site banner

Culture War Roundup for the week of January 15, 2024

This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.

Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.

We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:

  • Shaming.

  • Attempting to 'build consensus' or enforce ideological conformity.

  • Making sweeping generalizations to vilify a group you dislike.

  • Recruiting for a cause.

  • Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.

In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:

  • Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.

  • Be as precise and charitable as you can. Don't paraphrase unflatteringly.

  • Don't imply that someone said something they did not say, even if you think it follows from what they said.

  • Write like everyone is reading and you want them to be included in the discussion.

On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.

Jump in the discussion.

No email address required.

Colgan Air 3407 was a 2009 aviation disaster, where regional commercial airliner on final descent to Buffalo New York stalled and crashed, killing all on-board and one person on the ground. For those interested in more precise details, the NTSB report is here, while MentourPilot has a video breakdown of the timeline and personnel here.

The majority of US airline disasters leave an impact on airline regulation and aviation in general, if sometimes not quite enough, or only after a delay. While far from the most deadly western air disaster of the the time, or even the worst western stall-related crash of that year, CA3407 had an unusual impact. Where older historical reports are filled with tragicomedic disasters and near-misses, the majority of recent accidents tend to rest on extremes, where either mechanical problems coincided with areas well outside of training focuses encountered pilots who made heroic efforts, or where pilots operated with long periods of outright disregard for safe operations, with perhaps a scatter of situations where bad practices mixed with mechanical error.

Instead, CA3407 fell due to a few seconds of phenomenally bad judgement by its captain encountering a spurious stall warning. Stalls are caused by insufficient lifting force on the wings, usually due to insufficient airspeed or high pitch (more rarely, extreme icing). Stalls and especially low-altitude stalls reflect fundamental aspects of flight, and pilots will do simulator (and sometimes real-world) drills both directly focusing on them, and on weather conditions that can induce them, such as microbursts. They're probably just up there with single-engine-out procedures for matters that a pilot should know by heart.

The normal behavior for a low-altitude stall is to give as much engine power as possible, with wings level to the horizon, until airspeed recovers. Instead the captain instead increased speed to a lower threshold and then pitched up well above the horizon -- a behavior that would have doomed a stalled aircraft had that warning been correct, and in normal landing profiles initiated an extreme stall. Worse yet, the time between the initial spurious warning and impact with the ground was less than half a minute; the aircraft was probably only recoverable in the first five to ten seconds, making takeover of effective command by the first officer impossible. While the first officer may not have been physically able to override the pilot's erroneous control behavior, if she had been willing to defy protocol and procedures, her response of lifting flaps was likely not harmful but neither procedurally correct nor helpful.

Like most incidents, there was no one single cause. The NTSB mentions pilot fatigue (and first officer illness), weather, flight manual inconsistency, and the combination of a first officer new to commercial flying with a pilot-in-command who had just moved to a new aircraft, among others. Training at the time would emphasize maintaining altitude during recovery, which made historic sense when terrain and obstruction information around airports was not always great and aircraft power profiles looked different, but was increasingly outdated around modern airports and unrealistic in modern aircraft.

But pilot capability was the big one. CA3407's pilot-flying and pilot-in-command had many hours experience, but also had a spotty training and especially checkride record. Checkrides can be considered the 'tests' for aviation certification, where the pilot flies along with an FAA-registered designated pilot examiner to undergo certain practical tests. The pilot in command here had four checkride disapprovals (effectively failures) across his career. While individual disapprovals at checkrides are not uncommon and do not necessarily indicate serious problems if corrected, these tests are neither convenient nor inexpensive to set up, and a pattern of first-attempt failures can, to quote the NTSB report:

"However, the captain’s established pattern of first-attempt failures might have indicated that he was slow to absorb information, develop skills, and gain mastery or that the training he received was not adequate. This pattern might also have indicated that the captain had difficulty performing required skills while under the stress conditions associated with a checkride."

((Though the FAA does not necessarily agree here; it holds that there is almost no correlation between checkride failure and later citations... which is a bit streetlamp-examination. But where CA3407's pilot would be in the bottom 5% of commercial pilots by simple count of checkride disapprovals, I absolutely agree that this isn't proof he was in the bottom 5% of commercial pilots by ability.))

This issue gained additional poignancy in public awareness due to the voice recorder conversations shortly before the crash, where the first officer discussed her unfamiliarity with icing conditions before being hired to Colgan, and the pilot mentioned his own employment history with the company.

