This paper concerns the SBCG representation of five English constructions that combine special grammatical form with a special discourse pragmatics.  The grammatical features include (optional or obligatory) agreement between the verb and an argument that appears in what is usually thought of as direct object position. The discourse pragmatic features are presentational illocutionary force and a postverbal focused argument. The constructions are:

Although some of these constructions have received extensive treatment in the literature (especially PP Inversion and Existential there), to our knowledge no previous analyses have attempted to account for both the formal and interpretational similarities among all five, as well as the details of their formal and interpretative differences. Some of the interpretive differences are subtle: we will claim that Deictic there, unlike Presentational there, is a stative construction, despite its ability to combine with the dynamic verbs go and come. We will use frame-based representation to capture such aspectual properties.

Based on interactions between these constructions and constructions like Raising, Cleft, Nonsubject Wh-interrogative and Auxiliary Initial (SAI), we propose a lexicalist approach to this family of constructions, in which they are licensed through a combination of listemes, derivational constructions and inflectional constructions (Sag 2012). One virtue of this approach in our view is that it permits us to separate the External Argument (XARG) role from that of agreement ‘trigger’. While agreement features are included in referential indices (as per Pollard and Sag 1994: CH 2), subject behaviors like raising and control of a tag subject are a function of XARG status—a status occupied, e.g., by the prepositional argument in Prepositional (‘Locative’) Inversion. We defend the view that subject diagnostics discern the XARG properties of the oblique preverbal expression against recent attempts to debunk these diagnostics, in particular Postal 2004 and Bruening 2010. We further demonstrate that a derivational approach to Prepositional Inversion enables us to capture the distinction between complement and non-complement locative PPs, as discussed (in our view less than fully adequately) by Bresnan 1994, among others. This account provides a simple description of agreement variability—attestation of both canonical and ‘backwards agreement’ patterns in equative clauses (Corbett 2006: 63-64). We posit an equative be listeme in which both arguments in the identity relation are nominative, leveraging the fact that an inflectional construction requires only that some argument in the ARG-ST list of the daughter lexeme be nominative, not a particular one. When both arguments are nominative, as in the case of equative be, the inflectional construction can monitor agreement features of either the XARG or the non-XARG argument.

We observe that previous approaches to Prepositional Inversion, for example, have been forced to adopt special theoretical apparatus to account for the fact that properties of canonical subjects are distributed over two different constituents.  We show that SBCG, by contrast, is able to accommodate this grammatically unusual state of affairs succinctly, in a uniform analysis that accounts for all of the formal and interpretational commonalities and differences among the five constructions.

References

Bresnan, J. (1994). Locative inversion and the architecture of universal grammar. Language 70, 72-131.

Bruening, B. (2010). Language-particular syntactic rules and constraints: English locative inversion and do-support. Language 86, 43-84.

Corbett, G. (2006). Agreement. Cambridge: Cambridge University Press.

Pollard, C. and I.A. Sag (1994). Head-Driven Phrase Structure Grammar. University of Chicago Press.

Postal, P. M. (2004). A Paradox in English Syntax. Skeptical Linguistic Essays. Oxford: Oxford University Press. 15-82.

Sag, I.A. (2012). Sign Based Construction Grammar: An Informal  Synopsis. In Boas, H. C., & I.A. Sag, (eds.), Sign-Based Construction Grammar. Stanford: CSLI Publications. 69-202.