Berkeley 2001-2003"it's cool to discover someone
it's hard to support them
everyone is playing life
like it's some stupid sport"
-- Ani DiFranco
As for most new Ph.D students in the US, my first year at Berkeley consisted mostly of coursework, and that was what I was used to, so for the most part it went smoothly. At the end of the year, it should have been a warning sign when nobody wanted to be my advisor. One professor I talked to -- the one I'd mentioned in my statement of purpose as who I wanted to work with, and who encouraged me to come to Berkeley when I visited during prospective student day -- said "no" outright, saying he wasn't interested in what I wanted to study (functional programming languages). Another one didn't say no, but had a reputation of being someone who didn't answer email; I was hoping for someone who actually seemed interested in having a student. I ruled out two more professors who seemed close to retirement, and one more because she did scientific computing and that pushed my "I went to a liberal arts school and don't know anything" buttons too much. I ended up with an advisor who told me he was willing to advise me, but given what I was interested in doing, he wasn't going to be very involved and he would basically just be there to sign paperwork. At the time, I thought that was fine. Remember, I didn't like talking to people. I thought I would just work on my own, and that would be easy. Easier than getting up the courage to talk to somebody, anyway.
Later on, I saw it as a personal mistake to have chosen this advisor rather than looking harder for a more involved advisor, or even changing research areas. But part of why I made that decision was structural. I was socially shut out, as I'll discuss, which meant that I wasn't getting any tacit knowledge that would have helped me understand that I did need an advisor who was involved. I know this is a structural factor and not a personal issue because Barbara Lovitts talks about it in her book Leaving the Ivory Tower. That is, she discovered that a major component of grad students' success or failure is the extent to which they can use informal social networks to attain the tacit knowledge that's essential to completing almost any graduate program; faculty and staff rarely communicate this knowledge to students in any systematic way. Official lists of graduation requirements stick to course requirements and the specifications for what constitutes a dissertation -- they don't talk about the unofficial things, like having an advisor you can work with (and who has time for you) and which advisors are likely to be compatible with which kinds of people. Thus, people who find themselves misfits and outsiders in the (figurative) lunchroom in any particular department tend to get pushed out, even if they're just as able as the insiders to complete the academic requirements.
So here's where my impostor syndrome really began. My second year, I failed prelims. And then I failed prelims again. You needed to get special permission to retake after failing twice, but I did, and failed them a third time my third year. Of course, one fail was enough for me to start seriously feeling like I didn't belong there, which more or less guaranteed my failure the second two times. The third time, I really didn't even prepare. And no one appeared to tell me I did belong, that I was valued, or that it was worth it to try to pass. My advisor said "let me know if there's anything I can help with", which meant I didn't ask for help since I didn't know what to say, and anyway, didn't want to be judged for my questions.
After I failed the first time, I went to ICFP (the International Conference on Functional Programming) and realized that there were actually people in the world who were interested in functional programming. I started thinking that maybe I should leave Berkeley even if I didn't get kicked out, and I started trying to make arrangements to be a visiting grad student at another school. At some point, going to another school for good started looking like a better idea than just visiting for a semester. While I was casting about in this way, my advisor (the one who told me up front that he didn't care about my proposed research) actually did try to get me to stay at Berkeley for a while. But after I failed the prelim for the third time, I guess he changed his mind -- I was applying to other grad schools, and when I asked him if he was willing to write recommendations for me, he said that he didn't think there was any point in me applying to other top-tier computer science programs like Berkeley, because he thought the prelims were a good test of ability to complete a Ph.D.
At Berkeley, prelims were completely different for each area group (e.g. programming languages, systems, theory, AI, and so on). Here's how the programming language prelim worked: You got handed a piece of paper with a list of maybe fifty research papers on it. You were told you had to be familiar with the contents of all of these papers, but the professors who were running the prelim in any given semester were also free to ask about papers that weren't on the syllabus (and to not ask about any of the papers on the syllabus). Each semester, two professors would be responsible for prelims, and they would examine all the grad students who were taking the prelim that semester. Each student got to spend an hour in the office of one of the two professors, with those two professors and nobody else. The professors would ask two questions and the student would have to answer them on the spot. Other area groups did their prelims totally differently: if I recall correctly, the theory prelim involved writing a research paper on a topic of your choice, and possibly an oral part consisting of answering questions about the paper.
