|
Size: 713
Comment:
|
← Revision 8 as of 2009-06-01 17:02:57 ⇥
Size: 1782
Comment: YATC (Yet another time change)
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 1: | Line 1: |
| ## page was renamed from HlpLab/Labmeeting/Su09w4 ## page was renamed from HlpLab/Labmeeting/Su09w3 ## page was renamed from HlpLab/Labmeeting/Su09w2 ## page was renamed from HlpLab/Labmeeting/Su09w1 ## page was renamed from HlpLab/Labmeeting/Su09w2 |
|
| Line 16: | Line 21: |
| Also, this person's name is *Max Bane* (at Chicago, in fact, with Floseph), which is reason enough for me to read anything he writes that isn't unreasonably long (this paper is short). |
We'll also be reading a paper by D. Gildea and D. Temperley that Ben VD mentioned in lab meeting today (5.12.09). The idea, I think, is that they construct a number of different possible grammars that represent solutions to the problem of minimizing dependency lengths between words, and compare the grammar of English to these other solutions. Why would these two papers be chosen for the same lab meeting? Well, I think they both deal with the question of why languages look the way they do (although this is more implicit, maybe latent, in the Bane paper), and suggest that this question can be answered by appealing to general principles of efficiency (optimality?), rather than being satisfied with saying that language just has arbitrary structural properties that are innately specified. Of course, whether the constraints on language structure outlined in these two papers are themselves innately specified is an independent question. (not sent from my iPhone--alex) [[AttachList]] |
Morphological Complexity
(Via Alex)
I recently ran across a proceedings paper by Max Bane that I found pretty interesting. It represents an attempt to quantify morphological complexity, which is a notion that gets thrown around loosely in linguistics. Having a precise definition for what that means could turn a lot of interesting, fuzzy claims into interesting, falsifiable claims.
We'll also be reading a paper by D. Gildea and D. Temperley that Ben VD mentioned in lab meeting today (5.12.09). The idea, I think, is that they construct a number of different possible grammars that represent solutions to the problem of minimizing dependency lengths between words, and compare the grammar of English to these other solutions.
Why would these two papers be chosen for the same lab meeting? Well, I think they both deal with the question of why languages look the way they do (although this is more implicit, maybe latent, in the Bane paper), and suggest that this question can be answered by appealing to general principles of efficiency (optimality?), rather than being satisfied with saying that language just has arbitrary structural properties that are innately specified. Of course, whether the constraints on language structure outlined in these two papers are themselves innately specified is an independent question. (not sent from my iPhone--alex)