Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

You've obviously looked into the history of the system to a deeper degree than I have. I not only appreciate the effort, but I respect your opinion as to my point.

Do you figure I'm making a good point, or do you disagree? If I'm wrong, what is the point of school? Why does every country do it?

Again, curious to hear your thoughts.



Based on Rothbard's description... The educationists who pushed for it were not a completely homogeneous group, although they were reasonably well organized.

Some simply thought education was good and weren't bothered by the coercion.

Some subscribed to the absolutist State view going back to Luther and Calvin—people saying things like "in these schools the precepts of morality and religion should be inculcated, and habits of subordination and obedience be formed", "that a teacher must lead his students to accept the existing government", "the children belong to the State and not to their parents", etc.

And some had egalitarian aims along the lines of what you describe. "The vigorous championing of the public school's leveling role appeared again and again in the educationists' literature. Samuel Lewis particularly stressed that the common schools would take a diverse population and mold them into "one people;" Theodore Edson exulted that in such schools the good children must learn to mingle with the bad ones, as they will have to do in later life", and others are cited too.

Although you may be surprised at how far they go. Rothbard says that Owen and Wright—Wikipedia identifies them as utopian socialists—had very extreme plans involving the State raising all children away from their parents in a uniform environment, as the logical conclusion of making the childhoods of the rich and the poor be totally equal; those are the only two he cites advocating this, but he claims they were very influential, the essays published with approval by a great many newspapers, etc. Could be.

Back to what you said: The "including teaching it's boredom, authority, and pointless busywork"—well, authority yes, but the boredom and busywork seem pretty clearly not part of the original plan. I think that's an accident which has developed since then. Which brings me to this point:

Supposing that the above describes the state of affairs in the mid-1800s, a great deal of time has elapsed since then. Private companies with clear goals develop cancer (that is, pockets of people acting selfishly at the organization's expense and subverting the organization's usual methods of enforcing its objectives) on smaller timescales than that. You should expect some ... evolution.

Generally, I expect incentives to direct the evolution of an organization. If we consider the "customers" of an organization to be those who decide if it gets funding, then the customers of a public school are primarily the voters or the politicians they elect, not the students or the parents—except insofar as those can influence voters. (Also, sometimes a school is allocated funds proportional to the number of students, so students voting with their feet could be a force; on the other hand, switching schools involves hassle, losing one's entire social group, and likely a longer commute, so it's likely not used often.)

We might expect this to mean that schools would tend to implement policies that make things pleasant for school employees at the expense of students, as long as they can't be turned into public relations disasters. Continually increasing the homework load, starting school extra-early for teenagers despite decades of research about sleep schedules showing that's a bad idea, and assigning busywork (either in the sense of "literally to keep the students busy while the teacher can sit and relax", or in the sense of "make all 20 students do an extra 5 problems because of the 1-2 people who need the practice") are all examples of this. (Note that, for all these, opposition can be dismissed as "the complainers are lazy bums".) I've had a case where a teacher—a nice guy and a smart guy, I liked and respected him—told me that, if he made an exception to implement my request, he'd have 100 parents beating down his door with their own requests—so, keeping his negotiating position with parents was more important than designing a better program for his student (well, from his incentives, it was more important).

It could be worse. The book "Little Soldiers"—based on what I've read[1] about it—depicts what is apparently considered good practice in modern Chinese education. Students starting at age 3 are required to sit very still and do a delicate and pointless task over and over, and are yelled at whenever they mess up—all to establish "discipline" as early as possible. Parents are expected to fawn over the teachers and, under the table, to buy them gifts. This seems pretty clearly the end state of "optimized for school employees at the expense of students and parents" (well, I guess I could imagine it going even farther...). It seems American parents are more unruly and have a better negotiating position than that.

But generally, beware of ascribing purpose to any particular part of school. If the school was laid out in a coherent design by one person, and it's relatively young and small, and has good competition—then maybe you can expect most of it to be part of the plan. But public schools that have existed for a long time... When you experience busywork, it's not the taste of egalitarianism; it's the taste of organizational cancer.

