Richard Hamming, in one of his talks about research, said that scientists have to put up with stress, and he talked about his ulcers (and he was amongst other things a mathematician). I thought mathematical research should be mostly thinking / living a life of the mind and, thus, while involved, it shouldn't be anxiety-producing.

Is mathematical research typically stressful? Are there good researchers who don't have stress?

  • 10
    I find it less stressful to focus on understanding rather than results, but one needs to get some results too. Often, results come naturally as a consequence of understanding, but that's not always the case. No matter where one's focus is, there's always the possibility that something that ought to work refuses to work, and the search for corrections, improvements, or even completely new approaches can certainly be stressful. Aug 9, 2016 at 1:48
  • 10
    Uncertainty is stressful. Research is uncertainty (most of the time). Also see How should I deal with discouragement as a graduate student?
    – ff524
    Aug 9, 2016 at 2:16
  • 12
    Doing mathematics is itself not stressful for most mathematicians, I don't think. But your working environment (e.g., pressure on getting a grant, publish-or-perish culture, and review for tenure) can make it stressful in various ways. Sometimes, a referee's report can boil your blood, too. Your personality may also matter, e.g., pressure on proving big results. Math is fun. But living as a mathematician involves more than just math. Aug 9, 2016 at 2:17
  • 5
    @Insulin69: Yes, at Caltech and everywhere else. It seems like you might have an unrealistic idea of what academic life is like. Aug 9, 2016 at 16:31
  • 10
    @Insulin69: Whether or not Caltech professors are "obsessed" with math, they still just don't get to sit in an ivory tower and ponder math all day every day. They teach (though perhaps not as much as at your university), they apply for grants, they serve on committees, they supervise graduate students, and in the time that remains, they write and publish papers. And if they don't do all these things to the satisfaction of their colleagues and dean, they don't get to keep their job. Aug 9, 2016 at 18:55

5 Answers 5


I'm not a mathematician, but being an academic in a tenure-track position may be stressful itself if you're on a publish-or-perish program. I.e., either you generate (mathematical) results year to year in six or seven years or you get fired. The stress may not be in the work itself but in the built up angst thinking about how achieving each new result quickly is, and it can be so bad that sitting down to work on a new problem or even thinking about it can be stressful.

  • 11
    It can (unfortunately) be even more stressful than your answer indicates. It is not enough just to generate mathematical results at a steady pace. The results must be valuable enough to your superiors in order for them to keep you (and even more, to have hired you in the first place). That last part is murky enough to fry just about everyone's noodle at some point. (I upvoted, BTW.) Aug 9, 2016 at 17:17
  • I guess your answer does not address the literal question in the title, though it responds indirectly by suggesting that the question should really refer to a broader version? Aug 9, 2016 at 20:50
  • @paulgarrett, yeah that may be, though I'd like to get the feedback. I've never been a mathematical researcher, though I do have a couple of applied math papers. However, I didn't find producing them stressful. I didn't go to find Hamming's talk to find more context for my answer, but there are other embedded questions than the titular one which I think I'm allowed to try to answer. I don't think math research is, itself, stressful, but I do think that certain research careers can be, as reflected in my answer.
    – Bill Barth
    Aug 9, 2016 at 23:21
  • 2
    (I think you have no reason to worry about the general legitimacy of your answer...) But, possibly, "real math people", or, anyway, a certain segment of the population, does get very stressed ... maybe in not an entirely bad way... from their work: the stages of irresolution. As in my own answer, not sleeping is a pretty serious symptom of ... something... is stressful, and/but I can't help it. Not a sign of normality or equilibrium, no. And, no, I'd not encourage craziness as an avenue to progress. Dunno... Aug 9, 2016 at 23:25
  • 4
    @PeteL.Clark: "The results must be valuable enough to your superiors in order for them to keep you" You know, this is a problem that everyone in every other field of employment deals with as well....
    – Tom Church
    Aug 10, 2016 at 3:29

(I write from a U.S. perspective.) The stresses of the job of academic mathematician include many that have little to do with mathematics itself, but are due to various commodified forms of mathematics. I'll get back to the stresses of research itself at the end.

