Why No Science Of Nerds?

Different groups navel-gaze in different ways. Broadway musical writers write a bunch of musicals about what it’s like to be on Broadway. Poets write a bunch of poems about writing poetry. Philosophers speculate on how philosophy may be the most truly virtuous activity. Psychoanalysts analyze the heck out of the inner mental experience of psychoanalyzing someone.

All this leaves me a little surprised that there isn’t more scientific study of nerds.

And yet there is not. Typing “nerd” into Google Scholar brings up only a series of papers on desert plants by one Dr. A. Nerd, who must have had a very unpleasant childhood. The field remains strangely unexplored.

“Nerds” seem to share a bunch of seemingly uncorrelated characteristics. They’re generally smart. They’re interested in things like math and science, especially the hard sciences like physics. They’re shy and awkward. They’re some combination of bad at getting social status and not interested in getting social status. They’re especially bad at getting other people to show romantic interest in them. They’re physically unimposing and bad at sports. They don’t get in physical fights and are very unlikely to solve problems with violence. They’re straightedge and less likely to drink or smoke to excess (according to legend, “nerd” derives from “knurd”, ie “drunk” spelled backwards). Sometimes even very specific physical characteristics make the list, like a silly-sounding high-pitched voice.

A scientific study of nerds might begin by asking: why do all of these things go together in the popular imagination, form a single category?

Of these nine classic characteristics, we can imagine people scoring either “nerdy” or “anti-nerdy” on each. If that were true, there’s only a 1/2^9 = 1/512 chance that any given person has all of these characteristics. Our null hypothesis might be that “nerd” is just a made-up category used to describe this totally coincidental group of 1/512th of the population, or those people who are sufficiently close (maybe 6 or 7 out of 9?). Could be.

The other possibility is that in fact these traits are all correlated for some reason, and people who are bad at sports really are more likely to enjoy math, less likely to drink or use violence, et cetera. “Nerd” would then be a natural category, in the same way that, for example, “bird” is a natural category pointing out that animals with feathers are more likely to have wings, beaks, et cetera rather than a totally random distribution of traits. Why would that be?

Have there been nerds across different times and cultures? The term was only coined in the mid-20th century. But Isaac Newton seems to have been a nerd. Whatever else he was, Henry Cavendish was at the very least a nerd. On the other hand, some of the Europeans I’ve talked to say that the experience of nerdiness on their side of the Atlantic is very different from the American experience, so much that it’s hard to interpret them as having a “nerd” concept at all. From my time in Japan, the experience there is different as well.

One can sort of imagine how certain of the correlated characteristics might cause others. Young people who are small and weak and bad at sports might lose social status as a result. People who are good at math might be so transfixed by the mysteries of the universe that things like sports and socializing and dating pale in comparison. People who are very smart might, for the usual neurological reasons, also have high impulse control, explaining the lack of drug use and aggression. People who are bad at social skills might not get invited to sports games and so have no opportunity to improve; they might take up unpopular but solitary pursuits like math as a result.

Or we could take the fun route and go full biodeterminist. The connection to autism and Asperger syndrome is so commonly cited (despite a lack of any real scholarly investigation) as to be cliched. Anyone familiar with the full-fledged syndrome understands it’s something very different from everyday nerdiness, but the possibility that nerdiness is some very mild form or related condition probably shouldn’t be ruled out.

Or we could go a different route. Consider:

In men (but not women) low testosterone is related to increased mathematical intelligence and increased likelihood of being gifted.

Testosterone is associated with extraversion, alone of Big Five characteristics.

Higher testosterone men have higher social status, and the cognitive role of testosterone has been described as “best understood in terms of the search for and maintenance of social status”.

As anyone who has watched the controversy over steroid use in professional sports knows, testosterone improves muscle mass and athletic performance.

Men with lower testosterone have lower mating success.

High testosterone is correlated with increased drinking behavior with p < .001.

Especially in prison populations, high testosterone is closely linked to violence and aggressive behavior.

Men with lower testosterone have higher-pitched voices (the media bills this as Deep Voiced Men Make Bad Mates: Study).

So at least in men, low testosterone seems to cause most of the characteristics associated with nerdhood.

But I’m being unfair. There’s a lot of counterevidence as well.

There are so many conflicting studies on testosterone and intelligence that I despair of getting anything coherent. Many studies show increased testosterone increases intelligence, especially in tasks where men generally outperform women, like spatial rotation (which tends to correlate with math) – for example, see here. I respect this research, but I would naively expect, let’s say, brilliant mathematicians to have lower testosterone than the general population, just based on my stereotypes that testosterone is associated with aggressiveness, popularity, athleticism, etc. I’m not sure how this squares with the data.

Testosterone doesn’t actually make male faces more attractive, according to Testosterone increases perceived dominance but not attractiveness in human males.

Nerds are traditionally viewed as having high libido – think watching pornography. But of course high testosterone is associated with higher libido. I don’t know if there’s a cultural thing going on where nerds have normal-to-low libido but are stereotyped as having high libido to make fun of their lack of romantic success – or even whether the pornography connection is just that nerds are better with computers. I also note with interest that testosterone is said to affect sexual libido but not desire for “sensual touch”, and a lot of people have mentioned how anomalously some of the nerd communities I’m in tend to value cuddling compared to sex relative to the general population.

I could probably find other traditionally nerdy characteristics that correlated negatively with testosterone. For example, acne is associated with higher testosterone levels.

So I don’t think this is the whole picture.

But I still feel like it should be some of the picture. Sex hormones are really complicated – for example, there seem to be different effects from in utero sex hormone exposure compared to pre-pubertal sex hormone exposure compared to pubertal sex hormone exposure, and these aren’t necessarily correlated with each other in the same people. Estrogens can have effects ranging from very similar to testosterone to exactly the opposite. Testosterone is hard to get a good measurement from, especially with the no-fuss salivary measurements I bet most of these studies used, and its effects or lack thereof would depend on lots of stuff like how much of it gets converted to dihydrotestosterone. Also, some hormones have totally different effects if they come in short spikes versus constant gradual release. So there’s a lot of room to improve our understanding of sex hormones and start distinguishing between unlike constructs.

I am reminded of an observation common among transsexuals – and brought up in Ozy’s last post – that there is a distinct cluster of transwomen who have certain very traditionally-considered-male-gendered characteristics and are very nerdy. This seems like another example of some strongly male and strongly female characteristics anomalously going hand in hand.

And what about female nerds? If we’re trying to make this about testosterone, it sounds like they need a separate explanation. Is female nerd a distinct cluster in the same way male nerd is? I don’t know. I don’t, for example, have strong feelings about whether female nerds are more or less attractive/athletic/whatever compared to non-nerdy females.

Mostly I just feel like this field is strangely under-explored. Especially when you consider what the sorts of people who explore fields are like.

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

239 Responses to Why No Science Of Nerds?

  1. Anonymous says:

    >Our null hypothesis might be that “nerd” is just a made-up category used to describe this totally coincidental group of 1/512th of the population.

    Or “nerd” could be anyone with a certain amount of these characteristics; ex. 6 or more is nerdy, then [edited, wow i don’t know what i was thinking when i did that, pretend I said “some X%”] of the population is nerdy. Or maybe characteristics are weighted differently — I’d say “interested in math and science” rates MUCH higher on the nerd scale than “straightedge”.

    • Scott Alexander says:

      Yeah, I should have mentioned that. Added.

    • Dude Man says:

      Maybe this is why you don’t see the navel gazing Scott talks about; it is much harder to determine who is or isn’t a nerd compared to who is or isn’t a poet, philosopher, or psychoanalyst. It appears that we can define nerds as people who possess certain characteristics, but not all of the characteristics are necessary. This creates a scenario where a value judgement is needed to determine which characteristics are vital to nerdery and which are more incidental. Another concern is that many of these characteristics do not have a clear brightline for when they become applicable. For example, how do you define “interested in math and science” in a way that can be used to divide people into “not interested” and “interested” categories for the purpose of a study?

      Meanwhile, it’s easy to tell if someone is a poet or not. Do you write poetry? Congratulations, you’re a poet.

      EDIT: Another possible reason you don’t see the navel gazing Scott describes might be that writing poetry or psychoanalyzing patients are things that you do while being a nerd is a cluster of traits.

      • Evan Gaensbauer says:

        For a long time, ‘poet’, ‘philosopher’, or ‘psychoanalyst’ have been clusters that, while not the most celebrated, or mainstream, of groups, had a formal place in society that everyone acknowledged. These are nerd-like people who fell into niches that have been normalized for at least a hundred years in English-speaking countries.

        Being a nerd, and only a nerd, on the other hand, may have only been normalized recently. To the extent that being a nerd has been accepted by popular culture, it’s only for adult nerds, such as those in Silicon Valley, as Internet celebrities, at SDCC, and on hit sitcoms. I doubt American culture has hit a point where classically nerdy twelve year-olds are as accepted equals by all their peers in middle school.

      • AndR says:

        > Meanwhile, it’s easy to tell if someone is a poet or not. Do you write poetry? Congratulations, you’re a poet.

        Are you? I’d usually call someone a poet if writing poetry is either their source of income, or at least a significant part of their life. Or maybe if they were published.

        There’s a difference between writing poetry as a hobbyist, and being a poet. Similar to how there’s a difference between playing soccer and being a soccer player, etc.

  2. Matthew says:

    Even with the subsequent caveats about non-US contexts, I feel like this definition of nerd is highly overspecified. Or to put it more memorably, I think we may be dealing with the “Typical Nerd Fallacy” here. (Possibly related — people actually used to think the nerd/geek/dork distinctions were worth arguing about, but I don’t see nearly as much of that any more.)

    I see intellectualism (valuing learning for its own sake) and social awkwardness as highly correlated but distinct characteristics (Feynman seems to have been fine on the latter, for example).

    The not drinking or smoking seems noncentral to me (I’m actually extreme in that I’ve never smoked nor been drunk, but I’ve known plenty of people I would classify as nerds who do). “Interested in physics” is characteristic of the sort of nerd that is drawn to LessWrong, but I’m far from certain that all nerds would be drawn to LessWrong. (Also, I assume this doesn’t actually need remarking on, but the majority of nerds are not autistic even if the majority of autists are nerds.)

    I’m also not sure how porous the nerd/non-nerd boundary is. I’m actually not sure if I lost my nerdiness at some point. I’m uncertain whether other people would classify me as one or not (introverted, intellectual, like sci-fi and board games, but very athletic, very high testosterone, shy but not particularly socially awkward any more).

    • Protagoras says:

      Yes, I wonder if the kind of research Scott is wondering about has been done under other names, with slightly different but related clusters being investigated. As he says, it seems like the sort of thing that would get investigated.

    • Anonymous` says:

      I was actually just re-reading Surely You’re Joking, and while Feynman *ended up* being not too socially awkward, that was suggested to be the result of drilling it in college.

    • Evan Gaensbauer says:

      The lines between nerds, geeks, and everyone else may have become blurred over the last few years as being a nerd has become less stigmatized. Thus, perhaps people are either more willing to be open about their nerd-like characteristics, or they may be more prone to trying nerdy things than before.

      • Cauê says:

        My impression was always that the nerd/geek/dork thing was at least partially about trying to shed low-status connotations (“I’m not like those dorks, I’m just a geek”). Not as necessary nowadays.

      • lambdaphage says:

        Cognitive sorting has increased significantly over the course of the last half-century. Perhaps an increased perceived tolerance of nerds is due to the average nerd winding up in a nerdier environment.

    • The alcohol/drugs thing is definitely non-central, at least in my experience. As a data point, the social group that I thought of as the “nerd clique” at my college (a bunch of students from the Honors Program who shared a house, wore neckbeards, LARPed together, etc.) also was heavily into drinking and low-level drug use.

      Also, like you I feel like I became less nerdy over time, by virtue of acquiring/honing my social skills and gaining some general maturity. In particular, I used to actively pride myself in my geekiness, to the point that I deliberately exaggerated some of my geeky attributes. But now, I have started to see the general social utility of dressing nicely and being socially adept, and so I come across as lot less nerdy than I used to be, even though my interests haven’t changed.

      • nydwracu says:

        In particular, I used to actively pride myself in my geekiness, to the point that I deliberately exaggerated some of my geeky attributes. But now, I have started to see the general social utility of dressing nicely and being socially adept, and so I come across as lot less nerdy than I used to be, even though my interests haven’t changed.

        The nerd-memeplex contains things that damage social skills (“clothes should be purely utilitarian and people who worry about how they look are inferior”, for example), so some of it must be socially determined.

        I suspect that a lot of it is, and that most of the rest is due to the practice-at-social-interaction/difficulty-level thing mentioned downthread. I notice that the smart people I know who are from big cities are less nerdlike than the ones from other places. (Nietzsche talks about the ‘advantage of the ordinary’; in terms of social interaction vs. isolation, cities solve that by having more people in one place, and therefore more of most types of people in one place.)

  3. suntzuanime says:

    It could be that you don’t see metanerdery like you see metapoetry because nerds are deficient in the instinct to rhapsodize about their specialness to increase their status?

    I dunno, it seems like there’s plenty of non-scientific own-horn-tooting going on in the nerd community. Maybe it’s just all being done by the detestable colonialist Fake Nerds, who can’t do enough real science to produce the metanerdery you’re looking for.

  4. Andrew says:

    If there’s one thing nerds don’t like it’s looking in the mirror.

  5. Derek says:

    Many people claim that nerdiness is associated with autism, but such arguments are overly reliant on stereotypes and woefully deficient in their use of hyperlinks. Here I present a new explanation, one firmly grounded in the latest research:

    *Men with high testosterone tend to have deep voices
    *Men with high testosterone are violent
    *Men with high testosterone are muscly

    • Auroch says:

      There is a real pattern by which people who are unusually intelligent relative to their surroundings when young develop mild autism-spectrum symptoms. Or at least, so claims my therapist; he didn’t have a DSM entry so it might be his erroneous opinion.

