I read this fantastic article on Astral Codex Ten. The primary thesis being that the role of school was motivating kids to learn - despite it not actually being the most effective way to learn. But that motivation structure is needed to actually put in the work to learn - which is hard to do with the alternative forms of education that have been tried.

The core design of our schools – age-graded classrooms where all students are expected to learn more or less the same curriculum – are the worst form of motivation we could invent…except for all the others. While school is not particularly effective at motivating students, every other approach we’ve tried manages to be worse

A jumble of my thoughts and takeaways on the topic.

No-structure, low structure and high structure

I liked the division of students into : no-structure, low structure and high structure kids.

No-structure kids (5%) : learn effectively with minimal or no external guidance of accountability. They have high intrinsic motivation, can be self directed (find resources, keep themselves accountable). They don’t need teachers and often learn despite school. They are therefore bored in traditional classrooms because they already know the material or learn it faster alone.

Low-structure kids (75%) : students who need the basic motivational structure of school to get through learning the topics. They need the assignments, routine, ‘social pressure of everyone doing the same thing’. The quality of the teachers matters less, and it’s more that a structured map is provided. They may not learn on their own but they’ll learn whatever the class learns and keep up with peers without struggling too much.

High structure kids : students who struggle despite the standard school structure and need explicit support. They have low intrinsic motivation for academic work, avoid schoolwork, need constant check-in’s reminders, scaffolding. The quality of the teaching and support matters a lot and they will fall through the cracks with mediocre instruction.

He gives the example of someone studying computer science. The no structure kid is always programming on their own and learning from stack overflow outside of class, and occasionally showing up to class to ensure he gets a degree.

The low structure learner shows up to class, with the coursework and lectures being enough motivation to learn, but they aren’t learning much outside of that on their own. The high structure kid has a tough time, and they are using all the support - the extra office hours, the tutoring etc.

I would also add that people move in between these categories depending on what one is studying. For reading and programming, I was a no-structure kid, often just learning whatever I needed to on my own - the intrinsic motivation was high.

But for mathematics (and I think many other subjects), I was a low-structure kid - I needed the classroom structure, the regular assignments and accountability to keep going. I recently tried Math academy as a way to self-learn mathematics, and it sadly failed. I got through about 1 month before I quit.

Motivation structures

Motivation is a hard problem. It’s no wonder why MOOC’s have a ridiculously high attrition rate - they just don’t work.

Learning things is hard. There was a sister post to this called ‘Alpha school : a review’. In it he touches on deliberate practice looking at elite students, asking why did some students practice more than others? An initial conjecture was that they enjoyed it - but that was false - “The future elite soloists of the music world all hated practicing”.

So if the best students all hated practicing, but deliberate practice was how one improved, why did some students practice more than others?

Because they had stronger more developed motivational structures that kept them coming back to practice and put in the painful hours. He talks about motivation evolving through three stages

Parental and authority approval - initially kids practice because they get praise from parents when they do and are reprimanded when they don’t
Peer approval - then young musician start to care less about parents and more about relative status games amongst peers. You play status games with your peers to motivate one to improve.
Self actualisation - the best musicians eventually internalise their desire to be great - they see themselves as a musician and do the hard work of practicing because ’that’s what a great musician does’. i.e. You tie your identity to it.

School does 1 and 2 well, and allows for ‘3’ to potentially occur.

School provides basic scaffolding, accountability, motivational structures so that most students (who are low structure) leave school with a basic understanding of maths. The kids however who were no structure for maths, would often sit at the back of the class and doodle, having already taught themselves the lesson.

An argument is that those kids could be moving faster on their own, without school. I agree. But I also think that the kid is not like this with all the subjects. For example : someone might be brilliant at maths but not care for history or struggle with English.

Having school means that they keep on plodding along on those ’low structure’ subjects, that they wouldn’t be motivated to go into otherwise.

Any argument is then : why have them learn that?

Another role of school I think is to provide some form of ‘common knowledge’. To impart the idea, that there is depth to this subject, that other people know. i.e. that there can be experts in this field, and we can trust some of what the experts say.