Flight certification is heavily controlled by a new pilot's logged flight hours, with different licenses and certifications requiring certain thresholds or conditions of hours in a pilot's logbook. These hours aren't all literally flying, with some complex rules about what simulator hours can be logged in different grades of gear. Since an hour of flight can cost 100-300 USD/hour (and even BATD/AATD simulator time isn't free), including fuel and aircraft maintenance, optimizing hours someone else is paying for matters a lot.

Before 2009, one common route for new pilots involved self-funding their way through the private pilot's license and commercial license with instrument cert, which usually meant 150-300 hours, then zooming off to whatever regional airline needed first officers. While those first officers would not (and could not, legally) be pilots in command for the next 1200-1350 hours of flight, they would still get experience as pilots-flying and have time with various airline training and currency checks. Only then could they apply for an Airline Transport Pilot license, necessary to operate as a pilot in command. In theory, this would give a lot of experience in a variety of environments, most closer to 'real' pilot operations and some of which (like icing or flying near New York City) general aviation avoids like the plague, while still having the eyes and hands of an experienced pilot nearby to watch, and to catch any obvious faults.

After 2009, that was illegal. In response to CA3407, Congress passed the Airline Safety and Federal Aviation Administration Extension Act of 2010, which along with mandating a system for employers to more readily learn about applying pilot's past records (not finalized until 2021!), mandated that both the pilot and copilot of a commercial scheduled operating have an ATP. While a few exceptions were carved out (a new license, the ATP-r, was made available for pilots with a four-year pilot's degree, or two-year pilot's degree, or military flight training, at 1250 hours, 1000 hours, 750 hours, respectively), this rule remains today, and it has had no small effect on both the availability of airline pilots and their possible career paths. That's not as vast a change as it appears at first glance -- almost every airline had stricter hour minimums for hiring -- but it still significantly increased the number of hours a pilot would have to get on smaller aircraft first.

The argument is that many first officer roles would look to have a lot of varied flight experience, while not actually flying a plane most of those conditions. Pilots had to get a certain number of hands-on-stick landings to maintain currency, but a pilot-in-command would and often should take over landings and takeoff from a first officer in bad weather or awkward conditions. Especially in recent years, a lot of time would be flying the computer to set autopilot controls and monitor instruments mid-flight. To the extent small problems might show up, it would be very hard for documentation of those problems to show up if a pattern of. By contrast, flying a contract plane or as the flight instructor leaves you responsible for the safety of flight, and even recovery of a serious incident can and often is recorded.

And that argument is controversial. For CA3407 specifically, both pilots had significantly more than 1500 hours at the time of the crash, and while the pilot had earned most of his pre-Colgan hours at a flight school emphasizing bigger birds, the first officer had earned a lot of her hours as an instructor in a flight school. It's not clear that the theoretical argument applies, and there are some arguments against it. While not all arguments against are all well-founded (the rule is a little more complicated than allowing you to just log a thousand hours of tethered hot-air balloon time, even before considering that even overseas airlines would laugh you out of the office: pilots everywhere hate ballooners), some are more reasonable (flight instructors and flight schools will avoid many of the critical conditions and some aren't great about required incident reports, general aviation equipment in even newer aircraft is vastly different than even old airlines). While some groups like airline pilot unions have been strong advocates for the rule, there's reason MentourPilot and AOPA argues against it -- and there's more general-aviation pilots wanting in than airline pilots pissing out. Funding 1500 hours or a 4-year-degree in aviation is ludicrously expensive, and while some commercial operations remain legal for commercial-equipped pilots, the whole ecosystem is a mess, with a glut of flight instructors and contract pilots mixed with shaky demand for training and contract work. On the other side, when airline pilot demand is high, this has lead to flight 'clubs' or 'schools' that exist solely to burn hours and gas at the bare minimum of familiarity, or even airlines 'hiring' near-threshold pilots to build their last fifty or hundred hours in a rush. These markers are (and even pre-2010, were) disfavoured in airline hiring practices, but their increased prevalence makes them harder to filter out. The limited availability of pilots has even lead to consideration of tradeoffs against other forms of fitness, such as boosting the mandatory retirement age or decreasing flight medical rules. I'm generally against it, albeit not very strongly.

Ultimately, if this rule is a test of merit, it's a weak indicator, and selected more for convenience and politics than as the best option.

