130 110
English Pages 383 [393] Year 2023
Studies in Language and Social Interaction
Responding to Polar Questions across Languages and Contexts Edited by Galina B. Bolden, John Heritage and Marja-Leena Sorjonen
35
John Benjamins Publishing Company
Responding to Polar Questions across Languages and Contexts
Studies in Language and Social Interaction (SLSI) issn 1879-3983
Studies in Language and Social Interaction is a series which continues the tradition of Studies in Discourse and Grammar, but with a new focus. It aims to provide a forum for research on grammar, understood broadly, in its natural home environment, spoken interaction. The assumption underlying the series is that the study of language as it is actually used in social interaction provides the foundation for understanding how the patterns and regularities we think of as grammar emerge from everyday communicative needs. The editors welcome language-related research from a range of different methodological traditions, including conversation analysis, interactional linguistics, and discourse-functional linguistics. For an overview of all books published in this series, please see benjamins.com/catalog/slsi
Editors Sandra A. Thompson
University of California, Santa Barbara, USA
Elizabeth Couper-Kuhlen University of Helsinki, Finland
Editorial Board Peter Auer
Barbara A. Fox
Galina Bolden
Makoto Hayashi
Arnulf Deppermann
Marja-Liisa Helasvuo
Paul Drew
K.K. Luke
University of Freiburg, Germany Rutgers University, USA Institut für Deutsche Sprache, Germany University of York, UK
University of Colorado, USA Nagoya University, Japan University of Turku, Finland Nanyang Technological University, Singapore
Volume 35 Responding to Polar Questions across Languages and Contexts Edited by Galina B. Bolden, John Heritage and Marja-Leena Sorjonen
Responding to Polar Questions across Languages and Contexts Edited by
Galina B. Bolden Rutgers University
John Heritage UCLA
Marja-Leena Sorjonen University of Helsinki
John Benjamins Publishing Company Amsterdam / Philadelphia
∞
TM
The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences – Permanence of Paper for Printed Library Materials, ansi z39.48-1984.
doi 10.1075/slsi.35 Cataloging-in-Publication Data available from Library of Congress: lccn 2023035332 (print) / 2023035333 (e-book) isbn 978 90 272 1432 4 (Hb) isbn 978 90 272 4929 6 (e-book)
© 2023 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Company · https://benjamins.com
Table of contents Acknowledgments chapter 1. Introduction: Polar questions and their responses Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
vii 1
chapter 2. Repetitional responses to polar questions in Russian conversation Galina B. Bolden
40
chapter 3. Responding to polar questions in Brazilian Portuguese: É-responses and repeats Katariina Harjunpää & Ana Cristina Ostermann
76
chapter 4. Responses to polar questions in Polish Matylda Weidner
109
chapter 5. Three practices for confirming inferences in French talk-ininteraction: Repetition, voilà, and exact(ement) Rasmus Persson
139
chapter 6. Complexities of responding: Confirming responses to pseudo-tag questions in Korean conversation Seung-Hee Lee
179
chapter 7. The division of labor between the particles jah and jaa ‘yes’ as responses to requests for confirmation in Estonian Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
210
chapter 8. Code-switching, agency, and the answer possibility space of Spanish-English bilinguals Chase Wesley Raymond
239
chapter 9. Post-confirmation modifications: Trajectories of un-initiated responses to polar questions in Japanese Kaoru Hayano & Makoto Hayashi
272
vi
Responding to Polar Questions across Languages and Contexts
chapter 10. Responding to polar questions without a polarity item ‘yes’ or ‘no’ in Finnish Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen chapter 11. Renewing a social action in US primary care: One sequential context when actions formatted as polar questions do not require polar answers Jeffrey D. Robinson & John Heritage
301
328
chapter 12. Do English affirmative polar interrogatives with any favor negative responses? Elizabeth Couper-Kuhlen, Sandra A. Thompson & Barbara A. Fox
350
appendix. Transcription conventions and symbols for glossing
377
Subject index
381
Acknowledgments This book originated in a panel on responses to polar questions at the International Conference of Conversation Analysis at Loughborough University, UK in summer 2018. We thank all authors for their enthusiasm on the topic of the book, their perseverance with the editorial process, and their stellar contributions. The series editors Sandra Thompson and Elizabeth Couper-Kuhlen have carefully read and evaluated the contents of the volume, and we thank them for their insight and support. We are very grateful to Aleksandr Shirokov for his highly efficient, reliable, and prompt help in formatting the chapters and creating the appendices. Isja Conen at Benjamins has been infinitely patient throughout the process, and we are grateful for her steady hand and expert navigation throughout. We would also like to thank Susan Hendriks, who walked us through the process of transforming the documents on our computers into a printed book. New Brunswick, Los Angeles, and Helsinki, August 2023
chapter 1
Introduction Polar questions and their responses Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
Rutgers University | University of California Los Angeles | University of Helsinki
Question–answer sequences are arguably among the most basic building blocks for sequences of action in interaction and are ubiquitous among the languages of the world. This chapter reviews and synthesizes current interactional research on polar questions and answers across languages in order to contextualize the work assembled in this volume. We introduce grammatical and pragmatic resources for forming and recognizing polar questions, discuss principles of polar question design, and overview different types of polar responses. We conclude the chapter with a preview of the included studies, a discussion of their contributions, and suggestions for future research directions. Keywords: affirming, confirming, epistemic stance, epistemic gradient, polar answers, polar questions, preference, question-answer sequences, question design, repetitional response, response design, response particle
This book is about one of the most fundamental action sequences found across human societies and socio-cultural contexts: polar questions and their responses. Question–answer sequences are arguably among the most basic building blocks for sequences of action in interaction and are very nearly, if not actually, ubiquitous among the languages of the world (Dryer 2013a; Sadock and Zwicky 1985). They form the interactional backbone of many forms of task-focused workplace interaction, including the courts, education, medical practice, and interviewing of all kinds (Heritage and Clayman 2010; G. Raymond 2003), and they serve as vehicles for many types of other social actions (Schegloff 2007; Steensig and Drew 2008). Empirical research suggests that polar questions generally are the most common types of questions, occurring with greater frequency than question word questions (Enfield, Stivers, and Levinson 2010). In the last thirty years, polar questions have received increasing attention, particularly in conversation analytic and https://doi.org/10.1075/slsi.35.01bol © 2023 John Benjamins Publishing Company
2
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
interactional linguistic research (e.g., De Ruiter 2012; Freed and Ehrlich 2010; Steensig and Drew 2008). Different forms of response to polar questions have similarly attracted increasing interest, and it is these that are the focus of the present volume. The goal of this Introduction is to review and synthesize current interactional research on questions and answers across languages in order to contextualize the work assembled in this volume. We conclude the chapter with a preview of the included studies and a discussion of their contributions.
