Life isn't about getting to the right answers; it's about discovering the right questions.

Crossposted from Curious Human.

In school, we're constantly being asked questions and instantly rewarded when we find the right answers. We're praised in class for saying "25" when we're asked what "5x5" is. We're given perfect scores on tests when we answer every question correctly. We get good grades for repeatedly getting good scores. Students who give the right answers when asked are "good students".

We're also instantly punished when we propose the wrong answers. We get scolded, lose points, and get poor grades. Students who don't give the right answers are "bad students".

For 16+ years (12 grades + 4 years of higher education) we're literally conditioned, in artificial learning environments, to find the right answers to other people's questions. How often do students finish a paper assigned to them in a class and want to write more on the subject? Rarely, if ever. Because the questions - the essay prompts - were never truly ours in the first place.

It's the same for every subject. Take the way math is taught in schools. General concepts are broken down into extremely specific questions. Students are then given a list of methods to answer each specific type of question, which they memorize for the sole purpose of using them to quickly find answers on tests. This process only makes sense for basic, fundamental concepts (times tables, finding the areas of common shapes, etc.); anything more complex and it quickly becomes unnecessarily complicated. Even worse, it detaches math from reality and ruins its purpose, which is to allow us to take first principles, generalize them, and apply them to as many real-world cases as possible.

We weren't designed to learn this way. Have you ever seen a child learn something new? They're playing, and suddenly something sparks their interest. They come up with a question. You see their eyes light up as their curiosity takes over. At that point, they're insatiably motivated to learn - it's only natural.

The things you learn by yourself stick; the things that are “taught” to you do not stick. - Nabeel Qureshi

This way of learning - by starting with our own questions - isn't just more effective. It also gives our lives meaning. I think the beauty in life is precisely in finding our own big questions - with no "correct" answers - that we want to spend our time solving. One of Elon Musk's is "How do I get to and live sustainably on Mars?". Albert Einstein's were "What the fuck is up with light and matter, the motion of particles in a liquid, special relativity, and energy-mass equivalence?".

Our questions don't have to (and realistically won't) start that big. I like the way Paul Graham put it: "The way to get a big idea to appear in your head is not to hunt for big ideas, but to put in a lot of time on work that interests you, and in the process keep your mind open enough that a big idea can take roost."

So the fundamental principle of education should be to give students an environment, and tools, where they can make discoveries themselves. Teach them the most basic foundations of knowledge - math and language - and give them space, time, and autonomy to explore. All human beings are born curious (how could we not be; the world is fascinating!). Don't destroy it by training people to search for answers. Life isn't about getting the right answers; it's about discovering the right questions.


New Comment
10 comments, sorted by Click to highlight new comments since: Today at 3:08 AM

To complicate this, many questions require knowing a lot of answers to even conceive of.

School isn’t about accumulating knowledge. We forget so much.

It’s about becoming familiar with the intellectual signposts put down by earlier explorers. Being able to find your way to the frontier.

Only a little of what you find will be generally useful. You’ll not learn much about how to learn. But for the minority of people who are especially curious, it offers a way to fit the answers together until you’re ready to ask - and answer - unsettled questions.

And that is important.

The real question is how we motivate more people to maintain that curiosity. And how we create support for them to explore. Can school be less of a maze and more of a map?

The value of Montessori and other alternative educational paths is that it creates a niche that appeals to a specific type of student. Much like having special needs and gifted classes, black colleges and international schools, vocational training and homeschooling.

The more options there are, the more individuals can find the right intuitive fit.

I’m not always a fan of unlimited options. I don‘t need 16 types of peanut butter or more than one romantic partner. But a society with many approaches to education, including more exploration-centric styles, seems very good to me.

I disagree because I think you're forgetting humans are also inherently pretty lazy. Evolution programmed us to conserve our energy, and that includes our brain power. In the ancestral environment, we worked because we had to, we learned because we were forced to learn to survive (ie. learn to plant crops or hunt prey so that you don't go hungry).

Children and teenagers are also pretty terrible at delaying gratification, as countless psychological studies have shown. Forcing students to study specific problems, upon threat of failure, mimics the evolutionary environment of "learn or die".

Personally, I think I would have learned very little as a child and teenager if I hadn't been forced to by my schools and parents.

Kids (and adults) are only lazy in the context of being made to do things they don't want to do. Kids who aren't subjected to school have lots of energy because they're exploring things they're excited about. Learning is playing for them.

I think we need first to understand what the purpose of schooling is. When you say "terrible at delaying gratification" and "learn or die", is it something that helps one to be an achiever, or something that helps one to live a happier life?

