On the position and meanings of epistemic complement-taking predicates in spoken British English

Research output: Contribution to conferencePaper, not in proceeding


This is a corpus study of the use and meanings of epistemic complement-taking predicates (CTPs) such as I think in (1) – (3):

(1) I think that we mustn’t worry too much about this.
(2) One of the real damages I think that this has done is that it shakes our faith.
(3) Because I said I wanted to go at four thirty I think.

As shown in the examples above, I think, like most other CTPs, can occur in different positions in a sentence: in an initial (1), medial (2) and final position (3). The positional distribution of epistemic markers and its effect on their discursive meanings has received a great deal of attention in recent linguistic research (e.g. Traugott 2012, Degand 2014). Typically, the left periphery is associated with speaker-oriented meanings and the right periphery with addressee-oriented meanings. The systematic occurrence of CTPs in different positions has raised the question of whether there is a direct correlation between the positional distribution of the verbs and the functions they serve in discourse. Aijmer (1997) automatically classifies initial positions of I think as displaying a greater degree of deliberation and authority, with medial and final positions exhibiting opposite values. Dehé and Wichmann (2010), on the other hand, have noted that sentence-initial CTPs often fulfil addressee-oriented functions, such as hedging. However, little attention has been paid to the role of context in the interpretation of position and function.
Couched in the framework of Cognitive Linguistics (Langacker 1987), this study makes use of both qualitative and quantitative methodological techniques, which is necessary for a comprehensive investigation of such multifunctional constructions (Pichler 2013). The data consist of spontaneous conversations between educated adults in university settings, retrieved from the London-Lund Corpus (Svartvik & Quirk 1980). The topics of discussion range from friendly exchanges of the use of sewing machines to highly critical evaluations of university departments. Also, the conversations represent interaction between various age groups, genders and social roles, allowing a close investigation of the relationship between socio-contextual factors and the choice of positional distribution of CTPs.
So far, the results of the analysis indicate that the position and meanings of CTPs are influenced by socio-contextual factors, such as the sensitivity of the topic discussed and the relationship between the speaker and the addressee. What is implied is that the addressee-oriented meanings of CTPs are in fact equally characteristic in initial positions and often preferred in negotiating sensitive topics by speakers with less authority, while medial and final positions tend to be used for reinforcement and assertiveness by speakers with more authority.


Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Languages and Literature


  • corpus linguistics, Construction Grammar, language change, clause periphery, discourse grammar
Original languageEnglish
Publication statusUnpublished - 2015
Publication categoryResearch
EventThe 13th International Cognitive Linguistics Conference - Newcastle, United Kingdom
Duration: 2015 Jul 202015 Jul 25


ConferenceThe 13th International Cognitive Linguistics Conference
CountryUnited Kingdom

Total downloads

No data available