Marcin Lewiński

This paper addresses the following question: Can one and the same utterance token, in one unique communicative context, perform a plurality of speech acts? And to do so in a non-defective, intentional and conventional way? While some of the recent literature has rightly entertained and partly defended such a possibility (Sbisà, 2013; Johnson, forth.), I will build a case for it based on a critical redefinition of conversational context in which speech acts are exchanged. Traditional analysis characteristically assumes a neatly dyadic model of one Speaker and one Hearer. Departing from this model, I will argue that plural speech acts are standard features of polylogues, dialogues which include more than two participants, and thus more than one hearer (Kerbrat-Orecchionni, 2004; Lewiński, 2014, 2017). If indeed, as amply reported (Goffman, 1981; Levinson, 1988), multi-participant exchanges are the norm rather than the exception, then speech act pluralism is a regular feature of our conversational business and in need of serious investigation. (More in particular, illocutionary pluralism: locutionary pluralism (Cappelen & Lepore, 2005) basically extends the classic discussion of semantic underdetermination, while perlocutionary pluralism is trivially obvious (Austin, 1962/1975; Sbisà, 2013).)

Consider the following dialogue fragment:

Ann, to Barbara and Chris:     What’s the time?

Barbara:                                  Chris has a watch.

Chris:                                       Three thirty.

Which illocutionary act has Barbara performed? Even assuming that her (simple and true) assertion is merely a means to perform another primary speech act (Searle, 1975), we still have at least two options: a) it is an argument for Ann, justifying Barbara’s rejection to fulfill Ann’s request; or b) it is a request to Chris, to tell Ann the time, which he does. I argue that Barbara might be intending to perform both forces with one simple utterance, and to do so in a recognizably conventional way – as confirmed by the responses from the hearers of her utterance – without any deliberate priority among these acts.

By contrast, traditional speech act theory (Austin, 1962/1975; Strawson, 1964; Searle, 1969) is largely governed by the assumption of illocutionary monism (see Johnson, forth.): a felicitous performance of an utterance properly carries one unique illocutionary force. Cases where more than one speech act is undeniably performed are treated as two-level performances of indirect speech acts: a question (secondary act) is used to convey a request (primary act), etc. (Searle, 1975).

Sbisà (2013) and Johnson (forth.) have convincingly argued for a nuanced pluralistic account. If – as originally outlined by Austin – the audience’s uptake is an essential element in determining an illocutionary force of a speech act, then a multiple ascription by the same respondent or multiple ascriptions by various respondents open the door for pluralism. My aim in the paper is to give a theoretically consistent account grounded in the latter possibility. Analyzing examples such as Barbara’s utterance, I first introduce the concept of polylogue (Lewiński, 2014, 2017; Lewiński & Aakhus, 2014). I then show how polylogical circumstances allow for intentional and conventionally recognized performances of plural illocutionary forces via a singular utterance; in doing so, I refer to older literature on multi-participant conversations which directly bears on the issue but is conspicuously absent from the current discussion (Clark & Carlson, 1982; Goffman, 1981; Levinson, 1988). I further argue my view is consistent with a hybrid, interactional account of the ascription of illocutionary forces (see Sbisà, Johnson, Levinson, Witek): hearers and speakers can negotiate their illocutionary achievements in the normal course of (polylogical) conversation through various argumentative means, with the view to publicly recognizable commitments (Lewis, 1979). As such, this account is not merely hearer-relative.