14

First of all, I am a second-year PhD student in pure maths in Italy, in case some of this information is relevant.

My situation: around a month ago, my PhD advisor posed me a question (related to my previous research, I also had this question in mind) and told me we could work together since he had some ideas. Two weeks ago I found a full solution to the problem, quite interesting also and I told my advisor. He told me he also is thinking on some ideas for the problem but I think they won't be needed.

Question: Would it be OK to ask him if he can withdraw from the project since I think I can finish it by myself with my ideas? In case it's OK, any advice on how to tackle such a complicated discussion?

Extra info: I'm very happy with my advisor. He does not demand authorship for all papers, many papers of his former PhD students have been solo. My solution to the problem does not use any of his current ideas. I have the feeling that he is quite interested in the problem though.

8
  • 18
    Does posing the question represent an intellectual contribution? Was the question just a curiosity, or did it actually require years of experience in the field to come up with this question and with the belief that it can be answered? If so, posing the question is an intellectual contribution and requires co-authorship. Oct 16, 2022 at 3:10
  • 1
    @WolfgangBangerth well, the question is a natural follow-up to my previous problem so I don't think that is the case.
    – MathAuthor
    Oct 16, 2022 at 10:11
  • 5
    You'll just have to discuss it with them in person and see what they say is the short answer.
    – Tom
    Oct 16, 2022 at 11:50
  • 3
    Maybe you could see if his approaches would lead to some novel areas, and would be enough for a second paper. Then after a few weeks of collaborating on his ideas, you could propose making it into 2 papers, 1 where you are the solo author, and the second paper where you are collaborating?
    – Issel
    Oct 17, 2022 at 2:24
  • 3
    What is your goal here? Perhaps it depends on field, but to me, a paper on your CV with yourself as first author and your advisor as co-author would have equal weight as a sole author paper. Oct 17, 2022 at 20:23

10 Answers 10

40

He told me he also is thinking on some ideas for the problem but I think they won't be needed.

You may be jumping the gun here a bit. Yes, perhaps you can finish the paper on your own and have a solo publication. It may also be that your advisor's ideas will result in a better paper or open avenues for further research that you might not notice on your own. It rarely happens that solving a problem is an open and shut case with no further avenues to build on the solution, or to look at it from different perspectives. Preemptively asking him if he can withdraw from the project without hearing out what his ideas on how to approach the problem are therefore seems premature to me.

Certainly if he has nothing crucial to add to your solution to the problem, it makes perfect sense to ask him whether you can turn your solution into a solo paper.

On the other hand, asking him to withdraw from the project so that a situation where he has something crucial to add to your solution doesn't have a chance to arise (and mess up your would-be solo publication) is a different story.

A solo publication would certainly be valuable, but also don't underestimate the value of a lively on-going collaboration with your advisor which can lead you to places that you might not foresee yourself. Especially if you're not in a hurry to amass as many solo publications as you can.

The way you phrase your question makes it sound like you want to say: let's not collaborate on this topic any further so that I can get a solo paper. I would consider thinking of it instead as: let's continue working on this topic together, and in the meantime I suggest that I write up and submit my solution as a solo paper.

4
  • 7
    @MathAuthor Well, you can tell him exactly that. It's not like you have to either ask him to stop this collaboration or commit to writing a paper with him. You can just say that you think this is a good opportunity for you to write a solo paper of your own and ask him whether he can agree to keep that option on the table as you continue discussing the topic. A transparent discussion about authorship should take place, but that doesn't mean that the authorship issue needs to be set in stone right here and now. Oct 15, 2022 at 23:03
  • 2
    Bingo. I think there's more to be gained (supervisor's experience in drafting and getting the paper published in a decent journal is surely worth co-author status) here than what's losable (the trust of someone who now realizes he has a student of capability on his hands) from all this. My advice: bring him the part where you broke the back of the problem - then let him tidy it up for publishing. You've got more important things to do with your own doctoral work. I hope your professor doesn't make a habit of donkey-working stuff to you after this "success".
    – Trunk
    Oct 16, 2022 at 13:00
  • To further emphasize the first paragraph: the student doesn't need to worry about the authorship question at all until the paper is written. Maybe the advisor will be needed for help along the way, maybe not. But it's a question for the future when most of the paper is done. Start writing and forget all else for now.
    – Jerome
    Oct 18, 2022 at 5:11
  • 1
    Note that in mathematics it is quite common that an advisor will collaborate with their PhD student on some paper and the paper will still be published with the student as the sole author. So definitely continue collaborating with your advisor on the paper and then ask them whether they want to be an author. They may answer either way and I would recommend following their wishes.
    – quarague
    Oct 18, 2022 at 13:28
12

In an ethical and humane world, yes, you can ask and he will say OK. But I don't know if that is your world. You have to make an analysis of what his reaction might be.