I think the schools have been long useful for 3 main purposes:

  • Being a (mostly unsafe) training ground for socialisation.
  • Producing "standardised" packages of knowledge, which is useful for economy, since everyone knows what to expect.
  • Encouraging people with a tendency to overachieve to excel at it.

Of these three, only the first one is strictly for the benefit of the student, and even at that it is implemented quite poorly. The thing is, the schools were never designed to maximally help somebody in living a happy life. Of course, we've never had the luxury to do that, but as we are progressing further and further from the matters of pure survival, it's something to consider. And that's what the linked article is really about.

PS I must also mention that the schools are not at the root of this problem. The culture that judges people by their achievement and success is deeply ingrained in the US (compared to, say, Europe), and the schools and parents simply train the kids to survive in it. See also a great (and lengthy) inspection of this phenomenon by The Last Psychiatrist: .

Isn't this what the Montessori program is about?

Yes, Montessori education is one possible implementation of this idea.

The more general term for "education that tries to make kids understand" is constructivism. Maria Montessori was among the first. Famous psychologists that contributed to constructivism are Piaget and Vygotsky. The latest development in my part of the world is Hejný method for teaching mathematics. In computer science, Seymour Papert designed the Logo programming language.

Different methodologies have different focus. Although Maria Montessori had some opinions on teaching students up to 24 years, the majority of her work focuses on pre-school education. This is why there is so much emphasis on toys that teach the basic concepts. Vít Hejný was a high-school teacher of mathematics, so his method focuses on teaching math within the framework of standard elementary and high school. Different authors prefer different learning tools, for example Montessori method illustrates the concept of number using beads, while Hejný method uses a "stepping belt". There are many other different details.

There is also criticism of constructivism, which sometimes focuses on individual implementations (and some of them were really crazy, usually by taking the concept of independent discovery too far), and sometimes addresses the central claims. Here is my attempt to steelman the central criticism:

  • constructivists claim that understanding improves learning, but there is little empirical evidence for this (at least in modern era, because Maria Montessori definitely did miracles in her era), kids taught using constructivist methods often do not outperform kids taught using the standard memorization;
  • "copy blindly first, develop understanding later" is how homo sapiens learns things naturally (e.g. you talk to your babies, you don't explain them grammar), and it works, which puts a huge burden of proof on people saying they can outperform the method we are literally evolved to use;
  • exploration takes more time, so even assuming it would ultimately lead to better understanding, it is not obvious that the cost:benefit ratio would be better (some of the crazy methods resulted in 12 years old kids who knew dozen different methods for addition, but didn't know multiplication yet).

There are also, sadly, more pragmatic arguments against constructivism, such as difficulty of finding enough math teachers who really understand math, or resistance of parents who themselves learned math by memorizing and now can't help their kids with their homework.

Should I ask what question you were asking when you decided on this position ;-)

I think you are creating an incorrect dichotomy. Neither schools, nor learning in general, are about either getting answers or posing questions. I think they are a bundle we take together.

To the extent you are arguing modern schooling approaches might tend to penalize a curious (and probably less focused) mind in the interest of conveying known facts and knowledge I would agree to some extent. At the same time, there are a lot of people that seem born with a lack of intellectual curiosity, or at least a lack of discipline and gumption to pursue that interest (a lot of us are lazy, though I suspect many like me just have to work a lot harder than others to get the the same place so lazy might just be another way of saying lack or energy to get there).

You just described Self-Directed Education.

Of particular interest, at least to me, are Peter Gray's optimizing conditions.

You mentioned a few of them: Ample time for play, access to tools. I think the others combine to outline a sort of litmus test for learning environments that conform to our natural ways of learning.

Do you have any strong evidence to back this up? Opinions on education are dime a dozen, studies are what I'd like to see for once.

Not OP, but I've done a bit of digging on this. Education research is really in a bit of a bind when it comes to looking at very-different models. The conventional model is so well established and broadly applied that it's really hard to get powerful studies of anything radically different from what you see in public schools today. Nearly all of the energy in Ed departments at various universities is focused there because that's where 95%+ of kids spend their time.

There are a handful of homeschooling studies that differentiate between "unschoolers" and more-conventional homeschooling approaches, but it's so niche that removing confounders is basically impossible.

You also run into the issue that making meaningful metrics for comparison is really challenging. Unschoolers are probably less likely to do well on a pre-calc test, but simultaneously more likely to earn an advanced math degree - the kids who have a proclivity for math get to dedicate a lot more time to it, while the others happily get by with perhaps only arithmetic and algebra.

I'm pretty thoroughly convinced that dramatically more freedom for school-aged kids is a great thing, but my conviction rests mostly on anecdote and personal experience spending lots of time in spaces that are run in a more self-directed way -- the research-backing just isn't there.

New to LessWrong?