Quote note (#220)
Scott Alexander in full-flood (in his own comment zone):
Okay, that’s actually a fair point. I don’t have all of this in one place yet, but I’ll try to give you a summary. And in honor of you accusing me of paranoid rants, I’ll try to present it in as deliberately conspiratorial and tinfoil-hattish a way as possible. I make no guarantees I will stand by any of this when I’m sober / when it’s not 3:30 AM. [I’m going to assume that’s an abbreviation for 3:33]
People naturally divide into ingroups and outgroups. Although the traditional way of doing this is by race or religion (leading to racism, anti-Semitism, Islamophobia, etc), in modern America this is gradually being replaced by a more complicated division based on social class and political affiliation. Rural working-class people have become a very different tribe (“Red Tribe”) than college-educated urban people in gated professions (“Blue Tribe”), with different food preferences, sport preferences, entertainment preferences, dialects, religions, mores, and politics. These two groups are vehemently opposed.
(if you only read one link in this piece, read that vehemently opposed one. The rest are just citations; that one contains an important piece of the story that’s hard to summarize).
While politics is about equally split between them, the media and academia are almost entirely Blue Tribe.
To make the point about the media: a 2008 study found that 88% of contributions by people in the media went to Democrats; a 2004 study with slightly different methodology that limited itself to journalists found an even larger bias. Here’s a survey that finds that if journalists were their own congressional district, they would be the most liberal district in the country, much further left even than Berkeley, California.
To make the point about academia: a recent analysis found that 91% of Harvard professors who donated to a presidential campaign donated to Hillary (with the remainder divided between Sanders and all eight GOP candidates). Jon Haidt’s does a lot of work on this at heterodoxacademy and finds that there’s a 14:1 ration of liberals to conservatives in the non-economics social sciences. Meta-analyses in psychology, psychiatry, and economics all find that the personal views of experimenters affect what results they get; the psychology study, which quantifies the results, finds a very large effect size – larger than most effect sizes actually discovered in social science, meaning we have no idea how much of what we know is real effect and how much is experimenter political bias. On a related note, only 30% to 50% of experiments in psychology persist after replication attempts (other academic disciplines are as bad or worse). On a related note, meta-analyses observe clear evidence of publication bias in politically charged domains – for example, this meta-analysis finds that papers are more likely to be published as opposed to file-drawered if they support the liberal position rather than the conservative one. Also, lots and lots of people in academia, even the very liberal people, will admit this is true if you ask them directly. Haidt, Tetlock, et al (see previously cited paper) have found lots of horrifying things like journal editors saying explicitly and proudly they’d refuse to publish articles that support conservative ideas, or professors saying that other academics whose research implies conservative ideas shouldn’t be hired or given tenture.
So given the fact that our knowledge of the world is coming from a 90-percent-plus liberal group that’s working hard to enforce orthodoxy, and then being filtered and broadcast to us by another 90-percent-plus liberal group that’s working hard to enforce orthodoxy, our knowledge of the world is … about as skewed as you would expect from this process. To give just one example, every number and line of evidence we have suggests that the police do not disproportionately target or kill black people compared to the encounter rate (see Part D here and this study) but the conventional wisdom is absolutely 100% certain they do and anybody who questions it is likely to sound like some kind of lunatic.
Once again, I think of these political differences as secondary to (and proxy for) more complicated tribal/class differences, and these tribes/classes really really hate each other and are trying to destroy each other (remember, multiple experiements – 1, 2, 3 – find that people’s party/class/tribe prejudices are stronger than their racial/religious prejudices). So imagine an institution that’s 90% Klansmen, with all its findings interpreted by and transmitted through a second institution that’s 90% Klansmen, and consider how useful (or not) the information about black people that eventually reaches you through the conjunction of those two institutions will be.
Because the Blue Tribe’s base is in education and the opinion-setting parts of the media, their class interest is to increase the power of these areas. I don’t want to sound too conspiratorial by making it sound like this is organized (it’s not), but classes tend to evolve distributed ways to pursue their class interests without organization. In this case, that means to enforce credentialism (ie a system where the officialness of your education matters more than your ability) and orthodoxy (whether you hold the right opinions is more important than ability). We see the credentialism in for example the metastatic spread of degree requirements. You need a college degree to have the same opportunities as you’d have gotten from a high school degree in 1960. This isn’t because jobs require more knowledge today; there are thousands of jobs that will take you if you’ve got an Art History degree, not because Art History is relevant to the job, but because they insist on candidates having some, any, college degree. The Blue Tribe protects its own and wants to impoverish anyone who doesn’t kowtow to their institutions. For the same reason, we get bizarre occupational licensing restrictions like needing two years of training to braid people’s hair, which have been proven time and time again not to work or improve quality, but which effectively lock poor people (and people who just don’t do well with structure) out of getting liveable jobs.