But if you approach it right it will help. "Dr. M, I'm confident that I can finish this on my own and get a sole author publication. It would do me a lot of good. What do you think?"

That is sort of a US approach, however, though normally US doctoral advisors in math don't assume authorship on their students dissertation work. From what you say about him, I'd guess he will go along.

7
  • 1
    I also wonder which one is my world, although my advisor has always shown to be both ethical and humane. I wouldn't like to strain our relationship though and I'm under the impression that in general it is not an OK thing to ask for.
    – MathAuthor
    Oct 15, 2022 at 22:06
  • 5
    @MathAuthor can you explain what gives you that impression?
    – Dan Romik
    Oct 15, 2022 at 22:12
  • 3
    @DanRomik My precise impression is that once you start collaborating you cannot back out and say I want to do it by myself. I don't have much experience though so I might be mistaken of course
    – MathAuthor
    Oct 15, 2022 at 22:15
  • 2
    @MathAuthor there’s nothing about the description in your question that suggests that you and your advisor have “started collaborating”. He is your advisor and posed you a research problem, then sent you off to work on it - that’s what advisors do, which normally does not imply a collaboration. Was there anything he said that implied that he meant that this would be a collaboration?
    – Dan Romik
    Oct 15, 2022 at 22:25
  • 3
    @MathAuthor okay, maybe there’s something to what you’re saying but it’s a bit too vague to be able to say anything definite about whether “there is a collaboration”. Most likely it’s one of those situations when there isn’t a well-defined binary answer to the question, just a set of vague unspoken assumptions. I honestly don’t see a reason not to simply discuss it with your advisor like the two adults that you are.
    – Dan Romik
    Oct 15, 2022 at 22:47
11

So you advisor gave you an idea and you want to publish a solo article based on that idea?

Okay, I am not a mathematician and not Italian. Every research field is different, and every country is different. Maybe this is totally fine for mathematics in Italy. Maybe your professor will say, "Yeah, sure, go for it!"

But that would definitely not go in Japan, at least in most research fields. Everyone would be appalled by the fact that this came to your mind. A solo article is when you come up with an idea yourself, do all the work yourself, and write the manuscript yourself. And if you don't want to find yourself in the grey area of ethics, you shouldn't get any help or advice whatsoever in relation to your article. Ideally, you don't even let anyone know until the paper is published - otherwise people may give you advice and you won't be able to unhear it. And it's really not nice if you publish a solo article while being funded from a grant secured by someone else.

(Update: a comment below says that in mathematics, merely asking a question without contributing to its solution does not normally justify authorship, and obtaining funding doesn't either. Lucky you mathematicians. But read on. There's a totally different thing to consider.)

Another aspect is that even if you come up with an idea yourself, do all the work yourself, write the manuscript yourself, are being paid not from anyone's grant, and are thus perfectly positioned from the ethics standpoint to submit a solo article, you might later find yourself truly regretting publishing it. It happened to one of my colleagues. He wasn't experienced enough. He wishes he could unpublish it.

Look, you are a PhD student and don't have a lot of experience. Are you really sure you can write a paper you won't later regret publishing? Are you sure you won't overlook a serious issue?

There are a multitude of ways to screw up. If you make an error or a wrong statement in your paper, someone may publish a comment, and it will be a huge blow. If you write your paper poorly in terms of the language, structure, clarity, etc., this will seriously affect how others see you. If you publish your paper in a subpar journal, others will think it's what you and your work is worth. Your first papers set the expectations of others - it's called the anchoring effect. if you submit your paper to a high-profile journal and the referees say that your work falls very short of meeting the acceptance criteria, the editor may well have prejudice against you next time - that is, when you submit another paper to that journal. Remember, the editor chooses the referees, so you don't want the editor to have prejudice against you. And this might make it almost impossible for you to publish anything in that journal later. Also, if you publish an article and no one cites it, it won't look good to anyone looking at the citation stats of your papers - and that's what many referees do.