1.
What is a polar question?
From the perspective taken in this volume, that of conversation analysis (Sidnell and Stivers 2013) and interactional linguistics (Couper-Kuhlen and Selting 2018), questions are initiating actions that request the recipient to provide information concerning something which the recipient is imputed to have knowledge about (Couper-Kuhlen and Selting 2018: 217–218). A central aspect in this process is that in the act of questioning, a speaker adopts a stance that the recipient will have, and will be willing to supply, the relevant information (Searle 1969: 64–71). While question word questions (such as, what time is it? or how much does it cost?) request particular information without anticipating a specific response, polar questions solicit information by expressing a proposition that describes a possible state of affairs, and invite the recipient to affirm or deny it. In this way, polar questions make relevant a response that, at the very least, either affirms (or confirms) or disaffirms (or disconfirms) the proposition. Different practices can be used to accomplish these responding actions, including response particles, such as yes or no in English, or ja or nej in Swedish, and repetitions of a part of the question, such as its verb as, for instance, in Estonian and Finnish (see Response Design section for a detailed discussion). Because questioners can take up different (epistemic) stances towards the proposition in question (see below), it is useful to distinguish between polar questions that request (dis)affirmation and polar questions that request (dis)confirmation. Thus, on the one hand, questioners can launch polar questions from a completely unknowing position (K- ‘knowledge minus’) (Heritage 2012a, 2012b). Following Sorjonen (2001a: 36), we will refer to responses to such questions as affirming/disaffirming the question’s proposition. On the other hand, when questions are launched from a somewhat knowing position (K+ ‘knowledge plus’), we will refer to their responses as confirming/disconfirming. Confirming responses can be produced via dedicated turn shapes (e.g. that’s right in English, niin in Finnish, genau in German etc.), but confirmation can also be understood on a
Chapter 1. Introduction
sequential basis without a dedicated confirmation token (for example, when a questioner repeats something said by a previous speaker, A: I’m leaving at ten, B: Ten?, A: Yes.) Another important terminological distinction is that between questions and interrogatives. In conversation analytic (CA) and interactional linguistics (IL) scholarship, questions as an action category are distinguished from interrogatives as a linguistic category. As we discuss below, requests for information – i.e., questions – can be implemented using both declarative or interrogative morphosyntax. Further, interrogative constructions may be deployed in the service of a variety of actions that do not enact questioning as their primary action. These include requests for assistance, offers, invitations etc.
2.
The use of polar and content questions in conversation
How common are different kinds of questions and what do they do in interaction? In addressing these questions, a valuable resource is a quantitative study of questioning in social interactions in ten languages representing a variety of different language groups (Enfield, Stivers, and Levinson 2010). Drawing on a formal CA-based coding framework, this study showed that, in many languages, polar questions form the majority of the questions asked1 (see Table 1). In fact, polar questions outnumber content questions in seven out of the ten studied languages by a margin of 2:1 or greater. Among the languages examined in this study, ǂĀkhoe Hai||om, with a predominance of content questions, appears to be the only exception to this trend. The study also showed that polar and content questions are clearly differentiated in terms of function. As Table 2 illustrates, while a majority of content questions were devoted to requesting information, in nine of the ten languages only a minority of polar questions were directed to this end. Correspondingly, while large proportions of polar questions were directed to seeking confirmation, almost no content questions were used for this purpose.
1. The study coded a turn construction unit (or a TCU, Sacks, Schegloff, and Jefferson 1974) as a question if it was a formal question (having interrogative markings) or a functional question, i.e. a TCU designed to elicit information or confirmation, primarily on pragmatic grounds (see Sections 3 and 4 below for a full discussion of these issues). These questions were found to implement the following action categories: requests for information, requests for confirmation, other-initiation of repair (Schegloff, Jefferson, and Sacks 1977), assessments (seeking agreement), suggestions/offers/requests (seeking acceptance), rhetorical questions, outlouds (not seeking a response, e.g. where are my keys), and others (Stivers and Enfield 2010: 2623).
3
4
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
Table 1. Distribution of polar, content and alternative questions in ten languages * Polar questions
Content questions
Alternative questions
41.5%
58.5%
0%
Heinemann (2010)
75%
25%
< 1%
Dutch
Englert (2010)
73%
20.5%
6.5%
English (US)
Stivers (2010)
70%
27%
3%
Italian
Rossano (2010)
59%
39%
2%
Japanese
Hayashi (2010)
85%
15%
< 1%
Korean
Yoon (2010)
70%
29%
1%
Lao
Enfield (2010)
72%
27%
< 1%
Tzeltal (Mexico)
Brown (2010)
74%
24%
2%
Yélî Dnye (Papua New Guinea)
Levinson (2010)
61%
39%
–
Language
Author
ǂĀkhoe Hai||om (Namibia)
Hoymann (2010)
Danish
* Tables 1–3 are compiled from the cited studies, which were included in the special issue edited by Enfield, Stivers and Levinson (2010).
Table 2. Distribution of primary functions of polar and content questions in ten languages Language
Author
Information seeking
Confirmation seeking
Repair initiation
Polar Qs
Content Qs
Polar Qs
Content Qs
Polar Qs
Content Qs
ǂĀkhoe Hai||om (Namibia)
Hoymann (2010)
62%
56%
2%
0%
24%
38%
Danish *
Heinemann (2010)
15%
63%
40%
1%
18%
25%
Dutch
Englert (2010)
7%
77%
51%
0%
21%
23%
English (US)
Stivers (2010)
27%
79%
30%
0%
36%
20%
Italian
Rossano (2010)
26%
75%
47%
0%
17%
29%
Chapter 1. Introduction
Table 2. (continued) Language
Author
Information seeking
Confirmation seeking
Repair initiation
Polar Qs
Content Qs
Polar Qs
Content Qs
Polar Qs
Content Qs
Japanese
Hayashi (2010)
19%
72%
58%
0%
4%
26%
Korean
Yoon (2010)
20%
58%
41%
0%
30%
40%
Lao
Enfield (2010)
33%
52%
58%
1%
2%
40%
Tzeltal (Mexico)
Brown (2010)
26%
75%
47%
0%
17%
21%
Yélî Dnye (Papua New Guinea)
Levinson (2010)
48%
59%
44%
0%
7%
38%
* The Danish figures presented here incorporate a correction to a typographic error in the original.