For example, there're the issues of getting a job after the PhD, getting a tenure-track job after that, and getting tenure after that. These are not research issues per se. Similarly, "getting grants" certainly helps travel to conferences to see other work and promote one's own (though this may be less critical nowadays with Internet), but/and is sometimes used as an "objective external" evaluation of one's research program, so that getting tenure in upper-tier places may be impossible if one does not have an external grant.

The game of "getting grants" is much as what a skeptic would imagine, apart from various degenerative "streamlined" aspects as time goes on. That is, one must propose a thing that one can approximately do, but it is not considered entirely ok to propose to do things one already secretly knows one can do, etc. And "do" has to mean "in a year or two, at most", or it won't help the final report on the grant, nor on the re-application. Also, nowadays, apparently NSF wants its "panels" to favor certain general directions, certain subjects, etc, rather than attempting to seriously evaluate all proposals, thus "steering" research insofar as funding affects it. In particular, it's harder to get grants if you're not doing an "approved" thing.

"Publication in peer-refereed journals" is somewhat less "coerced" than federal funding, "status" is very important for getting tenure, and journals will reject papers perceived as insufficiently high-status to match them. Status is the chief commodity journals possess, and squandering it would be a mistake. But, of course, not all topics of research have equal status. Trying to establish status for a particular topic, or spend one's personal status-currency to do so, is a game in itself.

The long history of mathematics, and the fact that (mostly) things don't "become wrong" in mathematics, mean that there is a looooong backstory, and much low-hanging fruit is gone. (I can see the substantial changes in the 40+ years I've been observing...) Lots of things have been done, and many of the ideas that pop into the heads of people have popped into others' heads before. Stress: how to think of a new thing, that is worthwhile, and, once observed, other people would wish they'd thought of it themselves... but they hadn't??? Sounds like there'd be a scarcity issue, and, in many ways, there is. People have to find "ecological" niches that will generate sufficient status in dept heads' and deans' worlds to get jobs, get tenure.

And, yes, for many of us there is stress in doing the work itself, but there is perverse pleasure in it, or at least no viable alternative. I might claim that it would be silly to try to make a living as an academic mathematician unless one did care about the details so much that confusion or frustration did provide significant stress. Maybe there are better and worse stresses... But, yes, I'm unhappy and bored if I'm not confused or provoked by some mathematical thing, and if I'm confused or provoked I don't sleep well, etc. Just great. :)

But, seriously, apart from the job/business-aspect stresses (which are often hard to overlook), I would indeed claim that there is genuine stress (mostly of a less venal sort) in doing serious mathematical research. Significant scholarship is a prerequisite for not reinventing wheels, and for not thinking that trivial exercises are "research".

Probably the conclusion is that unless the "positive" piquant stress of the confusion of research is sufficiently fun/gratifying, the ugly/bad stress of the business aspect is possibly expensive. But the job/business stresses are universal, apparently, simply manifest in different ways.

  • 1
    This is an excellent answer.
    – user58865
    Aug 10, 2016 at 0:50
  • 3
    +1 for The long history of mathematics, and the fact that (mostly) things don't "become wrong" in mathematics, mean that there is a looooong backstory, and much low-hanging fruit is gone. Aug 10, 2016 at 14:46
  • +\pi Excellent answer, indeed. Unfortunately, to some extent also a sad answer. The best work in math cannot be in a field planned by a committee. Jun 2, 2023 at 23:54

First of all, I think it is helpful to separate the stress and anxiety which arises as part of being a mathematician, and that which arises due to being an academic, the latter referring to all the stress surrounding the job, such as publishing, finding a permanent position, grant applications, dealing with students and supervisors and relationships and so on, none of which are unique to being a mathematician.

So I assume you're more interested in the former. You state that

I thought mathematical research should be mostly thinking/ living a life of the mind and thus while involved shouldn't be anxiety-producing...

But surely living a life of the mind, as you put it, opens the doorway to a unique kind of anxiety?

For me the notion of a low stress job is one where I have some concrete task or goal which I am able to carry out successfully each day, one that doesn't involve a great deal of effort, after which I am able go home in the evening and switch off.