      • nydwracu says:

        Wouldn’t surprise me, if these autism-spectrum symptoms are the ones that would be expected from people withdrawing from their social environment.

        My guess is that young children aren’t very good at modeling others, so they will have difficulty interacting with people nowhere near their intellectual level, which causes them to perceive themselves as low-status, social interaction as undesirable, and isolated exploration of their intellectual interests as the one preferable activity. This begins a ratcheting process: withdrawal from social interaction lowers perceived status further, lower perceived status leads to signaling of lower status and avoidance [and dislike, through a process Nietzsche probably described somewhere] of high-status activities [which, in America, includes athletic activities], and signaling of lower status makes social interaction even more unpleasant, leading to further withdrawal, and leads to even lower perceived status. The existence of the ‘nerd’ type in the general cultural consciousness only strengthens the ratchet.

        • Evan Gaensbauer says:

          This is the best hypothesis of the phenomenon of ‘nerds’ I’ve read in the comments thus far. Thanks for sharing.

        • Nita says:

          In my experience, I was bad at social interaction because my understanding and skills in this area were underdeveloped, not because the other kids were too dumb. There’s nothing smart about being so engrossed in your own thoughts that you end up oblivious to your environment.

          Sure, it feels nice to think that all of our deficiencies are rooted in our superior intellect, but let’s try to be a little realistic.

          • veronica d says:


          • nydwracu says:

            There’s nothing smart about being four years old, but four-year-olds are still smart enough to prefer interesting things to boring or painful things. If the people around are boring, or if interacting with them is painful (or even sufficiently difficult), then there’s going to be underdevelopment in that area, just due to lack of practice.

            I don’t think resorting to biology is necessary here, especially since I’ve seen nerds move to areas with better social environments for them and stop being nerds.

          • Lizardbreath says:

            People will use a strength as proof they’re worth something despite their weaknesses. That’s just human nature. It’s true no matter what the strength is or what the weakness is.

            This reality shouldn’t stop us from ever even considering the possibility that a given “strength” might be the cause or the result of a given “weakness.”

            The possibility of a (not-initially-social) difference from the norm potentially impeding the proper development of social skills…

            just follows from the (widely accepted, still disputed by some) theory that we predict others’ behavior by “putting ourselves in their shoes.”

            It’s a perfectly good theory. You’re just getting tripped up by the specific difference we’re discussing.

            I’d ask you to consider if you might be overvaluing IQ. High IQ doesn’t have to be unequivocally a good thing. We’re discussing a potential downside to it. Taking discussing that potential downside as being about bragging or self-soothing suggests an assumption that if a problem comes from high IQ, it must not really be a problem–as if “High IQ just has to be completely good.” It doesn’t.

          • I think almost everyone here believes that high IQ is unequivocally a good thing, so if you don’t consider that a valid assumption you’d better say so explicitly.

          • Lizardbreath says:

            @Taymon, I assume you mean the generic “you” since that’s what I just said. And by “said” I mean “spammed the thread about.” 😉

            Yeah, it seems like a lot of LessWrongers and SSCers

            * are really focused on “high IQ as unequivocally good”
            * show signs of having been damaged by some of the typical downsides of high IQ
            * have never picked up a book on the gifted and so never figured that out; even if they suspect, they have no confidence in this suspicion
            * are very quick to attack as “an obvious status move” even the feeblest theorizing in that direction…

            I dunno, as a kid I made the mistake of deciding I “didn’t believe in arrogance” and basically trained myself not to recognize any status moves (yes, yes it is crippling). So this might just be my typical overcompensation. But I definitely feel like LessWrongers are disturbingly(-to-me) quick to take everything as a status move.

          • We should draw a distinction between “higher IQ is better, all else being equal” and “in practice a lot of people with high IQ have some kind of other issues”. The former isn’t just an unexamined background assumption around here, it’s an explicit transhumanist ideal.

            The question is, does the causal structure of those other issues make it possible for a person to have a high IQ and not any of the downsides? I see no reason why such a thing shouldn’t be possible. But I don’t really know what’s going on there, so this might not be the case.

      • Lizardbreath says:

        Those who study gifted children have been publishing this observation for decades, from Leta Hollingworth to Miraca Gross.

        From Volume 3 of the Terman study, 1930:

        “Precocity unavoidably complicates the complexity of social adjustment. The child of 8 years with a mentality of 12 or 14 is faced with a situation that is almost inconceivably difficult. The higher the IQ, the more acute the problem…. If the [ratio] IQ is 180, the intellectual level at 6 is almost on a par with the average 11-year-old, and at 10 or 11 is not far from that of the average high-school graduate. The inevitable result is that the child of IQ 180 has one of the most difficult problems of social adjustment that any human being is ever called upon to meet.”

        In Gross’ still ongoing longitudinal study, started when her subjects were children in the 1980s (Terry Tao is one of them), she observed that among her subjects, being isolated from intellectual peers past the age of ten (IOW, no grade skips before then) was associated with long-term (still present when the group was age 25-33) social dysfunction. ERIC has a pdf of one of her updates.

        Here’s Gross’ summary of the earlier research (online reprint):

        “Hollingworth (1926) defined the [ratio] IQ range 125-155 as ‘socially optimal intelligence’…. She claimed, however, that above the level of [ratio] IQ 160 the difference between the exceptionally gifted child and his or her age-mates is so great that it leads to special problems of development which are correlated with social isolation….

        “DeHaan and Havighurst (1961), examining the differences between what they termed ‘second- order’ ([ratio] IQ 125-160) and ‘first-order’ (IQ 160+) gifted children, reinforced Hollingworth’s findings. These findings suggested that the second-order gifted child achieves good social adjustment because he has sufficient intelligence to overcome minor social difficulties but is not ‘different’ enough to induce the severe problems of salience encountered by the exceptionally gifted student.

        “Janos (1983) compared the psychosocial development of 32 children aged 6-9 with [ratio] IQs in excess of 164, with that of 40 age peers of moderately superior intellectual ability. The findings of Janos emphasized that the social difficulties experienced by this highly gifted group did not stem from a pre-existing emotional disturbance, but rather were caused by the absence of a suitable peer group with whom to relate. There are virtually no points of common experience and common interest between a 6-year-old with a mental age of 6 and a 6-year-old with a mental age of 12.”

        • Nita says:

          Thanks for the research! I do have two objections to the applicability of these findings:

          1) There seem to be a lot of nerds / geeks / weird smart kids. There are very few children with IQ over 160.

          2) As far as I can tell, “mental age” here means a certain level of performance on IQ tests. So, by itself it does not indicate anything about the child’s experience or interests (although the researchers seem to imply that it does, for some reason).

          And generally, the writings on “giftedness” I’ve seen so far tend to have a certain undercurrent of zealous adoration, which renders me unfortunately skeptical :/

  6. blacktrance says:

    I think alternative status hierarchies are part of the story. Someone who starts out as slightly more intelligent and less athletic and socially apt has a lower place in the main status hierarchy, but there’s also an available metric for an alternative status hierarchy (grades, and academic achievement in general), and they focus less on making up for their weaknesses and more on improving their strengths. Over time, the magnitude of the differences increases.

    This matches with what I observe of young kids and from what I can remember of being young myself. In elementary school, the nerds are less nerdy and the jocks less jock-like, in middle school these differences grow, and by the end of high school the two groups are largely segregated from each other and develop their own cultures.

    But this doesn’t explain why things like tabletop gaming and anime correlate with lower social adeptness and higher intelligence.
    Edit: On second thought, that fits too. Someone who likes those things has low-status preferences, and they leave the main status hierarchy for the alternative one.

    • Matthew says:

      In elementary school, the nerds are less nerdy and the jocks less jock-like, in middle school these differences grow, and by the end of high school the two groups are largely segregated from each other and develop their own cultures.

      This was not my experience at all. The nerd-jock distinction was quite apparent already in elementary school. Social segregation peaked in middle school. Things were actually somewhat better in high school.

      Notably, my middle school only tracked people by ability in math and English; you were forced into classes with everyone for science, social studies, and foreign languages. High school tracked everything. I think there may have been more willingness to voluntarily associate once the requirement to involuntarily associate was removed.

      • blacktrance says:

        Yes, there were jocks and nerds in elementary school, but this difference grows as elementary school continues. There are no jocks and nerds in kindergarten, at least. As for high school, what I remember is the various small and less strongly segregated factions from middle school coalescing into large groups strongly segregated from each other.

      • Evan Gaensbauer says:

        I believe part of this may be that youth become more mature, more conscientious, and overall nicer after passing through the far side of puberty.

        To quote nyan_sandwich in a conversation I had with him once:

        “People don’t become conscious until their 15. That’s when they realize that other people matter. Like, sometime when I was 15, I thought “wait, something’s different now.”

        The distinction between jocks, and nerds, may apparently peak between the ages of 12-15, when teenagers may be most likely to exploit differences between each other, and bully low-status figures, to gain status themselves. This dovetails with my experience in high school: from the age of 15 onward, everyone gradually grew disgusted with jerks, and bullies, or most of those bullies changed so they were accepted, or because they learned that empathy is important.

        In short, psychopathy, on average, becomes a costly stigma if noticed by the tribe, as people grow older. Teenagers are maturing, so they’re less aware, and more likely to be awful to each other, so they’re bad at disincentivizing being a jerk. Chalk up another coordination problem to Moloch.

        • suntzuanime says:

          I remember becoming conscious at 14, but I was always a prodigy like that.

        • Ghatanathoah says:

          I remember my mom explaining the Golden Rule to me when I was little. I couldn’t have been older than 8. I was remember being blinded by how much sense it made. I understood deeply that other people had feelings and mattered. So I’m not sure if 15 is the normal age for becoming conscious.

          • 27chaos says:

            If you think the golden rule makes sense, I’m not sure you’re conscious even yet. 🙂

          • Ghatanathoah says:

            What I precisely meant was that it made me understand that the justification for ethical behavior towards others was that other people have similar wants, desires, the same way you do.

          • anon1 says:

            The Golden Rule is fucking terrible as it’s usually explained to children. “Treat others as you’d like them to treat you.” OK, so then why do people get so upset when I actually act on that by not talking to anyone and generally pretending people aren’t there, even though I REALLY wish people would stop talking to me and pretend I wasn’t there?

            To a kid, it’s not at all obvious that it’s supposed to be interpreted in a more abstract way as “you’d like it if others found out the way you want to be treated and acted accordingly, so you should find out how they would like to be treated and also act accordingly.”

    • veronica d says:

      I think “alternative status hierarchies” is pretty on target and explains a lot. Surely it is better than the mantra (repeated here) that nerds “don’t care about mainstream status hierarchies” — which, folks, this community cares about those hierarchies a lot. I would go so far as to call it a preoccupation.

      • ADifferentAnonymous says:

        I think the pattern may be that nerds don’t experience mainstream status and thus never acquire a taste for it, except when puberty hits they notice that people with mainstream status sure seem a lot more successful at dating. This leads to the “weird-imposition-entrance-fee” view of mainstream status/normal social skills.

    • Anonymous says:


      Think about the first numbers, the tally system:

      I, II, III, IIII …

      Those are… drawings! Cartoons! Caricatures of an idea!

      Upon reading this, I thought “Oh, that would explain my inclination towards anime.”

      Armchair hypothesis: Anime’s audience mostly comprises the nerd demographic because nerds are (neurologically?) more receptive to abstract representations than the real deal.

      Let’s assume this were true. I’d expect to see a correlation between anime-fanaticism and the “iNtuitive” attribute on the Myers Briggs test. For starters, I am an INTP. Other anecdotal things I find significant:

      1. The guy I cited (Kalid Azad) is also an INTP (and presumably thinks in terms of cartoons).
      2. Azad’s abstract and intuitive thought processes regarding math perfectly mirrors my own.
      3. Feynman was likely an ENTP and thinks in terms of pictures too.
      4. ~90% of LW is iNtuitive. LW is also rife with anime references.
      5. INTP’s and INTJ’s don’t place much value in popularity and security.
      6. Children like cartoons the most. Younger people are better at old people at filtering out noise [citation needed].

      Item 5 sorta dovetails with the “alternative status hierarchy” theory. I’d expect that lots of nerds like anime to begin with, but additionally are less inclined to conform to the primary hierarchy.

      • Matthew says:

        Anecdotally, the only anime I’ve ever liked is Fullmetal Alchemist, despite being very definitely a nerd, at least in the past. Meanwhile, my MBTI is InTJ, with the lower-case n indicating that it’s very weak, unlike the other three.

        On the other hand, I really like abstract strategy games, so maybe pretty weak even as anecdotal evidence goes.

      • Nornagest says:

        I feel obliged to convey a theory I’ve heard that nerds like anime because it tends toward exaggerated emoting and characterization — big eyes and eyebrows, a well-developed repertoire of shorthand for emotional states, lots of fairly consistent stock characters, a penchant for slapstick and hamminess — and that makes it easier for nerds, being inexperienced and/or neurologically deficient in reading people, to follow the emotional content of a plot.

        (Note: I don’t actually buy this theory.)

        • Matthew says:

          This is a large part of why I don’t like most anime.

          I was never sure if the dialog/over-acting was as absurd in the original Japanese as it is in translation. I’m taking this as evidence that’s meant to be that absurd in every language.

          • Nornagest says:

            Little from column A, little from column B.

            Some does get lost in translation, and I get the feeling that the Japanese voice acting community is better developed than the American. But genre conventions really do lend themselves to slapstick and overacting and absurdist humor, and that’s true in any language.

            A contributing issue is that the anime most popular in the West tends to be, fundamentally, kids’ shows. Now, Western anime fans will tell you that animation in Japan doesn’t imply children’s media like it does over here, and they’re not wrong — but they are glossing over the fact that the most popular animated series on their shores, as on ours, are targeted at e.g. boys 11-13.

