Oh god here it goes I’m doing the education one

I was, in many senses, extraordinarily lucky in my education. The public schools I attended through the end of high school were very probably in the top 1% in the US; I’d estimate that my education cost the school system well over $15,000 per year. Something like 95% of my high school graduating class went on to attend college, a cohort which included myself. My undergraduate and graduate education were completed at one of the best schools in the country for my field. And, like a lot of people in my position, I think the school system is pretty awful.

It’s worth considering that I might be wrong about this. I did well in college and even better in grad school, and I have every reason to believe that I’ll be professionally successful too. By some measures, I’m already a success story for the system. But I don’t think my success, such as it is, is attributable to the system - or, in fact, actually a success. In this essay, I want to talk about school: what makes it good or bad, where we are now, and where I think we should be.

What is School For?

The school system (I would guess) considers my education a success; I don’t. This comes down to a difference in opinion on how success should be defined - in other words, what school is for. In my view, there are two kinds of school, the kind that everybody gets and the kind that not everybody gets, and they have different (and extremely broad) purposes.

School Everyone Gets

The kind of school that everybody gets is specifically for young people and should have a few goals:

  1. Take care of children during the day so their parents don’t have to
  2. Transmit general culture
  3. Teach “survival skills” like reading, writing, social fluency, and basic math
  4. Provide access to an emotionally nourishing and safe environment where students are free to explore and develop as people
  5. Help students establish suitable career paths/exit trajectories

I think most of the problems with primary and secondary education are that they try too hard to teach while providing almost no mentorship and optimizing for specific exit trajectories. The purpose of primary education has, for around a hundred years, been to prepare children for work in an industrial society. Children are taught to sit still, be quiet, and deal with their boredom non-disruptively because these are essential skills for industrial workers. Secondary education, meanwhile, is differently tuned in different places - some schools, like mine, assume that everyone is going to college and doing specialized knowledge work, while others assume that everyone is going to work in a factory or the gig economy. Both of these assumptions are pretty dumb. People who are going to work a trade shouldn’t have their time wasted by pointless academics, and people with unusually high abstract reasoning abilities shouldn’t be funneled into driving for Uber or something.

School Not Everyone Gets

The kind of school that not everybody gets (which, in the US, is typically college, though trade school also counts here) has two further goals:

  1. Indoctrinate people into specific subcultures (to include meeting other students and more established experts in their fields of study)
  2. Prepare people for specific jobs and, potentially, social roles

The failure point here, I think, is that some colleges and universities either don’t understand what they’re for or aren’t incentivized to do their jobs. However, of course, there’s a lot of diversity here (a statement that I’m sure applies to primary and secondary education too, though this is of course a hard statement to put a number on).

In the time since it was founded in 1636, Harvard University has produced 8 US presidents, countless CEOs and executives, and influenced, for better or worse, a large fraction of the American elite. For more than 200 years, Harvard’s educational objectives were explicitly moral - they had a normative idea of how a man (Harvard admitted its first woman in 1920) ought to be and what he ought to know, and their education was inextricably bound up with moral development. And this is a very good thing! I want our CEOs and presidents to be good people! I want the people they spend time with and ask for advice to be good people! I also want all of these people to be smart and practical. And if someone is not a good person, I want them to have less access to power. Harvard should be an explicitly moral institution - and today, it seems to me that they aren’t and don’t want to be. It’s both easier and less controversial to cash in on your status as gatekeeper to the American elite than it is to execute that role faithfully.

Optical engineering is one of the more underappreciated fields in the world of academia. Undergraduates and high schoolers typically don’t hear about it or consider it as an option, and it doesn’t get enough media attention to boost prestige or draw donors or grants. And so, the University of Arizona is one of the few schools in the US to have an excellent optical engineering program, despite the fact that optical engineers are absolutely essential to a number of critical manufacturing processes, including all semiconductor fabrication. Graduates of the University of Arizona optical engineering program can expect to be actively headhunted by companies that routinely reject MIT applicants, such as Intel and Nvidia - and this is especially impressive when you discover that students can expect this even when they only complete the two-year associate’s degree program. The University of Arizona, with its acceptance rate of 85%, can punch way outside of its weight division because it focuses on strictly relevant material and takes a highly practical approach to educating its students in an important field.