3.
Grammatical resources for forming polar questions across languages
The structural properties of questions have been investigated across languages in linguistic typology. Much of this research is based on descriptive materials, such as reference grammars (see Dryer 2013a, 2013b; König and Siemund 2007; Sadock and Zwicky 1985), and is available in the World Atlas of Language Structures (WALS; online https://wals.info).2 Dryer (2013a, 2013b) reports from an examination of 955 languages in WALS, that the three major methods for forming a polar question comprise question particles (in 63% of the languages), interrogative morphology (19%) and interrogative intonation (17%). Question particles (N = 600) are commonly placed at the end of sentences (59%), while 24% occur in initial position of a sentence. Morphological markings are typically affixed to the verb (Dryer 2013b). Languages that use particles (independent or clitics) and morphological markings for forming polar questions include, for example, Japanese (Hayashi 2010; Hayano 2013), Korean (Yoon 2010), Finnish (Sorjonen 2001a, 2001b), Lao (Enfield 2010), and Indonesian (Hamdani and Barnes 2018). Prosody is also commonly deployed in forming polar questions. According to Dryer (2013a), 17% (173/995) of the languages surveyed rely on intonation to
2. Initially Dryer’s surveys came out in 2005.
5
6
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
form polar questions, though Dryer also observes that such languages are underrepresented in the WALS survey. In addition, many but not all languages (Enfield 2010; Levinson 2010) deploy prosody alongside other grammatical resources in the formation of polar questions. A variety of prosodic contours, not exclusively a TCU-final rise, may be associated with polar questions. For instance, in Russian (Bolden 2016), some varieties of Italian (Rossano 2010), and a number of other languages, a rise-fall pitch accent is used as interrogative prosody. Some studies on the prosody of polar questions have questioned the relevance and import of ‘final rise’ prosody in question formation, and the multifunctionality of final rising prosody is widely recognized (Couper-Kuhlen 2012; Geluykens 1988; Levinson 2012; Ogden and Routarinne 2005; Walker 2014). A very infrequent method of forming polar questions involves the use of word order that differs from that of declarative clauses, for example subjectauxiliary inversion (in 13/955 languages or 1.4%). Interrogative word order is mainly found in European languages, including Germanic languages, for example Danish (Heinemann 2010), Dutch (Englert 2010) and English (Stivers 2010), as well as Romance languages such as Spanish (C. W. Raymond 2015) and Slavic languages, such as Czech, together with other languages identified by Dryer. Questioning tags are a further common resource for forming polar questions cross-linguistically (König and Siemund 2007: 296–297).3 Languages may have a rich repertoire of tags, both positive and negative, lexical (e.g. right? in English) and sentential (isn’t it?), deployed in a final or medial position of a turn constructional unit (TCU). For example, in Italian, lexical tags are the only available morphosyntactic resource for forming polar questions (Rossano 2010). Tags are very common in many languages (see Table 3), such as Danish (Heinemann 2010), English (Stivers 2010), Japanese (Hayashi 2010), Korean (Yoon 2010), and Russian (Bolden 2016), but there are languages where their use is not frequent, for example Finnish.
4.
Epistemic asymmetry as a basis for recognizing polar questions
While interrogative polar questions index a putatively unknowing (K-, K minus) stance towards the propositional content of the turn, this stance can also be attributed to the speaker of a declarative turn who will thereby be understood as seeking information (Heritage 2012b; Kamio 1997; Labov 1972: 254; Labov and Fanshel 1977: 100–101; Pomerantz 1980; Weber 1993). Heritage (2012a, 2012b) observes that the recognition of a declarative utterance (such as arrow 3 in Excerpt 3. Dryer’s (2013a) survey does not include tags.
Chapter 1. Introduction
1 below) as requesting information can be understood in terms of the epistemic statuses of the participants, where epistemic status is “an inherently relative and relational concept concerning the relative access to some domain of two (or more) persons at some point in time” (Heritage 2012b: 4). These statuses may be relatively enduring (e.g., knowledge, experience or expertise ‘owned’ by the participants) or quite temporary (one participant has recent information on the matter). This type of distinction also appears to be grammaticized as egophoric marking in some of the world’s languages (see for example, Floyd, Norcliffe, and San Roque 2018; Sandman and Grzech 2022).4 Thus different question designs can be arrayed on a cline that shows that regardless of grammatical differences, the epistemic asymmetry they instantiate makes them understood as questions: Excerpt 1. (DEC 1205) 1 2 3 4 5 6 7 8 9
Doc:
Pat: Doc: Pat: Doc: Pat: Doc:
1-> Right. Did you have any problems with your ba:by? (.) No: 2-> The delivery was alright, was it? Ri:ght, 3-> Didn’t have any operations, ˚No,˚ Ri:ght, it doesn’t sound anything toom- too serious
All three arrowed utterances index an underlying assumption that the recipient has primary access to the information in play. Due to this epistemic asymmetry, all these turns are interpreted as implementing a request for information or confirmation. However, the grammatical choices of the three sentences embody distinct stances with regards to the speaker’s certainty or uncertainty about the proposition (see Section 5.1 below for a discussion of different epistemic stances). Most, if not all, languages appear to allow their speakers to present questions with a plain declarative (see Table 3), and there are languages (for example, Chalcatongo Mixtec, Dryer 2013a; and Yélî Dnye, Levinson 2010) that have been found to have no linguistic resources (either morphosyntactic or prosodic) to form polar questions. Speakers of these languages must therefore rely on the pragmatics of epistemic asymmetry to be understood as requesting information. A study of question frequencies in conversations in ten languages from a variety of different language groups (Enfield, Stivers, and Levinson 2010) has shown that in six of the nine languages for which data were available, declarative turn designs are the most frequent form of polar question, although the extent of their disparity varies widely (Table 3).