I took the prelim three times, and panicked during it every time. I remember very clearly that during one of the times, I was so anxious about the exam that I gave precisely the wrong answer to a question I'd gone out of my way to study because I thought it might be asked about (the subtyping rule for function types). I had massive social anxiety growing up, and while that's gotten better over time, ten years ago when I was a second-year grad student, conversations with just about anyone I didn't know well filled me with so much anxiety that it took all of my mental effort to not run away, much less actually think about solving a problem in real time. In retrospect, I think that's one way in which my impostor syndrome manifested. Even just in everyday social interactions, I was always terrified of being asked a question I didn't know how to answer, as well as terrified of having someone not understand what I said. I'm not sure what I thought -- or think -- is going to happen that's so terrible in one of these two scenarios, but there you have it. Anxiety isn't rational. But I think one of the things I was terrified of, in conjunction with those possible scenarios, is that I would be discovered, unveiled as someone who really wasn't as smart as everybody thought I was, who had managed to lead everyone on. In that moment of being unveiled, I just wouldn't know what to do or say, and that's what made the scenario so terrifying.
For me, impostor syndrome became a self-fulfilling prophecy. My fear of being unmasked as incapable caused so much anxiety as to make it impossible for me to express myself in speech (already something that was often challenging for me), which led to my being perceived as... incapable. When my advisor told me not to bother re-applying to Cornell, that was almost precisely the scenario that I'd envisioned (never assume that the thing you worry about won't come true by virtue of having been worried about!): by telling me that he didn't think someone who'd failed the prelim repeatedly could possibly be capable of completing a Ph.D, he was basically telling me that I was an admissions committee mistake, even if not in so many words. There's no reason why the admissions committee would have admitted a student who they thought couldn't finish the program -- so the only explanation left was that I'd done a good job of looking impressive when I just really didn't have the competence or talent.
I'd like to be able to say that once my worst fear came true, I learned that it wasn't really that bad, but that would be a lie. It was that bad. Almost a decade later, I'm still not over it. Whenever I ask someone a technical question, I still worry that I'll say something that will reveal me for the ignoramus I really am. Overhearing conversations full of vocabulary I don't understand makes my brain shut down rather than making me want to ask questions and learn about these areas I don't know. When I go to a research talk, I usually can't listen because the minute the speaker says something I can't understand (within the first 5 minutes, usually) I'm consumed by worrying about how bad I am at absorbing spoken information. Usually, I'm not consciously aware of that worry, but it manifests itself as an intense desire to escape -- mentally if not bodily -- by opening up my laptop or reading things on my phone. The same is often true for me with one-on-one conversations (only when it's about design or problem-solving, usually not when it's about anything else): I get so consumed by the stress of being around another person and potentially having my deep, essential ignorance revealed, I can't have a meaningful interaction. When one person has already told you that you weren't good enough to do the thing you'd been dreaming of, what's to stop anyone else from doing the same thing? The result is paralyzing fear that continues to limit my ability to work with other people.
I think there are people who could have -- and did -- get through the challenges of grad school, and even of failing the prelim at least once, with more resilience than I did. And I think that's political. How much support you have, which people will reassure you that it's all okay, which people will actively take steps to make sure you do better next time, even whether you feel like you can ask for help at all, is partly a function of random individual brain differences but is largely political. Social capital is not distributed evenly.