The opposite of credentialism is meritocracy – the belief that the best person should get the job whether or not they’ve given $200,000 to Yale. In my crazy conspiracy theory, social justice is the attack arm of the educated/urban/sophisticated/academic Blue Tribe, which works by constantly insisting all competing tribes are racist and sexist and therefore need to be dismantled/taken over/put under Blue Tribe supervision for their own good. So we get told that meritocracy is racist and sexist. Colleges have pronounced talking about meritocracy to be a microaggression, and the media has declared that supporting meritocracy is inherently racist. Likewise, we are all told that standardized tests and especially IQ are racist and hurt minorities, even though in reality this testing helps advance minorities better than the current system. For the same reason, colleges are moving away from the SATs (an actual measure of student intelligence), to how well students do in interviews, how well they write essays, and other things which are obvious proxies for social class and tribal affiliation.
STEM culture and nerd culture is (was?) this weird alternative domain that had Blue Tribe advantages like education and wealth, but also wasn’t drinking their Kool-Aid – they took pride in being meritocratic, they didn’t care what college you went to as long as you were smart, and they were okay enjoying their own weird culture instead of following sophisticated trend-setters. The Blue Tribe was spooked, so they called in their attack arm, and soon enough we started hearing these constant calls in Blue-affiliated media and circles to destroy nerd culture (2, 3, etc, etc) because it is inherently misogynistic, racist, etc. It’s why we’re told that Silicon Valley is full of “brogrammers” and “techbros” (compare “Berniebro”, which everyone now agrees was a Hillarysphere attempt to smear Sanders supporters). It’s why we’re told that tech is “incredibly white and male” and “needs to get less white” and just generally has this huge and unique diversity problem – even though in reality it’s possibly the most racially diverse industry in the country, at a full 60% non-white. It’s why we’re told that there is terrible bias against women in science academia, when in fact anyone can read the studies showing that controlling for all other factors, women are twice as likely to be hired for tenure-track STEM positions as men [bad link] and academic science is not sexist at all. It’s why we’re told women fear for their lives in Silicon Valley because of endemic sexual harassment, even though nobody’s ever formally investigated if it’s worse than anywhere else, and the only informal survey I’ve ever seen shows harrassment in STEM to be well-below the average harrassment rate.
What’s happening at GitHub itself right now is actually a pretty good example. The old CEO was fired because of various accusations (later investigated and found to be false; the firing was not revoked). The new CEO has banned the term “meritocracy”, replaced workers managing their own affairs with a system of no-doubt-well-credentialled middle managers, and given lots of power to a “diversity team” that declares all remnants of the old company culture racist and sexist. According to Business Insider, there’s now a “culture of fear” and a lot of the most talented employees are leaving. People are saying GitHub made some kind of mistake, but I suspect all is going according to plan, the talented employees will be replaced with better-credentialled ones, the media will call everybody who left “techbros” who were suffering from “aggrieved entitlement”, GitHub will join the general Silicon Valley 2.0 landscape of open-plan offices and Pointy Haired Bosses, lather, rinse, repeat, and ten years from now bright-but-lower-class unsophisticated people without college degrees won’t be able to find a job in Silicon Valley any more than they can on Wall Street or anywhere else.
I am pretty darned Blue Tribe myself – I’m pro-choice, pro-fighting-climate-change, pro-gay, pro-transgender, non-religious, pro-higher-taxes-on-rich, pro-single-payer, anti-gun, ready-for-Hillary, etc – and after having watched the Republican debate tonight I can honestly say I’m terrified at anyone other than the Blue Tribe having power. But just as I can be proud of my Jewish heritage but also upset about the occupation of Palestine, so I can be proud of the Blue Tribe and not too happy about their project of crushing everybody else with an iron fist regardless of the collateral damage. Doing anything about this is a dauntingly large project, but my own comparative advantage is in picking apart some of the sillier studies they use to put a fig-leaf over what they’re doing.
Content aside — as a note on sheer blogging execution — this is a freaking (double-tap) comment. It’s terrifying.