-
Notifications
You must be signed in to change notification settings - Fork 3
CambridgeWhInfoStructure
Present: Olga Zamaraeva (OZ) Sanghoun Song (SS) Emily Bender (EB) Guy Emerson (GE)
Scribe: Emily Bender
OZ: I prepared some screen shots… as usual, what I’m concerned with is which parts of this are relevant to my library. Erteschik-Shir argues that wh words aren’t always focused, although the rest of the literature says they are.
OZ: NES observes that linguists think wh question words are focused, because the corresponding answers always are.
EB: But what evidence does NES adduce for what the focus is?
All: A bunch of stuff that seems to rely on the intuition that the first NP in the double NP frame for give can’t be focused (‘dominant’), but we disagreed.
EB: Interested in (24b): Who GAVE a book to Mary? (Context: Mary has two books, one lent to her by someone, one given to her by someone.) Can that be contrastive focus on give and non-contrastive focus on the wh word?
EB: Sanghoun — can we use other tests for focus on the wh words?
SS: One is that focus is always required. Can you drop the non-wh stuff?
All: The ‘gave’ is required because it is (also) focused.
EB: Hard to drop the arguments of give, but that’s orthogonal.
GE: Can make them (unstressed) pronouns? Who GAVE her one?
OZ: In Russian it sounds much better with focus particle on who — who gave it? who prt gave it? Who was it who gave it? If I want a lot of stress on gave, sentence sounds better if I put the particle on who.
SS: I really wanted to work on focus projection in my thesis, but couldn’t because I didn’t know how to include prosody and so didn’t.
OZ: What do I want from these patterns in terms of my library?
GE: Seems orthogonal to your library, because you can put the stress anywhere?
OZ: So what should I do?
EB: Make wh word (wh constituent?) always focused.
GE: If there are different kinds syntactic expression of focus with wh words, capture that.
EB: And if the wh words show up in a typical focus position, leverage Sanghoun’s library rather than reinventing it.
OZ: Let’s look at Hungarian then.
EB: With a big grain of salt.
EB: Does the info structure library rule out a focused element (e.g. a wh word) from showing up in a non-focus position, if you’ve defined positional focus?
SS: That was my intention, but a problem might arise with multiple wh questions, because they can’t both be in a single focus position.
EB: So what happens with multiple wh questions in a positional focus language?
SS: Hungarian, Turkish, Basque
OZ: Paper on Somali, Berber, & one other. Claims that multiple wh questions are not allowed because of positional focus.
SS: Hungarian quote says that ‘why’ doesn’t have to be pre-verbal. In my analysis, is ‘why’ is a focus-sensitive item, but not quite the same as other wh words. It can appear at any position in a sentence in many languages, unlike other why words. Suggestion: don’t include why in your library without special treatment.
[ Moving on to parasitic gaps ]
OZ: “What did Peter file without a reading?”
EB: None of this looks relevant to your work.
GE: But parasitic gaps are the kind of thing HPSG would be good at.
EB: Yes, fun to show off 🙂
Home | Forum | Discussions | Events