But enough about such culture war questions like pilot training requirements or FAA reporting guidelines. You know what nobody hates each other about yet? Race and (dis)ability!


The FAA promotes diversity, and has for some time. Insert the joke about autistic people and transportation obsession here. The matter, however, got some increased traction after A Certain Someone on X Twitter highlighted a few sections of a Boeing DEI statement, and this quickly turned into discussion about what exactly that might mean, especially as diversity might include air crew entirely of one race for a flight (Canada, not US). And to be fair, there are no small number of nuts to pick who blame DEI for every fault, or popular idiots who think everything the FAA does involves their eyeballs, or who are using the matter as a poorly-camoflauged way to hate black people.

I am and long have been skeptical of the racial explanations for entire continents, and especially given the selection effects present for pilots, I am skeptical of any claim that African-American (or female, or gay, or whatever) pilots are categorically different in skill. Especially in the modern day, the Damoreish arguments don't apply: no one falls into commercial aviation, and everyone who tries to work in the field is fascinated by it to a large extent. What does it matter, here?

Because I have seen people say things like "The worst case for DEI is ending up with the lower end of the top 1% of candidates - where the difference between the best performer and the lower performer is measured in tenths". And there's some fair discussion whether the pilot of CA3407 was merely the lower performer, or so low he should not have been considered.

But that's not the option on the table. All the children in Lake Wobegon can be above-average compared to the country; not all of the country can be above-average compared to itself. Individual businesses or (possibly) entire fields could, perhaps, attract the 1% of subpopulations, and still remain at 1%-level capabilities: there are enough African-Americans in the United States that the top 0.1% could fulfill all pilot demand, even though I expect the majority have better things to do with their time and abilities. Piloting and the FAA are not the only places looking to fulfill DEI objectives. They are not the only one of ten commercial-pilot-sized places looking to fulfill DEI objections.

To be fair, there are other groups discrimination in hiring DEI hopes to help. We might just be downscoring half of qualified applicants in a crowded field, on matters completely and totally unrelated to their merit, rather than six out of seven.

There are ways to credibly challenge whether this is a problem. Perhaps training or experience matters more than innate ability; perhaps structure . Perhaps eventually everyone becomes a minority in some way; perhaps the position of modern equality has minorities as most equal.

But to suggest that the difference can't matter is to overlook literal piles of charred corpses. To complain that one extant metric is not optimally tied to merit while glossing over a new one that is disconnected from it does not strike as serious engagement.

Relatedly, the FAA is currently defending a class action lawsuit ( by prospective air traffic controllers whose aptitude test scores were purged and applications summarily denied because they failed a "biographical questionnaire" that gave more points to applicants who had recently been unemployed than applicants with relevant aviation experience. All because the incoming cohort was deemed insufficiently racially diverse.

The claim basically is the FAA went out looking for proxies for African-American race, included those in the biographical questionnaire, and then purged candidates based on not meeting those proxies. Both disparate treatment and disparate effect. I'm sure the way the courts twist themselves and the law into pretzels to argue that no, it's really OK to discriminate against white people or how this isn't really discrimination against white people will be interesting, but I would not expect any relief. (This particular practice was later ended by an act of Congress)

While black Americans have a higher unemployment rate than white ones, do prospective black air traffic controllers have particularly high unemployment rates?

Air trafffic controllers are generally having a hell of a time. But my knowledge is mostly limited to COVID-era hiring freezes, and this lawsuit was refers to actions taken ten years earlier.

These are prospective air traffic controllers- applicants to the government’s training program- so most of the relevant job experience is ‘being a pilot, but not going to become an airline pilot for whatever reason’, and this is probably a very white and male group.

The guy I’m thinking of had significant training. I never asked if it was the FAA course or some sort of pre-admit credential treadmill. Either way, he wasn’t a pilot. He was in the queue for assignment to an airport when COVID hit and the government just…stopped advancing the line. Hope y’all didn’t need a salary any time soon! After a couple months of that, he shrugged and went into marketing.

I think so- prospective white air traffic controllers are mostly pilots of small craft(so relevant experience), whereas prospective black air traffic controllers are probably a lot more likely to go through aviation school.

I don't know, but I bet the FAA does. They also asked about things like number of drama classes and high school sports the applicants were involved in.