        • Elissa says:

          No but this is totally the only reason for Counselor Troi to exist.

      • suntzuanime says:


        technically incorrect: the worst kind of correct

      • blacktrance says:

        Abstract representations may have something to do with it, but I think it has more to do with the characterization and plot in anime. A young prince gains mind control powers and seeks to overthrow the empire that exiled him, while his best friend opposes him and tries to reform the system from the inside. An assassin struggles with the demandingness of utilitarianism, while opposed by a priest who gradually abandons Christianity for nihilism. A few nerdy college kids discover time travel and use it to change their own pasts. Western TV doesn’t have plots like that, and generally tends to be episodic rather than having story arcs. And even comedy anime is qualitatively different from the closest American equivalent – it also tends to have story arcs, and where American TV shows would have the main character be some moderately popular guy or girl, the anime will either have someone more nerdy or someone more generic (so viewers can insert themselves into his/her position).

        • suntzuanime says:

          I could totally see “nerdy college kids discover time travel and use it to change their own pasts” airing on western TV. It would surely end up looking a lot different in the specifics than Steins;Gate though.

        • nydwracu says:

          I don’t think I like anime so much as I dislike American television: at its best, American television is…

          …I don’t have a word for this and I don’t know if one exists, but it’s the thing that The Wire, Oz, True Detective, and Robert Anton Wilson’s fiction have in common. Jigsaw-puzzle plots driven by the interactions of different institutions/bureaucracies/secret societies/etc. and of different people within those institutions. Serial Experiments Lain is the closest anime to this, but I wouldn’t count it because it pays too much attention to one main character; the jigsaw-conspiracy genre rarely has one identifiable main character at all, and instead has sets of some small number of characters given the most screen time out of anyone in their institution.

          At its best, American television (American fiction in general?) is that, whatever it is and whatever you want to call it; at its worst, it’s 3edgy5me every-male-character-is-either-evil-or-an-idiot fuckity-realism that implicitly promotes and reinforces stupid and harmful beliefs and patterns.

          I liked the first season of The Wire (though not enough to watch the second season yet) and the first few seasons of Oz, though I wouldn’t recommend that anyone else watch them, on account of the Tumblrite metapolitical tics. True Detective tried too hard to be edgy, but it was alright when it laid off.

          (Aqua Teen Hunger Force is a notable exception. I like that. Most of Adult Swim’s shows just seem tedious to me. I’m not sure what’s going on there.)

          Anime tends not to adopt the grating edgy/ironic/~so realistic~ tear-everything-down-and-piss-on-the-remains stance that’s common in America [both in television and in the wider culture]; when it does (Gintama), it makes up for it by shooting straight elsewhere. I get the sense that it has values, a sense of the good and so on. And isn’t that one of the things art ought to be about?

          (It also has the advantage of being a lot easier to get without Netflix — I don’t have to worry about getting anti-piracy nastygrams. And I need subtitles for most American TV; I can’t understand what they’re saying half the time. ATHF is the exception there. Less background noise? True Detective was especially awful about that.)

      • nydwracu says:

        I don’t think in terms of pictures at all. ENTP. (Very weak counterevidence, of course.)

  7. Liskantope says:

    “Nerds” seem to share a bunch of seemingly uncorrelated characteristics. They’re generally smart. They’re interested in things like math and science, especially the hard sciences like physics. They’re shy and awkward. They’re some combination of bad at getting social status and not interested in getting social status. They’re especially bad at getting other people to show romantic interest in them. They’re physically unimposing and bad at sports. They don’t get in physical fights and are very unlikely to solve problems with violence. They’re straightedge and less likely to drink or smoke to excess (according to legend, “nerd” derives from “knurd”, ie “drunk” spelled backwards). Sometimes even very specific physical characteristics make the list, like a silly-sounding high-pitched voice.

    I would say that this is an accurate description of a number of stereotypical characteristics of a nerd, but not necessarily an accurate description of the actual characteristics prevalent in the nerd population. Of course, many stereotypes evolve from actual trends, and I think some of the above characteristics probably are probably significantly more common among nerds than among the general population.

    My main social group consists mostly of math graduate students, which I guess could be considered to constitute a different sort of nerd culture than, say, high school nerds, or the characters on Big Bang Theory (although I think probably all of us qualified as high school nerds back in high school). Of course, I realize that Scott and probably most of the people who write in the comments sections on this blog qualify as nerds in one way or another, and may have different experiences than mine. Based on my own experiences, here are my impressions:

    First of all, there is a big difference between “shy” or “introverted” and “socially awkward”. There definitely seems to be a tendency towards social awkwardness in my group (possibly for the reasons Scott suggests), but there is a wide range of introversion/extroversion. I myself am pretty socially awkward but have never been shy at all, and wouldn’t consider myself introverted (although I had one acquaintance tell me that I was clearly introverted simply by virtue of being a grad student — apparently he swore by this stereotype). That said, severe enough social awkwardness can lead to social anxiety, so I suppose there should be some correlation; it just seems pretty weak to me.

    If one judges by the trends in my group of math nerds, the notion that nerds are generally straight-laced and don’t get drunk is, shall we say, very, very untrue. In fact, socially liberal atheism seems to be the norm in math departments, and I would speculate that this contributes to a slightly lower-than-average rate of straight-laced behavior (at least in America). The stereotype of nerds being straight-edge probably comes from the association of high school nerdiness with obediently working hard and being too busy with schoolwork to get into sex or drugs.

    As for not getting into physical fights or resorting to violence, I have witnessed firsthand no fewer than three arguments between math people come to blows (all under the influence of alcohol). To be fair, the three incidents involved four people total, one of whom is really small and weak and lucky he was immediately pulled away both times.

    Is there really a correlation among men between nerdiness and having a high-pitched voice? I have a feeling that this is pure stereotype derived from how nerdy guys have been portrayed in movies or something.

    Nerds are traditionally viewed as having high libido – think watching pornography.

    I don’t know that this is portrayed as a sign of high libido. I always assumed that the porn-watching stereotype came from the notion that nerds are bad at romantic relationships and/or hooking up and are therefore sexually frustrated.

    • Matthew says:

      First of all, there is a big difference between “shy” or “introverted” and “socially awkward”….

      ….That said, severe enough social awkwardness can lead to social anxiety, so I suppose there should be some correlation; it just seems pretty weak to me.

      I’m the opposite — Introverted with severe social anxiety (CBT-d down to moderate social anxiety now), but not socially awkward. I don’t think these things are related at all.

    • blacktrance says:

      If one judges by the trends in my group of math nerds, the notion that nerds are generally straight-laced and don’t get drunk is, shall we say, very, very untrue. In fact, socially liberal atheism seems to be the norm in math departments, and I would speculate that this contributes to a slightly lower-than-average rate of straight-laced behavior (at least in America).

      In my experience, nerds are more likely to be socially liberal atheists and straight-laced. They’re not straightedge out of religious piety or adherence to traditional norms, but for some other reason.

      • Nornagest says:

        Wild speculation: nerds are more likely to take ideas seriously. Traditional norms generally don’t consist of explicit concepts, but anti-drug propaganda does.

        • Liskantope says:

          Hmm again I am speaking mainly from experience with my current large group of math nerds, but among them (as well in other socially liberal groups comprised of people in my generation), anti-drug propaganda efforts such as the DARE program are almost universally ridiculed.

          • Nornagest says:

            I’m not talking about specific propaganda efforts like DARE so much as the ideas behind them. DARE was indeed universally ridiculed, but at the same time, looking for authoritative information about drugs would usually lead you to the same knowledge base couched in more or less academic language.

            That probably grew less true in the mid-2000s with the rise of sites like Erowid.

      • Liskantope says:

        I guess I can come up with some suggestions as to why nerds may be straightedge while not religious. I feel like voicing them may expose my ignorance of how non-nerd social groups operate and may even come across as offensive to them, but here goes. I believe drug use becomes a more attractive activity in the absence of other engaging things to do. Nerds often seem to occupy themselves socially with games and very involved geekish pastimes. Meanwhile, non-nerds seem more likely to throw parties revolving around, say, watching football, which is easy to accompany with heavy drinking.

      • Levi Aul says:

        As a nerd, I thought I was “straightedge” for most of high school. In retrospect, I just didn’t like depressants (alcohol), psychotropics (mushrooms), or things that are obviously egregiously harmful to your body (tobacco), and that’s basically the set of what teenagers think of as “drugs.”

        The dislike of self-harm is self-explanatory (maybe? I wonder if that’s a typical mind fallacy too!) but the other two I tended to rationalize with “I enjoy spending my time in fully-conscious deeply-intellectual thought, and I don’t want to make my brain less good at doing it. Conscious thought is where all my fun gets generated!”

        Thinking back on it now, though, I think it might have been a sort of self-medicative-aversion: I had undiagnosed inattentive-type-ADHD. Imagining consuming depressants when your mind already has to fight to maintain executive control (and even stay fully awake, really) is sort of like imagining eating really gluey white bread when you’re already feeling constipated. Just kind of gross on a physiological level.

        Conversely, I would have gladly tried MDMA but for lack of availability in my area; and if cocaine wasn’t stigmatized to all hell, but was just a regular thing teenagers passed around, I probably would have ended up not thinking of myself as straightedge at all. I did consume tons of caffeine and chocolate, and still do (though now it’s more a habit than self-medication.)

        I’m not sure whether this was affected by the fact that I actually understood neurochemistry enough to know what was in each drug I considered, though, or whether there was actually some form of “neurotransmitter pica” at work. I would guess more of the latter applied to the high chocolate and caffeine intake, since I enjoyed those long before learning what they did.

      • veronica d says:

        Funny, in my experience there is a huge correlation between being nerdy and being kinky and polyamorous. (Which probably says a lot about me and less about nerds.)

        • Daniel Speyer says:

          I think straight edge mostly refers to avoiding drugs and maybe also avoiding rebelllion-for-its-own-sake. Kink and polyamory are a different thing. Maybe even negatively linked: since those depend on good communication and drugs interfere with that.

          • social justice warlock says:

            I think straight edge mostly refers to avoiding drugs and maybe also avoiding rebelllion-for-its-own-sake.

            Definitely not the latter; if anything it self-conceives as an instance of rebellion, and correlates with other rebel aesthetic choices.

          • Anonymous says:

            Yes, straight edge is punk, hence rebel, but it is against rebelllion-for-its-own-sake.

      • nemryn says:

        Hypothesis: Nerds tend to be ‘straightedge’ because drinking/smoking/getting high are often part of social or status-seeking rituals, which nerds feel little need to participate in.

    • Evan Gaensbauer says:

      As long as you’re protecting the privacy of your maths graduate peers, do you mind being more specific? Like, I would suspect that as socially liberal atheists, nerds would be more accepting of drinking, smoking, and promiscuity, but wouldn’t indulge in it personally themselves, per se. There comes a point where the acceptance of such “vices” will lead to higher rates of engagement, but I’m not confident most nerds hit that threshold.

      We might also acknowledge that there are different types of nerds. Also, as people enter college, being intellectual, or academic, becomes normalized, so the lines between nerds, and everyone else, become more blurred. So, as a former humanities/arts major, there are philosophy, or literature, majors who seem like nerds, but they seem likelier to drink heavily, or smoke marijuana, than math, CS, or physics majors. I attribute this to the even more socially liberal, outgoing bohemian ethos of the liberal arts.

      • nydwracu says:

        I don’t think there were nerds at my college (though I saw no signs of the existence of STEM majors besides math and psychology, so there could well have been); the people who would have been nerds elsewhere got absorbed into anime goth crew instead. Anime goth crew smoked and drank heavily, but didn’t do any other drugs, and they were about evenly split between polisci and art majors.

      • Liskantope says:

        I can describe the overall trends in my department, but I don’t want to imply any strong conclusions about nerds vs. non-nerds, because it’s been a long time since I’ve gotten real exposure to a social group not composed mainly of graduate students.

        Drinking habits range pretty much all over the spectrum. A substantial number frequently drink heavily, and a few seem to make a point of getting black-out drunk on a weekly basis. Beer pong and several variants are quite popular. I’m told that our math department drinks more on average than many departments, but that heavy drinking is more ubiquitous in some of the humanities departments (law school for instance).

        Pot smoking is also fairly common, and quite a few who don’t smoke now claim to have smoked frequently during college. Harder drug use is very rare. A fair number smoke cigarettes occasionally, if not regularly.

        Those who are romantically successful enough to get into relationships are assumed to be having sex. Some couples live together without being engaged or married; almost all the other couples are comfortably open about the fact that they are constantly staying over for the night. A number of guys at least talk about making serious efforts to have one-night stands, with very limited success. My impression is that, in the absence of religiously induced social norms, most are following their natural inclinations as far as sex is concerned (though careful about protection).

        Poker was pretty popular at one point, although none of us felt rich enough to gamble large amounts of money.

        In all of the above, I’m excluding most of the Chinese students, who tend to be more socially conservative and generally socialize among themselves anyway.

        • Evan Gaensbauer says:

          You’ve disconfirmed my sort-of hypothesis, so this interaction provided value of information. Thanks for responding.

  8. Hypothesis: For some traits that are normally distributed in the general population, nerds are more likely to be at either extreme, and less likely to be in the middle, than non-nerds.

    Some traits this might apply to:

    * Weight. Eric S. Raymond made this observation in Appendix B of the Jargon File, A Portrait of J. Random Hacker, which is probably worth reading if you’re interested enough in this subject to be commenting here.

    * Desire for sex (which I guess is a different thing from libido; I don’t know if this has any connection to the pornography thing). I’m pretty sure asexuality is strongly correlated with nerdiness, and while there are so many obvious confounding factors here that listing them would be pointless, I still think there’s something going on. Certainly asexuality is correlated with autism spectrum disorders. I have only anecdotal evidence for the opposite extreme, but I’ve seen plenty of evidence that lots of people in the Less Wrong-influenced memeplex are very interested in sex.

