Let's consider a language where the direct/inverse markers appear also in intransitive verbs. In these, however, the congruence marker for the subject makes it clear that the verb is intransitive. Here's the basic template slots, without any order information:
{verb stem, person1, person2, inv/dir, other stuff}
For an intransitive verb, we prototypically would have
{verb stem, person1, other stuff}
but what if we went for
{verb stem, person1, inv/dir, other stuff}
and had inv/dir encode different things in the intransitive, much like how erg/abs encode different things in split-S languages?
Consider then verbs that are single lexemes but mean things like 'to leave for a hunting trek'; the inverse could then signify 'return from the woods'. Here, direct/inverse could easily become a direction marker, which seems to reflect the name of the markers a bit too aptly. For verbs like 'sleep', the obvious thing would be 'fall/be asleep'/'wake up'. The 'fall/be asleep' pair might be conflated, or maybe other stuff contains aspect markers or the like that provide distinctions.
Here, having pretty few systematic rules and a lot of impressionistic or possibly just lexical stuff going on could be the most interesting approach, and give the most 'lived-in' feeling to the language.
No comments:
Post a Comment