[1] https://old.reddit.com/r/slatestarcodex/comments/cz48fp/litt... and the discussion at https://slatestarcodex.com/2020/01/22/book-review-review-lit...


Maybe if I limited my concept of the 'point' of school to more specifically describe the average incentives driving the average parent in the current decade to send their child through the formal education system, rather than homeschooling?

Would you then consider it possible to draw a link between that and a tractable concept of the 'point' of school.


I would probably refer to that as "the point of sending your kid to school". Note that this is different from "the point of running the school the way it's currently run" (which is how I would probably interpret the bare phrase "the point of school"), and also from "the point of maintaining compulsory education laws".

It may also be different from "why parents choose to send their kid to school", because parents' perception of their choices may be incorrect—I think it often is.

And, of course, it'll be different for different kids. You say "the average", but as some say, the average human has one ovary and one testicle; taking an average across disjoint groups may not make much sense. A high school diploma is one of the most cited reasons for going to high school (not K-8), but that would really only matter if the high school itself is a brand name, which is rare (you probably have to be rich, talented, or lucky to get in); otherwise a GED should work equally well for college. For the upper middle class trying to get into elite colleges, an upper middle class high school will offer plenty of extracurriculars with which students can pad their applications. For socially adept poor kids that manage to live in rich districts, mingling with the rich kids may be useful. For kids with terrible home environments, school may be a decent escape. For kids that fit well into the social environment and aren't bothered by all of school's problems, it may be fun, good for making friends, and even a decent educational option. I think most of the above groups don't overlap much (except the "socially adept poor kids" obviously "fit well into the social environment"), and forming a picture of school's "average target student" wouldn't look like a real student and would probably mislead more than enlighten.

For kids in reasonable home environments, who are irritated by school's problems, who have friends and sports or other group activities they do outside school, who have access to libraries and now the internet and Wikipedia and free online courses, which are getting better while school seems to get worse due to its cancer... it's probably not the best choice (the kid's individual traits like motivation and curiosity may be deciding factors here), and as trends proceed that will become more clearly and generally the case. Parents' impressions of the available options are probably decades out of date.

As for the incentives that face parents themselves: for many, public school is primarily publicly funded daycare; for some parents, there is no reason other than avoiding punishment from truancy laws. Honestly, I suspect the strongest motivators in practice are social conformity—not sending your kids to school would be shameful or weird, make you stick out and become a magnet for criticism—and ignorance of alternatives.

Most haven't heard of unschooling, and probably think homeschooling is for Christian fundamentalists who hate evolution, or for geniuses; those who know more still usually assume that it requires a parent to act like a full-time tutor, and haven't considered options like "set up books and textbooks and ask your child once a day what they learned and if they have any questions". Some have thought that far and think their child would never be productive without constant supervision; the homeschooling advocate's response is that your child has spent years being taught that "education" = "adults imposing boring unpleasant crap on you", and it may take many months for that to fade away, but their motivation should improve eventually; still, it seems possible to me that there's significant intrinsic individual variation in motivation and curiosity that would make the difference.

It is disappointing when adults assume the worst about their own children and knowingly condemn them to years of misery; they could be correct, but what a thing for a parent to believe... One can also ask: if you really believe that, then how do you expect them to do well in college or in a career? Maybe they assume the motivation will automatically improve as the frontal lobe finishes growing (which is possible), and that faking it until then will get them into a place where they might develop hitherto unseen qualities and do well. To which I merely sigh, shake my head, and say life should be so much better than that, and for many reasons it's worth the risk of trying it the right way.


I guess a counterexample would be that kids can leave if they want to. They have to be organized, and highly motivated, which is also kind of a measure of how prepared they are to function in the real world. They also have to prove that they have achieved the level of knowledge expected of a graduating high-school student.

Kids can take the California High School Proficiency Exam, leave high school at 16, do two years at a community college, and transfer out to a state school.

I wound up taking a bunch of community college classes at night, my high school accepted the credits, and let me leave a year early.

Only a handful of kids seem to actually follow-through with it, but it's totally doable if you're organized, willing to work, and really want out.