    * Tendency to use drugs. Scott mentioned the straight-edge phenomenon, and I offer myself as an n=1 sample. Meanwhile there are a whole bunch of psychonaut nerd communities. (This seems related to the “taking ideas seriously” thing proposed above. I abstain from all psychoactive drugs because I’m paranoid about my mental state and don’t want anything going in there and messing it up. Psychonauts, meanwhile, are taking seriously the more gradually disseminating ideas about how some drugs probably aren’t going to hurt you and in fact can lead to positive experiences.)

    * Tendency to use swear words. My own group of friends admittedly falls pretty much in the middle here, but I myself pretty much never swear (and my friends like to make fun of me for this), and I don’t think I’m alone. Certainly, the stereotype is probably on this end. Meanwhile, the most-frequently-swearing group of people I’ve ever met were the residents of a dorm at MIT.

    The second and third of those are notable in that different-from-the-norm attitudes to sex and drugs are characteristic of countercultures. This certainly explains a lot; people who don’t fall into the societal mean are more likely to join countercultures, and countercultures encourage explicitly talking about things that mainstream culture discourages talking about. (It could easily be the case that Less Wrong-influenced rationalists are exactly as interested in sex as everyone else, but talk about it more openly.) And geek culture, particularly the more “hacker culture”-type forms of it, certainly owes a lot to the counterculture of the 1960s. (Since my hypothesis is that nerds fall to both ends, I should note that there are conservative countercultures as well, even if we typically don’t use that term for them.) But this doesn’t explain the first or fourth point on the list.

    I don’t know what conclusion to draw from this, but it seems worth looking at. Man, nerd psychology/sociology is fascinating (and I’ve been interested in it for many years and frustrated that there hasn’t been more of it).

    Anyone think of any other examples? Or counterexamples?

    • suntzuanime says:

      Taking ideas seriously is the sort of thing that would tend to lead, in general, to extreme behavior.

    • Liskantope says:

      As far as swearing goes, I’m more or less like you in the fact that I swear very little, and my friends haven’t failed to notice this with some bemusement. Meanwhile, swearing at least a moderate amount seems almost universal in my math department (or in any socially liberal atmosphere I’ve encountered), and several of my fellow math nerds swear constantly and sometimes with very creatively profane turns of phrase. The general consensus among them seems to be that since swearing doesn’t directly hurt anyone or anything, there is no rational reason to avoid it.

      Side note: I would be curious as to whether swearing is as common among graduate students in, say, a linguistics department.

      • I actually tend to get angry when someone insists that other people not swear. Swear words are important; like all words, they fill a particular semantic role, allowing us to express certain things that we otherwise wouldn’t be able to. When someone tries to suppress swear words, they’re saying that the things which swear words serve to express, shouldn’t be expressed. Often, that they shouldn’t even be thought. Morally, it’s no better than the notion of thoughtcrime in Nineteen Eighty-Four. See also this by Paul Graham, about 2/3 of the way down the page.

        (One thing my interactions with the rationalist blogosphere have taught me is that this isn’t always what’s going on. Some people just have a strong negative System 1 reaction to swear words and consequently experience direct disutility from hearing them. Nowadays I try to be more respectful of this. See also this Less Wrong post written by Scott.)

        So why don’t I swear? Mostly because not swearing fits my self-image and is part of the general communication style that I try to cultivate. Plus, the less often you use them, the more powerful they are when you do.

        • blacktrance says:

          As someone who’s averse to swearwords, I fit the first description more than the second. It’s not that swearwords mean anything unique, because there are other ways to say what the swearwords refer to (when they’re not just exclamations), but that they indicate a certain mental state or feeling that’s bad to have, and that getting rid of that impulse is self-improvement that starts with catching yourself before you swear.

        • nydwracu says:

          I actually tend to get angry when someone insists that other people not swear.

          I don’t. No taboo, no usefulness.

          • Yes, but at the same time those who swear constantly/excessively wear down the taboo and thus eliminate the only thing that swearwords are good for. These days, it’s gotten so bad that you have to call someone an ethnic slur in order to get anyone to notice.

            (Extremely tasteless joke along these lines elided after my better judgement kicked in.)

          • nydwracu says:

            “Shit” somehow manages to serve double-duty as a swearword and a near-exact synonym for “thing”.

            In the long term, yes, but new ones will arise. I would be surprised if their existence is not a human universal.

        • Liskantope says:

          So why don’t I swear? Mostly because not swearing fits my self-image and is part of the general communication style that I try to cultivate. Plus, the less often you use them, the more powerful they are when you do.

          This expresses my attitude perfectly.

        • Matthew says:

          Quoting my explanation to the OKCupid question about foul language:

          I believe swearing should convey information, not just provide color. If I swear, I am conveying one of the following: 1)”Ow! I just experienced something very painful” 2)”That other driver just did something that almost killed me” or 3)”I am REALLY pissed off, and you should probably stay out of my way for a while.” 3 is a very rare occurrence. 1 and 2, maybe not so rare, since I am occasionally clumsy and most people are awful drivers.

      • AJD says:

        I have no particular sense of the grad students in the linguistics departments I’ve been affiliated with as either more or less likely to swear than the general population; but most of my non-linguistics social groups are pretty nerdy also, so I’m not sure I gave a good baseline for comparison.

    • Scott Alexander says:

      This kind of thinking worries me, because imagine if there were a trait nerds weren’t on either extreme for – for example, height. As far as I know, nerds aren’t any taller or shorter than anyone else. As a result, no one thinks of it and it never comes up. Then you start thinking “Wait, nerds are either on one extreme or the other for all the traits I can think of, this must be significant…”

      • Nerds aren’t on the extremes of all the traits I can think of, only some of them. And I suspect that those traits might have something in common that other traits don’t, and maybe this might tell us something interesting about nerdiness.

        But your objection is well taken and this kind of speculation is extremely fraught with availability bias.

    • Lesser Bull says:

      Alternate hypothesis: nerds end up excluded from popular groups in highschool and junior high, where popularity is correlated with and even defined by ‘adult’ behavior such as drinking, drugs, sex, and swearing. And either stick with the exclusion or overreact to it by going in for the “adult” behavior in big ways.

      Alternate, alternate hypothesis: drugs, sex, and swearing are attractive and dangerous/offputting for all sorts of reasons to all sorts of people and there’s lots of variance in approaches among nerds and non-nerds alike

  9. gattsuru says:

    Alcohol and drug consumption and overconsumption seem easier correlated to lack of opportunity : if don’t spend much time in large social groups with other drinkers or drug users, you won’t even see it to start with, nevermind be tempted to try it.

    If this theory is accurate, I’d expect to see the ‘straightedge’ phenona be less present where access to alcohol (over 21) and access to drugs (states with legal marijuana).

    That said, I think at least part of the explanation for the central issues is that nerd != scientist. Some nerds are scientists and most nerds are interested in science, but if you actually look at jobs, you’ll find much more technical support workers than folk getting published in academia. Many necessary attributes for academic scientists are directly opposed to nerd traits: some folk can write grants and get tenure while not enjoying social interaction, but it’s not the first option that comes to mind — and even if you could do it, the reward for surviving within academic sociology and psychology isn’t as great to folk uninterested in status.

    If you look outside of academia, though, there’s more than a fair share of navel-gazing. The correlation between nerd and Asperger’s Syndrome did not solely originate inside of psychology: it’s a common discussion on forums where nerds gather, to such a degree that a lotta folk have developed unfortunate amounts of slurs for self-diagnosed Asperger’s. There’s no shortage of specialized nerd jargon, nor use of traits like lack-of-aggression, disinterest-in-social-status, and intelligence to make common nerd desires not only personally valued by Obviously Correct.

    • Daniel Speyer says:

      If this theory is accurate, I’d expect to see the ‘straightedge’ phenona be less present where access to alcohol (over 21) and access to drugs (states with legal marijuana).

      Disconfirmed, in my experience. Nerds over 21 I’ve known in New York are only slightly more likely to use alcohol than marijuana.

      The both extremes effect is present here. Those nerds who do drink often maintain database-backed liquor collections.

  10. Anonymous says:

    Nerdiness certainly does exist where I live in Europe.

    Who says that it doesn’t exist in Europe?

    And it doesn’t exist in Japan? I wish Scott expanded upon this.

    • Scott Alexander says:

      I only worked in Japanese elementary schools, so my knowledge of middle and high schools is secondhand and to some degree based on TV/media. But my impression is that Japanese kids really value intelligence, to the point where somebody who can, say, do math way above their grade level is more likely to be looked up to as an amazing person than to be bullied for it. There’s also a lot more respect for people with single all-consuming interests (for a very stereotypical and terrible example, think of the classic sushi chef who devotes his life to becoming perfect at sushi-making)

      • Anonymous says:

        There is one Japanese anime called “no game no life”. The protagonists are two siblings, a teenage boy and a little girl, who are incredibly intelligent and ridiculously invincible at “games” (ranging from MMORPGs to chess), but are pathetically devoid of social skills, with no social life whatsoever. They remark that real life is the only “game” they don’t understand.

    • MugaSofer says:

      >Who says that it doesn’t exist in Europe?

      Well, I live in Europe, and I’m pretty sure I would have no concept of a “nerd” if it weren’t from seeing fictional depictions.

      I’m really not sure that nerd/non-nerd contains any information beyond “low-status + intelligent”.

      I know plenty of people who are reasonably intelligent, and plenty of people who are reasonably low-status, and even some who overlap … but I’m not really seeing a distinct cluster.

      And I’m in Secondary School (High School, to Americans.) That’s where the nerd/jock divide is supposed to be at it’s most clear-cut and brutal, right? Where we’re all supposed to be isolated and bullied?

      I’m not saying it definitely doesn’t exist, but nerdiness as an arbitrary stereotype invented by US highschoolers … would certainly fit with my observations.

      • Nornagest says:

        n=1, but when I was in high school nerds formed a distinct clique, but generally weren’t bullied or actively ostracized. (They were, however, usually dateless.)

        That’s not to say that I never saw any bullying directed at nerds, but it seemed to form a sharp peak a couple years before high school: age 11 to 13. Some of this might have to do with changing disciplinary policy between schools, though.

        • Anonymous says:

          Here in Italy, I was bullied in junior school.

          In high school, I and other nerds were dateless and without friends other than each other, but not actively ostracized.

          • Anonymous says:

            I meant middle school, not junior school.

            It’s difficult to match school types between different countries. I mean the school where students are 11-12-13 years old.

      • Anonymous says:

        Where in Europe do you live, MugaSofer?
        While many European countries may not have a well developed concept of nerd as they do in the English speaking world, the reality of nerds I believe exists.
        It’s like gay people didn’t exist as a category until recent times, but they certainly did exist in reality.
        My experience here in Italy has been that nerdiness certainly exists.

        If we describe the nerd thus:

        1) socially inept, no girlfriend, hardly any friend other than fellow nerds, has been bullied at some point
        2) high intelligence invested in weird, geeky, obsessive interest, which often revolve around computers, role-playing games, or books.

        This has been a visible pattern in my experience over here.
        Granted, there are few of these people, and it’s possible to miss the pattern simply because you didn’t happen to come across or get to know enough of them (which can happen easily precisely because they aren’t good at making friends). But if you’re one of them, as I was back in my school years, you’ll probably end up crossing paths with several others.
        Note that simply being a good student or good at math doesn’t make you a nerd. The essential requirement is to have the obsessive, geeky interests which are associated with social clumsiness and isolation.
        Also, the bullying thing seems to happen more often in elementary or junior school than high school. I was bullied in junior school, not high school.

        • Anonymous says:

          I meant middle school, not junior school.

          It’s difficult to match school types between different countries. I mean the school where students are 11-12-13 years old.

    • Intello says:

      In Belgian high school (the one where I went at the very least, and I have no reason to believe it works otherwise elsewhere), nerdiness is a thing, and you can be bullied because of it. The term used was “intello” (shorthand for “intellectual person”, with negative connotations of being too absorbed in intellectual pursuits, not sociable, a teacher’s pet, etc.).

  11. Precisely *because* the standards are different across time and geography, I would expect a more social-conditioning explanation for this. Consider, for example, the era of two-fisted pulp scientists in fiction, who were brilliant geniuses AND athletic, etc. It’s only later that we see the more scrawny nerd type in common usage.

    I think it’s probably a sort of attractor in the vector field that is social pressure. That and people pigeonhole themselves into roles and stereotypes.

    • nydwracu says:

      I’m surprised no one has written a genealogy of ‘nerd’.

      (When and where did it originate? Did anyone think it was remarkable that the people at Bletchley Park played tennis? A biography of Turing that ten seconds on Google turned up says he was interested in “rowing, mountain climbing, hockey, tennis, cycling and long distance running”. My guess is that it came from ‘mad scientist’ to some extent, but that’s only a guess and is probably wrong.)

      • Kzickas says:

        I’ve heard it’s an originally american concept originating in stereotypes of jews and asians, but I’m not direcetly familiar enough with american culture to judge the accuracy of that.

      • lambdaphage says:

        Once you spend any amount of time perusing academic homepages, you start to notice that practically everyone is into hiking/mountaineering. Explanations:

        1) Nerds really like mountain climbing.
        2) Mountain climbing seems like a tastefully wholesome, achievement-oriented hobby for an academic to have. Hence everyone pushes themselves up at least one mountain for the sake of keeping up appearances.

        I am genuinely torn between these two hypotheses.

        • social justice warlock says:

          (1) precisely because (2) seems like a perfectly likely possibility.

        • Matthew says:

          People who don’t like hiking are the strange ones. Nerds liking hiking is just a subset of people liking hiking, because modern people spend too little of their time exposed to greenery. It’s well-documented that walking in the woods is psychologically beneficial.