4. Central areas of egophoric languages are Himalayas, Andes, Caucasus and Papua-New Guinea (see for example, Sandman 2016: Chapter 7).
7
8
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
Table 3. Distribution of formats of polar questions in conversation in ten languages Declarative
Interrogative *
Tag question
41.5%
58.5%
0%
Heinemann (2010)
23%
61%
16%
Dutch
Englert (2010)
38%
34%
28%
English (US)
Stivers (2010)
63%
31%
6%
Italian
Rossano (2010)
–
–
–
Japanese
Hayashi (2010)
39%
31%
30%
Korean
Yoon (2010)
74%
11%
15%
Lao
Enfield (2010)
17%
83%
< 1%
Tzeltal (Mexico)
Brown (2010)
74%
24%
2%
Yélî Dnye (Papua New Guinea)
Levinson (2010)
65%
–
35%
Language
Author
ǂĀkhoe Hai||om (Namibia)
Hoymann (2010)
Danish
* Interrogative includes both lexico-morpho-syntactic and prosodic interrogative marking here.
5.
Principles of polar question design
Speakers of any language have (linguistically constrained) options in how they design polar questions: for example, what exactly is being proffered for affirmation or confirmation, how the propositional content is formulated (e.g., positively or negatively), how the interrogativity is conveyed, etc. Here we introduce two central aspects of question design: how questions convey epistemic stances and set up preferences for particular responses.
5.1 Polar questions convey epistemic stances When speakers pose polar questions, they will ordinarily convey a view concerning the likelihood or probability of some state of affairs and the degree of confidence or certainty with which they hold this view. We take up each of these epistemic orientations in turn. 5.1.1 What is the likely state of affairs Prior research has shown that polar questions are typically designed to invite affirmation of the state of affairs expressed in the question’s proposition (Bolinger 1978; Heritage and C. W. Raymond 2021; Robinson 2020a; Sacks 1987), and do so by presenting a likely state of affairs (Heritage and C. W. Raymond 2021; Robinson
Chapter 1. Introduction
2020a; Pomerantz 1988; Sacks 1987). Likelihood here can be understood as anything from a relevant possibility that something may be the case (e.g., a doctor’s question Have you had a fever to a patient with a cold) to a strong probability (e.g., Are you going out now to a partner who is putting on an overcoat). In general, therefore, grammatically positive questions invite affirming or confirming responses to a state of affairs that is presented as somewhat likely. The design of questions to address issues of probability can take on considerable significance. In the following case, a community nurse (HV ) interviewing the new parents of a first child has learned that the birth process was difficult because the baby became stuck in the birth canal. The interview has reached the point where the nurse is mandated by an official form to ask about the delivery. The parents’ earlier description raises the possibility that forceps might have been used, but the question is difficult to frame because the parents did not explicitly mention them: Excerpt 2. (1A1:14) 1 2 3 4 5 6
HV:
=So you had a- uh: (1.0) -> You didn’t- Did you- You didn’t have forceps you had a:= M: =Oh [no:: nothing. F: [( ) HV: An- and did she cry straight awa:y.
The nurse’s first question at line 1 was probably heading towards ‘normal delivery’ as its final noun phrase. It is abandoned and replaced with a negatively framed declarative question, that may have been the beginning of the final version of the question You didn’t have forceps. However, this frame is abandoned and replaced with a positively framed interrogative, that was probably headed to Did you have forceps. This too is abandoned, and the nurse returns to her earlier formulation You didn’t have forceps you had a:. Evidently, from the nurse’s point of view it is very important to arrive at an appropriate framing of this question. The dilemma she faces is this: In the event that forceps were used, a question like You didn’t have forceps is incorrect and might be perceived as over-supposing and perhaps playing down the mother’s experience. On the other hand, the question Did you have forceps would, in the event that none were used, also be incorrect, and could also be perceived as over-supposing, but in the direction of catastrophizing the birth experience. Here, in a ‘high stakes’ situation, the nurse’s overt vacillation on the design of the question evidences her stance as to its sensitivity and the importance of getting it right. Grammatically negative questions (as in, for example, he didn’t phone you) present a state of affairs as somewhat unlikely and, in English, are typically implemented in a declarative form (Heritage and C. W. Raymond 2021). This question format frequently indexes a departure from an expected or desirable state of
9
10
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
affairs and may be avoided for that reason (C. W. Raymond and Heritage 2021). Alternatively, the negative framing of questions can be managed through the use of negative polarity items in otherwise affirmatively framed questions (e.g. did he ever phone you) (see also Couper-Kuhlen, Thompson, and Fox, this volume). 5.1.2 Epistemic (un)certainty A second and a distinctive aspect of the stance displayed in a question concerns the questioner’s stance of certainty about the likelihood of the state of affairs proposed in the question. Interrogatives, tag questions, and declarative questions index different degrees of certainty; through these different forms, the questioner expresses distinct claims of knowledge, varying from an unknowing stance (interrogative format) to a relatively close degree of parity with the answerer (declarative format).5 The following case (shown earlier as Excerpt 1) illustrates this variation. Here a doctor questions a recently delivered mother who is complaining about unexplained back pain. The parties have not met before as the doctor is a locum.6 His first question, framed interrogatively, entertains the possibility that there have been problems with the baby. Its negative tilt (implemented through the determiner any) may reflect the fact that the mother has, to this point, not mentioned any such difficulties (Heritage and C. W. Raymond 2021). The question does not present its producer as strongly committed to the likelihood of a confirming response: Excerpt 3. (DEC 1205) 1 2 3 4 5 6 7 8 9
Doc:
Pat: Doc: Pat: Doc: Pat: Doc:
1-> Right. Did you have any problems with your ba:by? (.) No: 2-> The delivery was alright, was it? Ri:ght, 3-> Didn’t have any operations, ˚No,˚ Ri:ght, it doesn’t sound anything toom- too serious
Following the mother’s negative response to his first question, the doctor’s next question (line 5) explicitly inquires into the delivery using a [statement] + [tag] format. This format more strongly commits to the likelihood that the birth was unproblematic, and its design was likely selected in the light of the mother’s previous negative response. The doctor’s final question is framed as a minimized
5. See also Drake (2015) for the turn-final or in English and A. Lindström (1997) for the turnfinal eller in Swedish used for conveying the questioner’s reduced certainty. 6. A temporary doctor at the clinic, often replacing a doctor who is on vacation.