On a different note: was my advisor's statement -- that the prelim accurately measures who can complete a Ph.D -- true? Well, I guess I'm not qualified to answer, not being a faculty member or even someone who completed a Ph.D, but my guess is that it's not. I likely would have been able to pass a prelim that consisted of working on a research paper in advance, then answering questions about it: that's exactly the style of the Research Proficiency Exam (the prelim equivalent) at Portland State, where I went to grad school later, and I passed it. If my area group had had that style of prelim, I might have graduated. So that's a sign that the prelim isn't the universal litmus test it was touted as. I do think the PL prelim measures something, though: confidence and ability to think on one's feet. While these are traits that certainly contribute to success as a researcher, I don't see any particular reason why one has to prove that one has these traits by the end of one's second year. The PL-style prelim favored people who have been taught from birth to be confident and to have little fear about being alone in a room with two (likely white and male) professors: that is to say, white, cis men. Finally, the PL-style prelim was unaccountable: it was completely subjective and made it easy for professors to get rid of students who were from the wrong social groups. After all, departments can admit women (and people perceived as women) to boost their diversity statistics, but it's not as common for anyone to scrutinize attrition rates: it's always an option to say that if a student leaves, that's their fault. In fact, this is exactly what Barbara Lovitts discovered in Leaving the Ivory Tower: that faculty members attribute the success of successful students to their own (the faculty members') efforts in mentoring, while blaming the failures of students who leave grad school on the character traits of the individual students. When it comes right down to it, I'm not sure why standing up in a room and talking coherently on a question with no advance preparation is so important that it has to be used to weed out students during the first two years, why what might be phrased as "bullshitting skills" are more important than ability to do research.
So just like social capital is unevenly distributed, impostor syndrome is too. For any given person, coping is a fixed resource; when you go through life getting told in subtle and not-so-subtle ways that people of your gender or your race, or people whose bodies look like yours or who think and communicate the way you do, or people with your sexuality or your economic class, just aren't good enough (and must, if they do well, only be doing well because they were given a "handout"), you tend not to have a lot of coping left to sustain yourself in a challenging and often hostile academic environment. The ways in which different grad programs assess students vary a lot, and at Berkeley, it even varied from research group to research group -- so, as it happened, the PL prelim was a particularly impostor-syndrome-unfriendly exam in my opinion. The exam required not just confidence, not just ability to think on one's feet, but a certain set of social skills that are not evenly distributed. But having that particular set of social skills doesn't have a lot to do with ability to be a researcher. The closest part of a researcher's job that resembles the PL prelim at all is giving academic conference talks -- but even that isn't so similar, because giving a talk on a subject you've worked on for years and know intimately, with advance preparation time, to an audience that almost uniformly knows less than you about your topic, is very different than answering questions, on the spot, with no preparation, from your social superiors. Plus, many researchers are pretty bad at giving talks and they still get jobs.
A friend who's a teacher who read a draft of this essay commented that the PL prelim was "inauthentic": it didn't tell the examiners what they needed to know, because it was so slanted towards people with a particular skill set that didn't have a lot to do with what the test was supposed to measure. I agree, but of course, that assumes the goal of the exam was to measure something objective. Given the lack of accountability inherent in an exam that (in any given semester) only two people have authority over, it's also easy to see the PL prelim as a means of gatekeeping and putting a veneer of academic legitimacy over weeding out students who are a bad fit for the culture.
A lesson that got reinforced for me later on when I went back to grad school was that the people who survive a flawed system often believe the system works... because it works for them. In the case of grad school, the survivors are faculty members: they did the work, met the requirements, graduated, and got one of the jobs that the system promises to those who complete a degree. Because the survivors are comfortable where they are, they have little reason to look at those who got forced out of the system needlessly. And that's why it was that people like my advisor at Berkeley could say they thought the prelim measured what it should measure: it was indeed true that those who passed the prelim tended to do well later on, but logically, that tells you nothing about whether the people who were gatekept by the prelim and forced out of the program could have done well too, or even better.
Dichotomies and DisparitiesBesides the obvious reasons of belonging to socially devalued groups, starting with the group of people perceived as female, a lot of my impostor syndrome probably also arose from a childhood in which I was placed on a pedestal as a "gifted child", was taught to value myself only for my academic achievements, and was never taught that things might be difficult -- even for a "smart kid" like me. My story isn't the most frequently occurring one: I didn't go to school before college, so I didn't really have any sources of validation or reassurance other than my mom, who I knew from an early age didn't have a good grip on reality. I didn't believe almost anything she said, so why should I have believed her when she told me how smart I was? I imagine it's different for most cis girls, trans women, and trans men -- being in school, they probably get validation from multiple sources, and impostor syndrome might not set in until college or grad school. Nevertheless, I'm tired of people othering me by telling me my story couldn't possibly apply to anybody else, and I think I have the authority to talk about impostor syndrome firsthand even though my experiences aren't exactly like everyone else's.
