TY - JOUR
T1 - Functional compositionality and the interaction of discourse constraints
AU - Birner, Betty J.
AU - Kaplan, Jeffrey P.
AU - Ward, Gregory
N1 - Funding Information:
funded by the Klaus Tschira Foundation, Heidelberg, Germany. The first author has been supported by a HITS PhD. scholarship. We would like to thank Éva Mújdricza-Maydt for implementing the mention taggers.
PY - 2007/6
Y1 - 2007/6
N2 - We argue for the existence of functionally complex constructions whose elements composition-ally impose discourse-functional constraints on the use of the whole. In particular, we examine th-clefts (as in That's John who wrote the book), equatives with epistemic would and demonstrative subjects (as in That would be John), and simple equatives with demonstrative subjects (as in That's John). We show that, contra previous approaches, the latter two constructions need not be analyzed as truncated clefts. Rather, the properties that these constructions share with th-clefts can be straightforwardly accounted for as the sum of the constraints on their shared elements-that is, the equative construction, the demonstrative subject, and the presence of a contextually salient open proposition. The convergence of these elemental properties in each of these three constructions results in the possibility of the demonstrative being used to refer to the instantiation of the variable in the open proposition, which in turn predicts a complex of distributional behaviors shared by precisely the constructions that share these properties. Because these distributional behaviors can be straightforwardly explained in terms of this functional compositionality, the motivation for a truncated-cleft analysis disappears. These results support the view that not all functional properties must be learned on a construction-by-construction basis; instead, the discourse functions of an utterance are built up compositionally from those of its parts.*.
AB - We argue for the existence of functionally complex constructions whose elements composition-ally impose discourse-functional constraints on the use of the whole. In particular, we examine th-clefts (as in That's John who wrote the book), equatives with epistemic would and demonstrative subjects (as in That would be John), and simple equatives with demonstrative subjects (as in That's John). We show that, contra previous approaches, the latter two constructions need not be analyzed as truncated clefts. Rather, the properties that these constructions share with th-clefts can be straightforwardly accounted for as the sum of the constraints on their shared elements-that is, the equative construction, the demonstrative subject, and the presence of a contextually salient open proposition. The convergence of these elemental properties in each of these three constructions results in the possibility of the demonstrative being used to refer to the instantiation of the variable in the open proposition, which in turn predicts a complex of distributional behaviors shared by precisely the constructions that share these properties. Because these distributional behaviors can be straightforwardly explained in terms of this functional compositionality, the motivation for a truncated-cleft analysis disappears. These results support the view that not all functional properties must be learned on a construction-by-construction basis; instead, the discourse functions of an utterance are built up compositionally from those of its parts.*.
UR - http://www.scopus.com/inward/record.url?scp=41349117661&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=41349117661&partnerID=8YFLogxK
U2 - 10.1353/lan.2007.0056
DO - 10.1353/lan.2007.0056
M3 - Article
AN - SCOPUS:41349117661
SN - 0097-8507
VL - 83
SP - 317
EP - 343
JO - Language
JF - Language
IS - 2
ER -