Tall, Dark, and Mysterious


Everything I ever needed to know about grad school, I learned the hard way

File under: Righteous Indignation, No More Pencils, No More Books. Posted by Moebius Stripper at 2:30 pm.

Over at Toilet Paper With Page Numbers, a new-to-me blog that’s going straight into my bookmarks, some advice on picking your poison - that is, selecting an advisor for grad school. TPWPN’s taxonomy of advisors (Micromanager, Ambitious Geek, Absentminded Professor) is geared at science students, but anyone considering studying math at the graduate level should read it, too. Now. Seriously, git; TD&M will still be here when you’re done. If it were up to me, which it isn’t, that post would be included in every science grad student orientation booklet. I don’t remember what was in my orientation booklet way back when, but it wasn’t as useful as, say, this:

The Creative Genius: An older researcher who, at one point in the past, struck a home-run or two and is now flush with cash. Usually has more than one sub-group, and misuses time switching his or her attention back and forth between them as fancy strikes. Usually gives students a lot of room to pick their own projects, and has the money to let you develop your ideas. That’s great if you, yourself are creative, but a nightmare of a 7-year Master’s degree if you are not. (For real. I know of at least one 10 year Master’s). Be honest with yourself, and if you are just not that creative, go for the Micromanager.

Advantages: Money, money, money, and room to run free.

Disadvantages: Has a lot of ideas to start you out with in your first year. No one in the history of the lab has ever gotten any of them to work in the advisor’s entire 20 year career, so find a new idea quick, or prepare for Master’s hell.

Ah, that brings me back. Which reminds me, what is a certain student of my former advisor’s doing these days? Last we spoke, he was wrapping up his seventh year as a grad student and celebrating his 34th birthday, ostensibly making some progress on his thesis. Lifelong Student’s (15 years in postsecondary education and counting) graduate career had seen him switch departments twice, make enemies out of more or less everyone in all three of above, postpone his quals until Year 6, and finally get our mutual advisor (hereafter “Eccentric Genius”) to just solve his research problem for him so that LS’s Ph.D. thesis was reduced to a paraphrasing of someone else’s work. If Lifelong Student actually did manage to defend, then that would make him EG’s second graduate. In fourteen years. I lost track EG’s refugees sometime into my second year: in my three year stint in grad school, a good half dozen students had signed up, or seriously considered signing up, to work under EG and then came to their senses before I did. The other student of EG’s who obtained her Ph.D., did so just before I arrived; I heard through the grapevine (LS, in particular), that she hadn’t understood any of the first sixty pages of a paper she’d cowritten with EG, and that she had understood only approximately half of the last ten. After finishing her seven-year graduate school career, she took an unrelated job in industry.

But that’s beside the point I was planning to make, which is this: if you’re a Master’s student, it’s nigh irrelevant whether or not your supervisor is a leading expert in their field. In fact, I’d advise against working with leading experts, as I’d wager that they’re probably less likely than non-leading-experts to be able to communicate effectively with amateurs. Master’s students do not need leading experts. Master’s students need supervisors who are solidly grounded in their subject, yes, but nearly everyone with a Ph.D. in the appropriate discipline can be counted on for that. What Master’s students really need, and what are in considerably shorter supply, are advisors who can advise. Master’s students, particularly ones who’ve never done any thesis-like research before, need advisors who can identify, to the uninitiated, the steps of the research process and guide their students through them.

I wish someone had told me that when I was assigned to Eccentric Genius, a world-renowned researcher in Impenetrable Geometry; knowing that would have saved me more than two years of grief. Actually, I didn’t even need to be told that I didn’t need to work with a world-renowned expert; I would have settled for not being told, repeatedly, from professors and peers alike, that I was so lucky to be working with EG, that so many other students would kill to be in my position, that I would be insane to even consider working under anyone else, yadda. When other schools saw that I’d worked under EG, I was told repeatedly, the academic world would be my oyster, and this was truly the opportunity of a lifetime.

People kept telling me this, and I believed it. I believed it when, after our very first meeting during which EG told me to study up on a certain invariant, I read a paper and an appendix on the subject and tried, to no avail, to compute some examples myself. During our next meeting, I reported on my lack of progress, and asked EG if he could show me an example or two. For the next - no lie - FOUR HOURS, EG paced back and forth in front of the blackboard trying, and failing, to work through a handful of examples. “I haven’t done this in years,” he explained, “usually I use computations other people have done. But you can probably figure out from these papers…” Later that week, I pored through handful of papers he’d given me. Each used a different formula to compute the invariant; none of the formulas worked for the examples in the other papers.

