How to measure success of a ground-able compositional data structure

If I were able to build a data structure that can represent a set of core language primitives, how would I know if it can represent compositionality?

I want to write/think about if I implement a compositional base language, how do I measure if it works for other new words. The idea is that if I implement NSM words and engineer compositionality, I should be able to implement a full dictionary. This article is probably not a short article and will need more time. I should write out some example sentences from ltwf.

Leave a comment

Your email address will not be published. Required fields are marked *