> Kids can take the California High School Proficiency Exam, leave high school at 16

Yep. Though I didn't do any college after that. I messed around with mathematics and programming for about 5 years, then made a serious effort to find a programming job, and have been working since then.

I didn't know about the CHSPE until 9th grade, though, when my sister told me about it (because she had a friend who used it too). Before then, my impression was that there was no feasible alternative to school. Certainly no one at school mentioned it.

Also, you're not allowed to take the CHSPE until either age 16 or in the second semester of 10th grade. Which strikes me as stupid. Seems to me that, if you manage to pass it at a younger age, then that is a stronger proof that you're smart/hardworking and shouldn't be stuck taking 9th grade or 8th grade or whatever. (I imagine most of my peer group could have passed it in 6th grade; it is really a test of basic minimum competency.)


I personally wasn't as intrinsically motivated to pursue a given field. I just wanted to get out and do something more interesting, discover the real world, and explore career options. I picked up Morris Kline's Calc, aaaaand then I put it back down and signed up for a class.

What would you attribute your deep intrinsic motivation to? It's admirable.


Thank you, let's see. One part is probably that I was brought up to admire great scientists—my dad gave me popular science books by Stephen Hawking ("The Universe in a Nutshell" in particular) at a young age, and some Feynman books, and such. (I remember "Black Holes, Wormholes, and Time Machines" by Jim al-Khalili from the library.) For a long time I figured I should become such a scientist myself.

Another is probably high self-opinion and wanting to prove I was the best. In elementary school, I entered the chess club; there was a five-week tournament (one game per week), and I studied chess books my mom gave me and by the end of the tournament was the strongest player in the school. (I only tied for first place in that tournament, but I had improved enough by the end that I believe I won all subsequent rematches against the guy I lost my first game to.) Upon entering middle school, I took a Mathcounts test to determine who would be on the school team, and I got the highest score in the school (which was actually kind of a fluke, because I had a friend who generally outperformed me at arithmetic-based contests and did so at the actual Mathcounts events—but I tended to beat him at proof contests later on). I did lots of math contests, and generally did "rather well". The pinnacle of that was qualifying for the math olympiad camp in 9th grade, from which the U.S. International Math Olympiad team is selected (although people below 12th grade have a lower bar, and aren't in the running for the team unless they've met the higher bar; my achievement isn't that impressive). So my self-opinion had some justification.

Regarding calculus textbooks, in the summer after 7th grade I went to one of my sisters and said, "You know, I've heard of this thing called calculus, but I don't know what it is"; she gave me my oldest sister's calculus textbook (Ostebee and Zorn), and I read through chapters, and did problems out of the the chapters until I figured I understood it, and moved onto the next chapter. (I didn't go through all the chapters—towards the end it did multidimensional stuff that just got boring.) With this as my sole calculus education, I later chose to take a calculus round of a math contest, and got a "decent" score (IIRC it might have been in the top 10%). :-)

So, for quite a long time, I wasn't actually sure that I'd ever met anyone who was smarter than me. (My abovementioned friend was certainly close, and was better at some things.) At the math olympiad camp, I did meet people who were clearly significantly better than me at math contests. I figured that a certain amount of that was due to drilling, which was something I didn't do (I mean, I went to math club and did whatever they put in front of me, but I didn't practice contests at home), which gave me a way to suspect that I might still be at least as smart as them. :-) I knew this was a self-serving line of thought, but I figured that, as long as I knew it was specious, it was harmless to indulge it and let it motivate me. I think Colin Percival is the clearest example of someone reasonably close to my age that I'd have to bow down to. (Though even then, if my education had been properly arranged, instead of wasting most of my K-10 years... Well, that is an experiment I hope to carry out multiple times in the next generation.)

The olympiad camp actually gave worksheets that they recommended we do throughout the year. I did not want to pursue that, or pure math in general (although I thought I was good, I didn't think I had a good chance of, say, proving the Riemann hypothesis, and anything less than that didn't seem worth it), and refocused on computer science. (I also had discovered Paul Graham's essays around 8th grade, Scheme in 9th grade, and took an AP CS class in 10th grade, whose curriculum included SICP—yeah, that teacher was pretty cool.) After leaving school, I did over 100 Project Euler problems, and did other "programming for math" stuff, although I drifted towards pure programming stuff (implementing languages, specifically).

