2014 at 20:22
  • 9
    So any time a student contributes to research efforts, "the publication [is] needlessly delayed such that the student [can] use it as a learning exercise and write the paper themselves (manual labor in creating figures/text)"?? Why on earth do you assume that students must necessarily be incapable of making a genuine contribution to a research effort? You realize that students are your future peers, right? Commented Apr 2, 2014 at 21:34
  • 23
    I think the down-votes are misplaced. Down-voting is supposed to indicate that the question is useless, not that the assumptions behind the question are reprehensible. This is an INCREDIBLY useful question, even though (or perhaps because) the answer is unequivocally HELL, NO!
    – JeffE
    Commented Apr 3, 2014 at 5:11

5 Answers 5


With the way this question is framed, "retracting" the idea sounds both difficult and unethical. Based on the information given here, my advice, and the only ethical path, is for the supervisor to collaborate closely with the student on the work. If the idea is good, the pair should work toward a co-authored publication. This question states that, "since [the student] has more free time on their hands [they] could potentially have some results first." In this case, it sounds like collaboration with the student is very likely to help.

Working with students means teaching and mentorship. It means that supervisors are, "hassled by the student on learning the specifics" and it often means, "answering the students questions on the specifics of how to implement the said method [in ways that] take up more time/effort then just doing it themselves." This is part of the job of an academic supervisor.

Moreover, I completely fail to see why co-authorship and shared ownership of an idea between a supervisor and student is a problem.

That said, one of the most important jobs a supervisor has is trying to decide when and what to delegate. It was the supervisors job to be more thoughtful and they need to deal with the consequences — whatever they are.

  • 1
    Does a high number of authors not impact the rating of the academic for promotion purposes? I assumed this since I had read papers where there was only one author (the professor), and the graduate students were mentioned only in the acknowledgements section
    – xyz
    Commented Apr 3, 2014 at 3:05
  • 10
    It doesn't matter. Ends don't justify the means; especially not in this case.
    – Suresh
    Commented Apr 3, 2014 at 4:33
  • 3
    @James Two is "a high number"? Commented Apr 3, 2014 at 12:37
  • 4
    In the vast majority of fields, co-authorship is the norm. You might have a hard time getting promoted if you are only ever author 15 on papers with 18 authors. Being first author on a co-authored paper with a student is just not going to hurt almost anywhere. If you are supervising students, you should know this. That said, even if it did hurt, it still doesn't matter in terms of what is ethical and appropriate.
    – mako
    Commented Apr 3, 2014 at 16:47
  • 7
    Since the question is formed in a hypothetical question, an answer should probably contain advice towards the student, too: run, was fast and as far as you can, and get a supervisor that won't skrew you over.
    – Raphael
    Commented Apr 3, 2014 at 22:40

The stated scenario sounds awful, and the clarification makes it sound even worse. When you start a collaboration with a student you're advising (I'm using "you" colloquially), you don't get to "keep the good bits". That's grossly unfair to the student who is after all attempting to learn from you.

Yes, research is uncertain, and you can't always predict what ideas will become interesting.

Which is why rules for collaboration are set up first, and are not contingent on the quality of the (potential) results.

I understand that in this case, there's a worry about the student "slowing things down". But that's about convenience, not about doing the right thing.

And I should say that because this is a student and not a collaborator, all rules should be adjusted to give the student the benefit.


It is hard not to sense some kind of "foul play" by the phrase He does not want the student to claim partial ownership of the idea and the phrase "just doing it themselves". Who are they? The senior professor, the PostDoc or a graduate student? If the idea was not good, the poor student should bang his head to make it worthwhile. If the idea is good then you somehow want the student out of it.

The main question is why? If you write a seminal paper based on the idea, one more co-author would not hurt you. Why don't you want the student involved? He will provide help and you can explain to him that he is not going to be the first author and immediately all your problems are solved.

Once the idea gets out of your head and into somebody's mind (and possibly his PC) it cannot be contained anymore. What will stop the student to implement it on his own? If he is fast enough, he may even do it faster than the busiest, more senior team members.

I know I do not answer the actual OP's question but this is deliberate. I consider highly unethical the fact OP thinks he can freely change your mind, make people work on his ideas and when those ideas seem promising (which is partially due to the student's work - although he clearly denies this fact) he claims full ownership and decides who else should work on his idea. The only reason I see legit for this kind of behaviour is if there is a highly ranked conference approaching and the OP (by doing the work himself) might make it on time, something that will be impossible for the student to do. If it is so, then the OP may explain it to the student and implement the idea himself. All other possible reasons are either questionable or possibly unethical.

