Emma: but! hey! we will forge the way to more communication between cogsci and ling
Celeste: yeah
Celeste: like that movie
Celeste: where the cat is friends with the dog
Celeste: and they go on adventures together
Emma: milo and otis
Celeste: yes.
Emma: and when we get older our puppies and kitties will hang out and go on adventures
Celeste: while the credits roll
emma :: Apr.10.2007 :: misc :: 4 Comments »
[Note: this was cut from a much rantier post on PRO, the Carnie text and the “mainstream” view of syntax. Simon read it and said I sounded hysterical…so I decided to cut it down to this, which is much less ranty, but also much less explanatory if you haven’t seen any alternatives to the mainstream view. So, for now, here’s this, and when I’ve come off my coffee high hysteria, I’ll try to explain what this was all about.]
This past week in Polly’s syntax class we’ve been covering the topic of control, cases such as:
(1) Emma is reluctant to accept the theta criterion.
We were introduced to two ways of approaching these types of constructions. With our “Tuesday mainstream†hats on (on Tuesdays we cover the mainstream treatments of syntax, on Thursdays we cover the Categorial Grammar treatments) we looked at how transformational grammar (and descendants thereof) introduced the null, silent PRO and coindexed PRO with the subject of the matrix clause to give the following analysis for (1) at D-structure:
(2) Emmai is reluctant to PROi accept the theta criterion.
Carnie’s explanation of PRO:
Why do we need PRO? If we didn’t have PRO, then we would have violations of the theta criterion. Notice that what we are doing here is proposing a null element to account for an apparent hole in our theory (a violation of either the theta criterion or the Case filter). There is good reason to be suspicious of this: It seems like a technical solution to a technical problem that is raised only by our particular formulation of the constraints. Nonetheless, it does have a good deal of descriptive power. It can account for most of the data having to do with embedded infinitival clauses. Until a better theory comes along, the PRO hypothesis wins because it can explain so much data.
emma :: Apr.10.2007 :: misc, syntax, linguistics :: No Comments »
Simon Charlow, a precocious and talented undergrad here at Brown, has a blog now. He’s already posting some neat stuff about binding in CCG, Anna Szabolcsi’s approach to reflexives and some interesting consequences of her approach.
emma :: Apr.09.2007 :: misc, semantics, linguistics :: 5 Comments »
Both fascinating and hilarious, David Dowty presents “Great Moments in the History of Syntax: the Invention of the Trace (1847)“.
emma :: Apr.07.2007 :: misc, syntax, linguistics :: 8 Comments »
Every class, I find myself scribbling down at least two or three (but usually many more) great quotes by Polly. So I’ve decided to start posting each week’s worth of great Pollyisms to share with everyone and also to preserve some of the funniest things she’s said.
This week we’ve been talking about control verbs in class.
Re: the transformational grammar view,
But, excuse me, who said the theta-criterion was correct? […] Dump the theta-criterion.
Re: my timidity to agree to ‘John promised Mary to cut the grass’ being totally acceptable,
Are you worried about being accused of Valley talk?
Re: how arbitrary category names are,
Call it ‘grasshopper/NP’.
Our ‘grapefruit/NP–wait, was it grapefruit?
Re: how any theory will need to give words meanings anyway (so why not make the semantics do all the work w.r.t. control),
Words have meanings!
And, this week’s best:
GB is the theory where you always get what you want. In CG, you can’t always get what you want–but if you try sometimes…
emma :: Apr.05.2007 :: misc, syntax, linguistics :: 2 Comments »
So today, my officemate Justin was debugging Matlab code and asked me about the boolean and operator. I explained to him that you don’t write and but rather use &&. He then asked if he could do the following “a < b && !== c", where he meant that a is less that b and a is not equal to c. "No, Justin, Matlab doesn't do conjunction reduction." Not the funniest thing I've ever said, but given the day I've had, it was a lovely moment for me.
Also: Happy Passover! I just came back from my first seder which was so so much fun. I got to ask the four questions AND I found the afikoman. I reclined and somehow managed to have the four obligatory glasses of wine.
Lastly: Celeste is blogging again!
emma :: Apr.03.2007 :: misc, linguistics, programming, nerdiness :: 6 Comments »
This is delayed news, but I wanted to wait until I finished some work to post this :-P
I found out last week that I got the fellowship to attend this summer’s LSA Institute at Stanford. Palo Alto, I’ll see you in July!
emma :: Apr.03.2007 :: misc, linguistics :: No Comments »
…and more!
New set of notes out, covering some of the things I worked on over Spring Break while lounging in sunny (and warm!) Los Angeles.
The task at hand was to see if an NP-S view was possible in functional readings of NPs and then, if so, if such a view could also be taken in the “binding into heads” cases. So, could you take the NP-S approach in a sentence like “The woman who every man loves is his mother”? And if so, can you then extend that analysis to “The woman who he loves who every man invited is his mother”? The answer I propose is that given apparatus we’ve already assumed we needed in a Det-Nom approach (most crucially, Jacobson 2002’s m operator), nothing extra is needed to get an NP-S analysis to work. I need to work out the details of each with respect to the “binding into heads” cases, but I suspect that, actually, the NP-S approach (coupled with a solid RC stacking rule) might come out cleaner and simpler than Det-Nom.
Next up: more Chinese data! And! A comparison of these variable-free approaches with the variable-full ones. Stay tuned, gentle reader ;-)
Also: I forgot to link the papers in the references section which are available online. Two are available at semanticsarchive, except for the quantifier restriction one, which you can get if you e-mail Polly; her L&P paper is available through their site but you need a subscription or access to a library with one to get a full text version (grr, I know!). The two at semanticsarchive, though, are both excellent and great reads if you already know a bit of VFS.
emma :: Apr.03.2007 :: misc, semantics, linguistics :: 2 Comments »