Harvard stopped educating their students in classical literature (and many other areas) to instill moral virtue and I’ve criticized them for that, but I’d be dismayed to see the University of Arizona assign their students The Iliad. Why? Because Harvard is educating a generation of aristocrats and Arizona is educating a generation of engineers, and those roles are different. All else being equal, I’d want the optical engineers to be good people too, but it’s not vitally important to me.

Some jobs don’t require any school, and this suggests that lots of school isn’t for everyone. I’d guess that society currently needs about 25-30% of people doing knowledge work and another 30-50% doing skilled labor (varies a lot depending on how it’s defined). If this is right, we currently have way too many people going to college (a quick Google search gave numbers between 40% and 60% of young people enrolled in college). On the other hand, we probably need more people doing skilled labor, especially if the US is ever going to recover its ability to manufacture anything.

Also: about 200 US colleges have programs for intellectually disabled students; I don’t think an ideal system would be set up this way. Many of these programs are actually good and helpful in that they focus on teaching life skills and social fluency with the goal that participants are able to function independently, but these programs are a worthwhile investment for society to make for everyone who needs them, while academics beyond what everyone gets are unlikely to be particularly useful to an intellectually disabled person. In other words, I would want to see these programs set up independently of colleges and made freely available to anyone who could benefit from them, without prohibitive tuitions. And I want this because I think it’s the morally right and nicest thing to do, but also because I think it’s good value for money to avoid having adults who are completely dependent on their parents, who might lose the ability to provide for them.

Higher Education and Status

Every year, thousands of students pursue undergraduate degrees in theater, English, journalism, and communications. There are intangible benefits that accrue to these students beyond the financial and practical, but it is indisputable that these students pay a lot of money for an education with a pretty low expected ROI. It’s incredibly rare to be able to support yourself as an author, journalist, or actor, which means that colleges are taking money - sometimes a lot of it - in exchange for a service that they know is likely to be worthless (and in fact of negative value, when you consider opportunity cost). Similarly, thousands of students drop out of school every year - some of them due to unforeseeable life events, but many of them due to very foreseeable financial stress, which they should have been warned about before they enrolled. These students are even worse off, as they don’t even get diplomas in exchange for the thousands (perhaps tens or even hundreds of thousands) of dollars they give to their schools. There are words for selling something worthless for a lot of money: scam, con, fraud, modern art (kidding!). The question is: why do people fall for it, and how do colleges keep getting away with it?

I think the clear answer is status. Colleges aren’t, in fact, selling an education, or they wouldn’t be able to get away with not teaching you anything useful. They aren’t selling earning potential, or they wouldn’t be able to get away with abysmal ROIs. They are selling the status that comes with being college-educated. And you get more status for going to a better-known college. This is why US News and World Reports, which publishes college rankings in a number of categories each year, is so powerful: they substantially control public perception, and public perception is the product colleges are selling.

This creates a misalignment of incentives. Colleges are trying to maximize some joint function of profit and status (Malcom Gladwell has an absolutely insane interview with the president of Stanford, who tells him that there is no donation so large he wouldn’t accept it), but the best way to do this usually isn’t to effectively prepare people for their jobs and social roles. Harvard has the unusual problem that it’s no longer as socially acceptable to look too powerful or ambitious, so they have to teach people to go after money instead. Other colleges tend to have the reverse problem: they can get more prestige by wasting the time and money of their students under the guise of “teaching them to be well-rounded.” Naturally, they are never actually successful at this, because everyone involved knows that the college doesn’t actually care about how their students develop as human beings. This regression to some mean perception of what college in general should look like hurts everyone. Some college educations should look a lot more like trade school; others should look a lot more like the aristocratic education of the 1700s.

