Rules of variety

Link post

At some point in high school I noticed an interesting thing about my choice of essay topics: it was definitely not allowed to be the same choice of topics that anyone else had.

One reason this seemed strange at the time was that I had never explicitly noticed this constraint or intended it, even though it was doing a lot of work. It was such a deeply assumed part of the basic rules of behavior that I didn’t know it was there.

But it seemed extra strange once I noticed it, because it happened in the context of the rest of my classmates as one blithely ignoring this absolute law of reasonable behavior and all writing about the same hackneyed thing. Which probably wouldn’t have even occurred to me to do if I had set out to write the most surprising essay I could. So, apparently other people didn’t even have this rule, though it seemed so inbuilt in me.

And this wasn’t just a failure to understand the rules of that assignment—I realized then that I had been assuming this constraint for every essay, and um, perhaps everything.

(In retrospect ‘maybe you are trying to be different and other people aren’t’ looks like an obvious explanation for the perplexing fact that I was different and other people weren’t. But I was used to knowing about things I was doing intentionally, and I was only trying to be different in the sense that I currently try not to murder people—it would be so wrong that it doesn’t cross my mind as a possibility. But while unconscious, this is a very effective form of intention.)

Years later, I think other people actually do have this rule, or similar rules. They just vary by topic, and I happened to be unusual on the topic of high school essays. But I think implicit constraints like this are actually pretty common, and usually feel too natural to be noticed, even while they entirely warp our behavior. I don’t mean ‘assumptions that we don’t notice’ in general, but in particular ones about how similar or dissimilar out behavior should be to others. I rarely hear these things spoken of, except to remark when they are broken, without comment on what they actually are or consideration as to whether they should be there.

Some examples of actions I think you would avoid to at least some extent, or make an excuse for:

  • Showing up in the same outfit as someone else

  • Naming your children the same names as your friend’s children

  • Decorating your room exactly the same way as your housemate (a friend of mine actually moved to a different room in the same shared house, leaving his art behind for the appreciative incoming resident, and replacing it with identical art in his new room. This seems widely considered weird.)

  • Using the same unusual adjective multiple times in the same article without it making an intentional point

  • Answering ‘how are you?’ with the same contentful description of your state as the one you just heard, without comment

  • Getting the same unusual car as your colleague

  • Using a turn of phrase that has been used many times before

  • Doing a thing that is trite, hackneyed, cliche, or stale

  • Going on holiday to the same place your friend just did

  • Doing a project that is basically the same as one someone else did, without it being connected to theirs

  • Using the same stylistic touches that others use (e.g. even though xkcd is widely considered good, if I draw comics that look just like xkcd, it would be weird)

  • Copying too many of anyone’s personal habits when you are not trying to flirt weirdly with them

  • Showing up to prom in the same car as someone else

This may all sound pretty unimportant. Ok, society has to support more dress variety than would otherwise be optimal. Worse things happen. But I suspect this also shows up in intellectual activities and strategic decisions. And having random unacknowledged rules driving decisions in those places strikes me as more terrifying.

For instance, discussing how surveyed machine learning researchers expect human-level AI further out now than they did before the recent ML boom, someone pointed out to me that of course people are going to be pessimistic now, because the interesting thinkers a couple of years ago were optimistic, so optimism is now boring. If that person is right that the opinions of a field on a topic as important as how imminently they are bringing about the end of human dominion are mostly determined by the dynamics of fashionable distances in opinion-space, I say we have a problem.

Other places I’d expect to see this:

  • Aversion to working on too close a question to someone else in your vicinity, if you are not working with them

  • Aversion to just straightforwardly agreeing with another intellectual rather than emphasizing differences

  • Aversion to liking things that are too popular (contrarianism)

  • Aversion to strategies that are too popular, even if that doesn’t affect their effectiveness

  • Not discussing topics once they are too commonly discussed, even if they are not resolved.

It’s old news that opinions move according to fashion. So why is this interesting?

First, I think we usually think of this as a pressure for conformity—for a few thought leaders to choose ideas somewhat freely then all the thought sheeple to follow. I’m claiming there are also strong forces for variety. And these don’t just cancel and give us freedom—they lead to a narrow band of appropriate choices. The next step in the dance has to be a certain distance from the last.

Secondly, since opinions following fashion has been pointed out in the past, it is weird to point it out again. But human memory and salience probably require it to be pointed out sometimes, if we are to actually remember it.

I’m not very confident about all this, beyond the more basic observations. But it leads me to an image of culture evolving like a fractal river delta, every piece curling off into several pieces that are the right distance from it and one another. Which is kind of how culture seems.