If we didn’t have that - we wouldn’t know what we don’t know - we lose a common knowledge. As one of the commenters says in the post, if we didn’t have that kind of common knowledge, society would not be stable.

You just know enough to know that you don’t know, and that there is real depth to the subject.

If you become too specialised, you may leave a brilliant maths student, but have no idea what Democracy means, or that slavery was a thing. I’d argue that leaves you unable to thrive in society.

Hold on : what is motivation?

A criticism in one of the comments is that the author doesn’t really define motivation. He also has this implicit premise that having high motivation means that you automatically learn ‘better’.

I know people who have ‘high motivation’, yet consistently fail to accomplish their goals, and others who have a bit of motivation, but keep on showing up regularly and succeed.

So what do we mean by motivation?

I would argue that instead of using the word motivation, the author might mean that schools and universities provide ‘structure’ such that we keep on showing up to do the work.

Either because (i) you’ve paid for it (university) (ii) it’s societally mandated (school) (iii) you’re identity is in it, you like collecting imaginary points in the form of exams and degrees

It allows for the formation of habit loops.

If we’re lucky, we then move down that motivational structure (authority approval -> peer approval -> intrinsic motivation/self actualisation).

Do we choose what we are intrinsically motivated to do?

Is intrinsic motivation baked in? Or is it a trainable (meta) skill?

Scott Alexander in another post ‘Lottery of Fascinations’, argues that we don’t choose what we become intrinsically motivated in. I think I largely agree.

The best you can do as a parent is to expose your children widely to various fields, start inevitably with external incentives (that school provides - social status, authority approval), and remove barriers when they show interest.

Excessive external pressure may actually drown out the arising of natural intrinsic motivation.

It’s a delicate balance - too much pressure crushes emerging interest, too little means no exposure at all. The goal is creating conditions where intrinsic motivation might take root.

Learning as an adult

I was very much a low structure kid, as I think most people are, for most subjects. It’s hard to create motivational structures. It’s also why learning things as an adult is quite hard! It requires you to keep coming back, putting in the hours and grinding with deliberate practice - and that is difficult to do without accountability.

Obviously you can do it - I’ve done it, I know many people who self teach themselves as an adult. But I also have so many aspirational projects sitting in the graveyard right now - attempts to self study Mandarin and mathematics. We know this intuitively. Technically all the information we need to self study anything is online - it’s just a problem of motivation (or more accurately creating habit loops and motivational structures such that one keeps on showing up!). Online programs don’t really work in this regard:

The core problem with these online programs is having every student work independently, without any connection to what the students around them are learning. That just doesn’t motivate many students

So in the essay, the author says that school is the best method we’ve got (despite it having multiple problems)

In the absence of one-on-one tutoring for every student, conformity is the best tool we have to create the motivation necessary for learning.

Suppose you want to learn a skill as an adult. The truth is that the early stages of any skill development are disproportionately hard, and it can help to have external accountability and support in this time.

The idea is that you can move down those motivational levels with external accountability, such that intrinsic motivation arises, and you develop strong enough habits such that you keep on showing up to do the work.

After having self-learned a few topics, I think if you want to seriously learn a topic as an adult, the best you can do is

Use strong external structures (paid courses, coaches, study groups, or just sign up to a degree…)
Develop sustainable habits such that you show up
Hope that intrinsic motivation emerges (if you are lucky) - and that a virtuous feedback cycle develops

So to have a strong chance of success for learning something on your own

Pay for the thing
Do it with a peer group/accountability structure
Have a coach or mentor (more external accountability)
Do it daily or at least regularly

Conclusion : school aint going nowhere

The author ends with the prediction that the schooling structure is largely going to stay the same, despite all the advances in AI etc. Because it’s the best tool we have to motivate the largest number of students, despite it not being particularly great at ’educating’.

I came into the essay, disagreeing with the author. And I left, with my opinions and view on the topic altered. This is what a good essay does.

I find myself agreeing with the author that the basic structure of school is not going to change anytime soon. School also acts as a daycare for when the parents go to work, and that isn’t changing soon. They also play a social function. They ensure that your kid will come out with basic numeracy and literacy, even if some no-structure ’learning’ has to be sacrificed along the way.