Toward a Grammar of Danish Talk-in-Interaction: From Action Formation to Grammatical Description

ABSTRACT

Is it possible to develop a comprehensive grammar of talk-in-interaction for a specific language based on descriptions of social actions? This is the question we will try to answer in this article. The article is based on the work of the project The Grammar in Everyday Life, which aims to build a systematic grammatical description of Danish talk-in-interaction based on descriptions of social action formats within three domains: question–answer sequences, commissive–directive sequences, and the negotiation of participation during longer spates of talk. Our ambition is to build a grammar that takes into consideration how talk is used in the real-time unfolding of interaction to do actions and to negotiate relationships. Through a presentation of three formats, we discuss how a grammatical description can be organized, how granular it should be, if and how traditional grammatical categories can be used, and how prosodic and embodied features could be included. Data are in Danish.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Thanks to an anonymous reviewer and the editors of the journal for suggesting this term.

² We deviate from the Leipzig rules, in that morpheme division (with “-”) is not shown in the transcript line, only in the glossing line. In the Appendix, we list all the abbreviations used in the glossing.

³ Analyses suggest that proform-questions without det point to the knowledge discrepancy issue as concerning the subject of the previous informing, whereas in proform-questions with det, the issue lies in the event or the circumstances described in the informing (see, Jørgensen, 2021, 2023, for further detail).

⁴ We have used Praat for the acoustic analyses (Boersma & Weenink, 1992–2022).

⁵ The verb ve’/vil, when used as a modal verb pointing to a (possible) “willingness” of the person inhabiting the subject role in the clause (Hansen & Heltoft, 2011, p. 773), might be translated as “want to” (as suggested by reviewers). We have chosen to translate it as “will,” retaining its less subjective, “future” meaning, mainly because particles like godt or gerne (roughly “well” or “willingly”) would be added if the focus was on willingness, and because the response does not show any orientation toward the utterance inquiring into the recipient’s willingness. A response like det ve’ jeg godt/gerne (roughly, “I would like ((to do)) that”) would show an explicit orientation to the willingness aspect and would call for a (retrospective) translation of ve’ du as “do you want to” or “would you like to”. However, no such orientation is found with the requests for immediate action that we are considering here.

⁶ There is a difference between cases like this, in which the beneficiary is left open, and cases of “bilateral requests,” in which the requested action is “integral to an already established joint project” (Rossi, 2012, p. 428).

⁷ A reviewer suggested a comparison with the English construction “why don’t you” (WDY), which is used to offer advice and “forward a solution for what is treated as a problem for the interlocutor” (Thompson & Couper-Kuhlen, 2020, p. 100). The differences between this format and the one we are investigating are, however, larger than the similarities: Even though there may be problems in the environments of modal interrogatives, these environments are not “advice-implicative contexts” in the sense of Thompson and Couper-Kuhlen (2020, p. 105); advice-giving turns in Danish often contain modal particles that we do not find in this format (Heinemann & Steensig, 2017; see also footnote 14 about German and Dutch in Thompson & Couper-Kuhlen, 2020, p. 110); the responses do not orient to the directive-commissive action as doing advising. We have, therefore, chosen to see the instance in Excerpt 5 as lying between a *Request and a *Suggestion, and we see this option as an inherent feature of the social action format.

Additional information

Funding

This work was supported by the Independent Research Fund Denmark under Grant 9037-00072B; and the Finnish Academy of Science and Letters, Personal Grant.