#trouble talk

Pre-empting reference problems in conversation

By: Jan Svennevig

Published by: Language in Society
Volume 39
Pages 173–202

LL Abstract:

In this article, Svennevig identifies how conversationalists deal with problems of reference as they are producing their turns at talk. The article discusses two ways that speakers change their turns in progress in order to introduce a reference to an interlocutor or to check their understanding of that reference. The author discusses previously-known processes of modifying or expanding a turn in order to accomplish this reference-checking, and proposes a heretofore-undescribed practice where speakers minimize disruption to surrounding talk by embedding referent information before a referring expression has been produced.

LL Summary:

Svennevig begins the article with some data introducing how establishing reference is a joint accomplishment between speaker and hearer. He notes that problems of reference may occur at two levels: at the level of linguistic
form (using an unfamiliar linguistic expression) or at the level of identification of the referent (when a place, a person, or an object may not be uniquely identifiable to the interlocutor). He ends the introduction by arguing that the insertion of an existential clause into an incomplete syntactic unit is a convention for dealing with emergent problems of reference. Next, the author describes previous research on references from the perspective of interactional sociolinguistics, outlining studies on the practices employed to check recognition of a referent that create intersubjectivity between interlocutors. Research from Auer, Schegloff, and others suggests that speakers prefer minimal or single reference forms yet create opportunities for the interlocutor to respond. In the next section, Svennevig provides background on theories of common ground where interactants rely on physical copresence, prior conversations, and community membership as resources for assessing the common ground. The author distinguishes between two kinds of referents signaling whether a linguistic expression is expected to be in the communal lexicon: those that are uniquely identifiable (like pronouns, definite noun phrases, and names) and those that are not uniquely identifiable (indefinite nouns). He uses some data from Norwegian to show how particles mark upcoming referents as not accessible, further outlining how subjects are often identifiable referents that are placed initially in utterances or introduced in existential clauses. Svennevig ends this section by identifying how the most common reference problem in spontaneous conversation is speakers overestimating common ground. He distinguishes between practices of expanding a turn before or after a problematic term has been produced, and identifies three preemptive practices speakers engage in to avoid problems of reference: marking the expression as unfamiliar, inserting background information, or checking what the interlocutor knows about the expression or referent. Next, Svennevig examines post-positioned turn expansions in data from a corpus of interviews in job centers for immigrants and Oslo sociolinguistic interviews. Excerpts from interviews with social workers and immigrants show how the workers suspend activity during explanations of referents in a marked way via a turn expansion. He uses another conversation to identify an “apokoinou construction” in which a syntactic constituent is integrated prosodically and syntactically with two different clauses, one preceding and one following like “they inject this…catalyst/ / catalyst they call it” in the conversation. This construction involves a shift in footing, as in a TV interviewer shifting from addressing a guest to addressing the public audience. The article continues with excerpts where interlocutors might not identify the referent of an expression used, leading to a speaker expanding their turn via apposition and expansion. Various excerpts show how speakers use intonation and parenthetical insertions to provide information relevant to identifying referents if a mismatch in encyclopedic knowledge was suspected. Next, Svennevig analyzes excerpts where speakers check if interlocutors know a referent or referring expression (such as by inserting a second metacommunicative question after a first question or by gaze before entering an expansion). The final section of the analysis deals with a new practice wherein speakers expand a turn in progress by inserting a clausal construction before a potentially problematic referring expression, paving the way for an explanation of meaning. Multiple examples from excerpts show how this strategy allows for delayed self-repairs where beginnings are not recycled but rather signal referent identification. These examples demonstrate how speakers can simultaneously perform conversational actions like answering questions while checking for understanding within expanded turns. The article then concludes with a discussion of these preemptive techniques used as part of recipient design in conversation as they provide answers to questions in advance. 

LL Recipe Comparison:

This article reminds me of the recipe for Prosciutto-Chicken Pasta:

Much as this article discusses how people preempt trouble in conversation about references, this recipe preempts any demands you might have for delicious flavor and blended tastes! Svennevig identifies how people insert existential clauses to check if others understand their references, and this recipe’s use of broccoli, chicken, and linguine will check all your savory boxes as you check out this dish. Good Cooking!

MWV 4/2/18