I am a physics PhD student, currently in my fourth year, in a US university. During the last couple of years, I have been experiencing a decrease on the respect I have for my advisor, for a series of reasons. Besides all the stuff related to his role as a teacher in our institution (mostly his lack of effort to make students learn), there are several reasons related to his role as my advisor.

The Problems

He has a very hard time to explain his ideas clearly, and even harder time showing (I mean actually writing down on a blackboard) mathematical relationships containing the information that he's very verbose about. This makes me feel that his knowledge about the details of the topics he works on is somewhat limited. He doesn't offer much structured guidance as advisor. He jumps between quite different areas in short time scales, pursuing only hot topics that (in my opinion) could maximise the number of citations received per unit time spent on preparing a paper, while he's not an expert at all in many of those topics. The depth of our papers is thus not great, although the yield is (we've published ten papers).

When I mention that I'm not happy with my understanding about some part of a paper, or about the confidence I have on the claims we make, I usually receive a dissuasive answer suggesting me to look at it in more depth after the paper is published, because it's important to send the paper as soon as possible.

When I mention that my research work seems unstructured and it feels hard to think about how to put everything together for a thesis, he seems not to see that as a valid point. Indeed, I often receive a "this goes in the thesis" when we're starting to work on a new paper.

What I consider to be the worst is that we have very different ideas of what academic ethics means. His approach to our projects is like "we need to obtain a result X for the paper to be relevant, so let's try to take only data that is favourable, let's only explore the model to the range of applications that could give favourable results, and let's not mention clearly if we obtain an unfavourable result".

Of course, I am a co-author of those papers too, so I am responsible for it as well. My own (soft and not too confident) personality doesn't help in going against his will. He's quite pushy and not often open to changing his mind. He speaks with grandiosity, but usually doesn't back up his words with technical details. I am not in the best mental condition, so many times I just give up after my opinions or suggestions are dismissed a couple of times.

Facing the second half of my fourth year, I'm trying to decide how to continue. There is a topic I am interested in, but I have lost faith in the ability of my advisor to guide me through it. I have been studying, and I already have experience with many of the concepts, and the tools (mostly simulations and statistics for analysis) that I would need to actually do some interesting work. I have mentioned three times my interest on focusing on this project as my dissertation work, instead of collecting all the previous papers and making a dissertation out of them. He seemed to understand and even respect my idea when we talked about it, but his actions haven't changed. He keeps pushing me into new projects and not caring about my desires. I find it hard to go to campus and meet him, and I try to avoid it unless I have to teach my own classes. I have very negative feelings towards him these days, and I feel very disgusted by the idea of doing my dissertation in his style. I think if I did that, I would just not have any respect for myself or my work, and would end up getting out of academia.


I would appreciate some advice on how to deal with the situation, or comments about similar situations you have gone through.

The specific questions I want to ask are:

  • To which extent am I supposed to submit to his will in terms of what to work on for a dissertation?

  • How feasible is it for me to direct myself on this specific topic I want to work on? What alternatives could I use, such as contacting other researchers who could provide some guidance?

  • Do I make a dissertation with my advisor without feeling good about it, or do I make it without my advisor and feeling better about it?

  • 6
    At last, doing a PhD is a mental and psychological stress for most people. Consider getting some professional help -in my case psychotherapy was helpful. And publish or perish is quite sadly a world-wide reality. Academic work is very competitive, and somehow a solitary activity. Keep your PhD advisor, he deserves your respect but as every human is not perfect. Commented Mar 12, 2020 at 9:31
  • 30
    @BasileStarynkevitch, Are you oblivious to the OP's question, how on earth this supervisor is a good one, it doesnot make any sense if your advisor is oblivius and ignorant, I am shocked there are people like you think this is a good one. The OP can only push to finish since you invested four years, or find co-supervisor, or swallow it to finish.
    – user116038
    Commented Mar 12, 2020 at 10:49
  • 4
    How much longer to you still have before completion?
    – Buffy
    Commented Mar 12, 2020 at 11:13
  • 14
    "we've published 10 papers" It could be much worse then. Stick with your advisor, graduate and move on! I think it would be crazy to abandon your advisor at this point if you've already published ten papers together.
    – littleO
    Commented Mar 13, 2020 at 0:26
  • 4
    @Erik Stockholm Syndrome: since the one person claiming to that this must be a good advisor is one who had a similar advisor, that might be the reason he does so.
    – Jasper
    Commented Mar 14, 2020 at 17:59

5 Answers 5


Unless your field has a very high publication rate or your publications so far are mostly part of bigger collaborations (neither of which seems to be the case), you seem more than ready to defend your PhD. And this is what I recommend to do: Finish your PhD as soon as possible without getting into disputes with your advisor.