If you have your professor as a co-author, he'll ensure your paper won't have any serious issues. It's about his or her reputation. And having your professor as a co-author is likely to have a positive effect on the recognition of the paper by the referees and readers as well as on its citations.

My advice is: think twice whether you really want to proceed with this on your own and publish a solo paper. That's the very first thing to decide on. Make sure to take a well-thought decision. If you do decide to talk with your professor, do your best and utmost to ensure he or she won't feel bad about you. One of the key factors is whether you are a paid a salary or a stipend from a grant secured by your professor. If that's the case, he might feel as if you were trying to steal his money. Remember, you depend on your professor. The relationship is extremely important. Focus on that.

Consider this:

Hey boss, I've noticed some of your PhD students published solo papers. Is there any chance I can do the same and you'll be fine with it? If so, how?

That's a good start, and his reaction will show whether it's a good idea to talk about that particular paper.

10
  • 5
    This does not really apply to mathematics, where merely asking a question without contributing to its solution does not normally justify authorship (and obtaining funding certainly doesn't either). Oct 16, 2022 at 7:04
  • 5
    Whatever about technicalities of authorship, I think that, ethically and amorally, Mitsuko has it right. It's better this way and worse for your relationship with supervisor for your to whoosh him out of the way in your drive to establish a reputation. Academic patricide is not a good habit to start at your age.
    – Trunk
    Oct 16, 2022 at 13:08
  • 7
    This is a well written answer, but to add to what @EspeciallyLime said, two other reasons why almost everything you said does not apply to mathematicians are: 1. In mathematics it is very common for graduate students to publish solo papers. They are perfectly capable of producing good papers written to the standards of the discipline (and usually receive detailed advice on the writing from their advisors, without the advisors expecting coauthorship in return). …
    – Dan Romik
    Oct 16, 2022 at 14:56
  • 7
    … 2. In mathematics, being paid by someone’s grant does not make the PI of the grant feel entitled to be a coauthor of your papers. Again, it is very common for students and postdocs to publish without their PI, even if they are paid from a grant.
    – Dan Romik
    Oct 16, 2022 at 14:57
  • 4
    This certainly gives some interesting insight into the academic culture in Japan (not sure in which discipline) but is very very different than what I am used to -- and I'm not really sure it goes well with my personal preferences either. I am most surprised that you think any sort of help or suggestion deserves authorship - it is not uncommon in my department to ask a colleague for a read and provide some feedback when submitting a high-stakes manuscript, where the expectation for what they get in return ranges from a coffee to an acknowledgement. (This is in a computer science department)
    – penelope
    Oct 17, 2022 at 9:39
7

Would it be OK to ask him if he can withdraw from the project since I think I can finish it by myself with my ideas?

Yes, it’s okay to ask.

In case it's OK, any advice on how to tackle such a complicated discussion?

Your assumption that it’s a complicated discussion is part of the problem. It isn’t. Your advisor may be disappointed by your wish, and might even push back on it or outright say no — we cannot predict the outcome. But he is a grown-up and does not need you to tiptoe around his feelings. Your question is both reasonable and straightforward, so just go ahead and bring it up, and you’ll be fine. Good luck!

3
  • Thanks for your answer! I wonder whether the fact that I find it difficult and you don't might be due to some cultural difference. If that is the case, unfortunately my advisor is also italian.
    – MathAuthor
    Oct 15, 2022 at 22:30
  • 7
    @MathAuthor “unfortunately also Italian”? lol. The Italians I know are all super nice and reasonable people and make me think that it’s unfortunate that I am not Italian. (Admittedly, I have heard rumors that there are some non-nice Italians out there, but I have yet to meet such an exotic specimen.) Well, obviously only you know your advisor and your culture, so you’ll have to filter our answers through that knowledge. But, for what it’s worth, there isn’t anything about the description you’ve provided, or about my knowledge of Italian culture, that makes me reconsider my answer.
    – Dan Romik
    Oct 15, 2022 at 22:44
  • hahaha the "unfortunately" was conditional on there being cultural differences making the situation more thorny for italians.
    – MathAuthor
    Oct 15, 2022 at 22:46
6

My advice would be to ask not if you can publish this solo, but whether you should.

