A new way of investigating the semantics of constructions. Collostructional analysis and its application to causative constructions
(2005) 6th Annual Stanford Semantics Fest — Location: Stanford (11.March.2005)
Files
No attached file found for this publication.
Details
- Authors
- Abstract
- The notion of construction has always been important in the study of language. Over the past decade, however, it has been given a central role within the framework known as ‘Construction Grammar’ (Goldberg 1995), which views grammar as consisting of linguistic signs (just like lexicon) and so considers grammatical structures as meaningful units. Relying on these assumptions, the method of collostructional analysis (Stefanowitsch & Gries 2003) investigates the interaction between words and constructions and seeks to identify the words which are, statistically, strongly attracted or repelled by a particular slot in a construction, in an attempt to determine the meaning of the construction. Collostructional analysis includes three different techniques, viz. (i) collexeme analysis, which studies one slot in a particular construction (e.g. the V slot in the [X think nothing of Vgerund] construction, see Stefanowitsch & Gries 2003); (ii) (multiple) distinctive collexeme analysis, which studies one slot in two (or more) similar constructions (e.g. the verb in the ditransitive and to-dative constructions, see Gries & Stefanowitsch 2004a); (iii) covarying collexeme analysis, which studies the interaction between two slots in a particular construction (e.g. V1 and V2 in the [X V1 Y into V2gerund] construction, see Gries & Stefanowitsch 2004b). A multiple distinctive collexeme analysis will be applied to the verb slot of English periphrastic causative constructions. In other words, it will be seen whether constructions such as [X MAKE Y Vinf] (1) or [X CAUSE Y Vto-inf] (2) attract different verbs or whether, as is often assumed in the literature, they can be used with any verb and so cannot be distinguished on these grounds. (1) But on the other hand, the approach would also make learners realize (in the sense of recognize) the significance of grammar, and raise their consciousness of its relevance. <BNC:W:CBR 1562> (2) Heating collapses the blue spectral peak, causing the violet luminescence to shift towards red. <BNC:W:H9S 1964> Using corpus data from the BNC and the software Coll.analysis 3 (Gries 2004), it will be shown that the causative constructions display strong preferences for particular (groups of) verbs and that these are essentially different for each construction. In addition, it will be demonstrated that examining the list of verbs attracted to a construction (its ‘collexemes’) helps determine the typical meaning(s) of the construction. Thus, collexemes such as 'cut', 'perm', 'build', 'replace' or 'repair' in the [X HAVE Y VPP] construction evoke the frame of service, where a CAUSER commissions an (implied) CAUSEE to do something. The [X GET Y VPP] construction, by contrast, although it is often equated with the 'have' construction, exhibits no such collexemes (its strongest collexeme is 'sort out') and so must be described as a construction in its own right, distinct from the 'have' construction. Likewise, the distinctive collexeme analysis makes it possible to distinguish between the other causative constructions and assign each of them a specific meaning.
- Affiliations
Citations
Gilquin, G. (2005). A new way of investigating the semantics of constructions. Collostructional analysis and its application to causative constructions. 6th Annual Stanford Semantics Fest, Stanford. https://hdl.handle.net/2078.5/225872