Layers of Langlish Lift

By Lawrence J. Kramer, for the Grice Club

I wrote in The Langlish Lift:

"Efficient systems have three layers: a part that is always the same, a part that is the same for all members of a class, and a part unique to the current move. ... “

JL asked whether “components” would do for “layers.” I’m sticking with “layers” because I think there is an order to them. The counterweights on the Langlish Lift are physical components, but they are logical layers. The elevator arrives from the factory with the carriage counterweight. The database of user-based counterweights emerges over time, but in advance of any particular ride to which it might contribute a weight. The final weight is added when the ride occurs. This temporal ordering seems to me to justify a metonymous use of “layer” taking the ordering feature as its transferable component.

It seems to me that almost all creative processes can follow the same pattern. Think of Monet’s Water Lillies. What part of the process was always the same for him as a painter, what part always the same for painting water lillies, what part unique to each painting? You can bet that he did not shop for a new canvas merchant each time he sat down to paint. His canvas merchant was “always the same.” Maybe the canvas, too, and the brand of paints or source of pigments. What is the point of learning what works if we cannot employ it and build upon it?

My sense is that the part of language that is always the same is what we lay folks associate with Chomsky: a universal logic of language. That makes sense to me, because all language exists in the same logical space (is there more than one?), and it seems right that the nature of logic dictates the most efficient route for passing information, and that we evolve both physiologically and culturally to take that route.

Then there are classes of conversation – conversations between two specific individuals. When two people meet, the language they speak applies to all conversations between them. The participants will each assume that the other comes equipped with certain conversational tools and will use them in a certain, maximally efficient way, uttering only to add or subtract from what can otherwise be assumed.

The three layers I identify work only from a certain perspective. To the person adding the third layer, the division of the first two layers is irrelevant and so, as JL might put it, otiose. As to a participant in a given conversation, there is only the part of each move that is always the same – the language and the interpersonal context and the maxims and the rules of inference – and a part that is always different – the utterance. Much as I like Napoleon’s observation that on s’engage, puis on voit, a certain degree of disengagement is necessary to see that what looks to the speaker as a part that is always the same is, for the evolutionist, made of layers of decreasing generality. And I assume that the layer of things applicable to a class consists of layers applicable to each of the broader categories into which the class itself may be slotted. But that’s a detail, I think.