so here’s the skinny: i’ve been filling my days doing some work with mary on cooperation in behavioral game theory. we’ve been looking at a new way of segregating classes of social preference models, pinning commitments on them based on what information certain players have about the payoffs and about what the others know about that distribution of information. the facts are really stark, and bad news for some pretty famous people.

eat your veggies! doesn’t entail eat your veggies or eat some ice cream!

and (performative) free-choice permission puzzles:

you may eat your veggies or some ice cream entails you may eat some ice cream

craige roberts was in town to give comments and a good time was had by all.

(the mini-workshops aren’t regular, they’re occasional. and there’s a regular reading group associated with them - hence, omg. all of that is so i can send emails with subject-lines like “omg: this week’s reading”.)

anyway, we had our first mini-workshop last week: kai von fintel presented a paper, “must … stay … strong!”, that we’ve been working on lately. this is the fourth paper in our trilogy on epistemic modals, and is affectionately known as “shatner” to us. in it we take aim at the mantra that must p is weaker than p and give a new account of the evidential feel that must has.

chris potts and craige roberts gave us comments and provided lots of good stuff for us to think about. (so it was an invasion but a fun one.)

we’re mid-revision on the paper right now but we’ll get a draft up asap.

it’s odd, though, that it was published “online” in november but the issue is june 2007. But, wev, as the kids say.

one other thing: a few typesetting errors managed to sneak in after i corrected the proofs … nothing major, though. you wouldn’t have believed the corrections. seriously. makes me happy about s&p’s launch. those guys have some typesetting skillz. mad skillz.

Both of these papers develop a strict conditional story in way that is faithful to the Ramsey test — well, at least the version of the Ramsey test I learned on the playground. That means that conditionals are both index-shifty (they shift the point of evaluation) and context-shifty (they shift the context relevant for that evaluation). So if is both strict and shifty (twice over).

If you’d rather see the advertisement for the paper rather than the paper itself, you can check out the beamer slides for the talk. (I left the pauses in so you can recreate the whole experience…really.)