You seem to overly value your dissertation. While I totally get and it honours you that you want to deliver a respectable masterpiece here that is your and only your thing, this would be pearls before swine. Unless your field is atypical in this respect, your dissertation will be read by your supervisor and some co-examiners and that’s it. What a PhD certifies is the ability to perform and publish research. This already happened in your case and your papers already demonstrate this; the dissertation is just the formal finish. Also see: What's the point of PhD theses if nobody reads them? Moreover, you will hardly get any better at this with your current supervisor.

Pragmatically speaking, you are ready to defend when your advisor considers you ready: You cannot defend without the approval of your advisor; and when your advisor approves, you will pass the defence.

Your ideas for a new project are a great starting point for a postdoc. Contact group leaders that have expertise in this topics and ask whether they have a postdoc position for you or would be supporting you in writing a grant to get one. From your description, I would expect that your supervisor would even be supporting such plans, but of course you know him better.

As far as I can see, you have two choices:

  • Spend a few more annoying months while quickly finishing your PhD (your advisor’s way) and then have a hopefully great time as postdoc doing your own project while learning from a better advisor.

  • Spend one more year (or similar) trying to turn your pet project into a thesis without proper guidance, probably missing some of the opportunities it provides, and learning far less in the process that you potentially can.

  • 10
    (+1) For things like "You seem to overly value your dissertation". This was my thought also. The OP needs to get through the degree program with the view that, once done, he/she can then work on things he/she wants to work on, in the manner he/she wishes to. I remember when in high school my main thought was to get through all the stuff I didn't like so that I could go to college and study more interesting stuff (math) in an actual classroom rather than always out of books on my own. Of course, while things were better, I then yearned for graduate school when I could study just math . . . Commented Mar 12, 2020 at 13:39
  • 6
    This answer is good, but it doesn't address the obvious moral issues OP is facing. We all need to address more honestly the fact that some supervisors disregard scientific ethics and that this can be emotionally very difficult for people who have strong scientific values.
    – PejoPhylo
    Commented Mar 13, 2020 at 12:38

There are institutional rules when it comes to a PhD, which (almost) always includes that you can't get a PhD without an advisor. In that case, your second "option" (finish without an advisor) just does not exist.

You could get a different advisor, but since you are in your fourth year, you should be pretty much done by now. In that case, changing advisors could become problematic and you should consider the option of just finishing, and focus on getting a more rewarding post-doc position.

You could keep your current advisor and add a co-advisor. This depends on the formal rules in your institution and the relationship between the advisor and co-advisor (you really really really do not want to end up in the middle of a conflict between two advisors). But if the rules allow it, they get along well, and you value the input of the co-advisor, then this could work.

  • 4
    Actually, in the US, a fourth year doctoral student might be far from done. Some won't have started serious research yet, having spent the time passing comprehensives. There may be several years yet to go. The OP should clarify. This is quite unlike much of Europe.
    – Buffy
    Commented Mar 12, 2020 at 11:12
  • 10
    @Buffy: Given that the asker has published ten papers already, I consider this unlikely.
    – Wrzlprmft
    Commented Mar 12, 2020 at 11:21
  • @Buffy: you man that getting a PhD may be a matter of 8-10 years (4 + several years) after your MSc? That's indeed an awfully long time by European standards (where the hope is 3 years, the reality more like 4, 4.5)
    – WoJ
    Commented Mar 14, 2020 at 13:21
  • 1
    No, not after the MS. But after a BS. Usually 6 or 7 years now. In the US, most students enter doctoral studies with a BS so it takes a while. Moreover the BS is a generalist degree with only some specialization. At one time it was possible to do it in 4, but I don't think that is true anymore. But with an MS it depends on how much prep is required for comprehensive/qualifying exams plus the research. And usually some time for serving as a TA. But more than 8 is also a bit unusual here.
    – Buffy
    Commented Mar 14, 2020 at 14:46