  12. pwyll says:

    It’s only my anecdotal observation, but the older I get the more I suspect that autistic/aspergers-y tendencies and high intelligence are *not* that highly correlated. To a degree, both can give you success in STEM fields however, which might be why they’re so often conflated.

    As for the tendency to be less religious, I think that’s much more closely tied to the autistic side of things. This fascinating book, written as if it were a first-person narrative by an autistic boy, includes some typical autistic arguments against the existence of God: (Google searches for autism and atheism have more if you’re interested.)

    • Christopher says:

      [My intuition is that the concept of the nerd is entirely rooted in childhood and specifically school, and so] it might be that such kids as participate less in the social aspects of going to school are thereby less distracted from the actual education and undertake it more seriously, resulting in the overall apparent correlation between being a nerd and being smart? I’ve certainly known people labelled as nerds who weren’t strong academically, earning their name instead by being obsessively interested as a group in comic books.

  13. Richard says:

    On the other hand, some of the Europeans I’ve talked to say that the experience of nerdiness on their side of the Atlantic is very different from the American experience,

    People who are bad at social skills might not get invited to sports games and so have no opportunity to improve; they might take up unpopular but solitary pursuits like math as a result.

    These two sentences explained something that has been idling in the back of my mind for ages. I am a Norwegian nerd and I hang out with other Norwegian nerds. Among these nerds, there are precisely zero people who don’t run marathons, ride triple centuries on bicycles, or regularly go cross country skiing in the mountains.

    In fact, the Norwegian nerd would fit the description of being highly intelligent, introvert and athletic.

    The thing is that Norway has a practically zero tradition for team sports, but a very strong tradition for individual sports. (Just look at how our soccer team does compared to our skiers…)
    The whole notion of having a ‘track team’ does not even compute over here. If you run, you bloody well do it on your own or with close friends.

    So, as StephenMeansMe said above: I would expect a more social-conditioning explanation for this.

    • zslastman says:

      From germany, can confirm similar phenomenon. When I moved here I found that almost all the nerds were into solitary athletic things like cycling or running.

      • von Kalifornen says:

        Such of my university friends as were nerdy, about half of them scorned exercise beyond that needed to maintain health and vigor, and the other half were into solitary or communal but not team sports.

        • Fazathra says:

          Can also confirm. Although a lot of I and my nerdy friends profess to hate sports, that is only due to being forced to play rugby in high school. Now we’re at university, a lot of us take part in solitary sports like running or swimming or shooting while racquet sports like tennis, badminton, or squash are also popular. I think it is less to do with hating sport in general, but rather hating team-based physical contact sports which are the norm throughout highschool

      • Niall says:

        I like cycling AND running! I also used to play 5-a-side football but stopped due to the inconveniences of coordination and distance to travel. That’s maybe another reason actually, since there’s an element of independence in nerd-culture as well – if I do all this other nerdy stuff on my own, I am less likely to tolerate the inconveniences of organising a group of 10 when I could put my trainers on and be out the door in 30 seconds.

        I thought the big difference between Europe and US with sports was the lack of a dedicated sport scholarship track. In Scotland/UK at least, if you’re good at academic stuff you go to university. If you’re good enough to be a professional at some sport then your progression is going to be through sports clubs (eg youth teams for premier league football clubs) and college/university is optional or discouraged. If you’re just good at sports but not professional level then no big deal really – you’re cool amongst peers but not like it’s going to do much for you getting a job.

        Disclaimer: what I know of US jocks and nerds comes from popular culture rather than experience.

        • Niall says:

          Ah I had another point to make – it’s easier here to be successful academically without the extra effort needed by US colleges, eg there’s none of this AP stuff, and extra-curricular activities don’t count for much here. I remember reading about 5.0 GPAs in horror, like some arms race gone madly out of control.

          Example: I got 4 unconditional university offers based on high school grades alone, and most of my peers would say something similar. I didn’t join any clubs or do anything out of the ordinary for this, so you could excel without anything much in the way of sacrifice, and so that reduces the need for extreme academic effort and our nerd type reduces down to the poor social skills, poor fashion sense and not strictly linked to intelligence.

          So we don’t have jocks in high school, and we have different calibre of nerds, but it’s still hell.

  14. pwyll says:

    I am reminded of an observation common among transsexuals – and brought up in Ozy’s last post – that there is a distinct cluster of transwomen who have certain very traditionally-considered-male-gendered characteristics and are very nerdy.

    Steve Sailer has been doing the most work exploring and documenting this phenomenon, e.g. here: (there’s plenty more in his blog archives.)

  15. Zanzard says:

    Hi Guys. First time commenter here. Long time fan of the Blog.

    The discussion about “the science of nerds” brings to mind the writings of Nicholas Nassim Taleb. He’s the author of “Fooled by Randomness”, “The Black Swan” and “Antifragile”.

    In Antifragile in particular, Taleb makes a passionate case in many instances about “nerdism” being a terrible thing.

    Has anyone else around this blog also read “Antifragile”? Or, for that matter, any book by Nassim Taleb?

  16. CaptainBooshi says:

    Is being straightedge really considered part of being a nerd? It’s not something that would come to my mind as nerdy, and having been through 4 years of college and 2 years of grad school in physics, it’s definitely not supported by any of my real-life experience.

  17. lmm says:

    Those aren’t necessarily different theories, in that there’s a hypothesis that autism is caused by prenatal testosterone levels.

    I think a disdain for social science is a trait I’ve seen among nerds. Physics or stamp collecting and all that.

    Those females who are identified as nerds tend to have far fewer nerdy traits than a typical male nerd. This could just be that the threshold for identifying someone as such is lower because the whole phenomenon is much rarer in females generally.

    • suntzuanime says:

      I always saw “physics or stamp collecting” as a slur on the life sciences, not the social sciences. The “stamps” being different species of slug that zoologists were always getting excited to discover even though they looked to a physicist very similar to the past twenty species of slug that the zoologists discovered.

      • Peter says:

        Or even the chemists, who have long since learned that their role in life is to be ignored in skirmishes between physics and biology. The quip is Rutherford’s, and in revenge, they gave him the Nobel Prize for Chemistry. He was apparently irritated by this. Then he died, and they named an element after him – an entry on the Periodic Table, it even looks like a stamp!

      • Kzickas says:

        That and as Peter says on chemistry. But I’m not sure it can really be called a slur, but rather it was an accurate statement at that particular time. Before we knew about genetics or atomic theory those fields really were more about gathering data than doing science.

      • Matthew says:

        “physics or stamp collecting”

        Should I recognize this reference?

        Sort of apropos, the closest Russian equivalent of nerd is ботаник (literally, “botanist”).

  18. BFNA says:

    One thing of note is that there’s been a long term campaign by incompetent and stupid psychologists who can’t do undergraduate math to suppress the personality metric with by far the highest predictive validity of any system yet developed – the Myers-Briggs.

    • Predictive of what, exactly?

      (Obligatory link to Scott’s post on the MBTI, in which he argues that predictive value isn’t really the point. I’m somewhat sympathetic to this position, but it’s quite different from asserting that MBTI is better than the Big Five for scientific purposes.)

      • BFNA says:

        Pretty much everything any personality instrument anywhere has been considered to be predictive of, like atheism, crime, interests in various arts, careers.

        Of course MBTI also has high mathematical reliability compared to other personality instruments and it’s actually predictive of the explicit things people intuitively think a personality measure is supposed to be about, routine behaviours and social interaction. Being predictive of everything else is just impressive.

        • I have literally never heard anybody claim this before. Source?

          • BFNA says:

            You’ve never heard of anybody saying MBTI is predictive of people’s career choices, or anything like that? I certainly did not come up with that myself.

            I’m really a bit too confused where you are coming from to even start. Like, it’s completely understandable if you haven’t personal experience with people going on about the MBTI in real, non-Internet life, not an American for instance, what you’ve heard of but I understand asking people just to Google around is a little impolite.

            I just don’t know what to say that I don’t want to be blindsided against a position in an argument like the entire concept of introversion and extroversion doesn’t exist and is pseudoscience. There is lots of multidisplinary research on personality, what things are at least claimed to correlate with each other, anthropological work across cultures and such. Without understanding what you’re really asking I don’t want to jump into some philosophical discussion on what the purpose of psychometrics is supposed to be or why we should care about predictive validity outside of particle physics or something.

          • I’m familiar with MBTI; in fact, as an undergraduate I worked with a sociology professor who used it as a serious psychometric. I fully believe that it measures something, even if some of the theories that purport to explain it are likely bogus and some of the uses it gets put to are no better than astrology.

            The claim I’d never heard prior to your first comment was that, out of all the personality inventories that have ever been developed, MBTI has the highest predictive validity. Generally the Big Five model is recognized as being empirically better, notwithstanding the various reasons you might care about things other than predictive validity.

            Incidentally, extraversion/introversion is recognized as an explicit factor by both MBTI and the Big Five model, so it’s not a good example. A better one would be MBTI’s judging/perceiving category; again, I can believe that there’s something going on there, but I find it implausible that that particular metric really means everything that people say it does.

        • Anonymous says:

          What is “mathematical reliability”?

          • BFNA says:

            Anonymous commenter, I was trying to clarify that I mean measures in discrete statistics as commonly used in psychometrics in general, as opposed to a gut feeling or intuitive use of the word. Test-retest correlations and the like.

            Even the poorest and most critical assessments of test-retest on subjects with MBTI put it at 0.5 or above. It’s definitely and notably more stable with age than most other psychometric instruments in the literature, even if you could always cherry pick something out there for an unfavorable comparison including proprietary data which is unfortunately common in the field. Despite misperceptions that the general public may hold because psychometrics as a whole are notoriously unreliable MBTI is overall comparable with anything else in that sense; it’s not like childhood and adult IQ scores don’t only correlate at 0.5

            Cronbach’s alpha for MBTI items (not on made-up, unofficial internet questionnaires) is at least 0.70 through many iterations.

    • Scott Alexander says:

      I like MBTI more than some people, but you’re going to have to back that up or else the way you phrased that statement is pretty close to a bannable offense.

      • BFNA says:

        Ok, this is funny in that I could not for the life of me figure this out for the longest time. It’s entirely my bad as I can see how with this blog’s community it would not be implicitly understood whom I was referring to. I had entirely forgotten that Scott considers himself a psychologist! That this was coarse but would also run into some sort of professional pride thing. I know that this is some sort of previously unrealized personal bias showing in not personally thinking of MDs and PhDs as the same group in any field and assuming the norms of another internet community. Didn’t mean to go too far to malign other professions.

        What I definitely should have said is
        “psychometricians.” A subset of psychologists but I am referring to the Arthur Jensen types. And I know the audience here is familiar enough with the subject to know that plenty of people in psychometrics have spent decades long careers trying to show that, “IQ (g factor) is the only thing that matters,” often disparaging other psychometric work. I can be back and provide plenty of citations on that if you think it’s necessary, to be clear I think the point in the sentence above is very strongly supported.

        I don’t know what Scott wants discussed on the topic or open up tangential arguments with the neoreactionaries. I don’t expect anyone to take a position of some “third way” and not consider all sorts of those publications of IQ researchers “real” peer reviewed journals or anything. One can be assured the assertion is that personality metrics are a much better predictor of whether someone is a lawyer or an engineer than IQ scores, that complaining in a philosophical sense about the purpose of predictive validity for personality measures and not for IQ is in my view senseless and so on. I just felt it’s pointless to start a discussion on why the state of science is as it is on a subject like the OP without acknowledging that the IQ research crowd have been pretty effective at suppressing related cognitive and personality research for decades. (Some of this of course is second order feedback over prestige and public controversy for the field of psychometrics as a whole)

        Anyway if there’s going to be some concern about decorum and martyring and such, it can at least be shown to all the neoreactionaries here who complain both sides aren’t treated fairly and you’re easy on leftists. If that comes up I’d just ask that in the future people can cite this thread for someone suggesting that Arthur Jensen had to be, at the least, incompetent.

        • Douglas Knight says:

          If you’re interested in clarity, you should know that your original comment made it sound like you were accusing OCEAN supporters of being bad at algebra, etc.

        • Creutzer says:

          Wow. You’ve now managed to write three long responses to people without citing a single source for anything in any of them.

  19. memeticengineer says:

    It seems like many traits of nerdiness are likely to correlate to the high-Systematizing brain per Empathizing-Systematizing theory.

    This could explain:
    – Generally smart (IQ tests seem much more loaded on systematizing than empathizing)
    – Interested in things like math and science, especially the hard sciences like physics. (Plenty of systematizing there)
    – Deficit in social skills or interest (lower empathizing than systematizing)
    – Bad at getting other people to show romantic interest in them (same)
    – More likely to be male (high-S personalities more common among males)
    – Significant intersection with autism spectrum disorders (high-S personalities much more common among autistic people)
    – Take ideas seriously
    – More likely to be atheists (consequence of thinking things through systematically), or if religious, to take religion somewhat seriously (converts to Catholicism, etc)
    – Obsessive interests (especially in things that contain or are systems)
    – Interests less likely to be socially acceptable / high-status (lower empathizing)
    – Unusually low or unusually high interest in things like drugs, sex, swearing (consequence of conscious decision over going along with the socially normal level)
    – Interested in games
    – Interested in fiction that involves systems (e.g. of science/technology/magic)

    More speculatively:
    – Maybe high-systematizing implies better impulse control, which would explain less use of violence.
    – Maybe low-empathizing implies less interest in / ability to join group physical activities, thus worse/less interested in team sports.

    High-systematizing is proposed to be correlated with higher fetal testosterone levels. Maybe the neediness cluster is a combination of high fetal T, low T thereafter, and an evolved subculture which attracts people who partly or fully fit the stereotype.

    Postscript: I feel like Scott must have heard of the E-S theory at some point, so the fact that he didn’t mention it at all raises my prior that it is total BS for some reason I’m not aware of.