Chapter 1. Introduction
declarative question and even more strongly commits to the likelihood that the birth process was unproblematic (C. W. Raymond et al. 2021). Here the three questions ascend a cline of increasing certainty that the birth process was unproblematic and, by implication, that the causes of the mother’s back pain will have to be found elsewhere. These two aspects of epistemic stance – the likely state of affairs and the speaker’s (un)certainty about it – frequently function in combination. For example, a declaratively formed question, by indexing the questioner’s greater level of certainty, will also tend to reinforce the questioner’s stance that a state of affairs is likely. This is readily seen in Stivers’ (2010: 2776) observation that declarative polar questions in her American English data are overwhelmingly used to request confirmation or initiate repair. By contrast, only 4% (n = 5/136) of declarative polar questions were information requests. Polar information requests, on the other hand, were implemented with interrogatives 85% of the time (n = 58/68). Similar findings are reported for Dutch (Englert 2010), Japanese (Hayashi 2010), and Lao (Enfield 2010). As these findings confirm, interrogative formatting expresses the questioner’s stance of uncertainty about the state of affairs under question, and this formatting therefore aligns with the function of requesting information, rather than seeking confirmation.
5.2 Polar questions convey preferences In general, polar questions establish a context in which the recipient will affirm/ confirm the question’s proposition or disaffirm/disconfirm it. These two classes of responsive actions are performed in distinctive ways. In Conversation Analysis, the term preference (Pomerantz 1984; Sacks 1987; Schegloff 2007) is used to describe “a structural relationship of asymmetry between (typically) two alternatives, whereby one is said to be ‘preferred’, or privileged, over the other, the latter being consequently ‘dispreferred’” (Couper-Kuhlen and Selting 2018: 19, on-line chapter B). Thus, despite its connotations, the term preference does not refer to the psychological characteristics (or desires) of speakers, but rather to structural differences in how affirming and disaffirming responses are produced. Polar questions convey preferences in that affirmations and confirmations are typically produced promptly and straightforwardly, while disaffirmations and disconfirmation are often delayed, mitigated, and prefaced with a variety of turn-initial objects (Kendrick and Torreira 2014; Roberts, Margutti, and Takano 2011) and, where disagreement is at issue, may be accompanied by accounts or explanations (Heritage 1988; Pomerantz 1984). For example, in (4), in a conversation between two near neighbors in an apartment building, the question on line 1 references a topic they have previously
11
12
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
discussed (indexed by the particle ta; Bolden 2008, 2009b): chicken that both of them knew to be on sale. The question is checking whether Rachel bought the chicken, which paves the way for a question about its quality (line 3). Rachel’s affirming response (line 2) has the typical features of a preferred response: it is immediate, brief, and unmitigated: Excerpt 4. (Russian; GM3 1:00) 1
MUS:
=mut< Matti käy siel NEG-1SG be-STEM but [nameM] visit.3SG DEM.LOC {I} haven’t but Matti goes there työmatka-lla #koko ajan#. work_trip_ADE whole time on work trips all the time.
A’s disaffirming answer (line 4) is in the form of the (appropriately inflected) negative verb (en oo ‘{I} haven’t;’). In contrast, in response to requests for confirmation, negative particles are typically used. For instance, in (8), while holding up a photograph, Päivi suggests to Matti that the woman in the photo was his (Matti’s) godmother, using a B-event statement (Labov and Fanshel 1977: 100–101) that conveys a relatively knowledgeable stance (line 1). Excerpt 8. (adapted from Couper-Kuhlen and Selting 2018: 246–247) 1
Päivi:
↑tää ol-i su-n DEM1 be-PST-3SG you.SG-GEN This was your godmother
kummi#täti# godmother
((2 lines of concurrent talk omitted)) 2
Päivi:
3
Matti:
[nii-hän ] se ol-i. ((Matilta)) PRT-CLI DEM3 be-PST.SG3 ((asks Matti)) That’s how it was >ei-ku< Juka-n. NEG-CLI [1nameM]-GEN No, Jukka’s
Chapter 1. Introduction
((6 lines omitted)) 4
Päivi:
5
6 -> Matti:
[↑mä] luul-i-n et tää o-n I think-PST-1SG COMP/PRT DEM1 be-3SG I thought that this is su-n kum°mitäti° you.SG-GEN godmother your godmother ↑ei::::, NEG
No
In line 3, Matti disconfirms the proposal with a corrective negative marker eiku (Haakana and Kurhila 2009; Haakana and Visapää 2014), and subsequently (in line 6) with a mere negative particle ei (Couper-Kuhlen and Selting 2018: 205–207). Research on a variety of languages shows that polar questions may be answered differently depending on the action implemented by the question, which highlights the importance of sequential context. Particles are much more frequent in responses to questions that implement ‘non-primary’ actions, a term Keevallik (2010) uses to refer to questions initiating pre-sequences, repairs, requests for confirmation, and the like. In contrast (as we discuss below in Section 6.2), repetitional responses tend to respond to questions implementing ‘primary’ actions, such as base first pair parts (Enfield et al. 2019; Keevallik 2010). Observations of this kind have been made, for example, for Brazilian Portuguese (Harjunpää & Ostermann, this volume), English (Heritage and G. Raymond 2012; Stivers 2010), Estonian (Keevallik 2010), Finnish (Raevaara 1993; Sorjonen 2001a, 2001b), Korean (Lee 2015), Polish (Weidner, this volume), and Russian (Bolden 2016). Particle responses have also been noted for ‘low stakes’ questionanswer sequences, such as bureaucratic form filling (Heritage and Sorjonen 1994), prompting Heritage and Raymond (2012: 184) to observe that “little questions get little answers.” A collection of studies of ten languages discussed earlier (Enfield, Stivers, and Levinson 2010) showed that questions implementing ‘non-primary’ actions (requests for confirmation and initiations of repair) are much more common than information seeking ones, the latter making up only 28% (range 14%–62%) across the ten languages in aggregate (see Table 2 above). Among polar questions, the comparative frequency of these ‘non-primary’ questions relative to the much more infrequent information-seeking ‘primary’ questions likely contributes to the relative frequency of particle responses across most corpora. Together, these findings raise nuanced questions about the status of particle (vs. other kinds of ) responses to polar questions, both cross-linguistically and across sequence types. These questions are at the forefront of several contributions to
19
20
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
this volume (Bolden, this volume; Harjunpää & Ostermann, this volume; Weidner, this volume).
