University of Cambridge > > CILR workshop: Pragmatics in interfaces > Ellipsis, Dialogue Modelling, and the Grammar-Pragmatics Interface

Ellipsis, Dialogue Modelling, and the Grammar-Pragmatics Interface

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Napoleon Katsos.

ABSTRACT : In this talk I take up the challenge of providing a grammar formalism which can provide a basis from which the patterns of dialogue can be naturally reflected (Pickering and Garrod 2004). There are two major problems posed by dialogue patterns. First there is the so-called split utterance phenomenon, in which speakers and hearers can switch roles across any syntactic dependency whatever. This ability is wholly systematic and displayed even by very young children with emergent syntactic skills. This phenomenon is highly problematic for all formalisms which presume on a use-neutral formalism which production and parsing separately make use of. Secondly, there are widespread parallelism effects, displayed both in ellipsis construal and more generally in priming effects; and these have generally taken to be a primitive constraint on dialogue processing. In this talk I will give a sketch of a grammar formalism which has the intrinsic dynamics of a parsing device, Dynamic Syntax (Kempson et al 2001, Cann et al 2005), and show that this framework directly meets the Pickering and Garrod challenge, as follows.

First, the grammar is defined in terms of mechanisms for the process of parsing, and these, along with the partial structures which such processes give rise to, form part of the context relative to which interpretation is progressively built up. It is re-use of such procedures which forms the heart of parallelism effects, both in ellipsis and in lexical, syntactic and semantic priming. A happy consequence of this account of ellipsis is an integrated basis for explaining ellipsis effects, despite the diversity of available interpretations. Secondly, the phenomenon of split utterances is directly predicted: the production system, in having to use the associated grammar formalism, has to have the same incremental dynamics as the parsing device. The very tight coordination of parsing and production is ensured: and, as I shall show, the phenomenon of split utterances follows immediately.

A consequence of this system is that there is no concept of sentence-meaning, since the mechanisms for growth of interpretation interact with context-dependent choices as part of the process of constructing interpretations. In closing, I shall argue that, with this shift in perspective, the contextualism debate (as posed by Cappelen and Lepore 2005) is transformed. Far from being irrelevant as Cappelen and Lepore presumed, ellipsis phenomena demonstrate the force of the contextualist claim that natural language interpretation is essentially and systematically context-dependent.

Cappelen, H. and Lepore, E. 2005 ‘Insensitive Semantics’. Blackwell.

Cann, R, Kempson, R, and Marten, L. 2005. ‘The Dynamics of Language.’ Elsevier.

Kempson, R., Meyer-Viol, W. and Gabbay, D. 2001. ‘Dynamic Syntax: The Flow of Language Understanding’. Blackwell.

Pickering, M. and Garrod, S. 2004. Toward a mechanistic psychology of dialogue. Behavioral and Brain Sciences, 27, 169-225.

This talk is part of the CILR workshop: Pragmatics in interfaces series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2019, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity