It seems useful to address the question of why people create new languages - either DSLs or full languages. Would anyone know of work in such an area? The answe

It seems useful to address the question of why people create new languages - either DSLs or full languages. Would anyone know of work in such an area? The answer to the "why" is not the reasons people say they make a language. I mean the question as a probe into the psychological urges to "linguate" (is that a word?). One possible kind of answer could be that, given formal systems are powerful tools to model some interesting corner of reality, the urge to create a DSL could be related to an urge to understand a domain by creating an automaton that can be mapped to domain aspects.