While I generally agree with the advice given by the answers here (just finish the damn PhD and then go on doing something else, possibly working with someone who you think is thinking more like you), I think the answers here are sweet-talking an often overlooked, very problematic and sad fact in modern academia: Some big-shot PIs have an obviously unsound (scientific) ethic framework and mainly care about themselves. To me, it seems like you are a person that values scientific integrity very much and you have found such a PI. You're now having troubles coming to terms with the fact that your values (integrity, curiosity, excellence, and above all, an honest look at the world and the data) are not being mirrored by your boss. I've been in a similar situation and I would have greatly appreciated if just one damn person had the honesty to simply tell me that this can happen, that it's not your fault and that there's no point in dwelling over it. You obviously have three choices: (1) Quit and leave academia (2) Quit and restart/continue your work with another supervisor. (3) Stick with it and make the best out of it. Pragmatically, I think you should stick with (3) and then go on doing something you really like. Don't overthink it, you're almost there.


How to complete PhD?

Work on that extra idea/topic you want to finish up your Ph.D. with independently, or together/with the help of someone who's not your advisor.

This is:

  1. Allowed.
  2. Legitimate.
  3. Often looked favorably upon.

That's not to say your advisor will necessarily like it - he might and he might not - but you can always try and present it as the result of independent discussion and being a joint initiative of the other academic you've talked to; and argue that it's important for you to also have experience doing more independent work.


Stop chasing topics and simply put down everything into thesis. Your advisor is very likely could lose interest in you too, so in 1-to-1 situations you will be the losing side. Because you need him more then he needs you. Push for thesis faster to get to your Postdoc. Then you will be able to switch to the new subjects your advisor is not able to convey for you. We all are mortals and your advisor is too, so he is unable to know everything about everytopic.

Again, nobody cares why you failed your thesis. So you should not also care about whatever obstacle you got there. If it is your morals or "dignity", step over it and break yourself. Morality exists only for winners, not for losers. You will have plenty of time to discuss your previous papers and your mistakes in those papers when you finish your thesis and start working on Postdoc research.

Check out how programming development goes on. Same as in open source. If you are afraid to push new commits (pieces of code) to upstream, nobody will care whatever moral reasons you got. If you don't push things, you failed. Make mistake, then go back to fixing it, it will be much better then if you would simply do nothing. Development of scientific idea goes in the same style.

PS. Following to remark by @nick012000: Then you will find yourself in infinite loop of overthinking about whether or not some issue should be pursued, you are not a scientist and you become a philosopher or publicist. Scientist should not overthink and hence should not inflict self-censorship on himself. There is no such thing as "bounded scientist". You are either bounded person or you are scientist. And even worse thing is self-bounded scientist, who just overthink every time he got really genius idea and so he stuck in endless cycle of genius ideas which are forever lying on the shelf and never used. I know it because I have it at some degree and I fight my "overthinking problem" every day.

PS2. As long as existed word science, it was called unethical. Up to the point of making me think that this words are more intimately related then it could seem to casual observer. Of cause I am against barbarian methods in science. But think about it: simply looking inside human body, even if it is your own body, was considered a sin.

  • 5
    "If it is your morals or "dignity", step over it and break yourself. Morality exists only for winners, not for losers." Downvvoted for this statement alone. This is the sort of mentality that leads to bad science.
    – nick012000
    Commented Mar 15, 2020 at 9:11
  • @nick012000 you need seriously reread my answer, because it is more complex then your one-sided opinion
    – sanaris
    Commented Mar 15, 2020 at 18:35
  • 1
    Downvoted, both for the same reason as @nick012000 and for your response to his comment. Insisting on acting ethically/morally is not the same as self-censorship or overthinking.
    – JeffE
    Commented Mar 15, 2020 at 19:42
  • @JeffE You really need to read history books about how science was unethical for whole period of existence of word science itself. When looking inside human body was unethical and so on.
    – sanaris
    Commented Mar 15, 2020 at 20:17
  • 3
    Why should we continue doing an immoral thing, just because it has been done in such a way in the past? What kind of reasoning is this?
    – PejoPhylo
    Commented Mar 15, 2020 at 20:26

You must log in to answer this question.

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