Combining corpus and experimental methods to gain new insights into APPRAISAL in spoken discourse

Research output: Contribution to conferenceAbstract


The mainobjective of this paper is to challenge the treatment of first-person epistemicand evidential complement-taking predicate (CTP) constructions in Martin andWhite’s (2005) appraisal theory,and to offer suggestions for improving the model. Based on empirical evidencefrom a corpus-based analysis and a laboratory experiment, we demonstrate that CTPconstructions do not only serve to expand the dialogic context in which theyoccur, but also to put a lid on alternative views. The paper contributes to the refinement of appraisal as a corpus annotation tool,and provides a practical illustration of the usefulness of combining corpus andexperimental methodsto gain new and robust insights into linguistic phenomena.

In appraisal, first-person CTPs such as I think and I believe are classified as instances of dialogic expansion in that they make dialogic space for possiblesubsequent arguments (Martin & White, 2005: 98). They are set in contrast toexpressions of dialogic contraction (e.g.obviously, however, but), which areused in discourse to challenge, resist or reject alternative value positions. Inspoken discourse, however, CTP constructions do not appear to always perform anexpansive function. In (1), for example, Ithink is prosodically unaccented, serving as a starting point for theopinion expressed in the complement clause, which contains an evidential markerwith a falling accent – obviously –signalling a high degree of commitment (Cruttenden, 1997; Kärkkäinen, 2003).

(1) B: I think he was \obviously trying to steer us in that direction and

sort of

A: yes

B: dropping hints

Martinand White (2005: 103) recognize that the function of engagement expressions "may vary systematically under theinfluence of differentco-textual conditions, and across registers, genres and discourse domains."These conditions, however, are not discussed in detail by the authors, nor havethey been systematically investigated in the literature. In this study, we combinecorpus-based and experimental methods to test the effect of different contextualfactors on the dialogic function of CTP constructions.

The study is conducted in twophases. First, an exploratory analysis of CTP constructions in the London-LundCorpus (LLC) of spoken British English is carried out to identify factors thatmay play a role in determining the dialogic force of the constructions. The annotationof CTP constructions in the corpus is performed following Fuoli’s (forthcoming) step-wise method for annotatingappraisal, and is validatedthrough an inter-rater reliability test. Hypotheses derived from the corpusfindings are then tested in a controlled experimental setting. The resultsindicate that CTP constructions not only serve to expand the dialogic contextin which they occur, but may also function to inhibit dialogue. Interlocutorstatus, prosodic marking and the co-occurrence of a contractive marker are shownto have a significant effect on the function of CTP constructions.


Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Specific Languages


  • stance, complement-taking predicates, corpus annotation, inter-rater agreement, reliability
Original languageEnglish
Publication statusPublished - 2016
Publication categoryResearch
EventCorpora and Discourse International Conference - Pontignano Conference Centre, Pontignano, Italy
Duration: 2016 Jun 302016 Jul 2


ConferenceCorpora and Discourse International Conference
Abbreviated titleCADS
Internet address