Assuming a reasonable supervisor open to solo papers from their students, you have two options here:

  1. You carry on this project on your own and gain a single-author publication
  2. You and your supervisor work together on this problem, developing a potentially richer and more interesting paper in the process

The difference between the two is your supervisor's contribution, and it's up to them to decide how much they are willing to commit. Therefore, you should ask them whether they wish to spend time and effort on the current work (option 2), or not (option 1). This doesn't rule out collaborating in the future on the same topic, of course.

Note that it's really the same question but with a subtle difference in focus - instead of "I'd like you to not contribute so I can have a single-author paper", it's "I'm happy to go ahead and publish this single-author unless you tell me you want to contribute". However, presenting it as two options is less confrontational, makes you look more collaborative, and if your supervisor isn't actually happy to let you publish this solo, by picking option 2 they have implicitly committed to doing active work on it (which honestly may be more useful for you at this stage than a solo paper).

Disclaimer: I'm from a field where solo papers are uncommon and typically reserved for reviews and minor work. My impression, but it's coloured by my field, is that the value of solo papers as proof of independence can be overestimated, especially by very early career researchers - I'd be more impressed by a collaborative paper that does not include your supervisor (for example, together with other PhD students).

2
  • Thanks for you answer @Ottie ! The only problem is that we are collaborating by default since this was raised before I obtained a full solution.
    – MathAuthor
    Oct 16, 2022 at 14:54
  • 1
    @MathAuthor that's exactly the point: by explicitly presenting the problem as a choice between options 1 and 2, both of which are a good outcome to you, you are trying to rule out the undesirable option 3 (you do all the work and still get your supervisor's name on the paper).
    – Ottie
    Oct 16, 2022 at 14:57
4

My advice would be to be generous.

In the long view of life, it doesn't really matter if your paper has 1 or 2 authors, particularly if that author is in a position to contribute to the work, either now, or at the time of manuscript revision, or both. So, by erring on the side of generosity, you have much to gain (including learning new things from your advisor).

On the other hand, if you chose to go solo, that's fine too, but this attitude can quickly become about "I/me/mine" rather than the far more sublime attitude of the service of math/science. There is also the very real possibility that over time people will be reluctant to discuss ideas freely with you, because there is the possibility of an impression that if you cross the "least publishable unit" boundary by yourself, they get cut out.

7
  • 1
    Thanks for you answer! It will not matter if my paper has 1 or 2 authors in the long view of life. But it will definitely matter for getting postdoc positions and similar, and that's what I should worry now I think.
    – MathAuthor
    Oct 16, 2022 at 10:14
  • @MathAuthor By the time you get to sit down before any interview panel you'll have a few sole authored papers too. But do you think you would get even a co-authorship if the professor didn't implicitly acknowledge a major contribution by you to the solution ? Also, seasoned academic hiring committees can reliably adjudge the extent of your contribution from the nature of the solution in this paper (the approach, the choice of support theorems, the flexibility, etc) and will make up their own minds on it. They ain't stupid: they went through the PhD process themselves.
    – Trunk
    Oct 16, 2022 at 13:15
  • 1
    @Trunk Why wouldn't I get a coauthorship if the idea is mine and I can carry it out by myself? Also, if I keep doing like this how I am going to get sole authored papers? Finally, if the result is good (hopefully) but the coauthor is a big name, then by default it will be assumed that it is mostly his contribution.
    – MathAuthor
    Oct 16, 2022 at 13:31
  • 4
    I am not saying you can't do that. I am saying you shouldn't seek sole authorship (because it's "bad form") so that you grab all the credit for solving a question presented to you by a professor who is your supervisor. Your final thesis paper will be your first-name paper at least. Any questions occurring along the way that you solve will be your papers alone. Your fellowship papers afterwards likewise. Please read my previous comment again. You are a young man in a hurry with sogni d'oro . . . Your professor is good to you. You try to be good to him.
    – Trunk
    Oct 16, 2022 at 13:45
  • 4
    @MathAuthor your question seems to be drawing a lot of advice from people who work on different disciplines or aren’t even from academia. Without implying anything bad about those people, who may be very smart and generally people who give sensible advice, this is something for you to be mindful of. Mathematics has a very different publishing culture than a lot of other disciplines, in ways that aren’t widely understood by non-mathematicians.
    – Dan Romik
    Oct 16, 2022 at 15:11
4

I refer you to lots of good answers here that advise whether or not you should do this, but if you are set on publishing solo, I just have one piece of advice: write a manuscript first, then have that discussion. If you put a fully written paper in front of someone, they are more likely to see that they have not contributed to it and might decline authorship on their own (if they are reasonable).