If you treasure raw research ideas that much (and most of us do), I would suggest next time be careful who you share it with. When most people get an excellent research idea they only share it with the ones who will work on it. Once they share it, they should stick with these people, unless something unpredictable (work-related arguments, work relocation) happens. But even then, plan changes should be fully explained. And since from your previous questions I believe that you are possibly a graduate student, please do not start your academic career by questionable practises.


The only time I can imagine retracting a project from a student, is if continuing with it would be detrimental to them or to the university.

Retracting in other circumstances is unethical. It's a breach of trust. And it's a fundamental failure at the task in hand, of being a supervisor of research students.

It is my job, as a supervisor of research students, to enable them to become better researchers than me.

The first time I was told that, by a very wise colleague, I was shocked. I felt threatened. My students would go on to take my research, my grants, my future jobs away from me.

Then I took a deep breath, had a think about what my role was, and realised he was completely right. And I accepted it, and now I revel in it.

If it ever got to the point where I didn't want to do what I practicably can to make them better researchers than me, then it would be time for me to step down from that role.

++(yes yes, I know, it's my job to enable them to be the best researchers they can be, insofar as it does not ask for impracticable demands from me. The phrasing above isn't perfectly accurate. But it does have impact, and it does punch home an important message)

  • 1
    Personally, my experience is that somehow there is no scarcity in problems that wait to be solved. The bottlenecks are rather finding good students and thein their time as well as my time... I'd think it would help tremendously if by some magic I could make them good and experienced researchers immediately ;-) Commented Apr 3, 2014 at 18:50

Too late, you should have thought that more carefully beforehand. You cannot undo.

Like it or not, this student is now your collaborator for this idea, so I'd suggest to get the best out of the collaboration that you can get. This means no races, etc. Try to give him chunks of work that won't require too many questions and interaction (if that's a problem) and try to publish asap, because you are collaborating with this person, but probably you are "competing" with other people in other places, who may be able to produce results faster than you.

Next time, think about it twice before you share your next idea, specially if you are going to be so possessive about them. Now it's shared.

If the problem is being hassled, you can postpone investing time specifically on that until the deadline for publication is met. Priorities are priorities. The student may not care after that or may have no questions. By your wording the problem doesn't seem to be that, though.

We can only hope there are micro-contributions in the future so that we can trace exactly how much did each person contribute to something and have the provenance of everything with a personal granularity. Either good or bad, it's not happening anytime soon. I'd think this would be good, specially if it helps to collaborate smoothly, there are many open problems that need to be solved, the more we can collaborate (if done correctly), the more problems we will be able to solve in less time, in theory.

“We are masters of the unsaid words, but slaves of those we let slip out.” ― Winston Churchill.

  • 14
    'think about it twice before you share your next idea' I disagree. Share your ideas with your students, the chances that they'll 'steal' them from you are practically nonexistent while they might be able to help you test them.
    – Cape Code
    Commented Apr 2, 2014 at 18:06
  • 1
    I agree with Jigg. Although you're completely right that's it's too late at this point, the fact that someone might think there is a competition or race between you and your own student is a sign of deep dysfunction.
    – mako
    Commented Apr 3, 2014 at 0:15
  • @Jigg 1) you are taking the sentence out of context, please don't do that. 2) that's a very naïve POV, you overlook that people may be a burden. Sharing ideas is always good in theory but it may not be so good in practice. It's well known in software projects that more people involved means more communication channels between those people (they increase in a factorial fashion), more people may imply a delay in delivering the software. Sharing an idea and answering an indeterminate and potentially infinite number of questions about it may take longer than simply implementing and publishing it.
    – Trylks
    Commented Apr 3, 2014 at 10:18
  • @BenjaminMakoHill same answer as for Jigg.
    – Trylks
    Commented Apr 3, 2014 at 10:19
  • 1
    I agree with "think twice about sharing an idea". In the circumstance of the question asked because if/as the supervisor is extremely possessive, (s)he should at least take care that this is not lived on the back of the students. But more fundamentally also: if the supervisor does not think enough about the idea to recognize a good idea (which IMHO should be the case with an idea given to a student for a thesis), I doubt that students could rely on the supervisor not handing out bad ideas. Commented Apr 3, 2014 at 18:53

You must log in to answer this question.

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