constraining equations question

[Helge Dyvik]

Hello Ash,

The XLE platform allows non-integrated f-structures, and I think that  
they should be allowed. I find them useful (and theoretically  
acceptable) e.g. in the analysis of selected PPs, as in ‘I am talking  
about the problems’. Here (or actually in the corresponding Norwegian  
construction ‘Jeg snakker om problemene’) I would like to say that the  
preposition originates from the regular lexical entry for ‘om’, PRED  
and all, but that its PRED is not contributed to the global f- 
structure. The point is that when something is selected, and hence in  
a certain sense predictable, it does not carry its normal amount of  
information. In this case the information contributed is that we are  
faced with the predicate ‘talk*about’, taking the OBJ of the  
preposition as its own OBL-TH. The result is that the f-structure of  
the preposition itself is left dangling. (The interested reader can  
inspect this on the page http://decentius.aksis.uib.no/logon/xle- 
mrs.xml , parsing the sentence ‘Jeg snakker om problemene’ with the  
grammar ‘Norwegian Bokmål (new morph)’, unchecking ‘Packed  
representation’ and checking ‘Include non-top F-structures’. The  
relevant reading is the first one.)

As for Steve’s question, I don’t see why we shouldn’t be allowed to  
state constraints on orphaned f-structures once we allow them.

Best,
Helge

Den 27. april. 2011 kl. 16.16 skrev Ash_Asudeh:

> Dear Steve
>
> That's a difficult question, but my understanding is that it should  
> be satisfied: since the constraint has existential force, it should  
> at least logically be satisfied if
> there is some f-structure that satisfies it and, on your scenario,  
> there is some f-structure that satisfies it.
>
> In the scenario you describe, I think the final f-structural result  
> would not be well-formed, because it would be disconnected, but that  
> is for reasons other than the constraining equation in question  
> (i.e., it would be due to a general, global constraint on f- 
> structural well-formedness).
>
> Ash
>
> On 2011-04-27, at 8:36 AM, Stephen Wechsler wrote:
>
>> Hi all,
>> Can anyone answer this question for me?  Is a constraining equation  
>> (f
>> ATT) =c VAL satisfied, or not, in a case where the f-structure for  
>> the
>> root node does not contain the function f, but the 'orphaned'
>> f-structure for a lower node in the tree does contain f with the
>> feature [ATT VAL] ?  This is possible where a c-structure node fails
>> to pass up some information.
>>
>> Steve

--------------------------------------------------------------------
Professor Helge J. Jakhelln Dyvik
Institutt for lingvistiske, litterære og estetiske studier
Faggruppe for lingvistiske fag
Universitetet i Bergen
Sydnesplassen 7  	Tel.: +47 55582261
5007 Bergen              	E-post: helge.dyvik at lle.uib.no
--------------------------------------------------------------------



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lfg/attachments/20110427/ec646d84/attachment.htm>