This thread to discuss the idea of adding inquisitive semantics to Toaq, both the “meta” of it (what goals should be achieved, and whether it should be done) as well as technical implementation details.
As far as I know, the goals are to
Preserve every declarative sentence, both in form and meaning, with a possible exception for ones that interact with questions
Whatever changes are allowed must be very weak and non-disruptive for Toaq. We’re trying to keep it as backwards compatible as we can!
Not add new declarative sentences, both in form and meaning, except for the ones that interact with questions
Add a well-defined notion of questions that largely matches the current understanding of questions in Toaq and matches the way that questions are asked
I haven’t followed Magnap’s work in a while (we’ll have to wait for them to make an account!) but I’m sure they’d be happy to post updates on the matter whenever they can
I don’t have much in the way of updates at the moment. The two main open problems in terms of getting the compositional semantics to work out nicely are domain restrictions in quantifiers, and the big one, tense. Thinking about controlling scope wrt. tense was what lead me to the idea of having a prefix for “disjunction-of-distributing-over”, and that is something I think could be quite promising, in that it could lead to inquisitive semantics being essentially entirely “opt-in” while also potentially allowing for the asking of questions where the questioning quantifier is in a sub-clause. Having that be the only way to ask questions other than what’s in the refgram would be quite clumsy, but the idea is to use it to identify the pain points so that they can be individually resolved.