Now, don't get me wrong. I'm not some kind of work-maniac like Isaac Newton. I've spent plenty of time playing computer games, and still do. But I have some reflexes that have served me well. (1) Whatever you're doing, do it well. If you're doing it well, see if you can do it even better. (2) Keep an eye out for opportunities to practice your skills. When I got irritated at repetitive Youtube comments, I used curl/egrep/sed/sort/uniq to count how many of them matched a particular format. When I felt frustrated about limitations in EV Nova or Civilization II or other games, I did some work towards reimplementing them myself—which, in the case of EV Nova, led me down a rabbit hole of researching real-time garbage collection, which occupied my productive efforts for some years. When I set an alarm clock, it is a shell command of the following form:

  ~> sleep (math "3600 * 8 - 600"); pelimusa
where "pelimusa" is a command that sets the computer volume (in case I muted it) and then plays a song[1] in a loop until I control-C it. (3) You should know how to do everything. If there's something you don't know how to do, it should irritate you a little or hurt your pride. (4) Don't do unnecessarily stupid things. Being smart means you can do very stupid things (which other people wouldn't even try, or think to try), and if this actually leads you to cause permanent damage to yourself or other terrible outcomes, then you really weren't smart in the ways that mattered, which is unacceptable.

I was in the weeds of implementing my idealized real-time GC in x64 assembly (which may have been a bad idea, but I had my reasons, and I didn't have anyone whose opinion I trusted advising me against it; at least I learned a lot), and eventually my family started putting pressure on me to start earning money. So I did. I got a referral to Google—which turned me down, by the way, which... lowered my opinion of them :-P—and then to a few other places. I've been working at a medium-sized Silicon Valley tech company since then.

Today, in my job, the things that motivate me are pride, curiosity, and lolz. Pride could be summarized as proving to myself that I'm the best (however I might choose to define it), and that those who have invested in me made the right choice. Curiosity is self-explanatory. Lolz is when I use absurd means (like overwriting executables in /usr, or piping "yes" into installation programs, or running "kill -STOP; kill -CONT" in a loop to slow down a program) because they're actually a very effective and quick path to what I want.

Beyond my job, I want to fix things for future generations, perhaps to prove that I'm right that there was a better way. John Holt tried to reform school, ran into the cancer (as well as the astonishingly common view that children are no damn good), and concluded you had to work outside school. Seems about right to me.

Civilization II has taught me that exponential growth is the most powerful strategy ever. If I can create several people like me, each of whom creates several more, and so on, then that can achieve a lot more than I might do myself. (I am being 70% serious.) Even if they're not as smart, raising them properly (i.e. no goddamn school, plus find them some cognitive peers, and generally have an intellectually rich household) may more than make up for that: the improved education and non-isolation (my social development was certainly hampered by growing up 99% around kids I couldn't relate to, which may have made the difference in my squandering some opportunities) and possibly improved motivation (having 6+ hours a day forcibly wasted is very demotivating). In the worst case, they'll still carry my genes and can try again with the next generation.

I encourage smart and motivated people to try this too. At some point we may figure out how to directly increase intelligence (that would be one of the projects I would hope a kid of mine tackles); for now, the most obviously workable approach is to find a suitable genius (with no severe mental health or personality issues) who is willing to give you their genetic material. That may take the form of marrying them, if you're lucky. Once you have the kids (plural; one of the easiest ways to have peers is to have them at home, although you can't guarantee they'll be friends), the default approach of unschooling seems best: buy good books and intellectually stimulating toys, help them meet friends, maybe experiment with 1:1 tutoring if you have the resources and they like it, support them in activities they seem interested in, but fundamentally just make sure they are fed and watered and loved, and trust to their natural curiosity and to luck. (There is going to be a large luck component. Another reason to have multiple kids.)

[1] https://www.youtube.com/watch?v=XKs5mkEpWuU




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: