HELP!

[bresnan at csli.stanford.edu]

Many of the questions you raise about lfg are issues addressed in my
forthcoming text, LEXICAL FUNCTIONAL SYNTAX, and I will be previewing
the issues of phrase structure typology and
structure-function correlations in my course in Barcelona ESSLI this August.
Till the text is generally available (and others are in the works by
Dalrymple, and by Austin), you might look at some of the recent
literature on the linguistics of lfg (articles and dissertations).
Several excellent dissertations (Tara Mohanan, Paul Kroeger, Miriam
Butt) have recently been published by CSLI Publications in an
attractive paperback format, and more are in the pipeline.

> My first problem has to do with functional-structure.  Here's how I think
> f-structure for a given sentence is developed, please correct me if I'm
> wrong.  A typical verb might be characterized in f-structure along the
> lines of the following, taking a classical example (I know where the up
> arrows go, but I don't know how to represent them in this text format):
> 	'HAND<(^ SUBJ)(^ OBJ)(^ OBJ2)>' 
> It seems, however, that there is a representation more basic, namely
> one in which the entry seems to correspond to the theta grids of GB,
> along the lines of the following (I don't know if capitals should be
> used here to indicate a semantic notion or not):
> 	hand (agent) (goal) (theme)
> It is principles of grammatical function assignment that then associate
> these (universal) thematic roles with the (universal) grammatical functions.
> I am thinking that because this mapping is well defined, it is rarely
> referred to, and the representations using grammatical functions is used
> instead as a sort of shorthand.

There is a literature on the topic of mapping from predicate argument
structure roles to grammatical functions:  See Bresnan and Zaenen
1990, Bresnan 1994 (Lg paper), and the references cited therein to the Lexical
Mapping Theory. 

  Is this correct?  (One of the major
> important insights of the theory, of course, is that such grammatical
> functions are universal, with the details manifested in language-specific
> ways in different languages.  I have to leave discussion and clarification
> of this point to a later inquiry.)
> 
> OK, so is it possible to take the position also that it is the verb's lexical
> entry that constrains the remainder of the f-structure for a given sentence?
> I.e., given that this instance of HAND requires SUBJ,OBJ, and OBJ2 to be
> present, those grammatical functions must also appear in the f-structure of
> the sentence, specific manifestations selected arbitrarily in any given
> instance (as is the verb itself)?  I.e., the principles of Completeness and
> Coherence insure that the proper number and type of arguments is associated
> with the predicate (not unlike the Theta Criterion, it seems).  However,
> some questions persist:  Is it required that SUBJ, OBJ, OBJ2 be NPs?  Where
> is that information found?

See Bresnan 1995 (Category Mismatches) and Bresnan 1994 (Lg article)
for some discussion of this.  Constraints on structure-function
associations are proposed there, and in much fuller detail in the text
I mentioned above.

  Do NPs and other constructions have their own
> c-structure rules?  I would expect so, but I have not come across any
> examples in my reading.
> 

C-structure rules are derivable from generable principles of X'
theory.  The version of X' theory used in lfg differs somewhat from
that used elsewhere; see Austin and Bresnan (forthcoming, NLLT) for a
synopsis and illustration with respect to Australian nonconfigurationality.
I believe that a number of current papers and dissertations are listed
in the lfg bibliography available from PARC.  

> If I am correct, the development of the f-structure for a given sentence may
> be characterized as a progression along the lines of the following:
> 	predicate and its thematic roles selected
> 	thematic roles associated with grammatical functions on universal
> 		grounds
> 	grammatical functions established within f-structure and "filled in"
> 		appropriately
> 

There is no derivational order specified in the formal architecture of
lfg.  Everything is done by constraint satisfaction based on recursive
unification over equality.  However, for expository linguistic
purposes it is natural to think of each lexical predicatory
"projecting" a skeletal f-structure by means of the lexical mapping
theory, which must then be matched up and unified with the surrounding
syntactic context.  Principles of structure-function assocation (of
which X' theory provides the most familiar type) provide partial
f-structure information from the context, which is unified with the
lexical f-structure information.

> Your help is very much appreciated.  I realize that this is so basic that
> all of you on the list could respond and a lot of time would be wasted by
> the redundancy.  I don't know what to do; perhaps you have a designated
> guide?!
> 

You can obtain a number of papers by looking at the lfg bibliography
and requesting copies from Mary Dalrymple at Xerox PARC or from me (at
Stanford).  I hope to have a draft of my text available by the end of
this summer.

Joan Bresnan