It's possible that I would have had less social anxiety if I'd gone to school as a kid, but I don't think so. For one thing, I might have been labeled in some pathologizing way (pick one of ADD, autism spectrum, or just "noncooperative" -- not that it's bad to be any of these things, just that kids who are labeled in any of these ways are frequently treated badly by school systems). For another thing, I probably would have been bullied -- my transsexuality, which I was able to ignore (somewhat) blissfully until I got to college, probably would have been an issue much earlier, and I likely would have faced wrath from girls for being someone who was assumed to be a girl but didn't know how to do girliness (I'm guessing that would have happened in primary school because it's what I faced as a student at a (historically) women's college). For a third thing, I likely would have been really bored and would have been stopped from doing things that would have engaged me (for example, I actually did go to the first half of first grade at a parochial school -- and got in trouble for reading books at times when such activity wasn't allowed). On the plus side, it's also likely that I would have had adults other than my mother -- that is, reliable people -- encouraging me and telling me I had value. On the whole though, as wistful as I feel when other people are talking about their school memories and I have nothing to share, I can't say I really think I would have been better off in school.
What is common across the experiences of lots of people, not just people who didn't go to school as kids, is that being told you're smart is apparently hazardous to children. Kids who are told they're smart are much more likely to be anxious and perfectionistic, as they see difficult tasks not as chances to grow and learn, but rather, dangerous situations in which their essential, core level of smartitude might be exposed and judged by others not to be good enough. That's impostor syndrome right there.
A closely related issue that contributes to impostor syndrome, I think, is the cultural value we attach to being "smart" or "intelligent" and the stigma we attach to being "stupid" or "idiotic". We insult people by calling them "stupid" or "idiotic", even when we know perfectly well that the person we're talking to probably has an IQ above 30 (which was once the definition of "idiot"). We even call ideas "stupid" or "idiotic", presumably to suggest those ideas could only have come from a stupid or idiotic person, and thus can't possibly be valid. Why is it good to be smart and bad to be stupid? I haven't figured out a logical reason for that yet (except within an ideological framework that defines a person's value as their ability to generate wealth for others in a capitalist system, which I reject). My logical mind knows it's ridiculous, but my illogical mind is still deeply attached to the idea of "smart" as "worthy" or "valuable", which is why it provokes so much fear in me whenever I think of someone maybe judging me as not so smart. Being "smart" has nothing to do with being kind, caring, or compassionate, and yet our culture teaches us to value it so much. This leads directly to massive insecurity, as there is no way we can ever convince ourselves that we are "smart enough". Valuing smartness rather than, say, effort (or anything else on this list) is essentially encouraging an attitude where "intelligence" is an innate quality one is born with and can never change, only observe. Behind impostor syndrome, I think, is the fear that someone else will notice that this innate quality in you is lacking, and then, there will be no hope for you; you'll just never be welcome in the circles reserved for those who Can Do Things. If we valued those capabilities in ourselves that can get stronger as we learn and grow, on the other hand, there'd be less reason to be afraid. Someone seeing where you fall short would just be an opportunity to grow.
And I think that valuation of "intelligence" as a homogeneous thing is deeply political. Who benefits when everyone thinks what matters is being "smart", or at least convincing people you're "smart"? Well, for one thing, a belief in the value of "smart"ness makes it easy to believe that we live in a meritocracy, or at least that particular organizations, or professions, are meritocracies. Belief in meritocracy is a way of willfully not seeing inequality and injustice, and it's obvious who benefits from that: the same people who always benefit from unfair distribution of resources. Moreover, even if society was a meritocracy (though it quite obviously isn't), would that be fair? If intelligence is some single quantity that you're born with (not that I think it is, but that's how it's usually understood by people who value being "smart" a lot), what's fair about assigning rewards and punishments based on its presence or absence? (For more on this line of thought, see William Ryan's book Equality.)
To be continued!