Students also value status; their incentives are not necessarily aligned with the best interests of society either. This is why we have way too many people in college: even though you’ll probably do much better financially and deliver more social value by becoming a plumber instead of an English major, plumbers are accorded less status, so the job market doesn’t allocate labor efficiently. I think Bernie Sanders’ free-college-for-everyone scheme is a bad idea in that we don’t want or need everyone to be in college, but it resonates similarly for a lot of people to ideas like “everyone deserves respect.” The better solution here is to respect people even if they don’t go to college. But lots of people don’t do this, so we end up with many students feeling deeply betrayed by their college educations, even if they would never admit this to themselves. Sadly, these students don’t even get to keep what they paid for - most of the jobs we all called “essential” during the COVID pandemic are not high-status. Your barista at Starbucks can never admit, to you or to himself, that his degree in communications was a waste of $100,000 that will keep him trapped under a mountain of debt for decades, but there’s a good chance he is nonetheless incredibly angry at the entire system.

Tangent: Money, Power, and Culture

In human history, it has generally been much more socially acceptable in most places to seek to accumulate power than it has been to seek to accumulate money/resources. In some ways this makes sense: it’s safer and easier to redistribute resources than to disperse power. The most notable exception in the ancient world was the city of Athens, which famously banished (ostracized) anyone who was becoming influential enough that they might threaten democracy. Given this, it’s sort of interesting to see that it’s currently more acceptable in the US to pursue money than power, even though this hasn’t always been the case. I’ve been frustrated for a long time that many of the brightest minds of our generation work in advertising and finance, but it also makes sense to me: you would have to be either extremely disagreeable or extremely ambitious in order for it to make sense for you to pursue power in today’s environment.

This is a cultural problem, and, as I’ve said, schools at all levels are institutions responsible for transmitting culture. I think what goes wrong here is a perhaps understandable desire for schools to present as value-free or apolitical. In some ways, this is a practical decision: if you take a stand on some particular value and some political group doesn’t like that, you become a target in the culture war. Also, taking a stand as an institution means you have to do work aligning your institution, so being value-free is in many ways the default. The problem is, there’s no such thing as “value-free” in that there’s always a default/dominant set of values that get adopted in absence of intervention. Those values might be different in different places, but I usually see many elements of neoliberalism in the default set, where people adopt the logic of markets and competition in their attitudes towards themselves and others. I don’t think this is particularly good; I would rather see almost any set of values explicitly promoted and offered for examination than neoliberalism unconsciously absorbed as the natural or only relationship you can have with yourself and the world.

Primary/Secondary Education, Unschooling, and How People Learn

One thing you might have noticed in my list of purposes for school everyone gets is that it doesn’t involve teaching. This isn’t because I think people can’t, don’t, or shouldn’t learn a lot when they’re young - just the opposite! I think most children are more capable than they’re given credit for, and traditional schools, as an artifact of their Industrial Age design, tend to unnecessarily play to children’s natural weaknesses instead of their strengths, making school an emotionally challenging environment that even adults usually find aversive. You probably have a lot more patience and self-control than a child; would you enjoy sitting through a day in elementary school? My list, then, is intended as a set of goals that I think should be common to all students. Especially in primary education, I’m not sure an agenda should be set beyond this one.

In How Children Learn, John Holt makes what I think is a very important objection to traditional schooling: if we set out to teach children based on our own agenda, what are we telling them about the things we try to teach? Children know that they learn on their own, because children spend most of their time learning things on their own. If we have to teach them something (let’s say, for example, reading), there must be some reason they can’t or won’t learn it in their own time. There are two possibilities:

So, it is natural for children to conclude, I am being forced to read either because reading is bad or because I am bad. And so on for almost everything. Children, as they so often tell us, want to do it themselves. And parents who choose to “unschool” their children (that is, non-coercively educate them, usually at home) report that they do, in fact, do it themselves. That isn’t to say that the children learn to read entirely without help, but they do invariably choose to seek help learning to read. If a child is raised to view herself as smart, capable, and agentic, it turns out she naturally wants to learn all of the things it’s really important for her to learn. It is only when we assume that she is stupid and lazy and act from that assumption that we are proven right. Fortunately, children are resilient and damage can usually be undone.

