Mfcttrf

I was asked by this (tenure-track) assistant professor X, with whom I'm in very friendly terms, to collaborate on a project together. X and I work in the same broad area of pure math (topology) but our subfields of expertise differ considerably. At the beginning I was very excited but I've now realized that X is at best extremely inefficient and at worst incompetent. We have been working on the project for some months now and 90% of the things that X says are wrong for very elementary reasons and the remaining 10% are borderline tautological. I'm getting very frustrated because I spend more time correcting his "proofs" (of very elementary claims, not even high level stuff!) than doing actual research. So far all we have is stuff I had proved before starting the collaboration with him which forms the base of the project we are working on. On the other hand he seems to be very keen on us working on a project together and as I've said we're friends. I don't know how to deal with this.

Should I give him some more time to actually contribute in some way to our project or should I tell him that I no longer want to collaborate with him? In the latter case would it be ethical if I worked on my stuff that I shared with him? He literally hasn't added any value whatsoever to it so far.

This may be too personal for a good answer here, but let me make a couple of suggestions. The big issue, I think, is how important to you is the personal relationship. If it is important enough, then you might want to just help him with his education rather than the collaboration, per se. If you are colleagues at the same institution, it might even be necessary to do something like that for your own protection.

One way would be to found a research group with a few faculty and some graduate students where the goal is just as much about pushing people up the scale as it is the actual research. If your colleague agrees to this, he might improve his own grounding in the subject.

But if you are willing to let the personal relationship go, then you can also let the collaboration go as well. It might be difficult to step away from it, of course.

But you don't really say what you mean by "collaboration". If you mean "I'll do this and you do that", then it probably isn't going to work - ever. But if you work closely together (one desk) or in the research group idea, then it might work better, as long as you don't feel used with "giving" joint authorship for things you think are really yours. But I don't really consider splitting the work as collaboration.

One way to split up is to ask your colleague how he thinks it is working out. If you think you could go faster alone, then it isn't wrong, just uncomfortable, to say so. Egos can be bruised, of course.

Ethics would only enter in if you took work of his and didn't provide the appropriate credit, whatever that might be.

I guess I should also note that in many collaborations, someone is dominant. For example, many of the people who are Erdős-2 likely felt like the junior partner. But one that is uniformly unbalanced can be frustrating as you say.

Not all attempted mathematical research succeeds. Sometimes you work on something for hours, days, weeks...and realize at the end you have too little to show for it, or that you made a crucial error that ruins everything. This applies in particular to collaborations: just because you get together with someone (or someones) and propose to work on X, it doesn't mean that you will end up writing a paper on X or even making any progress on X. (This has certainly happened to me.)

Hows is this relevant to your present situation? Well, on the one hand you seem worried about disappointing your friend, which I certainly understand, but actually you delivered on the deal: he wanted to collaborate with you, and so you have. Surely you didn't agree to collaborate until you achieved some specific goal, no matter what, right? So I don't think you should feel too bad about breaking off the collaboration: all collaborations are parted sooner or later, and most by mathematicians who remain active on other projects. On the other hand, you use the word incompetent, which seems harsh to me. Most mathematicians lack the competence to solve (or even to superficially understand!) most mathematical problems, of course. Just because your friend can't make correct arguments on this topic doesn't mean they are not competent in something else. Also collaboration gives you a chance to see someone at their worst as well as their best: some mathematicians make a lot of mistakes that no one ever hears about, but people hear about their remarkable successes. (We generally respect and admire these mathematicians.)

If you are describing your friend as "incompetent," it sure sounds to me like you want to be out of the collaboration, so I would suggest that you do so, by which I mean ending the research sessions with your friend. In my opinion the "classy move" would then be to write up what you have so far and ask your friend how they wish to proceed. They may well decline to be an author on the paper. Or maybe they will accept to be an author, either because they think they brought more to the table than you do or because professional exigencies make them feel that they cannot turn down an authorship on something they spent this much time working on. This kind of thing used to bother me. It doesn't really anymore. I don't think your friend is going to get tenure "unfairly" because of this one project with you.

Of course once you have decided what to do about your joint work and that you will not collaborate anymore, you are completely free to do further work as you see fit.