I believed it the next week, when, in passing, one of EG’s colleagues at the grad school told me that he was impressed that I was reading those papers, which were high-level research papers, didn’t I know, certainly not the sort of thing that most graduate students get weaned on. I pointed out that there was nothing impressive about being assigned high-level papers to read. If I were able to understand them, that would be impressive. Still, though, I blamed myself for not living up to what I foolishly thought were reasonable expectations.

I believed it when I finished the summer research term with nothing to show for my work. I was lucky, I knew, to be working with Eccentric Genius. The reason I wasn’t making any progress was because I wasn’t smart enough. I just needed to work harder.

I started to grow doubtful a few months later, when I noticed that EG’s solution to my complete lack of progress and utter inability to make heads or tails of any of the papers he gave me, was to give me more papers to read. After a year I had some thirty high-level research papers in my office, and I’d started working with Lifelong Student on one of them. We were able to make some sense of it, but neither of us could see how on earth it applied to our assigned research topics. Still - I was lucky to be working with such an expert, and I knew I should be grateful.

I was frustrated and upset, but still grateful in some perverted way, when my questions about how the subject matter was motivated were brushed off. “That doesn’t matter,” he said, “just use the axioms.” It was true that those axioms had been developed in response to some other research problems, rather than being handed down by God Himself to his prophets in Impenetrable Geometry, but the fact of the matter was that I had a thesis in Impenetrable Geometry to write, and that it behooved me not to worry about what all of this stuff meant. There just wasn’t any time for that kind of thing. Meanwhile, had I managed to compute those invariants yet? I explained - again - that each of the research papers computed the invariants with a different formula that seemed to come out of nowhere, and perhaps I could make some sense of the formulas if I had some context, or something. None was provided; just trust the formulas, I was told, and don’t worry about where they come from.

It was at this point that I began to see clearly the roots of my frustration: I had decided to study math specifically because I thought I would never be expected to trust in tradition, to take any theory for granted. But I also knew that if I was going to accomplish anything under EG, it would have to be at his level, so far removed from the more elementary material that I couldn’t possibly learn all of the background in the time I had and that I would consequently have to start by accepting some results that I had neither the background nor the intuition to comprehend deeply. I had been working - or, more accurately, “working” - under EG for nearly two years at this point, and had gotten nowhere with him.

He realized it too, and decided to try something different. He started meeting with me and Lifelong Student together, and assigned us to dig through yet another collection of papers to see if there was anything useful there. I am not omitting details here: he handed us three hundred pages of mathematics with the explicit instructions, and I quote, “See if there’s anything useful here.” Anyone who’s ever taken a serious math course knows how absurd this is: getting through even a twenty-page pager will often take an entire weekend,

And I, beaten into submission and resigned to never being able to earn a Master’s degree, set to it with LS. We pored over the three hundred pages of mathematics, meeting periodically with EG. I couldn’t for the life of me tell what on earth anything in these three hundred pages of mathematics had to do with my research problem (which I still didn’t really understand, but whatever), but I plodded through (a subset of) them anyway. After all, I knew why it was that I wasn’t understanding anything: it was because I was stupid. And how could I squander the opportunity to work with a leading expert in Impenetrable Geometry?

A few months into this, LS commented in passing, “You know, I can sort of see what these three hundred pages of mathematics have to do with my research problem - however, from what Eccentric Genius told me, I don’t see what they have to do with yours.”

I suddenly felt ill. The next day, I composed an email to EG, asking, not in so many words, that question: how was the research problem I didn’t understand related to these three hundred pages of mathematics that I also didn’t understand? EG wrote back, calmly, “That’s a good question” (!!!) and then explained that it seemed I’d “lost interest” in my original research problem, so he’d taken it upon himself to give me another one, one related to what LS was working on.

He’d never bothered to tell me this.

This was two weeks before I was to take off for my summer job, which I spent variously distracted and in despair. I knew that I would not get a Master’s degree under Eccentric Genius. Classmates of mine who’d struggled through graduate level classes and had routinely solicited help from me, were getting their diplomas in two years. I didn’t know if I’d be done in three.

The next fall, fortunately, luck was on my side. Two things happened. One, I attended a conference of six talks, five of which I didn’t understand even slightly, and the sixth of which had me excited about math all over again; two, my grad school had just hired a young professor whose area of expertise was related to the topic of the sixth talk. I started working with him, first unofficially, then officially. It took me a few months to work up the courage to abandon EG, but finally I said to myself, out loud, you will never graduate with him. If you want to avoid hurting his feelings, keep working with him. If you want to graduate, jump ship, NOW.