I am skeptical of any claim that African-American (or female, or gay, or whatever) pilots are categorically different in skill. Especially in the modern day, the Damoreish arguments don't apply: no one falls into commercial aviation

Nobody falls into being a doctor either, yet you'd probably prefer to be treated in a majority-white hospital to a majority-black hospital.

If the administration of airlines, air traffic control, pilot training and so on was ruthlessly meritocratic, then I'd agree with you that there'd be no difference in skill between black or female pilots and white male pilots, since they would all have passed the same tests and be above a certain benchmark. But this isn't so, not informally and now not even formally. If it were meritocratic, then we wouldn't see DEI rules and so on.

And the only reason we'd depart from formal meritocracy is because influential decisionmakers don't want meritocracy. Presumably they've already been informally advising on hiring decisions and making their expectations known. I have a friend who did anti-plagiarism work at a university. There are strict de jure rules against plagiarism, academic integrity is supposedly very important. De facto students=money and so they were told to slow down, don't be too efficient, make sure to let them appeal (and so be it if we can only do one or two such appeals per day, the other 90% of cases will be deferred into the never-never). Hence the anti-plagiarism unit has been churning through staff for some time, nobody seems to want to half-do their jobs.

The RAF and USAF seem eager to lower the number of white male pilots via an informal hiring freeze and 'aspirational' diversity goals, respectively. Informal methods already do a lot of work.

Nobody falls into being a doctor either, yet you'd probably prefer to be treated in a majority-white hospital to a majority-black hospital.

Are you making the HBD-IQ('skill')ish argument, or the Damore difference-in-interest one? I think the latter is non-applicable and readily shown to be non-applicable, at least for current numbers of pilots. The former... I could be more persuaded for doctors, though I'll caution that places like King/Drew or individual cases like Patrick Chavis run into trouble isolating or controlling other variables, and the data may simply not exist.

((In turn, though, one of the other variables for King/Drew and Chavis was/is lack of willpower to enforce against bad actors when such enforcement would bring accusations of racism. Which leads to problems even presuming equal capability on average.))

But I think airline pilots run into, or at least close to, the same issues as I listed for the continent-level scale question:

My main problem is that it's easy to come up with an X-factor broad enough to explain the wide berth of problems and disparity present, or closely-defined enough to match the traits we see in the real world, but that trying to achieve a reasonable synthesis gets rough. These traits will necessarily be motioning around the shadow of the thing rather than the true borders, but I'm hard-pressed to believe you'd get the same borders when starting from American test scores, Haitian test scores, Brooklyn politics, or Haitian politics.

Are airline memory items closer to free recall or serial rote learning? Reading landing charts closer to symbolic manipulation or to reading symbols? Reading an weather radar as pic-vocab or inspection time? Not getting overwhelmed by alerts or alarms closer to odd-man-out or reaction time? You can come up with a hundred combinations, and there's certainly some number that will have a racial component in some studies simply by chance, but it's easier to come up with ones that are either nonsensical or wrong or both.

If the administration of airlines, air traffic control, pilot training and so on was ruthlessly meritocratic, then I'd agree with you that there'd be no difference in skill between black or female pilots and white male pilots, since they would all have passed the same tests and be above a certain benchmark.

You can have a pass-fail test and still some subgroups that are more skilled than others, even if they're all above a certain benchmark, though as long as that benchmark is set reasonably it is less concerning, as others have already gone deeper into the math. And while aviation has (thankfully!) moved away from the sort of heroics that saved part of UA232, there are still a lot of incidents and near-misses with current (and I'd argue pretty aggressive!) benchmarks.

But I don't find any of that nearly as disturbing as people who believe, enough to say publicly, that ability won't make a difference.

If the administration of airlines, air traffic control, pilot training and so on was ruthlessly meritocratic, then I'd agree with you that there'd be no difference in skill between black or female pilots and white male pilots, since they would all have passed the same tests and be above a certain benchmark

Actually, even with the same minimum standard for each race you would expect the white to be better on average.

If X is N(100,15) and Y is N(85,15) then E(X|X > 120) > E(Y|Y>120)

If the administration of airlines, air traffic control, pilot training and so on was ruthlessly meritocratic, then I'd agree with you that there'd be no difference in skill between black or female pilots and white male pilots, since they would all have passed the same tests and be above a certain benchmark.

Nitpick - this still won't be true as long as there is noise in the tests or variability in skill above the bar. The groups with higher average performance before the cutoff will also dominate the top percentiles after the cutoff, and be less likely to be a false positive under noisy tests.