Pollenfest
That’s the name of the write-up in progress, which is at a nice enough point now that I feel pretty confident to post.
[The html version has a couple of dropped images that are just syntactic trees of the Bach & Cooper NP-S analysis. These appear just fine in the pdf, so if you really really want to see them, I suggest that you look there.]
The working name of the project is “Pollenfest”–in honor of the terrible seasonal allergies I had when I started writing this.
Included in the document is some background into the NP-S vs. Det-Nom debate (as well how to interpret relative clauses with and without variables), the Bach & Cooper analysis, my revision of their analysis, how my revision can be used for pied piping & binding into heads, and BONUS! how to get the von Fintel semantics for exceptives to work with my system. Make sure you check out the part on exceptives at the end; I think there are some very nice consequences of my approach coupled with von Fintel’s semantics for exceptive constructions. (I have a whole bunch of stuff on exceptives from my fall semester last year and also a paper that I did for Elena Guerzoni and Jim Higginbotham’s seminar in semantics/pragmatics that I should get around to posting as well–in this document, I assume Kai’s analysis because a lot so very nicely falls out of it.)
Coming soon: extraposition, the syntax of all this stuff, more work on functional domain restrictions (taking inspiration from Jason Stanley’s work), and…I think that’s quite enough for now, yeah?
This weekend I’ve been reading/re-reading extraposition-related stuff: Bhatt (2002) The raising analysis of relative clauses: evidence from adjectival modification, Sauerland & Hulsey (2004) Sorting out relative clauses, Fox & Nissenbaum (1999) Extraposition and scope: a case for overt QR. The Sauerland & Hulsey is particularly good in that it fleshes out exactly what the raising/matching analyses are and how they fit in with copy theory and late merge. However, I’m still pretty skeptical that we should go this route. And, furthermore, I’m not certain that the raising/matching analyses couldn’t be adapted to fit an NP-S approach. In fact, Sauerland & Hulsey argue that you need a matching analysis to account for idiomatic expressions (like “Simon praised the headway that Emma made”)–while in the RC you might want to have a copy of “headway” so that at LF there was a “made headway” constituent, you also need to do something to force the determiner to be “the” and nothing else, right? “Simon praised every/no/some/a/a lot of/many headway that Emma made” is bad bad bad. I’m not really sure what do to with the stuff on idiomatic expressions. It seems like there’s so much we don’t understand about the quirky behavior of such expressions that to base (or partially base) hypotheses on idioms feels a little unwarranted. There is some cool stuff involving intensionality (”the longest book that Simon believes Montague wrote”) that Bhatt and Sauerland & Hulsey talk about, but I really really need to brush up on how to work intensions.
Oh, and something that I realized this weekend: in the Pollenfest document, there’s a footnote where I discuss the constituency of stacked relative clauses. One consequence of my approach is that stacked relatives clauses are considered constituents, and I worried that perhaps that wasn’t an appropriate claim to make. But I realized that you can extrapose stacked relative clauses quite naturally: “Simon gave me the book yesterday that he read last summer that he thought I would like.” And, even cooler, you can get an extraposed relative clause & exceptive construction: “Simon gave me every book yesterday that he read last summer but Direct Compostionality.”
emma :: Jun.24.2007 :: misc, syntax, semantics, linguistics :: 3 Comments »
Ponder, ponder. I’m actually surprised by one fact: when I read the “he who” data, my thought was, “Yes, but you can only do that with subject pronouns, even in object position, which suggests that the ‘he’ is somehow heading the relative clause, and not external to it. For instance: ‘Let he who is without sin cast the first stone’, in which ‘he’ can’t be the object of ‘let’.”
What surprised me was the sheer number of Google hits for “him who”, and the fact that there was even a pretty even split on whether it’s “let he who is without sin” or “let him who is without sin”. (Along with a lot of “He who is without sin, let him cast the first stone.”) It’s odd, because I have such strong intuitions about the awkwardness of “I’ll give this book to him who asks first”, as opposed to “…to he who asks first”, but I may be alone in this.
Well, will continue to ponder.
Lance, it’s interesting the variation in case among some closely related IE languages for that phrase: cf. the bible, though I admit bible translations are not really ideal sources for what most people will tend to say in actual life.
(And I was surprised myself that the KJV doesn’t say ‘let he who is …’; I remembered the phrase that way too.)
Ah, now I look more closely, the variation in case is explained completely by the verbs used; neither of the other languages needs ‘let,’ they both use subjunctive forms.