6.2 Repetitional responses Repetitions are perhaps the most frequently studied alternative polar response type after particle responses. This research differentiates between different kinds of repetitions, particularly on the basis of how much material is repeated (full vs. different types of partial repeats). In his paper Confirming Allusions, Schegloff (1996a) observed that full repeats of preceding questions could function as confirmations in contexts where the repeated elements had already been conveyed allusively. For instance, in (9), from a radio interview, Susan Shreve, a well-known children’s author, explained how she came to write children’s books: Excerpt 9. (Schegloff 1996a: 183) 1 2 3 4 5 6 7 8
Shr:
Shr: IR: Shr: IR: Shr:
.hh I started writing: (.) juvenile books fer entirely pra:ctical reasons, .hh (.) [u- u[Making money::. Making [money [Yes ((laughter)) that- that practical reason hhh
Shreve’s allusive reference to practical reasons for writing children’s books (line 2) is explicated by the interviewer (IR) with a candidate understanding making money (line 5), and this is confirmed by the author with the repeat of making money at line 6 and an elaboration that practical reason at line 8. Schegloff observed that, in this context, the repeat functions as a confirmation, and as a means through which the repeating speaker (re-)secures authorship of what was previously stated allusively. He also noted that the “matter of who is agreeing with whom may seem petty, but parties to conversation do care about it, and it can be interactionally quite consequential” (Schegloff 1996a: 82–83). Schegloff ’s analysis was confined to full repeats of the prior turn (itself an explication of a previous allusion) as the whole content of the repeating turn or as its initial part (see also Sorjonen 2001a: 76–80 on Finnish). In a subsequent paper Heritage and G. Raymond (2005) examined full repeats that did not address a previous allusive remark, as in the following case where a mother and daughter are talking about the daughter of a third party (Ann). The mother is explaining that Ann’s daughter wants to follow a new fashion for multiple earrings in a single ear, provoking a disagreement with her mother:
Chapter 1. Introduction
Excerpt 10. (Field 1.1:8) 1 2 3 4 5 6 7 8 9
Mum:
Les: Mum:
.hh We:ll. (0.2) Uh you know (.) there’s a cra:ze with the girls now to have (.) a secon:d. (1.1) ring ih- a secon:d uh earring in on[e ear. [Oh: it’s very chea:p isn’it. It’s very cheap yes’n this is u- this is what Ann said. An’ Ann said (0.3) sheshe’ll haf (.) tuh have another. (0.5) .hh (.) hole in’er ear…((continues story))
Commenting on the daughter’s stance, Lesley observes that the fashion is very cheap (line 5), producing the observation in the form of a [statement] + [tag]. Mum’s response to that takes the form of a confirming repeat of Lesley’s turn followed by an agreement (Yes) (line 6). She subsequently underscores her confirmation by reference to the child’s mother’s opinion (this is what Ann said). Here Mum privileges confirmation over agreement, in the process leveraging her access to Ann (a party to the dispute) as an element of epistemic primacy in the opinion. On this analysis, and even without the allusive context (as in 10), full repetitional responses can be used to provide a more committed or agentive response than a particle response would convey. In some languages, repetitional responses may be used in cases where the recipient of a question is claiming a greater relative epistemic access to some state of affairs than the question had conveyed (e.g., Heritage and G. Raymond 2005; Bolden, this volume; Weidner, this volume). For example, declarative questions, especially, may claim an access to a state of affairs in the question recipient’s domain of knowledge that can prompt the recipient to respond with a confirmatory repeat. In the context of polar requests for action (e.g., via polar interrogatives) where deontic rights and obligations are in play, repetitional responses may be understood as expressing a stronger degree of commitment to fulfill the request (Bolden, this volume; Houtkoop-Steenstra 1987; A. Lindström 2017). In both of these action environments, repetitional responses are understood as having greater assertive force than particle responses. The current theoretical understanding of the distinction between particle and repetitional responses rests on the following notion, perhaps most fully expressed in Enfield et al. (2019). Considered at a purely propositional level, a particle response affirms the questioner’s proposition without its speaker voicing that proposition in so many words. In other words, a particle response is indexically dependent on the question to which it responds and cannot be understood independently of it. With repetitional responses, by contrast, speakers repeat at least some part of a question’s propositional content, and to that extent both affirm and assert the proposition as their own (cf. Goffman 1979). Depending on the extent of the repetition, the question’s proposition may be understood independently of the question to which it responds. Thus, in English, in a case such as: Q: Is John
21
22
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
coming? R: John’s coming, the response can in principle ‘stand-alone’ as an assertion of a state of affairs, and as embodying its speaker’s independent commitment (whether epistemic, deontic, etc.) to its representation of that state of affairs. In examining repetitional responses, it is crucial to distinguish between different types of repeats: full clause repeats, verb repeats, including modal and auxiliary verbs, verb repeats with some other parts of the question (e.g., the object), and repeats of other sentence constituents. Much research on repetitional responses has focused on partial repeats and especially repeats of the verb in the preceding question. Studies of these kinds of repetitional responses have been made for Brazilian Portuguese (Harjunpää and Ostermann, this volume), English (Heritage and G. Raymond 2005, 2012; Stivers 2022: 122–146), Estonian (Keevallik 2010), Finnish (Laury 2018; Raevaara 1993; Sorjonen 2001a, 2001b, 2018), Greek (Alvanoudi 2022), Japanese (Hayano 2013; Hayashi 2010), Korean (Lee 2017), Polish (Weidner, this volume), and Russian (Bolden 2009a, 2016, this volume). A key finding arising from this line of research is that the use of verb repeat responses is shaped by the question’s epistemic stance and its sequential and action context. For instance, Sorjonen (2001a: 37–44; 2001b; see also Raevaara 1993) found that in Finnish a simple verb repeat is used as an affirmative answer to a particular kind of interrogative construction (with the interrogative particle attached to the finite verb to indicate that the entire proposition is the focus of the question, termed V-interrogatives). The verb repeat claims to offer new information, creating shared knowledge, and suggests that the information is part of an in-progress activity. For instance, in (11), Simo asks Pekka about his recent trip to Finnish and Swedish Laplad (lines 1–3), receiving a verb repeat as an affirmative answer (line 4). Excerpt 11. (adapted from Sorjonen 2001a: 43–44) 01
S:
No
↑mite-s (.) ol-i-ko ne hurja-t (.) mite-s (.) how-CLI be-PST-Q DEM3.PL wild-PL how-CLI Well ↑how (.) were they wild (.) how (.) PRT
02
vaikutt-i-ko ne Norja-n lumi-myrsky-t sinne affect-PST-Q DEM3.PL Norway-GEN snow-storm-PL there.to Did the snow storms in Norway affect there (.)