Under my new supervisor, I saw for the first time what the grad student/advisor relationship could be. My new advisor gave me bite-sized pieces of work to do. Each week I’d read over a small section, work on some problems, and come up with questions. Later on, I’d work on more involved questions, and get stuck; during my meetings with him, he’d give me just the push I needed to get unstuck. I’d give him drafts every few weeks, and he’d make copious notes explaining what I needed to correct or clarify. I submited my thesis six months after I officially started working under him.

A year after I snagged the Master’s, I don’t blame my former advisor. Not every professor can be all things to all people; EG was an expert, suited to working with other experts, and perhaps some postdocs. A handful of Ph.D. students might have thrived under him, but I’m skeptical: when I left the school, his most promising Ph.D. student was frustrated by his lack of progress and was thinking of leaving. (EG’s response was to tell this guy to “think it over” at a conference in Europe that he could attend that summer, at EG’s expense.) I do blame the school, though, for assigning me to work with him, even though I’d never expressed an interest in his particular flavour of Impenetrable Geometry. I wish I’d been given an explicit idea of my responsibiltiies to my advisor, and his to me, as a graduate student doing research , so that I wouldn’t, in the absence of any clear guidelines or expectations, conclude that the reason for my lack of progress was because I was too stupid to do graduate level mathematics. I wish I’d been given explicit deadlines and tractable goals against which I could measure my progress. Instead, I was given one goal and told, “Once you can do this, you’ll have a good thesis.” And I couldn’t meet that goal.

I wish I’d been told that my job as a Master’s student was to build a solid background in my chosen field - not to build a reputation in it. I wish I’d been told that I did not need a big-name mathematician to help build a solid background in a respectable field - and that many a big-name mathematician wouldn’t help me get any closer to that goal.

In retrospect, all of this seems so obvious, but it wasn’t to me at the time. When I was in grad school, each failure of mine, contrasted with each success of a student working under a different advisor, eroded further my ability to look clearly at my situation. It blinded me even to the simple possibility that I could ask my peers about how they did their research; when I did find out some details, I wondered what I would do if I quit working under EG. By the end of my second year I was virtually paralyzed with self-loathing, at least as far as my work was concerned. I couldn’t even contemplate an advisor switch; I wasn’t worthy. I couldn’t see past my own inadequacies in assessing the sorry state of my research. It took me two years in grad school before I realized that I could, and should, demand better than what I was getting. It took year away from grad school before I was able to put everything in focus, before I could articulate a clear answer to “why did you leave before you got a Ph.D.?” that didn’t sound to me like an excuse.

Still, though, I can’t help but think that I could have learned at least some of this before. Would have made for a more helpful orientation than the “let’s meet the faculty and then have some cookies” one I attended, that’s for sure.


  1. Your advice is duly noted. Thank you!

    - didier — 5/15/2005 @ 3:40 pm

  2. Ah, I gotta say that I felt much better after I recycled all the paper I generated in my research haul to non-dissertation-hood. It can be very good to realize “Hey, I’m never gonna do this crap again. Yee-haw!”

    Anyway, now I spend my days trying to get grad students to come over to the dark side…. business… You’re not necessarily better-treated, but you’re better-paid and the goals given are usually very concrete and achievable.

    - meep — 5/15/2005 @ 6:04 pm

  3. The post comes at a time when I am trying to decide between the two graduate school admits I have. Couldn’t have come at a better time.
    Thank you

    - Punit — 5/16/2005 @ 4:50 am

  4. I keep looking over that list trying to figure out which one of those *I* am going to end up as. :) I’d say my advisor was the “Absent-minded professor” type.

    Unfortunately that guide doesn’t have that much value. Really, I think you only really find out what your advisor is like after a year of working together, and it’s not like everyone has a ton of choices for who’s willing to advise them.

    But it’s definitely worth knowing what you’ll have to deal with.

    EG, for instance — I have the feeling he might best have been handled by, after doing lots of work on tough papers, asking him for something a little more mundane so you could actually write a paper or two — that kind of person probably has their head in the clouds and after being reminded that YOU have pressing needs, they might snap out of it enough to help you graduate.

    Maybe. Who knows.

    - Moses — 5/16/2005 @ 11:20 am

  5. MS, thanks for the link. I think one thing I failed to make explicit over there was that in choosing an advisor, once you know which type you can personally work with given your own temperament, talk to the grad students first. Don’t waste your time with the advisor, talk to his or her students. Find out how many have to teach regularly because funding runs dry too often. How often do they talk to their advisor one-on-one? Does he or she take attendance at 8:30 every morning? How many papers does a Ph.D. student in that lab usually publish, and in which journals?

