Our words bind us, as in the slogan “my word is my bond” which is the theme of this year’s DGfS. This talk explores two senses of the ways our words bind us. The first of these is the familiar sense of commitment. Here I focus on one aspect of commitment and how it arises: the notion of sincerity, in two aspects: the reliability of the speaker with respect to truth and to communication of their social positioning, and the broader notion of trust, which can be viewed as arising from reliable communication about values and may not be linked in a strict sense to truth at all. The second kind of binding I discuss is less visible. Many parts of natural language come grounded on assumptions about social and metaphysical the world, and about what kinds of agents can be centered as the locus of subjectivity. I consider several such grounds, taking honorification, politeness, and the anthropomorphic nature of perspective-taking as my main examples. The result has implications for the ways in which we can use language to discuss social issues and problems of the nonhuman and climate change.

This presentation argues that the formalization of a certain class of sociological theories into game-theoretic models can help researchers in formal semantics, pragmatics and analytical philosophy of language get a handle on an area that has long been resistant to formal study: the social world. More specifically, we study the `Pragmatic Sociology' framework of Boltanski & Thévenot (1991) and argue that, when applied to language, it can be integrated with current models of language use and interpretation that are based on game-theoretic principles, such as Franke (2009), Frank & Goodman (2012) and Burnett (2019). Although Boltanski & Thévenot use language that is suggestive of mathematical models used in economics, the fine details of framework that they lay out in their book remain underspecified. Furthermore, although Pragmatic Sociology has spawned a wealth of sociological work in the past 30 years (see Lamont & Thévenot 2000 for a sample), the influence of these ideas in linguistics and philosophy has been minimal. We argue that this framework (and other more "practice" oriented social theories) has particular properties that make it a good candidate to be integrated with current linguistic theories, and with cognitive science more generally. We therefore give a formalization of Pragmatic Sociology's main units: 'polities', 'worlds' and
'orders of worth', and show how they can be extended with ideas from formal game-theoretic pragmatics to provide a general model of how language influences and is influenced by the social world. We argue that such a formalization has two major benefits: firstly, it will result in a better understanding of major aspects of Boltanski & Thévenot's theory, which could benefit sociologists, and, secondly, it will help provide better foundations for linguists and philosophers studying linguistic meaning from a mathematical perspective.

References: Thévenot, L., & Boltanski, L. (1991). De la justification. Les économies de la grandeur. Paris: Gallimard. Burnett, H. (2019). Signalling games, sociolinguistic variation and the construction of style. Linguistics and Philosophy, 42(5), 419-450. Frank, M. C., & Goodman, N. D. (2012). Predicting pragmatic reasoning in language games. Science, 336(6084), 998-998. Franke, M. (2009). Signal to act: Game theory in pragmatics. Amsterdam: Institute for Logic, Language and Computation. Lamont, M., & Thévenot, L. (2000). Rethinking comparative cultural sociology: Repertoires of evaluation in France and the United States. Cambridge: Cambridge University Press.

Linguists treat language as referring to the social world, perhaps expressing things in the social world, but still maintaining language and the social world as separate structural objects. This has been as true in sociolinguistics as in semantics and pragmatics. While this objectification of language and of the social can be analytically useful, it misses the fact that language is a social practice – that all meaning is constructed in a changing world, in the minute construals that are the center of every semiotic act. And as change is fundamental to social as well as biological life, one could say that at a very elemental level, this process of construal is our bond. As we move from less volatile semantic content ‘out’ to the performative extreme of sociolinguistic variation, meaning construction engages increasingly with the here and now in the speaker’s navigation of the social landscape. At this point, one could say “my style is my bond.”

From their very beginning, pragmatic theories have considered the fulfillment of normative obligations as essential preconditions for the felicitous performance of speech acts (Austin 1962, Searle 1968). More recently, ‘commitments’, in particular with respect to assertions, have moved center stage in pragmatic theorizing (Geurts 2019).  While these approaches adopt a normative perspective focusing on isolated speech acts, I propose to look at how participants in naturally occurring social interaction hold each other accountable for pragmatic obligations and normative expectations that they have created by their verbal actions. In social interaction, two issues concerning normative expectations inevitably arise whenever a turn-at-talk is complete: Is there a response due from co-participants? And if so, what kind of response, i.e., which type of action is normatively expected as a next action, given the type of action that the current turn has implemented?

In Conversation Analysis, but also in similar ways in other approaches as the German Dialoganalyse (Hundsnurscher 2001), these questions have been answered by proposing models of sequence organization. At the heart of these models lie concepts of projection, conditional relevance, and preference (Schegloff 1968, 2007), which account for the ways in which first actions (like questions) make second actions (like answers) expectable and their absence accountable. More recent research, however, has shown that many action-types do not fit this picture. Stivers & Rossano (2010) suggest that specific responses are not necessarily made relevant by specific action-types per se. Instead, responses are mobilized to different degrees by virtue of specific linguistic and nonverbal resources. (Stivers/Rossano 2010).

Building on this research, I will zoom in into a specific kind of practice, namely declaratives of trouble (Kendrick/Drew 2016; Fox/Heinemann 2021). In line with prior research, I will show that the declarative format in many cases rather invites than strictly normatively requires a response (cf. Gubina i.pr.). More specifically, declaratives of trouble are responded to by a wide range of different actions, whose occurrence hinges on pragmatic factors related to the nature of the trouble, speaker’s responsibility for the trouble, speaker’s vs. recipient’s competence and availability for remediating the trouble, etc.

As a conclusion, in many pragmatic contexts, obligation and normative expectation concerning next actions rather seem to be an object of negotiation within a scope for choice and agency of the responder. Instead of strict expectations concerning the occurrence of responses and specific actions types, many actions allow for a wider range of possible responses. For the responder, this means that often it is not important to produce a specific action, but that s/he is able to make their actions accountable as acceptable next moves within a cooperative context.