03
(.) tei-lle. you.PL-ALL you. ((PL))
04
P:
05
S:
.mt Vai:kutt-i. [°heh .heh heh .heh° ] affect-PST [ [Et sie-ll ol-i kova tuuli ja] PRT DEM3.LOC-ADE be-PST hard wind and [So there was a strong wind and ]
Chapter 1. Introduction
06
(.) lun-ta [(sat-el-i), ] snow-PAR rain-FRE-PST (.) it was snowing
In line 4, Pekka responds with a repetition of the finite verb of the question only (vai:kutti; without using any of the other material from lines 2–4 or the anaphoric pronoun ne ‘they’ as the subject). This provides an affirmative answer to Simo’s question. After the affirmation, Simo presents an understanding of the implications of Pekka’s affirmative answer, topicalizing the response further.12 As mentioned above (Section 6.1), the importance of sequential context for understanding response options was also highlighted by Keevallik’s (2010) work on Estonian, which showed that verb repeats (and other longer responses) are used to respond to ‘primary’ actions (such as, base first pair parts), forming a minimal sufficient response to a question that initiates a sequence. Additionally, epistemic stances conveyed by the question have been found to shape response options. For example, Hayano (2013: 209–231) found that, in Japanese, when questioners display that they do not consider the proposition as the likely state of affairs, answerers orient to repetitional responses as more relevant for conveying a committed, more assured confirmation, thus aligning with the question. Mesoamerican languages represent an interesting contrasting case to the argument for the relative markedness of repetitional responses. In these languages/cultures repetitional actions have been found to permeate socio-cultural and conversational practices (Brown 1998, 2010; Brown, Sicoli, and Le Guen 2021). One of these langauges, Tzeltal, was found to constitute an exception to the numerical prevalence of particle (vs. repetitional) responses in a study of 14 languages (Enfield et al. 2019). In this culture, the practice of affirming by repetition is motivated by a cultural desire “to share a single position on matters, to be as one in relation to pragmatic matters of knowledge and stance” (Enfield et al. 2019: 297–298; Brown, Sicoli, and Le Guen 2021: 269). As the above discussion suggests, there is a considerable variety of repeat response types that show massive variation related, in particular, to the typological affordances of the languages concerned. While it is difficult, if not impossible, to draw generalizations from this diverse literature, it points to several aspects of repetitional responses: they may, and often do, have their home in particular action environments that might be different from those where particle responses are normatively expected; they convey distinct (vis-a-vis other response types) epistemic and deontic stances; and their action import may vary depending on what is being repeated (and how, prosodically). Some of these 12. For more details about the use of various types of repetitional responses in Finnish, see Sorjonen (2001b, 2001a) and Raevaara (2001).
23
24
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
issues are addressed by several contributors to this volume (Bolden, this volume; Harjunpää & Ostermann, this volume; Weidner, this volume). In general, more investigation is called for.
6.3 Beyond particles and repetitions In this section, we briefly consider other forms of response – both minimal and expanded – that display a stance towards the question asked. Most languages have resources with which responses can be upgraded, for example, of course, por supuesto, bien sûr, and downgraded, for example maybe, kanske, vielleicht, and so on (Betz and Deppermann 2018; Bolden, Hepburn, and Mandelbaum 2023; König and Siemund 2007; Persson, this volume; Sadock and Zwicky 1985; Sorjonen 2018; Stivers 2019). Complex refinements of epistemic stance can be managed through these means. In the following example, when responding to Nancy’s question at line 5, Hyla treats it as obvious that the boyfriend is Jewish (an inference from his family name) by responding o:f cou:rse (Stivers 2019: 198–199). Excerpt 12. (HG:II) 1 2 3 4 5 6 7 8 9
NAN: HYL: NAN: HYL: NAN: HYL: NAN: HYL: NAN:
W’ts iz last name,] =Uh:: Freedla:nd. .hh[hh [Oh[:, [(‘r) Freedlind.= =Nice Jewish bo:y? O:f cou:rse,= =‘v [cou:rse,] [hh-hh-hh] hnh .hhhhh= =Nice Jewish boy who doesn’like tih write letters?
Contrasting this upgraded response with unmarked particle responses such as yes, Stivers (2019: 198) observes that the contrast “lies in a problem not with the question’s design but with it having been posed to this recipient at all”. If upgraded responses problematize questioning, Stivers correspondingly argues that downgraded responses, such as possibly, convey uncertainty and difficulty in mounting a response, and thus “problematize answering” (Stivers 2019: 201), as in the following case (line 2): Excerpt 13. (TG) 1 2 3 4
Bee: Ava: Bee:
°hmhhh .hh So yih gonna be arou:n this weeken’?, Uh::m. (0.3) Possibly. Uh it’s a four day weeken-I have so much work t’do it isn’ funny.