    • Viliam Búr says:

      From the Wikipedia article:

      children with autism are delayed in their development of a theory of mind, that is, the ability to understand the thoughts and feelings of themselves or others

      Seems to me there could be two different causes creating similar effects, contributing to the confusion. The effect is, approximately, being bad at predicting other people, and all things that happen as a consequence of it, such as being bad at politics, or problems with dating.

      One cause is autism: somehow, the “modeling people” part of the brain does not function properly.

      Another, completely different cause is extremely high intelligence. In this case, it is not the brain function per se that doesn’t work; it is a wrong environment. More specifically: an environment full of people who are not like you.

      To understand this, we have to look more closely at how do people understand others. The natural way is to use our own brain as a black-box model, and simply ask ourselves: “What would I do if I were in such and such situation?” The quality of prediction depends on how much you understand the situation of the other person… and how much the other person is similar to you.

      If you are surrounded by people similar to you, understanding their situation makes your predictions better. If the correct prediction brings rewards (either by better cooperation with others, or by better abusing them; social skills are value-neutral), you are naturally motivated to try understand them more and more; thus, developing social skills. But if you are surrounded by people too different from you, your modelling based on “what would I do?” will fail repeatedly, regardless of how good understanding of their situation you develop. So you will either give up, or learn to rely on explicit models… which is complicated, because they take more computing time, and are often not available, or even worse, you are under a strong social pressure to learn wrong models. (The average people strongly compartmentalize here; they profess belief in these models for status reasons, but they don’t really use them. The models are not optimized for being used seriously.)

      So the difference is that when an autistic person meets more autistic people, the person still remains autistic. But when a highly intelligent person meets other highly intelligent people, it’s the “oh, finally someone who makes sense!” moment, and the social interactions may start to develop naturally. However, if this happens too late in life, they have already lost a few decades of time, and they have acquired a few bad habits. (And likely, so did the people they interact with. Some of those bad habits may translate to bad social norms.) So the problems are not immediately magically fixed.

      People often suggest that highly intelligent children should spend their time with average children, because they will have to deal mostly with average people in their life, anyway. I believe this is completely wrong and actively harmful. You start learning skills on easy level, not on hard level. For a highly intelligent child, easy level is interacting with other highly intelligent children. (Sometimes older children can be a substitute, but it’s not the same.) Starting with this easy level, and only then progressing to the hard level would in my opinion bring better social skills (even towards the average people). But of course, this suggestion is completely politically incorrect, so… the next best option is probably groupping children by something that acts as a proxy for intelligence.

      • memeticengineer says:

        Interesting theory. Some observations that it doesn’t seem to explain:

        * There are people who show many traits of the nerd cluster, but are neither of unusually high intelligence nor diagnosably autistic. You can often find many of them in media or gaming fandoms. So by the theory there would have to be a third reason for being bad at modeling people which is yet a third explanation. Note though that many people in this category do get a sense of having “found their people” at events such as science fiction conventions.

        * Children of high intelligence who are tracked into special programs do not seem to end up less nerdy. I guess since nerditude as a phenomenon is not very well studied, it is possible that there is a statistically significant difference, but it’s not so dramatic as to be obvious to outside observers.

        * Autism spectrum disorder appears to be a continuum, and further there is not a sharp discontinuity at the boundary. Many people who would not be diagnosed with ASD show significant autistic personality traits, sometimes referred to as Broader Autistic Phenotype. This can include such traits as aloofness, rigidity, anxiousness and hypersensitivity. These traits seem in line with nerd stereotypes. Some studies show that parental BAP is correlated with autism.

        An alternate explanation that may fit the evidence better: instead of multiple independent causes, perhaps some of the heritable personality traits that comprise Broader Autistic Phenotype are a common factor in nerdy traits for autistic nerds, non-autistic highly intelligent nerds, and non-autistic average intelligence nerds.

      • Nita says:

        I think you are conflating intelligence and weirdness. Are all high-IQ children really behaviorally different from average kids?

        Also, I’ve seen some autistic people say that they feel pretty comfortable around other autistics, better than among neurotypicals.

        • Matthew says:

          I’m not sure if this metaphor will stretch to incorporate social situations, but anyway….

          When I used to play Go a lot, I made the following observation. When someone was 1 or 2 stones weaker than me, I could see them make identifiable mistakes as we played, and punish them. When someone was 1 or 2 stones stronger than me, I could understand their strategy, and see (just a bit too late) how they made slightly better choices at the margin.

          But when someone was 3 or more stones weaker than me, it almost seemed like they were putting stones out at random — it didn’t look like they were slightly worse at executing a strategy; it looked like they didn’t have one. When someone was 3 or more stones stronger than me, it seemed like they were winning by magic, because I couldn’t follow their strategy either.

          People far enough apart in intellectual capacity* may seem to each other not to be playing the same game, if you will. Large difference in intelligence is not really distinct from weird if it leads to mutual incomprehensibility.

          *Intelligence is actually worse; you can get better at go.

        • Lizardbreath says:

          There’s lots of research that yes, they are. Quick summary from a presentation by Miraca Gross:

          Intellectually gifted children differ from their age-peers of average ability not only in their cognitive capacities but on virtually every socio-affective variable yet studied (Gross, 1993; Silverman, 1993).

          * Their play interests tend to be those of older children

          * Their reading interests tend to be those of older children….

          * They often display an early and quite passionate concern with ethical and moral issues, which more usually appears in children some years older (Hollingworth, 1942; Silverman, 1993; Gross, 1993)

          * Their conceptions of friendship are those more generally held by children some years older….

          [A]t ages when children of average ability are still choosing friends on what Selman (1981, p. 251) calls the “fair-weather-friends” basis of similarity of sporting or play interests (when the shared interest fades the friendship cheerfully dissolves), their intellectually gifted age-peers have already moved on to conceptions of friendship in which friends are perceived as people who will understand the way they feel, people to whom they can talk about their deepest feelings, and people who will accept them as they are rather than expecting them to adopt social masks (Gross, 1998).

          These are the friendship conceptions more usually held by children some years older and it is one of the reasons why gifted children very often seek older students for companionship and friendship.

          IOW, as kids, they’re more like others of their “mental age” (that is, level of cognitive complexity easily handled) than others of their “chronological age.”

          But they don’t stop growing and changing when, as chronological kids, they reach the “mental age” of the average adult. They keep changing until they’re actually chronological adults…by which point they’re weirdos 😉 (that is, they’re different from typical adults).

  20. Sniffnoy says:

    It’s come up several times in the comments now, but I’m surprised you managed to get through all this without ever mentioning the notion of “taking ideas seriously”.

    I mean, there are a number of nerd traits/stereotypes you haven’t mentioned, simply out of necessity; there’s no way you could list everything, and your prototypical nerd is presumably different from my prototypical nerd is different from etc. But this is an important one!

    When Robin Hanson, Razib Khan, Nassim Taleb, Sark Julian talk about “nerds”, this is what they are talking about! This is perhaps a nonstandard use, and also at least with Taleb and Sark (I haven’t read Taleb, but AIUI Sark Julian largely echoes what he says) there is a bit more to what they mean by “nerd” beyond just “takes ideas seriously”, but it’s an important notion, and it does seem to be at least somehow related to “nerdiness” in the more usual senses. Robin Hanson also refers to it as “Smart Sincere Syndrome”. (Search OB for “nerd” or “smart sincere” for more. Also, here’s a relevant Razib Khan post.)

    (Note in particular that in this model, the social parts are largely a consequence of nerdiness, and taking seriously particular common ideas.)

    There’s a comment I remember seeing on Less Wrong, I think by PJ Eby, though I can’t find it now, about the common nerd notion that it is wrong to attempt to make a good first impression. (A notion that nerds independently reinvent, not one they copy from each other — while I certainly immediately recognized myself in the description, I don’t think I’d ever seen anyone other than myself mention that idea explicitly before.) I feel like this is (in our culture, anyway) a good one-way test. 🙂

    …I will write more on this later, I had a whole blog post I had been intending to write on the subject, but I wanted to get at least one comment out of the way quickly before going into more detailed speculation…

    • Sniffnoy says:

      Note, by the way, that if you go ahead with the assumption that “taking ideas seriously” is a key feature, then you’ll see a lot of variation betwen nerd-clusters, based on what particular ideas they happened to take seriously.

      (I guess I said this implicitly above, and suntzuanime kind of mentioned it as well, but I thought it was worth stating more explicitly.)

      • Zanzard says:

        I’ve seen people criticizing nerds by saying that “they take ideas way too seriously”.
        Emphasis on the “way too”.

        I’ve never seen anyone refer to them as “people who takes ideas seriously”. This phrase, without the “way too”, sounds positive, sounds like something I’d hear from someone defending or praising the nerds.

        • Sniffnoy says:

          I’m not sure what your point is. For what it’s worth, if you take a look at the links, Razib Khan is pointing out that taking ideas seriously can lead to religious fundamentalism, and Robin Hanson makes the note that terrorists also fit the nerd-pattern. His other posts about nerds are often of the form “here is an example where nerds take an idea seriously even though it isn’t meant to be, leading them to do stupid and disadvantageous things while other people laugh at them.” And while I didn’t link them, Sark Julian certainly has little nice to say about nerds!

          But now let me say why (as a nerd) I find the idea of taking ideas “way too seriously” to be kind of infuriating. It’s basically just a matter of “say what you mean” — if you didn’t mean the idea to be taken seriously, why did you say it? Well, no, that’s not the truly annoying thing. The truly annoying thing is the idea that one can just reply with “You weren’t supposed to take it so seriously” or “Well of course you shouldn’t do it then“. An honest response (assumign the other person is in fact right) would be modifying the idea to correct the problem, or, since doing that is likely to be hard, just noting that yes there seems to be a problem but admitting that you don’t know the fix for now.

          (Scott Aaronson once contrasted bullet swallowers with bullet dodgers; but in my experience, most people don’t honestly swallow or dodge bullets, they say unhelpful things like “you weren’t supposed to take it so seriously”.)

          Saying “you weren’t supposed to take it so seriously” is like “no true Scotsman” — you’re redefining the terms, moving the goalposts, in a way that’s opaque and unchallengeable, and completely unhelpful for anyone who wants to actually act on what you say.

        • The phrase “taking ideas seriously” is a Less Wrongism that originated in this post, which presents it as a good thing. However, I suspect that a lot of people using it in this thread learned it from this essay by Scott, in which he argues that it’s not always good.

  21. Crowstep says:

    I feel like Paul Graham’s ‘Why nerds are unpopular’ essay can help shed some light on this.

    His hypothesis is that intelligent kids in high school are so distracted by smart-kid stuff that they cannot devote enough time and energy to zero-sum popularity games.

    • von Kalifornen says:

      That’s probably partly accurate, but it still sort of suggests the formation of a sub-hierarchy.

      That happened among the arts nerds I knew (who were all popular among the intelligent, beautiful, capable of finding love but not choosing to actually do so, able to relate to the true Populars but uninterested in them).

      OTOH the science nerds I knew were kind of a Brotherhood Of The Outcast who went so far as to tolerate extremely unpleasant people even though they had no reason to do that. Tended to have inferiority complexes, tended to scorn the true Populars.

      • subforum says:

        This distinction is worth exploring. Armchair hypothesis: the “status detection” module is comparatively stronger in the arts nerd — there’s a systematizing impulse common to both types, but while a formal system like programming, physics or math can be mastered without getting tangled up in questions of social status, the impulse to collect and classify aesthetic experiences is inevitably affected (tainted?) by status considerations, since shared aesthetic experience is an important coordination mechanism for primate social groups.

        I’d identify as an “arts nerd” — I was a high-IQ loner who spent my teenage years reading literary magazines and collecting indie rock records, so that by the time I arrived at college I was already largely self-socialized into that subculture.

        But I wasn’t prepared for the bruising social status games. I’d been a more “traditional” nerd in my childhood and still kept up an interest in anime and tabletop roleplaying alongside my literary and musical interests. (Hardly anyone at my high school was into either.) During freshman activities fair, an upperclassman who I’d already befriended (he worked at the college radio station, liked all the same bands I did, and has since gone on to a respectable career in the art world) noticed me swinging a sword at the LARP table and told me in no uncertain terms that I was never to do that again if I wanted to keep hanging out with the cool kids. I took this advice, which I now regret.

        A non-sociological and (relatively) non-pejorative definition of “hipster” might be “someone whose sense of aesthetic merit is tightly coupled to their perception of social status.”

    • alexp says:

      I never got that popularity in high school was a zero sum game. It’s more of a positive sum game.

  22. Scott F says:

    On the confusion regarding low-testosterone linked with gifted/math intelligence but increasing testosterone increases some cognitive skills – high testosterone individuals might have the raw power to do very well at math / be generally gifted, but their high testosterone also makes them focus on climbing status hierarchies (which are weakly anti-correlated with gifted/math intelligence) and so they have little interest in developing theses skills. IIRC those cognitive boosts show up most clearly in people whose testosterone levels are adjusted – handing more raw power to someone who already has developed the cognitive skills.

  23. Peter says:

    On the subject of the partial overlap between nerd/geek etc. and the autistic spectrum, there are definitely people on the spectrum who are not particularly nerdy, I’ve heard people talk about it as a stereotype they want to get rid of. OTOH the area of overlap is quite large.

  24. Nita says:

    Why No Science Of Nerds?

    Because it would be social science. Ew, ew, ewwww! What are you, some kind of soft-headed, woolly-brained muggle, to even think of wasting your time on bogus bullshit like that? It should be plainly obvious to anyone with a non-embarrassing IQ that a few hours of armchair theorizing by worthy minds will lead to much better results.

    On a more serious note, perhaps by the time someone can do science, both their “nerd” identity and the issues surrounding it have become less important to them, displaced by other (higher status? more useful to humanity?) concerns.