Mathematics is anything but this. At a given point in time a researcher will probably have a set of vague goals, problems they want to solve or areas they want to get to know better. They might be in the process of writing up a new result, tidying up the details and watching it develop into something publishable. This is can be quite satisfying. But for the most part progress in mathematics is extremely painful, involving long periods of frustration where you might be going nowhere. This in turn often leads you to doubt not only your methods and approach but sometimes even your own ability and value as a mathematician. Moreover it is rarely easy to switch off. I often find myself 'relaxing' with family or friends, when in reality the whole time I'm unable to ignore the research problem which is persistently gnawing away at me.

Don't get me wrong, I love my profession and I wouldn't change it for the world. But like any creative pursuit it sucks you in, and while there are brief moments where you see something and feel elated, there is a lot of time where you're banging your head against a wall day after day.

Of course, anyone working in a vaguely scientific subject might have broadly similar experiences. But I think that these experiences are particularly acute the more towards the theoretical end of the spectrum you get, and maths is right there at the apex. When you're stuck there's typically little input from the real world to help you, no experiments you can run to test your approach, and once you're past graduate level often only a handful of people in the world who even understand your work.

As a result, it can feel like you're living alone in a big dark room with no idea where the light switch is, all the time worrying that actually there might not be a light switch after all, and that maybe you walked into the wrong room in the first place! So we have our fair share of stress and anxiety in mathematics. But to avoid slipping into self-pity as well I think I'll stop here!

  • 3
    :) Indeed, now that many of us can work at home, the perverse seduction of "doing chores" is very great! Achieving tangible, reachable goals within predictable time frames! :) Laundry, dishes, vacuuming, mowing, weeding... :) Aug 9, 2016 at 23:04
  • 1
    For me mowing is not achievable within a predictable time frame. It's even worse than maths.
    – Rhidian
    Aug 9, 2016 at 23:07
  • 3
    Ah! As a manifestation of some (to-me-useful) methodological ideas, a bit later in life I acquired a very large yard, so that any inefficiencies or sub-optimal ideas about mowing would have their failures violently demonstrated. :) Sort of a testing-to-destruction, or logical-extreme methodology. So now I know how to do it efficiently... ?!? :) Aug 9, 2016 at 23:12
  • 2
    I think peoples' internal notions of stress are very different. I see for instance maximum lifting as not "stressful" but fun (as long as I don't drink coffee or max out repeatedly). Many people who do manual labor have low stress even though they exert/effort themselves very hard throughout the day. Aug 10, 2016 at 0:15
  • 1
    ...I think I did laundry at most 3 times over the past year and a half, while studying for math phd qualifying exams (as a masters student). I lived at the library all year long, and I wouldn't trade that 'stressful' experience away for anything else...
    – user58865
    Aug 10, 2016 at 0:57

All people, everywhere, need food, clothing, and shelter. 99% of us or more gain those things by working. Work is guaranteed to cause stress at some point in your career, regardless of the career.

I've spent my entire life trying to avoid stress, without success. It's a stupid endeavor considering the fact that I'm an engineer that works in a production environment. Bean-counters are ALWAYS watching my productivity. It brings back memories of when I used to think that the practical application of science and math would be a utopian existence. LOL! I personally don't know of any engineer who would choose to do it again.

Here is a list of supposedly low-stress, high-paying jobs. I believe very little of it.

  • 1
    Well it is not very stressful if you keep savings and have confidence in your abilities. Aug 11, 2016 at 3:32

I'm no longer in acadamia and work as a software engineer (US).

One thing I've found is that thinking about the technical problems brings me joy, and if I'm walking my dogs and think about them, it is a happy situation.

Thinking about office politics, however, is quite the opposite. For math in academia, this would include a larger world mentioned by others: publishing, pressure, getting tenure, etc.

Therefore, concentrate on the math itself, if you can.

To me, one thing that would help is a back-up plan: What would happen if you didn't get tenure? Can you work in industry (as I did), do programming, web-tutoring, teaching at a community college (in US, usually an education degree is required for K-12).

One engineer I knew ran a bakery for a few years, so you can think outside of the box.

You must log in to answer this question.

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