A brief personal side note on this: when I was young, my parents signed me up for piano lessons, which I hated. My piano teacher barely lasted two weeks, and though I briefly taught myself to play a little in college, I never learned to play fluently. But I think this is because, despite the fact that I had a piano in my house, I never once saw my parents play it. Why should I have spent time learning something that no one around me seemed to think was worthwhile? At it turns out, I’m exactly the sort of person you’d expect to learn to play the piano, and at different points in my life I’ve learned to play a number of instruments purely for fun (though not terribly well). I think that if I grew up in an environment where people regularly played the piano for fun, I would have learned it enthusiastically.

Unschooling isn’t the only right way to educate a child, and I think it’s very valid to point out that traditional schooling gives kids much more exposure to a diverse peer group. If you’re doing it alone, unschooling totally fails to accomplish the objective of taking the burden of childcare off of parents who may both need to work. But unschooling doesn’t need to be done alone! I think primary and secondary education would ideally much more closely resemble a learning cooperative than traditional school.

Learning cooperatives partially incorporate a concept that dovetails nicely with self-directed/non-coercive learning: aristocratic tutoring (recently popularized by a Substack post by Erik Hoel). The tl;dr on Erik’s post is that aristocrats used to educate their children by essentially hiring graduate students to individually coach them in a diverse range of subjects; it wasn’t uncommon for an aristocratic child to have 10 or more tutors. Erik thinks that the disappearance of this practice is responsible for a decline in the number of exceptional thinkers relative to the past. I happen to think he’s wrong about the second part, but the first part is very intriguing. Like I said earlier, we have a ridiculous surplus of college-educated people. I have to think a lot of them would much rather tutor than wait tables; hire your kid a grad student! Not to prepare them for tests or anything, just to talk about whatever your kid thinks is cool.

Tangent: Test Scores

I feel like I hear about average test scores a lot. There’s an average test score gap between the US and [insert Asian country here] and that’s a BIG PROBLEM that needs to be fixed RIGHT NOW. Except… it isn’t? Average score is such a terrible metric when the skills that allow you to be good at, for example, standardized math tests are completely irrelevant to 60% or more of your population. John Holt has another amusing point in How Children Learn: when was the last time in your day-to-day life that you divided one fraction by another fraction?

Even test scores for individuals aren’t really that good for the things they’re used for: maybe it makes to track progress with individual test scores, but scores capture way to much confounding information to compare people with each other. With that said, though, I do understand the argument for standardized tests in the college admissions process, in that standardized test scores are some of the best datapoints in the incredibly low-quality dataset available to colleges. Most of the more “holistic” criteria are hopelessly skewed towards wealthy students.

Conclusion and Practicalities

If I have a child, my first choice for primary and secondary school would probably be to give them the option to choose between a learning cooperative and traditional school. If a learning cooperative wasn’t an option, I would hope to be able to sort of roll my own by collaborating with 10-20 other families in my area (assuming my child preferred that over traditional school).

I think that in some disciplines, college isn’t as important as practical experience and it’s probably worth going to an easier/cheaper/less prestigious college so you can get the piece of paper HR departments look for while you do actually relevant things on the side. But this also isn’t always the case - for example, Yale Law and Harvard Business School really are the best law and business schools, not because they teach you law and business really well, but because law (some types in particular) and business rely on building social networks and Harvard and Yale attract people you want to have in your social networks. There are also plausibly some fields (maybe chemical engineering?) where self-study is really hard and high-quality mentorship is really important. So I would encourage my child to figure out what kind of field they’re in and make their decisions based on relevant signal rather than marketing materials (written by people who don’t know what they’re talking about) or US News and World Report (written by people who don’t know what they’re talking about and also probably extremely Goodharted at this point, even though they’re smart enough to keep their scoring algorithm secret).

Basically, my conclusion is that the entire educational system is a mess of misaligned incentives and unclear goals. I think people should take their kids seriously and be fair towards them and try to avoid making them feel like they have to earn approval or affection. I think the middle class has been pretty effectively duped into competing for status that won’t help them. But also, I think that if you start by being clear about what you want and what you’re going to avoid, there are pretty good ways forward, and things will probably work out OK! I’d hope that in a hundred years, people will look back on modern schools a lot like we look at schools of the 1920’s: oppressive and outdated. And I’d hope that the children educated in those schools of the future grow up to be kinder, freer, and more open people than we are today.