  25. Evan Gaensbauer says:

    We nerds might be rationalizing both our successes, and misgivings, with convenient just-so stories. The inane but ever-insightful webcomic Saturday Morning Breakfast Cereal has touched upon this with Uncomfortable Truthasaurus to dispel such myths:

  26. I’m probably not a full-fledged nerd, but I don’t use psychoactives because I have a strong preference for the way my mind feels when it isn’t chemically altered. As far as I can tell, this is an intrinsic preference.

  27. A linguist’s take on science fiction fandom

    A sample:

    “We accept corrections on matters of fact and of pronunciation; when I asked her about whether fanspeak might be related to Asperger’s Syndrome, and mispronounced “Asperger’s”, I was corrected in mid-sentence by the man sitting next to me, corrected myself, thanked him, and finished the sentence. One Doesn’t Do That in Mundania. Fans understand that mispronouncing words one has only read is very common in fandom, and not mortally embarrassing.

    When we make a joke, we don’t do a little laugh in the middle of a word to signal that it’s funny; we inhale and exhale a very fast, short breath at the end of the sentence, rather like a suppressed beginning of a laugh, or a kind of a gasp.

    She didn’t get much into why this is all the case (I think she was surprised at the laughter when she suggested diffidently that we might be a bit under socialized. No, really?? ), and turned away questions about possible pathology. While more comfortable with us now, I suspect she was probably still worried about offending us. She did suggest that many of the common features of fanspeak seem to be related to thinking in “written English”.”

    • AJD says:

      A few things in that seem obviously wrong (or perhaps, misremembered or misreported by the person writing the article): “rounding the lips somewhat even for ‘ee’ and ‘ih'”, “she pointed out that I had pronounced the ‘k’ on the end of ‘talk’. Mundanes, she said, wouldn’t.” Many of the other things seem quite correct, and are supported by Bucholtz’s sociolinguistic research on nerds, linked above.

    • Nornagest says:

      Link’s broken for me.

    • lambdaphage says:

      Second, fans articulate more than mundanes.

      I’ve noticed this before, and it comes close to a one-sentence summary of the nerd accent.

      When we make a joke, we don’t do a little laugh in the middle of a word to
      signal that it’s funny; we inhale and exhale a very fast, short breath at
      the end of the sentence, rather like a suppressed beginning of a laugh, or a
      kind of a gasp.

      I found this painfully accurate, though more strongly associated in my mind with ‘geeks’ than ‘nerds’.

      • jaimeastorga2000 says:

        more strongly associated in my mind with ‘geeks’ than ‘nerds’.

        Do not start that discussion. It never ends well.

  28. J. Quinton says:

    It looks like testosterone is correlated with low empathy, as in possibly both autistsics and psychopaths. Maybe whatever causes the differences between psychopathy and autism is also why high T is associated with both jocks and nerds?

    • Zoe says:

      No. Empathy is two different things. Being able to read other people’s emotions, and being influenced to feel a certain thing by seeing somebody who feels it. Psychopaths don’t have the second, though they might have good empathy in the first sense. Autistic people have reduced ability to read others, but are easily affected by other people’s emotions when they do see them.

      • Nornagest says:

        I’ve heard this broken down into “empathy” (the former) and “sympathy” (the latter).

        • Zoe says:

          I have two responses to that

          1) That’s not the most common usage of sympathy. You have more or less empathy in general, but you have more or less sympathy for specific people or causes, and having sympathy for something is related to liking or agreeing with it.

          2) If what you describe is how you talk in general, then for you “psychopaths have no empathy” is false. Psychopaths lack what you call sympathy but have varying levels of empathy, including some that have quite high empathy.

  29. Quite Likely says:

    Wouldn’t the logical conclusion of your series of professions turning their professional talents on themselves be “a science of scientists”?

    Also, I think an important aspect of modern nerdiness is ‘nerd culture’ – being into video games, comic books, fan conventions etc.

  30. moridinamael says:

    Let me restate your condensed list of nerd-traits affected by testosterone.

    1. Low testosterone -> increased intelligence.

    2. Low testosterone -> introversion.

    3. Low testosterone -> low social status.

    4. Low testosterone -> muscly.

    5. Low testosterone -> low mating success.

    6. Low testosterone -> low drinking/impulsivity.

    7. Low testosterone -> nonviolent.

    8. Low testosterone -> high-pitched voice.

    Let us take Subject A as our example.

    Subject A is genetically misfortunate. He expresses has at least one recessive genetic trait that makes his posture hunched and his body very weak and frail-looking. He wears thick glasses starting in elementary school. His teeth are more crooked than average, perhaps due to the aforementioned genetic issues. Subject A is also more intelligent than average, but he’s no genius. So far, he physically seems “low testosterone” in the sense of (1) and (4).

    However, Subject A also has an unusually deep voice. He has low social status (1) *because* he is a physical poster-child of nerddom. He is introverted (2) *because* he has low social status, and his life has taught him to hide from the abuse of other children. He has low mating success (5) *because* of all of the above, regardless of his impulse to mate. He doesn’t drink or do many impulsive things (6) because he never gets invited to, because of all of the above, but he would probably do wild things if somebody did invite him. And he’s nonviolent (7) because he knows he’s weak and will be crushed in any conflict.

    Subject A eventually grows up and goes to college, gradually gains the wherewithal to disguise or compensate for his overtly nerdy traits and by the time he’s started his professional career you would never have guessed what he looked and acted like as a child and adolescent.

    A lof of people I know professionally are a manifestation of Subject A, to one degree or another. Competitive, hard-charging adults who were formerly picked-on kids. In short, normal- or even high-testosterone men who had to eat shit as children/adolescents and now, if they have any visible issues, those issues are mainly associated with the huge chip on their shoulder they carry due to being viewed how they were viewed as children.

    So anyway, to summarize, my hypothesis is that just looking nerdy *causes* all the other nerdiness qualities.

    • Anonymous says:

      Even if physycal inferiority is the immediate cause of a number those nerdy traits, if physical inferiority itself is caused by low testosterone, then this doesn’t actually contradict the hypothesis that low testosterone is the cause of nerdiness.

      We know that there is a relationship between low testosterone and a number of traits such as low social status. It’s possible that the mechanism through which low testosterone gives someone these qualities is by causing them to be physically weak, which in turn causes them to have low status, low mating success, low propension to violence and low assertiveness.

  31. peterdjones says:

    I can assure everybody that UK IT nerds are not known for abstemiousness.
    In fact we don’t even have a word for straightedge, like we don’t have one for “Bon appetite”.

  32. orangecat says:

    The other possibility is that in fact these traits are all correlated for some reason

    It seems intuitively obvious that subsets are, whether or not nerdiness is a thing. “Interested in math and science” is likely to be correlated with analytical intelligence. “Shy and awkward”, “bad at getting social status”, and “bad at dating” are somewhere between strongly correlated and literally the same thing (for men at least). Ditto for “physically unimposing” and “unlikely to get into fights”, and intelligence likely also feeds into not fighting.

    So I’d speculate that you really only have around 5 bits of entropy, which makes the existence of a large number of people that shows most or all of the traits less surprising.

  33. Anonymous says:

    Some of the racist / racialist crowd have similar ideas – they think that non-africans are less masculine relative to africans, leading to more male nerds and ultimately a more peaceful and advanced civilization.

    I don’t know. The general form, “People who are extremely gifted in math and science also are physiologically atypical in SOME way, and therefore have a host of disadvantages…and also tend towards hobbies that engage the same cognitive functions as math and science do” seems right.

    You can fill in SOME way with low/high testosterone, aspergers, or anything, really, and get similar effects. Almost all neurological weirdness will lead to social deficits because social skills recruits a very large number of cognitive facets. That doesn’t explain the impulse control, but then I’m not sure this is actually a feature everyone associates with nerds rather than just you – I know a lot of non-conscientious slackers who’d be called nerds due to working in IT, tabletop gaming constantly, not having social skills, and so on..

    I suspect nerds are all dis-similar from each other and are mostly bound together by being high-functioning atypicals of various stripes, with some atypical variants being more common than others and defining the Archetype.

    • Anonymous says:

      There is quite a bit of science concerning the personality traits of scientists, by the way…research which is done by psychologists, which is arguably the single least nerdy and most social-skill involving science. Also a lot of research into IQ, which is probably something scientists identify with more than “nerd”.

      Via my experience in the scientific community, the “nerds” I’ve met gravitate towards fields like physics, computers, or entomology … biologists (except for those studying plants or bugs or whatever), psychologists, sociologists and so on are pretty much neuro-typical.

      If anything, I’ve experienced psychologists and sociologists as MORE socially oriented than the average person, let alone the average scientist. They’ve got a light geek touch – still very intellectual and if American they can often catch all the nerdy culture references – but they don’t fit the physical and social dimensions of the archetype.

  34. Quixote says:

    I don’t see the same cluster you do. In particular I see almost no straight edge association with nerds. Most nerds I know fall into either “heavy drinker” or “get high and watch animie” clusters. Very few seem straight edge or even in that direction unless it’s for culturally exogenous reasons.

  35. a person says:

    I’m curious to what extent you guys actually had nerds in your high school / college. I know plenty of people who are smart and don’t have great social skills, but it still seems weird to call them nerds if they don’t have the obsessions, the strangeness, the sense of belonging to a different culture entirely. People who actually are like that seem pretty rare and not something that merely arises out of just one factor.

    In my first high school, there was a very loose spectrum of popularity that people freely intermingled across. If there was a group of nerds I was probably in it – I hung out mainly with a close group of high-IQ people who played a lot of video games and talked about weird shit most people weren’t familiar with, but some people in this group were cool and also hung out with e.g. the kids on the football team, and some were less so. There wasn’t really a sense that we were a distinct weird unit, we had plenty of friends outside of the group as well. People might have called us nerds because in this environment the word was usually used to describe people who studied hard and got good grades. But no one in the group would self-identify as such. We also had a lot of idiosyncratic obsessions that wouldn’t be considered nerdy, e.g. Kanye West. However, in the class below us, there seemed to be a fairly sharp divide between the nerds and the “cool kids”. There was a pretty large group of uncool people who would hang out in the computer science lab during lunch and talk about computers, and I definitely thought of this group as being nerds.

    In my second high school, there were definitely nerds. Everyone hung out with everyone except there was a weird group of like twelve kids several of who drenched in fedoras and trenchcoats (yes, really) and expressed almost no desire to hang out with anyone other than each other. They would be constantly conversing animately about something nerdy, and “normal people” often remarked that they couldn’t understand the nerds’ conversations and it was like they had their own language. I could follow them pretty well – they talked about Homestuck a lot.

    In college, where I currently attend, I see very little signs of nerd culture, despite the fact that I go to a STEM-oriented school for highly intelligent people. There are a lot of people here (maybe a majority) who are smart and kind of awkward but almost all of them still manage to fit into the dominant culture and don’t openly indulge in weird hobbies. However, I went a few times to the game developers’ club student group, and it was absurdly nerdy. Also sometimes you see people LARPing on the lawn in front of the dorms. So these people definitely exist, but I rarely encounter them on a day-to-day basis.

    • I was going to ask whether you went to the same school as me, but then it occurred to me that pretty much every polytechnic university in the U.S. is probably like this.

    • memeticengineer says:

      As an n=1 counter-example, my college was a STEM-oriented school for highly intelligent people and, at least at the time I attended, had high levels of visible nerd culture and generally many folk there proudly self-identified as nerds. Nerditude was often expressed through academics and related projects in one’s field, not necessarily through weird hobbies, though those existed as well. My attendance was a number of years ago so I can’t say for sure it is still that way.

  36. I’m starting to suspect that much of what’s going on here is attributable to imprecision of language.

    The baseline assumption under all this speculation is that “nerd” is a natural category, that has some kind of underlying sociology. But I often use hear different people use the word to refer to very different kinds of attributes. I don’t simply mean that nerdiness is composed of more than one factor; I mean that the underlying phenomenon that one person uses the word “nerd” to refer to, is different from the one that someone else uses it to refer to.

    For example, while nobody in this thread has yet mentioned Freaks and Geeks by name, several people have made clear that the phenomenon depicted in that show is what they’re referring to: socially awkward low-status people developing alternative status hierarchies based on particular non-mainstream interests. Meanwhile, someone else has mentioned Richard Feynman: someone who had a particular way of seeing the world and deriving fulfillment from it, intellectual but also with a certain irreverence. I think that each of these phenomena, along with a few others (this is not a “there are as many categorization functions as there are people” thing), is a distinct definition of the word “nerd”, and has its own psychology and sociology.

    Each of these notions of nerdiness is itself composed of multiple different factors. For instance, most notions of nerdiness incorporate intellectualism, and many of them also incorporate social awkwardness (if not as a necessary condition, then as something associated). Because some of these factors are shared in common between multiple definitions, there are correlations, which explains why the same word is used for multiple different phenomena. But I don’t think these correlations are interesting; they aren’t multiple manifestations of a single underlying psychological/sociological phenomenon, but rather of distinct phenomena that happen to have some factors in common.

    That’s part of why the concept of nerdiness is hard to discuss; it’s not one concept, and the word brings ambiguity. Thoughts? Does this make any sense at all to anyone else?

    • subforum says:

      Yes, I think this is correct.

      I find it unbearably grating when people start debating “nerd” vs. “geek” vs. “dork” as if these words are objective taxonomical categories, but that discussion does capture a desire to distinguish between different phenomena.

      Another salient meaning of “nerd,” or at least of “geek,” is as a niche (and not so niche any more!) marketing demographic. In some contexts, “geek culture” just means “the demographic which spends its disposable income and leisure time on videogames and SF/F rather than, say, pro sports tickets.”