I did a similar thing. I sent my postdoc advisor a manuscript (though it was completely my own idea). She gave me useful feedback on the manuscript, then I asked her if she wanted to be a co-author. She said that she didn't think she had contributed enough, and I ended up publishing it solo, only mentioning her in the acknowledgment. That paper landed me an academic position. Don't underestimate the power of single-author papers (note: I'm not a mathematician, so this may not be as important in your field).

1

Might be useful to wait for review to see if their contribution is necessary in the opinion of refs/ed. If after review their contributions aren’t necessary then you have a solo paper, else you have helped your advisor hook into the topic.

If you accept their contribution and it makes publication that will be something they won’t forget. If the contribution is thought superfluous in review then you have a solo paper.

I don’t really know the advantages one way or the other, just consider your long term goals and know that you might find yourself on the other side of the symbiosis in the future.

2
  • 1
    In mathematics, as far as I know, the paper is sent as a single unit to review. The referee does not discriminate between authors.
    – MathAuthor
    Oct 16, 2022 at 10:18
  • 2
    Professors are more sensitive when the matter in dispute is going for publication. How is it going to look if OP acknowledges that his professor brought the question to his attention but he alone found a solution ? I have a feeling from the professor's demurring-ish response that he's been working on an approach of his own and will want at least a share of the authoring laurels. It would be very insensitive - and professionally unwise, maybe relationally harmful - to use a third party to horn out the professor from this trough. As some others suggested, it is better to be generous here.
    – Trunk
    Oct 16, 2022 at 13:25
1

Reminds me of a paper I wrote in grad school, it was a term paper, and I did all the work upfront. All of the ideas were mine from start to end. I wrote the paper up. I made all of the work and effort. Turned it in, and was advised via a note at the top of the paper, that with a bit of polishing, it could be made publishable. I took the paper to my advisor with the notes and all. He agreed with the other professor's assessment and he helped me add some shine to the paper, but otherwise didn't contribute much to the substance of the ideas I put forward. We submitted it as coauthors.

Why?

Without his contributions to the actual writing to get it to an acceptable state for a journal, the damned thing wouldn't have been published, or at least if I did manage to get it published, it would have been published in a lower-tiered journal. I learned a lot about the publishing business in academia by allowing my advisor to help me and share credit along the way. Lessons that were invaluable later on.

I would suggest being generous on this one and learning from someone that has been in the trenches for some time on how to get the publishing done. You might actually learn something different than you think about the business of a career in academia that you wouldn't learn in any other way.

0

On all my PhD papers I included my supervisors as authors, not because they contributed but because they stood behind me, guided me and I benefited from fruitful discussions with them.

If you are first author then everyone who reads your paper will know that you are the primary contributor and you will received due recognition. Bear in mind you will awarded your PhD on the basis of the Supervisor acknowledging your contribution - not because you are sole author on a particular paper. Take the time to read some papers by some of our great scientists and you will see that none of them were afraid to recognise the contribution of their supervisors.

Have confidence. Asking to be a sole author will prove nothing and your risk being seen as churlish.

2
  • 1
    This answer does not apply to the OP's field, which is pure maths. Authors are ordered alphabetically in pure maths, so there is no concept of a "first author". Courtesy authorship for PhD advisors (as suggested in your first paragraph) is frowned upon in this field. Regarding your last paragraph: it's not about "proving something", it's about having a single authored paper. (For instance, some members of hiring committees tend to be more skeptical towards candidates with no or only very few single authored papers). Oct 20, 2022 at 16:43
  • Thanks for picking me up on the "alphabetical listing" in mathematics. I was not aware of that, but I don't accept that during your PhD work you do not benefit from your Supervisor, but I suppose "acknowledgements" may be a way out of the dilema.
    – Alexander
    Oct 21, 2022 at 0:56

You must log in to answer this question.

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