Can we scale cultures that support learning?
new essay in Asterisk
TL;DR I have a long piece, called “Culture Studies,” in Asterisk today.
In February 2023, I published an essay called “Childhoods of exceptional people” where I looked for patterns in how various people who went on to do exceptional work were raised. There were several commonalities, but the key pattern, as I saw it, was that nearly all of them were raised and educated in cultural milieus that were very supportive of learning and achievement and were they had access to exceptional role models. Before school age, von Neumann, for example, would sit with his father at meetings where they discussed the management of his father’s bank.
This week, I have a new piece in Asterisk that picks up on this theme. The animating question is: can you scale these types of cultures? I explore it by looking at how Montessori and early Jesuits approached scaling specific cultures in their schools. This pairing of examples is interesting to me because Montessori and Jesuits are in many ways opposites when it comes to educational philosophies, but they seem to have converged on similar strategies for scaling cultures. They have also, arguably, been better at this than most other educational initiatives.
My working thesis for the future of education is that the curation of cultures that support learning and growth is the main bottleneck right now, and scaling better cultures a promising path to give more people the opportunity to live fulfilling lives. As I wrote about in “AI tutors will be held back by culture,” most of the technical problems of pedagogy are increasingly becoming a solved problem—we have good resources and tools, and AI is becoming a powerful substitute to 1-on-1 tutoring. But most people don’t have the cultural scaffolding to use these resources well. I suspect we will see an increasing gap between people who have the motivation and know-how to leverage the tools at their disposal and the rest—many of whom use the same tools to distract themselves and harm their learning.
It is unclear to me how far we can scale cultures that provide better scaffolding for learning. But I’m pretty sure we can make progress compared to status quo, at least for a subset of the population. This week’s piece in Asterisk is a small step in my thinking about this topic, by highlighting some design properties that might be useful for curating and scaling cultures. Read it here.
For those who are more nerdy about this topic, let me also make a few rough and technical observations about where these reflections about learning cultures point onward for me, in case you want to bounce ideas:
I think it is possible to think about this much more systematically than the Jesuits and Montessori did/do. We can, to some extent, model culture as a property of a social graph. I think we have at least some of the tools to reason mathematically about which topologies and network properties will have desirable effects on the cultural flow of the graph. I know a few of you research networks and cultural evolution—if you have any pointers or ideas on this topic, please email me by replying to this email (or comment)!
I also think that AI systems and zero knowledge proofs will provide us with interesting new design possibilities that could be useful when building better cultures.1 Those of you who work in these areas—I’m interested in learning more about what you think, and what opportunities you see.
Also, anyone working practically scaling new kinds of educational cultures / schools—if you have any new observations, I’d love to hear from you too. I know there is a lot of interesting work going on, and I’d be interested in hearing from those of you that I’ve discussed this with before (how are your challanges, and your thinking evolving?) as well as others that I’m not aware of.
In case I get several interesting answers, it might take me some time to reply because I’m overwhelmed with work right now. But I’ll get to you.
Anyway, here is the essay in Asterisk, and here are the two previous pieces in the same series: “Childhoods of exceptional people” and “AI tutors will be held back by culture.”
I’m thinking for example about AI systems that filter and moderate communities; steering cultural visibility; matchmaking; or using AI to let people interact in more precise ways by having their AI systems interact, etc; and I’m thinking about using zk-proofs to more fluidly sort people into high trust communities, etc.
I've some firsthand experience with Jesuit education and can confirm that was the most learning-positive environment I have ever been in it was the best school I attended, easil, in terms of pushing me to develop but also giving you room to range & grow towards your own interests.
What I am currently working on / trying to move forward is a way to link ADHD people in communities to hold each other accountable / collectively work on goals. For me, I struggle to do anything alone that I don't 'want' to do, but add another person doing it with me / holding me accountable, and that supercharges my throughput.
Which is to say - I think a key piece of this, where it scales to as many people as possible, is finding ways to make the framework adaptive to the needs of individual learners. Those lucky enough to be fully self-driven with no need for external aid - well, y'all won the lottery.
For me, the most interesting problem is hacking a solution to those of us who /want/ to learn, but have to fight our own brains every step of the way.
A Friend and religious historian pointed me to Christ's Churches Purely Reformed, which is an academic study of how Calvanism scaled its own culture. I was originally looking at it through a lens of spiritual communities, but I imagine the lessons are very similar. I have yet to read it, but it's on my list.