  37. Johannes says:

    I believe that “nerds”, the corresponding culture etc. are largely an artefact of the US educational system of the last 40 years or so and its subcultures. In most of Europe, sports are almost irrelevant at school and university. In the schools kids get sorted earlier according to academic ability, so they are less diverse in the more demanding Gymnasium/Lycee etc. School may be also hell for some kids, but it is not the sports, sex and status crazed hell the typical American high school apparently is.
    Languages and humanities are important subjects as well, so there is not such an assortment or isolation of computer/science/maths types, but a diverse spectrum of bright kids.
    And until a few decades ago, video games, comics and scifi did not even exist. In the 50s and 60s European kids discussed Sartre and Jazz rather than video games and rap music.

    And if one looks at famous scientists, there are quite a few “renaissance men” not really fitting the nerd stereotype. Not only with respect to broader knowledge, abilities and interests, but also with respect to wine, women and song…

    • Nornagest says:

      School may be also hell for some kids, but it is not the sports, sex and status crazed hell the typical American high school apparently is.

      I can’t speak for all of the US, but my high school experience didn’t match the media stereotype all that well. There were plenty of zero-sum status games going on, sure, and high-performing athletes were generally well-liked; but the football team was just another clique, and the cheerleading team wasn’t even that.

      If any group was characterized by consistently high levels of academic success, it wasn’t the nerds so much as ambitious middle-to-upper-class college-bound types (“preps”). There certainly were nerds, and the ones that cared to do well in school generally did; but many of the nerdiest kids I knew were less interested in classes than in hacking or gaming or, in a few cases, drugs, and a number of them took equivalency tests instead of graduating conventionally. (This isn’t equivalent to dropping out, but is lower-status than graduation with a high GPA.)

  38. subforum says:

    Commenters offering anecdotal evidence on whether nerds are or are not straightedge: at what age did the nerds in your social circle start drinking/drugging? High school? College? Once they’d hit the legal age? These seem like useful distinctions to make.

    There’s a Vladimir M. comment somewhere (not sure if it’s at LessWrong or elsewhere) where he suggests that the relative prominence of “nerd” culture in North America relative to Eastern Europe might have something to do with teen drinking habits. A culture which tolerates teenage drinking and has a public transit system gives the average adolescent a chance to party (and learn socialization/romance outside of the school enviroment.) Whereas in suburban America, with its zero tolerance policies and risk of DUI, teenage hard partying is something that only the risk-taking cool kids do.

    So for the nerd, even more than the average high school student, college is their first chance to experience partying. (Source: personal experience.) It’s a not uncommon pattern for an eighteen-year-old to go from “fuck alcohol, that’s for meatheads” in February to “I love alcohol! Alcohol is great! Wanna get a drink?” in November.

    Come to think of it, this might also be an explanation for the observation upthread that smart kids from cities seem to be less “nerdy” on average than smart suburban/rural kids.

    Side note: my anecdotal impression from hanging around local punk/hardcore scenes is that self-identified members of the straightedge subculture (as opposed to people for whom “straightedge” is simply a synonym for “teetotaler”) tend to be high-T jocks, often blue collar guys breaking away from alcoholic parents.

    • Multiheaded says:

      A culture which tolerates teenage drinking and has a public transit system gives the average adolescent a chance to party (and learn socialization/romance outside of the school enviroment.)

      This is needlessly specific.

      When I was 14-15, me (the “nerd”) and my classmates (a couple of high-achieving nerds, a couple of slackers and troublemakers) hanged out at shady beat-up LAN gaming centers. Some of the older adolescents there drank, but I don’t remember seeing any underage drinking, although undoubtedly it happened. (Teenagers borrowed cigarettes from people, though.) LAN gaming was for everyone where I grew up… exclusively male though.

  39. Sebastian says:

    n=1: I’m a nerd in the midwest, and my friends and I are into cycling, sailing and table tennis.

  40. pxib says:

    Ten years ago, Paul Graham wrote a brutal little essay about nerds and popularity. The general thrust is that nerds are kids (and ultimately adults) who prioritize being smart over other traits. Navigating social trends in the cauldron of public school takes an enormous amount of time, attention, and energy which the nerds devote to what they see as the actual purpose of school: learning.

    It doesn’t explain all the stereotypes, but it has a lot to say about the mentality. A good read, regardless.

    • veronica d says:

      PG strikes me as one of those feel-good cheerleaders that his community loves because he tells his community just the nicest stuff they want to believe about themselves. He is so glorious, the biggest, most-bestest voice in the echo chamber.

  41. AJD says:

    “Nerds are traditionally viewed as having high libido”—this isn’t congruent with my stereotype of nerds; the “cuddling > sex” feature you mention is more in line with what my stereotype is. One of the Bucholtz papers I link in another comment says that nerds “opt out of the heterosexual matrix of the high school, in which pressure to engage in sexual activity is paramount.”

    With regard to nerds as teetotalers, I’m reminded of this article from the Harvard Crimson some years ago, which discusses in detail the Harvard science-fiction club as an example of a social environment not centered on drinking.

  42. Lucia Dremsly says:

    Nerdy trans woman here. I was also told I had absurdly, confusingly low testosterone before starting treatment so…

    Just going to share some experiences as data points here:
    I was/am actually considered conventionally attractive and used to be good at sports before health issues ruined everything.

    I think the “Nerd gets immersed in nerd stuff. Stops caring about sports and social events.” fits my experience, but I think the drugs thing was not necessarily impulse control for me, I just never actually got the impulses in the first place because the idea of smoking, doing meth or even drinking alcohol horrified me from the start for reasons I’m only partially sure of.

    It did have to do with the consequences of it but it also just didn’t look like much fun.

    P.S. I’m not really as intelligent or knowledgeable as most nerds either.

  43. veronica d says:

    One thing I have not seen much discussed is the ex-nerd, which is the person who during high school was trapped in nerd space, but by the time they are out of college has escaped the “socially awkward” trap and is now — not socially awkward (at least not the way nerds are).

    One flavor of this is perhaps the tech “brogrammer” type guy, who was a computer nerd when young but decided to trade his big software engineer salary for some kind of hyper-frat-boy lifestyle. (I’m personally not very fond of this personality type.)

    But also there are plenty of still-kinda-nerds, who maybe still play role playing games (or whatever), but seem to have figured out social spaces to a fair degree, at least enough to operate in the world. (I think Scott probably falls into this camp. As do I — I’m married and poly and go to clubs and parties and meet strangers and make friends easily. I make jokes about my job and my math obsession, but people don’t mind because I change the subject to makeup and shoes.) Do such people still count as nerds? Or have they become something else?

    Regarding trans nerds, some nerdy trans woman needs to do a cool webcomic (or whatever) about her scene. I would read the fuck out of that.

    • John says:

      One flavor of this is perhaps the tech “brogrammer” type guy, who was a computer nerd when young but decided to trade his big software engineer salary for some kind of hyper-frat-boy lifestyle. (I’m personally not very fond of this personality type.)

      I’ve worked in tech in Silicon Valley for several years and I still have yet to meet any “brogrammers”, FWIW. I think stories make them out to be more than they seem.

      • no one special says:

        Midwestern programmer: brogrammers are missing here too. Also, no unicorns.

        I think the brogrammer is a rhetorical device used to transfer hatred of the older archetype of the successful socialite onto the newer archetype of the successful tech entrepreneur.

  44. Viliam Búr says:

    What a small world! Scott linking to a research of my former boss, Ms. Laznibatová, director of the school for gifted children where I taught. 😀

  45. Mike Johnson says:

    I’m not sure how to weight this, but one additional possibility would be that humans vary in terms of how “self-domesticated” we are, and nerds are on the high end of this.

    By “self-domesticated” I mean the same sort of domestication process that happened with wolves/dogs, but applied internally. The proto-humans that couldn’t play nice, or couldn’t contribute, got kicked out of our ancient communities. Nerdish behavioral tendencies cluser in the “play nice and contribute” area…

    A possible test of this is analyzing neural crest development in nerds vs. jocks. Are you familiar with the siberian fox domestication experiment? One of the big findings was that the process of domestication delayed neural crest development, such that the friendlier foxes ended up with a less-developed adrenal gland, and were more “perpetual adolescents” in terms of development (and behavior).

    Nerds are said to be late bloomers. Maybe there are differences in specific developmental processes (i.e., neural crest?) that support this.

    • I’m pretty sure this is evolutionarily impossible. Among many other reasons, while nerdiness may be somewhat heritable, nerds aren’t a distinct subpopulation as far as evolution is concerned. There’s no way we faced different selection pressures from other humans in the ancestral environment that caused us to turn out differently.

      (I’m assuming a definition of “nerd” that’s primarily about individual psychological traits, because under any other definition this idea makes even less sense.)

      • Mike Johnson says:

        My statement claims much less than I think you think I’m claiming. There are differences in how domesticated dogs are, even within the same litter. (This within-population variation is specifically how the fox experiment worked, in fact.)

        On the other hand, it might be interesting to explicitly take this idea into the context of distinct subpopulations. Do nerd parents tend to have nerd kids? (often…) Do some distinct subpopulations have very few nerds, based on Scott’s definition? (maybe…)

        “There’s no way we faced different selection pressures from other humans in the ancestral environment that caused us to turn out differently.” — this statement seems awfully strong and a little vague. It’s clear to me that yes, some groups of people probably *did* face different social selection pressures, and it caused their gene pool to turn out differently (proportionally more or fewer ‘nerdy’ genes in the gene pool). Of particular interest, different populations will have experienced different game-theoretic realities of selection– some supporting nerdishness better than others.

        Maybe I’m missing some part of your objection. You say you have “many other reasons” why this idea is “evolutionarily impossible” or “makes even less sense” — do you care to go into them?

        • I owe you an apology. I didn’t (and still don’t) understand what exactly you were suggesting very well, but it sounded a “human behaviors always have simple evolutionary causes that analogize closely to animal breeding” kind of thing that I sometimes hear from people who don’t understand how evolution works. So I was more dismissive than I should have been.

          The Wikipedia article talks about self-domestication in the context of both animals and humans. But its discussion of self-domestication in animals sounds like a completely different phenomenon from its discussion of self-domestication in humans. So I guess I’ll just acknowledge that I don’t understand what’s going on at all and bow out.

          • Mike Johnson says:

            I think were largely arguing past each other. It might be useful to ground any discussion of ‘domestication’ in terms of a specific mechanism– e.g., delayed maturation of the neural crest (which should vary both within and between populations?). At any rate, cheers, and thanks for the kind reply.

  46. Mike says:

    Great post, this is something I have thought about a lot, given that I self-identified as a low-T nerd during my high school years. I’ve always been mathematically and scientifically gifted but never truly excelled at sports or romance (the nerd stereotype).

    However, I’m currently a biochemistry senior in college, and after taking up a weight training program (consistently training 3x a week, eating 500+ calories above BMR) for the last 13 months I’ve noticed significant changes in both my intelligence, social status, and physique.

    As silly as it is to say that I feel more testosterone in my body, I do. I’m more extroverted, aggressive, dominant, I’ve put on 35lbs of muscle/fat, and can interact with the opposite sex much more fruitfully. However, I do feel like my intelligence has somewhat decreased. This can’t be correlated with grades, as I’m still top of my class, but I just *feel* dumber.

    For reference, I started out doing Starting Strength, and have since moved onto a modified Texas Method strength program.

  47. Sniffnoy says:

    I am going to pull a weird multicolored sun and comment not with a statement but with a noun phrase. (Sort of.) I think I want to make a statement with this noun phrase later, but, one thing at a time. (Actually, I guess I do the “here’s a useful concept, make of it what you will” thing a bit…)

    Basically, Scott has described the “nerd” cluster with something of a prototype — well, not a prototype exactly, but one particular set of nerd characteristics. But we could point out other characteristics and get a different image with different connotations. People have done this to some extent above; I did it above with the “taking ideas seriously” bit.

    Let’s take this further. What are some other things we associate with nerds? For one, bluntness, or frequently just straight-up rudeness. An insistence on correcting other people even when inappropriate. Basically, an ethic of “substance over style” — and a disdain for those who waste time on the latter or get in the way of the former. Nydwracu has touched on this above, but it’s often not “bad social skills” but “a rejection of social skills”; the nerd’s response to the idea that they need to learn social skills is something like “You mean learning to lie?” Also, a strong belief in thinking for oneself and not going along with what other people do.

    That’s the noun phrase. Perhaps I’ll make an actual statement with it later… 🙂

  48. Shenpen says:

    I think to be a nerd is to a large extent to hate yourself, and make an imagined self in the form of RPG (like AD&D) heroes, computer game heroes, fantasy heroes, anime heroes…

    I will expand upon it later, but to a large extent it explains things like how nerds tend to have robotic movements, poor perception skills – they try to distance themselves from their actual sense, “I am not here guys, this sack of shit is not me, I am in my dreams being a super handsome 20th level paladin”.

    Self-hatred can have many reasons, but your low-T hypothesis is good: low-T boys are perceived as weak and cowardly (they often are) and thus perfect targets for sadistic bullying. This bullying creates the self-hatred.

    This is a quick summary only, obviously this topic would deserve a book.

    Begbie gets one aspect of it right:

    Basically for a normal STEM guy, who accepts himself and is genuinely interested in it, if you correct a mistake of him, you will be glad: he learned something and got better. But the nerd basically wants to die as his own self and REBORN as a new identity, such as The Programmer or The Engineer. If you criticize anything about his knowledge, if you point out a mistake, you just challenged his whole identity and he will be obsessive about proving you wrong. Also, he can hardly talk about anything is than his subject. He is not John anymore, he is The Engineer and as such he can only talk about engineering.

    This is why nerds tend to obssess.

    And so on it is really a good topic.

    Homework: what aspect of the self-hatred of nerds do various Dragonlance Chronicles characters represent?

    • Matthew says:

      This is wrong. Like, seriously, seriously wrong. Escapism is not necessarily a sign of self-hatred; I certainly never hated myself. It was more about finding the mundane things that interested other people boring.

      Also nerd != STEM type. Plenty of us (including our blog-host) have stronger verbal skills than math skills.

  49. Pingback: Lightning Round – 2014/10/01 | Free Northerner