On the elusiveness of Uppercase Things

by James Somers, March 2, 2011

A few days after I loaned the excellent The Man Who Loved Only Numbers: The Story of Paul Erdős and the Search for Mathematical Truth to a friend, I received a note highlighting a sentence that he said contained “a preposterous number” of some of his favorite words:

In the ninth century, nomadic Magyar warriors from the steppes of Eastern Europe crossed the Carpathian Mountains and, renouncing their peripatetic way of life, settled in the middle Danube basin at the heart of what is now Hungary.

His comment surprised me, not because I don’t equally enjoy the sounds and flavors of that sentence, but because I almost certainly ignored them all the first time I read it. In fact, for me this was probably one of the least interesting sentences in the book, so uninteresting that my glossing over it was an unconscious non-event. It’s just a part of the way I read.

Here’s another example: if I’m reading a law review article and the name of a case comes up, and the case isn’t something I’ve heard of before like “Brown v. the Board of Education,” I don’t put any effort into mentally assigning the new name X to the content of the paper, claim, context, etc. I treat the name X as a kind of local variable. That is, the only sense in which you could say I’ve cognized it is that if two sentences in different parts of this paper both said something about X, I’d recognize that they shared their referent; but the moment I move on to something else, that nascent <word, object> pair dissolves.

The same thing happens when I read a long technical blog post about, say, tax policy. I pay zero attention to basically anything in capital letters: the names of various plans, their proponents, the states where certain pieces of the plans are in play, etc. I treat these strings as happy little warp-slides I can safely zoom past.

Maybe more alarmingly—and tellingly—the same thing happens with mathematics. Unless I’m specifically reading an article as a way of learning a piece of math, or the math in an article is so simple that it can be parsed at something close to regular speed, I skip it.

So now we have our first obvious explanations: (a) James is lazy, and/or (b) James goes into these articles at a certain level of abstraction, and both the Capitalized Names and mathematics are below that level. So James ignores them.

That probably does most of the explanatory work. But there are other possibilities.

One is that I have pretty bad recall. It probably hurts my recall to think it’s bad (apparently recall performance has been pegged to self-perception of recall performance), but whatever, my memories just don’t seem to be indexed by names or dates the way they are for some people.

So one theory is that because I have such low expectations about remembering things like names and dates, I simply don’t pay attention to them. In other words, because my brain wants to put strings like “Magyar warriors” or “Ashcroft v. Iqbal” into the to-memorize queue—this it does because Capitalized Strings are often just names, vs. concepts or narratives or ideas which can fold their way into the brain in a more natural sort of way just by nestling into the appropriate associative context—and because my to-memorize queue is run by a bunch of idiot half-a-loafs, I discard them as a way of saving energy.

Example: I often ignore lists. When a writer says that some economic theory has been tremendously successful in explaining behavior in Germany, Austria, Japan, and Norway, I think my eyes literally jump over the List, Of, Capitalized, Country, Names much faster than over the stuff explaining the theory, or the stuff explaining what elements of that list have in common.

In the short run that shouldn’t hurt me because with any luck, the more conceptual-narrative-imagistic-flowcharty lowercase kind of knowledge does stick with me and is indexed analogically against other concepts of the same kind, etc. But in the long run, I’ve lost the information that lowercase concepts x and z are tied to Norway and Japan.

Note also that this strategy is path-dependent: if early on you develop the habit of discarding names and dates, you won’t have a lot of them swirling around, which means that when you later run across new instances in new contexts, you won’t have a rich index to cross-reference against, and so you won’t be able to build organized stores of knowledge about particular things. I think that effect could turn out to be pretty important.

The normative upshot is that I could probably do with a ton more history, where this time around I actually pay attention to the dates, because timelines are a tremendously useful organizing concept–knowledge template, and I have this black hole where I should have the context that most of my friends seem to carry around.

It also means that Sporcle and Quiz Bowl and all this nonsense, in which I have essentially zero interest because I’m so terrible at it, could turn out to be cognitively really important just as a way of ordering what would otherwise be loosely overlapping fact miasmas.

To circle back a bit, I want to ask this question about my friend’s Erdős sentence: could it be rewritten such that I, with my current readerly quirks, would want to underline it? Initially I thought it could, maybe by extracting the conceptual/lowercase content into its own sentence and doing a sort of “for example” to introduce all the names and dates. But in this case there’s really nothing to extract. The lowercase content is that “peripatetic warriors crossed mountains to settle near a river.”

Maybe that’s the upside to my handicap: a mind unencumbered by Uppercase Things can focus on meat and mechanisms. But then again, maybe that’s just Wishful Thinking.