It's Not a Telephone Game

Originally published December 21, 2014
Sometimes really smart people, perhaps because they are harried or busy, help perpetuate badly flawed models of important ideas. Memes that get traction because they are easy to repeat, not because they are right.—Venkatesh Rao
In context, the quote may well be spot on. But in general, it is far too optimistic.

When people spread degenerate versions of important ideas, it is usually because the version of idea in their head is only slightly less degenerate than the version they are spreading.

With each passing from one ear into another, an idea will randomly mutate, and creep closer toward preconceptions, cliches, less nuanced, and more viral versions. Why do ideas creep toward certain impoverished versions of themselves when mutating supposedly randomly? The answer is complex, and sure to be mostly misunderstood. I will give it anyway:
  1. The more precise an idea is, the less space it takes up in the cluster structure of thoughtspace. So when a very small, precise idea changes slightly in some random direction, there is no reason that subsequent changes will be exact reversals of this first change. This is a very general principle and closely related to the reason why genetic mutations are almost universally deleterious.
  2. People can't really remember the entire content of what they learn, so they (must) employ compression heuristics which naturally bias them toward thinking in terms of the worldview they already have. It is easier to remember something if you relate it somehow to the things you think about every day.* We are seeing through a lens of preconceptions.
  3. Related to the previous reasons is that both speakers and audiences prefer counterproductive oversimplications and worthless speculations over nuanced construals. If this isn't obvious to you, consider how many non-physicists think they can talk about quantum mechanics.
  4. Also, minor misunderstandings do happen. 'Understanding' is not a binary variable. There's a lot between bellyfeeling and grokking. A smart person with a headache is slightly less able to understand things than the same person otherwise.
  5. Most important is the not-quite-tautological observation that more viral forms of a meme spread and less viral forms disappear. And lower-fidelity copies of a meme are more viral than higher-fidelty copies.
Now then.

You would think the internet would stem the tide of memetic mutation, by preserving the original ideas. In fact it has had the opposite effect by allowing the less nuanced more viral things to propagate more freely, with progenitors helpless against increasingly disastrous misunderstandings.

I don't want to give examples, because then this essay will be about them instead of the more important general point. I will give one anyway: behold the censure of The Bell Curve, a sober, neutral book examining the nature and consequences of variation in intelligence. It does not make any strong claims about whence cometh the variation; to the contrary, it concludes that any confidence thereof the reader might have is misbegotten. It doesn't emphasize race. No one thinks of it that way. People think of it only as "that racist book about how whites are genetically superior."


People I talk to are sometimes frustrated or confused when I openly try to pre-empt miscommunications where someone is inevitably going to convey a mistaken understanding to someone else. They wonder why I'm trying to stifle conversation. "It's an interesting topic," they'll say. "I want to hear what he has to say. I think it'll be interesting."

It is at this point that I grab them by the shoulders, ferociously shake them, and scream, "THERE IS AN INOCULATION EFFECT WHERE MISTAKEN IDEAS TAKE THE PLACE OF THE CORRECT VERSIONS OF THOSE IDEAS."

Whereof one can only speak incorrectly, thereof one must remain silent. It is better to give no idea rather than the wrong idea. A hole can be filled; one that has been filled with the wrong stuff must be painstakingly dug up. It is harder to undo these mistakes than it is to make them.

It is not a telephone game. It is telephone in real life. It is not some toy academic principle that only appears in the lab. It actually happens, in real life, all the time, everywhere.

I have never seen anyone other than me try to do anything about it. It's not like nobody cares. All the time I see genuine experts lament the idiocy of laypersons who think they understand. I feel a kinship with Douglas Hofstadter in this talk, because throughout it I sense an attitude of, "Please try to actually understand the things that I am saying instead of rounding it all down by superficial analogy to ideas you already hold." Maybe experts mostly despair of the possibility of making the situation any better. And maybe they are not wrong to do so. But more likely is that it never occurs to them that they can minimize the likelihood, and furthermore the impact, of misconstruals.


When I first read this post, I didn't understand it. How did I know I didn't understand it? Venkat does several things right:
  1. He lists common misconceptions and explains why they are wrong. I had trouble distinguishing these from his correct conception, which wouldn't have been the case if I understood.
  2. He gives pop quizzes. These are annoying and therefore reduce virulence. But they are a useful tool for the reader: I couldn't answer the questions, so I knew my understanding must have been poor.
  3. He fluently navigates the ladder of abstraction. I don't know whether this generic writing virtue increases audience meta-comprehension. It might. (Pop quiz: what is meta-comprehension?)
These things happen naturally there because it's a post about a common misconception, and its correction, in a topic tangled with the whole abstraction hierarchy. But these things can also be done in writings not so directly about them.

There are other things you can do. When speaking, you can ask the people you're talking to whether they understood what you meant by some phrase. Last time I did this, the whole group of five said no (I expected most of them would get it). Even when people realize they don't understand, they seldom seek understanding.

As an audience, you should ignore anything that is framed as, "You need to be outraged at this thing." You should be especially wary of, "You need to be outraged at the people who aren't outraged at this thing," which is something that people actually say nowadays. Outrage is almost the opposite of understanding.

I am sort of breaking character by thinking seriously about practical solutions to a problem instead of just complaining about it. I really want to make a dent in this one.

This is all preliminary and unfocused. I don't know where to go from here. Maybe I should read more urticator; he seemed to think carefully about memes before disappearing.

Now is a good time to reread Wiio's Laws.

No comments:

Post a Comment