5 min read14th Jan 20181 comment

6

Status: fun to write, very derivative, ambivalent about content.

This post responds to Why and How to Name Things.

In the evolution of language, new words are constantly produced, bubbling in and out of primordial concept-space like quantum foam. Most of them are awkward, unevocative, and noncentral, falling out of fashion faster than you can say "Planck time." A few gems cut out unexplored territory and grow in popularity until they take their rightful place in the common parlance. As they do so, however, they fall prey to concept creep, being diluted to incorporate such a plethora of valences as to become useless. Essential words like "love" have been dragged through the mud or mouthed insincerely so often that it feels affected to even utter them.

As old concept handles die or creep, a constant creative drive is required into invent new words for old and new phenomena. This places a heavy burden on writer and reader alike. Improper names are like programmer debt - let a weak name stick early on and you'll pay for it your whole life.

This babble outlines the shape of this problem and what we might do about it.

First, I argue that as far as names go, only length matters. Shorter words are exponentially more important. I call this the Inverse Teenage Boy Model (ITBM) of Names.

Having built an ITBM, we need a nuclear weapon for it to deliver. I propose a new usage of the word syllable: a verb, meaning "to rename with fewer syllables." I give examples of savvy syllabling in math, computer science, and rationality, and try to syllable some names I hate.

The Inverse Teenage Boy Model

The Teenage Boy Model (TBM) is a universal theory of value developed in the most advanced locker rooms in the history of our civilization. Its thesis: only size matters. Carl
Sagan was one of the champions of the TBM:

Look again at that dot. That's here. That's home. That's us. On it everyone you love, everyone you know, everyone you ever heard of, every human being who ever was, lived out their lives. The aggregate of our joy and suffering, thousands of confident religions, ideologies, and economic doctrines, every hunter and forager, every hero and coward, every creator and destroyer of civilization, every king and peasant, every young couple in love, every mother and father, hopeful child, inventor and explorer, every teacher of morals, every corrupt politician, every "superstar," every "supreme leader," every saint and sinner in the history of our species lived there-on a mote of dust suspended in a sunbeam.

Upon reflection, TBM is not a good fit for my values. All things considered, human beings are the most interesting and least understood objects in the known universe; I would rather study the inner workings of a single human being than that of a neutron star.

In fact, TBM is exactly inverse to how I value names. Off the top of my head, the salient rationality concepts from the last year are Slack, Zeroing Out, Hard ModeDouble Crux, and TAPs (props to Zvi, God of names). I guarantee you I wouldn't remember TAPs if nobody had bothered to abbreviate it. Shorter names (in letters or syllables) are more important. That's why I'll call my model of names the Inverse Teenage Boy Model (ITBM).

People understand that shorter words are more important, but people are scope insensitive and don't grasp how much more important. It's not merely that shorter words are easier to remember. Shorter words are exponentially more important, by the brain's own metric. If the brain is anything like a computer, this has to be technically true because of how compression algorithms work.

The vocabulary of a language functions is something like a Huffman tree collaboratively designed by its population to communicate thoughts as compactly as possible. More common and important thoughts are stored with fewer letters and syllables. The exact details of the compression algorithm don't matter - what is universally true is that a word stored with one less bit should be used twice as frequently:

$latex \text{Importance} = 2^{-\text{length}}$

Thus, our languages are built around the proposition that linear changes in length correspond to exponential changes in importance. This is ITBM.

ITBM predicts more than the obvious fact that shorter names are easier to remember.

ITBM predicts that our brains have learned the heuristic and a shorter word is felt instinctively to be more important.

ITBM predicts that if I reduce the title of this post by a single letter or syllable, the reader will viscerally feel its importance double.

The evolution of language is a laborious crowd-sourced attempt at maximizing compression ratio. This inexorable compressing force acts by abbreviation and overloading. Concepts creep beyond their original valence and are mushed together, sometimes haphazardly. Shiny new words are like the proverbial hammer: everything looks like a nail.

Opposite the compressing force, there are several stretching forces. Linguistic real estate is hard to come by. Longer words are prolific in jargon because they retain that protective layer of precision unavailable to words abused by the layman. These polysyllabic monstrosities also signal novelty and irreducible complexity to ITBM, and provide a barrier-to-entry to the outgroup.

To Syllable

The picture of compression I described has another layer of nuance: language is the global and public Huffman tree, but each individual applies a private layer of compression based on her own preferences and usage needs. This private compression allows longer concepts like "existential risk" and "inadequate equilibria" to feel more important and natural to a particular person or subcommunity than shorter ones. But if these value judgments get propagated successfully, it's only a matter of time before the important long word gets abbreviated and slangified: existential risk => x-risk, inadequate equilibria => Moloch.

At every step of the process, there's luck, time lag, and divine inspiration. To jump-start the evolution of language, I propose the following definition:

syl·la·ble

/ˈsiləb(ə)l/

  1. (verb) To rename with fewer syllables. If he ever wants to publish that book he'd better syllable it.

Syllabling is a delicate art. Starting with a big important concept, we need to find a short, evocative word with the proper connotations. Geometry is laden with examples where shrewd minds must have syllabled awful names:

A sheaf is a system of data attached to the open sets of a topological space such that the data on overlapping sets agree. For example, the sheaf of continuous functions on the real line attaches to each open interval the set of all continuous functions on that interval. Given an interval $latex I$ and a subinterval $latex J$, every continuous function in $latex I$ restricts to a continuous function on $latex J$.

Blowing up is the process of replacing a subspace in a surface with the space of all tangent directions from that subspace. For example, the blowup of the plane at a point is a kind of spiral around a vertical line.

A pair of pants is the surface constructed by cutting three holes in a sphere.

There are also plenty of names that deserve to be syllabled. Sometimes, I think the only reason I'm not a topologist is the word cohomology. Imagine if we syllable "homology" with simply "hole." The hole groups of X vanish if and only if X is contractible. The hole ring of $latex X \times Y$ is the tensor product of the hole rings of X and Y.

I'll end with one example of an important phenomenon missing a name. Jessica Taylor wrote in Against unreasonably high standards:

Consider the following procedure:
Create unreasonably high standards that people are supposed to follow.
Watch as people fail to meet them and thereby accumulate “debt”.
Provide a way for people to discharge their debt by sacrificing their agency to some entity (concrete or abstract).This is a common way to subjugate people and extract resources from them.

Inspired by the 4X game, I dub this the 2X model of exploitation:

  1. eXpect too much.
  2. eXtract resources.

Somewhere in there must be some unresolved anger against humans of the 2X variety.

New to LessWrong?

New Comment
1 comment, sorted by Click to highlight new comments since: Today at 2:33 PM

The babble layer is lively and it makes the post very enjoyable to read.

However, while the central claim contains a grain of truth, I'd be wary of turning it into advice. How much your cognition depends on the choice of words (and their length) is to a significant degree a matter of how you design it. It's a learnable skill to think mostly with non-verbal concepts, and translate them into words only at the latest possible time.