Everything worth inventing has not, in fact, already been invented

There are good reasons to believe that education could be a lot better than it currently is in most school systems. Freddie deBoer is, as he often is, confused about this.

Unrelated picture of me with my son, since it seems like that motivates you people to actually read my writing!

I’ve been reading a lot lately, and writing some, but haven’t published anything since my most recent post, which I wrote a couple of weeks after almost committing suicide. I’ve wanted to write publicly more, and actively participate in the increasingly engaging cultural sphere that is Substack, but I haven’t really figured out what I want to write about.

Is Orca & Fishbowl a personal blog? Maybe, and I suppose it’s true that the more personal content I’ve written so far has gotten the most engagement, so maybe that’s what the people want. But it definitely isn’t what I really spend my time thinking about.

If I don’t spend my time thinking about my own life, you might ask, what exactly the fuck am I spending my time thinking about?

I’m glad you asked.

I spend a lot of my time thinking about when artificial general intelligence will really shake up the world, and what that is likely to look like. I spend even more time thinking about consciousness in non-human animals (insects and cetaceans, in particular), and a good amount of time thinking about the potential consciousness of machines and digital systems, including but not limited to large language models.

And as a relatively new father, I spend a lot of time thinking about parenting.

Also education.

And “the alignment problem”, borrowed from the AI safety community, but applied to the problem of ensuring that my son grows up to be happy, well-intentioned, and curious about all that confuses him.

And as someone who really likes to be right, I spend a lot of time thinking about how we know what we know, the history of science, statistical methods and common cognitive fallacies, etc.

But where does that leave me, as someone who wants to write a Substack blog? Well, for one thing, it leaves me with about ten-thousand Claude conversations in a project labeled “Intellectual Curiosity”. And it also leaves me with writer’s block.

So I decided to follow the age-old instruction for how to treat writer’s block: read, read, and read more. And I would even add one more specific piece to that recommendation, which goes something like, “read the writing of authors you generally disagree with and whom you think have weak arguments, especially given how many followers they have”.

And this, time and time again, has led me back to Freddie deBoer.

This time, I encountered a piece he recently wrote, in which he was uncharitably arguing against a writer I DO tend to agree with and usually find to be exceptionally thoughtful, Kelsey Piper. So I figured this piece, titled “There Are No Miracles in Education”, would be a good place to start.

He makes a few reasonable points, like how schools account for less than 10% of the variance in student academic outputs (agree) and the No Child Left Behind policies were a complete disaster and manipulated statistics to try to paint a picture of meaningful improvement (agree).

But he then makes a peculiar jump to something like, “Therefore, any time someone thinks they have found something that can be implemented widely across school systems and improve education outcomes, it is bullshit and must be either selection bias, fraud, or both.” (My words, not his, but that’s pretty damn close to what he says — just much more succinct).

This is, of course, a VSC (very silly conviction) — to believe that all ways of delivering education are equally effective and none could prove to be genuinely better than any of the others. It is, in fact, entirely absurd. It’s a form of relativism that plagues much of deBoer’s writing, but I think is especially ludicrous in this case.

Could it really be true that teaching students to read by showing them only picture books, for all of elementary school, is just as effective as more traditional methods? No, of course not. Could it be true that teaching students algebra by slapping them in the face and shouting “figure out algebra” is just as effective as traditional methods? No, of course not.

So why do we believe that there could not, in principle, be something much better than the currently used methods in most public school systems? Is there some universal law that dictates that all good pedagogical methods must have been discovered before the year 2025, and from then on it’s all gonna have to be just selection bias and fraud?

This strikes me as a massive failure of imagination. Much like the people who think believe there’s no way something could be orders of magnitude smarter than humans (read my related satirical piece here), I think it is naive and myopic to think that all the low-hanging fruit in education has already plucked, and that there are therefore no meaningful improvements that can be made to the science of teaching.

It seems, actually, that there have been several *major improvements* made in teaching literacy, including replacing “whole-language” reading methods with “phonics”-based approaches, as Kelsey Piper accurately reports. For Freddie deBoer to recall that George W. Bush had some major policy failures and then jump to the conclusion that all attempts to improve education must also be policy failures is, at its core, a misfire of imagination and a lack of curiosity about what’s actually possible.

Adam Mastroianni wrote a great piece about how ideas are not getting harder to find, contrary to popular belief. There are still *plenty* of valuable discoveries yet to be made.

So to close out this piece — which I am trying to write in a single, 30-minute sitting while my son sleeps on my chest — I am inspired for some reason to share a core memory of mine. It is a memory I have from circa 2007 when a neighborhood friend of mine told me, with 100% seriousness: “We’ve got TV’s. We’ve got microwaves. We’ve got phones. Everything’s already been invented!”

Freddie deBoer may suffer under the illusion that everything (in education and pedagogy) has been invented already, but you certainly don’t need to. We may have microwaves, but we could probably still figure out how to teach people better.

—Alex

p.s. here’s that great post about ideas not getting harder to find:

Ideas aren’t getting harder to find and anyone who tells you otherwise is a coward and I will fight them