    I bet that talking to LS before you were assigned an advisor would have saved you a lot of grief. In the sciences, even the Master’s level student choose their own fate. We got tremendous leeway in choosing an advisor, and unless the advisor was obscenely famous, there was a good chance that a good student would get his or her first choice. However, a lot of advisors do not want to take the time to train someone who is going to leave, just at the point where they are becoming useful in the lab, with an MS.

    - John — 5/16/2005 @ 11:48 am

  6. This sounds almost exactly like my physics graduate school experience. Except my EG was certifiable, as in, he now thinks the the universe will eventually be run by a giant computer who will resurrect us in virtual reality. I had 2 years of being told “concentrate on your classes” and another year of feeling sick every time I forced myself to go see him, because he had excoriated me in the physics office (behind my back) for not being able to find the intersection of 2 straight lines. For the record, I was expected to find the equations for 2 curves and their intersection– I didn’t know they were straight lines, much less how to derive the equations…at any rate, when I finally approached the chair of the dept about switching advisors (after hearing the story of how EG’s previous student had suffered a nervous breakdown and locked himself in his office for 3 days, before being allowed to submit a 56 page diss) the response was “what took you so long?”

    My next advisor was fantastic, and I finished up in 2 more years for a respectable total of 5. I hear they’re not allowing EG to have any more grad students. I’ll be an assistant professor starting this fall; your blog is already terrifying me. :)

    - Jess — 5/16/2005 @ 12:55 pm

  7. That’s really terrible, and I’m glad you’re sharing your experience. In general grad students are too much on their own, and don’t get the advice they need… they don’t know how to be productive, or what to expect from their advisors.
    But I think advising tends to be even worse in Europe (not counting Britain)

    Do you think you would want to get a PhD had you been assigned a decent advisor?

    - Gustavo Lacerda — 5/16/2005 @ 3:32 pm

  8. I just finished filling out the confidential exit questionnaire for the UC Davis office of graduate studies. With a little luck, I will submit it next month with my dissertation and leave the university with a Ph.D. in math education. The wonder of it all is that I had to navigate among the following: The doctoral program was sponsored by the Graduate Group in Education, a cross-discipline group with members in many different academic departments. My graduate advisor, who chaired the Group, was a professor of mathematics, although the program was administered by the education department. During my time in the program, the university created a School of Education that subsumed the activities of the education department and brought in a new dean. My math ed prof was appointed associate dean of the new School of Education and my math ed cohorts and I no longer had a prof. We managed with an adjunct prof for a couple of quarters. Eventually a new math ed prof was appointed, although by that time I had pieced together the core courses required for my option with the help of the adjunct prof (and with one course formally waived for lack of qualified faculty to teach it). I racked up 134 quarter units while maintaining my full-time status in pursuit of a degree that required 60 units. Yes.

    My graduate advisor has been very encouraging, but his field is math itself and my research is math ed, so he naturally defers to my ed profs in evaluating my dissertation although he is the chair of my committee. The filing deadline is June 6 and I have yet(!) to hear from one of the ed profs on my dissertation committee (my dissertation has been sent out in multiple drafts to all parties since the first of the calendar year). My advisor is checking with his colleagues across the campus in ed to determine whether I’m really graduating next month.

    It’s exciting.

    And as for that exit questionnaire: “Did your professors return your work promptly?” Uh, no.

    - TonyB — 5/16/2005 @ 4:11 pm

  9. Moses - that might have helped, but honestly, I just don’t think that EG should have been assigned to Master’s students, period. He was clearly so out of his element working with us. As I said, there were other Master’s students in my department who were actually making progress under their advisors, and didn’t have to pull teeth to do so. And - at the risk of sounding arrogant - many of those students didn’t have my background or my creativity. The only differences were that 1) they weren’t working in Impenetrable Geometry, and 2) they had advisors who guided them.

    Jess - ah, the dread of meeting with the advisor. I don’t think mine bad-mouthed me behind my back - my EG was a man of few words, whose
    MO was to stare at me for long periods of time whenever I asked a question. He might have been thinking that I was an idiot; he might have been thinking about his (unrelated) research. Hell, he might have been thinking about what he was going to have for dinner. Who knows? I sure didn’t.

    Gustavo Lacerda - Do you think you would want to get a PhD had you been assigned a decent advisor? Honestly, I don’t know. Definitely not in the field in which I got my Master’s - I had lost interest in it well before my thesis was finished, and this was even with a competent prof. Part of my problem is that I have no interest in being a specialist. If I could cobble together a Ph.D. program that focused on connections among ostensibly unrelated fields, and that allowed me to work on distilling interesting postsecondary-level mathematics to a semi-general audience (think Martin Gardner et al), I’d do it in a heartbeat. And I’ll be the first to admit that my bad experience in grad school went well beyond my advisor: I felt that there wasn’t any room for anyone with my particular interests. And at my particular grad school, that was likely true.

