Skip to main content

Shadows of Ideas

I think it’s an important fact that people can become capable of things they weren’t previously capable of, and making this happen more could have huge societal benefits. Maybe that’s not a great way to put it—vague and wordy at the same time. How about this? I think education is really important. There, now you know exactly what I’m talking about.

Well, not exactly. The word “education” calls to mind this broad idea as well as a particular way of implementing it: schools, courses, teachers, books. I didn’t mean to refer to any of those things specifically; in fact, I suspect there are much better ways to learn. But education is such an established and closely related concept that it’s tempting to use the word anyway. Andy Matuschak and Michael Nielsen describe this problem in Quantum Country, an experimental “textbook” on quantum computing:

When we talk with people about the mnemonic medium, they sometimes described it as an educational experiment. This is a limited ambition, as education is commonly conceived. Imagine taking a month of public speaking lessons in school. Now compare that to a situation where you’ve been asked to present onstage at a major Apple product launch, and will have a team of people from Apple helping you practice for a month. Which do you think would be a higher growth environment? This is a fanciful example, but illustrates the point: there’s tremendous scope to develop extraordinarily high-growth environments.

I wish there were a well-recognized term for what Matuschak and Nielsen are getting at. Though it may seem like splitting hairs, the distinction is potentially very significant. If someone sets out with a goal like “promoting high-growth environments”, they’ll naturally start looking around for related ideas. And there’s no way to miss it, standing like an ancient redwood in the most frequented region of idea-space: EDUCATION. Perhaps they could start a scholarship fund, or become a teacher. Everyone knows what those are. Lots of institutions would support and praise them. And the world would continue to miss out on the seed of a brilliant idea that they passed along the way. It’s too close to the tall one. The light will never reach it.

Here’s a similar situation that’s been on my mind lately. I think it’s a great idea to use evidence and analysis to help others as much as possible. If that phrase reminded you of Effective Altruism, that’s no coincidence. It’s almost a verbatim pitch. And a pretty compelling one too. “Great,” say the new recruits, “Where do I send the check?” Fortunately, there are expert-managed Funds for the most important cause areas: global health and development, animal welfare, the long-term future, EA infrastructure… wait a minute—only four?!

Of course, official definitions of Effective Altruism will present a vision much broader than those four funds. They’re not wrong. But pragmatically, Effective Altruism is also a specific attempt at realizing that vision, one that does tend to center on a few causes. There’s a message—sometimes explicit, sometimes quite subtle—that if you want to help others as much as possible, this is where you should look.

I find myself spending a fair bit of time around the Effective Altruism community, and it really is wonderful. It’s catalyzed a ton of important work while remaining extraordinarily cause-neutral as social movements go. Even so, I can’t shake the feeling that half of the good ideas are going to be missed. That elevator pitch, precisely because it’s so compelling, casts an enormous shadow.

We tend to like names that are concise and communities that are vibrant. “High-growth environments” and “Effective Altruism-adjacent” are neither. There just isn’t enough conciseness and vibrancy to go around. Faced with this constraint, it often makes sense to work with ideas that are imperfect or incomplete. They’re close enough. Looking out at the empty space nearby, one can only wonder what might have taken root there.