    TonyB - congrats; hope your degree survives the process ;)

    - Moebius Stripper — 5/16/2005 @ 5:32 pm

  10. Here I am in year 2 of a Master’s Program, working under the Chair of the Math Dept. here at TechU.

    He’s been trying to get someone else involved in our research (Projective Geometry–probably something related to Impenetrable Geometry, now that I think of it). But I’ve been having to bring Other Discrete Math Prof up to speed on the details. Even worse, Dept. Chairs suffer from random insertions of meetings, off-campus speeches, etc. into their schedules.

    And here I am, promising myself that I will push ahead, doing everything in my power to get a Masters-level at the end of the next academic year. Because teaching isn’t all I thought it would be, and I wonder what else I can do with a Masters in math…

    - karrde — 5/16/2005 @ 6:43 pm

  11. Wow, that’s a great post. I’ve known a lot of miserable math students in my time, and some of the same as happy happy students, and a good advisor can make all the difference. In general I think the academic system does not reward good advising type qualities very well, resulting in a spate of professors who are good at publishing papers but not so good at nurturing students. Since we have more students than we have slots in Academia, Academia doesn’t care that it chews up its own young and is happy when people quit/move on to industry or another field. But in the long run it seems like a big waste of mind.

    - Saheli — 5/16/2005 @ 11:06 pm

  12. Really nice post. Though I’ve had problems identifying any of the potential advisors I’m facing with any of them (well, except for the emeritus-in-training; one of the profs I’ve spoken to about PhD advisorship warned me that he will be emeritized (?) in a year, and that contingency plans for senility are in place)

    On the other hand, I’ve already had a few .. interesting advisor-student relations. The one professor who knows Geometry waaaaaaay too well for anyone to follow, and who has a rather weird grading on what facts are non-obvious and need proof, and what facts are completely obvious and need not be proved. Or the other assistant professor who is rather good at the advising process as such; but completely unable to do things in decent time - if you want him to actually DO something, you may want to setup automated reminder scripts to keep reminding him about the task you need. Otherwise the relevant papers will drift down through the 1/3 m pile of papers on his desk and be lost for ever and ever, and your task will be forgotten. On the other hand, he -is- clear, enthusiastic and good at giving advise once you’ve actually managed to pin him in one place.

    Professor management skills seem not entirely bad to develop.

    - Mikael Johansson — 5/17/2005 @ 1:35 am

  13. Interesting - I’ve received roughly the same number of comments on this post as I get on most of my others, but I have received far more private email about it than I’ve gotten on the others. Frustrating experiences with advisors seem very, very common; and, for obvious reasons, people don’t want to put their names to them. As it is I fear I’ve said too much.

    In general I think the academic system does not reward good advising type qualities very well, resulting in a spate of professors who are good at publishing papers but not so good at nurturing students. - Bingo. Advising is rewarded only marginally more than teaching in academia, which is to say - not much. It’s publish, publish, publish (quantity over quality, too, which is its own post), which probably gives rise to the vast number of students I know who have “cowritten” papers with their advisors - papers in which the prof pretty much got the student to do all of the grisly computations.

    - Moebius Stripper — 5/17/2005 @ 4:39 pm

  14. As someone rather nervously going back to school *mumblety-umph* years after his BA, I will try to keep your advice in mind — as another commenter said earlier, your timing couldn’t have been better!

    - ThatTallGuy — 5/18/2005 @ 9:10 pm

  15. I posted a story mentioning this at the Young Mathematicians’ Network.

    - Hack — 5/21/2005 @ 2:41 pm

  16. There are some interesting parallels, MS, between your dealings with Eccentric Genius and your students’ dealings with you depicted on this blog over the past few months. I’m thinking mostly of the seeming incomprehensibility, to the student, of the material the mentor/advisor/teacher presents.

    Also, you’re not still mad at me about the MS-photo fiasco, are you? Or were you ever?

    - wes — 5/22/2005 @ 10:24 pm

  17. I wish that you were older than me so that I could have learned from your experience. I loved undergrade but I walked away from my NSERC grant after one term at UBC.

    - Peter — 6/15/2005 @ 4:14 pm

  18. A friend of mine wrote an excellent guide to finding an advisor. It’s math-specific, but I’m sure much of it can be generalized to other technical fields. This is the advice that should be handed out to every incoming grad student.

    - Moebius Stripper — 8/10/2005 @ 12:05 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.