Here the doubt or equivocation of Ava’s response at line 2 is, as Stivers (ibid.) notes, reinforced by the ‘hesitant’ delay that precedes it, and conveys that, in the event that the question was designed to set up a request or invitation, a denial
Chapter 1. Introduction
or rejection is a real possibility (see our discussion of Preferences in Section 5.2 above). In the case of expanded responses, it is clear that even polar questions that are answered affirmatively may frequently be expanded. In a study of responses to Danish polar questions, Steensig and Heinemann (2013) found that nearly 40% of yes-type responses were expanded, many of which occurred in sequential environments that were ‘ripe’ for expansion or in response to question-types, e.g., specification requests, that invited it. In a paper on yes/no responses to polar questions, Raymond (2010b) demonstrates that ‘flatly’ intoned versions of yes and no tend to project that the respondent will not expand or elaborate the response. Expansion of the responsive turn, by contrast, can be projected with slightly rising intonation on these particles. Elements of the production of the particles themselves, including sound stretching, or terminal plosives (as in Yep and Nope) also project that a response will or will not be expanded (G. Raymond 2010b; see also Bolinger 1946; Heritage and Sorjonen 1994; Hayashi & Hayano, this volume; Lee, this volume; Robinson and Heritage, this volume). Other elements in the pre-beginnings of turns (Schegloff 1996b) including in-breaths and multimodal aspects of early responses (Deppermann 2013; Deppermann, Mondada, and Doehler 2021; Deppermann and Schmidt 2021) lend themselves to projections of this kind. Beyond expansions, respondents may also, of course, find ways to adjust their responses so as to depart from the terms of questions put to them (Walker, Drew and Local 2011). One such method is to produce what Stivers and Hayashi (2010) term ‘transformative answers.’ In the following case (in Japanese), Noboru asks Hiroshi about the distance between Boston and New York, framing his request as inviting an opinion about how ‘close’ the two cities are (line 1): Excerpt 14. (JAPN4573; Stivers and Hayashi 2010: 5) 1
NOB:
(eh) nyu- bosuton tte nyuuyooku kara chikai wake. Boston QUOT New.York from close reason Is Boston close from New York.
2
HIR:
ee::to kuruma de y- yojikan gurai ssu kedo ne:. well car by 4.hours about COP but FP Let’s see, (it)’s about 4 hours by car.
3
NOB:
a yojikan ka.= oh 4.hours Q Oh 4 hours.
4
HIR:
=a ha:i. Yes.
5
NOB:
a soo: hu::n. Oh is that so. I see.
25
26
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
In response, Hiroshi offers an objective measure of distance which, while allowing Hiroshi to decide whether the distance is ‘close’ or not, nonetheless involves, according to Stivers and Hayashi (2010: 5), a “retroactive transformation of the question from a question about relative proximity to a question about absolute distance” (see also Bolden 2009a; Stivers 2022: 147–178; Vepsäläinen et al. this volume). A variety of turn-initial objects produced in response to polar questions can serve as harbingers of expanded responses, departures from the agendas of questions, or both. In English, these include address terms (Clayman 2010, 2013), interjections such as look or listen (Sidnell 2007), and well (Heritage 2015). Turn-initial objects serving similar functions have been observed for many languages, including Finnish, Estonian and Swedish (Hakulinen and Seppänen 1992; Hakulinen, Keevallik Eriksson and J. Lindström 2003), French (Persson 2020), Garrwa (Mushin 2018), Japanese (Hayashi and Hayano 2018), Korean (Kim 2018), Polish (Weidner 2018), Russian (Bolden 2018) and Spanish (C. W. Raymond 2018), and there are good reasons to believe that these kinds of interjective particles projecting departures from the question’s constraints may be very general across languages.
7.
Overview of the volume
The chapters in this volume present qualitative analyses of responses to polar questions across a variety of languages, action environments, and social contexts. The languages featured in the volume – Brazilian Portuguese, English, Estonian, Finnish, French, Japanese, Korean, Polish, Russian, and bilingual SpanishEnglish – represent several language families, thus adding to a linguistic diversity of research in this area. The studies vary from providing broad overviews of polar response options (e.g., Chapter 4 on Polish) to being more narrowly focused on specific action environments (e.g., Chapter 5 on responses to inferences; Chapter 6 on responses to pseudo-tag questions; or Chapter 11 on responses to medical concern solicitations) or specific response types (such as repeats in Chapter 1; specific response particles in Chapters 3 and 7; code switching in Chapter 8; responses without initial polarity items in Chapters 10 and 11; the design of the answers to questions containing any in Chapter 12). While most of the chapters rely on ordinary conversation as their primary source of data, some include data from institutional, especially medical, contexts as their main source (as in Chapter 11) or an additional source (as in Chapters 3 and 4). What unites these chapters – beside their topical focus on polar questions and their responses – is their analytic approach: While some chapters include descriptive
Chapter 1. Introduction
statistics, the primary analytic approach shared by all authors is that of Conversation Analysis. The first three chapters of the volume examine the differences between repetitional and particle responses to polar questions across different action environments. In Chapter 2, Bolden analyzes Russian conversational materials to demonstrate that, compared to particle responses (such as da ‘yes’ and net ‘no’), repetitional responses constitute a relatively marked response option. The chapter examines the use of repetitions as a response to different actions accomplished by polar questions (including understanding checks; enactments of disbelief; requests for information, and requests for assistance), and shows how repetitional responses are deployed to problematize some aspect of the action and its implied epistemic or deontic stance. The chapter also argues that, rather than treating responding options as a binary choice – between unmarked particle responses and marked repetitional responses – a full repertoire of responses should be considered, as different kinds of repetitions (e.g., minimal vs. longer) take up different stances towards the question. In Chapter 3, Harjunpää and Ostermann examine repetitional and particle responses to polar questions in Brazilian Portuguese. Their particular focus is on é-responses, which can function as either a confirming particle or as a confirmatory repetition of the question verb. The chapter shows that é-responses (either as particles or verb repeats) unproblematically confirm old information. However, they are insufficient to affirm new information, and when used in such contexts, they show only a weak commitment to the veracity of the proposition or treat the issue as ancillary. To fully affirm new information and to embrace the action implemented by the polar question, other minimal repetitional responses are used. The study highlights the linguistic specificity of response options and their dedicated interactional jobs. Furthermore, the findings point to the need to recognize action specificity of what constitutes a ‘type-conforming’ or unmarked response. Chapter 4 examines polar response options in Polish. Weidner’s focus is on the contrastive deployment of particle and repetitional responses across a variety of sequential environments. The analysis shows that particle responses are used in subsidiary courses of action to unproblematically confirm the proposition in question and to align with the action it implements. In contrast, repetitional responses enact more agentive stances and are deployed in special environments, such as following a request for assistance. Overall, the chapter highlights the importance of sequential and action contexts in the analysis of different types of polar responses.
27
28
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
Chapters 5 through 7 focus on a particular action context for polar responses: responding to requests for confirmation. In Chapter 5, using French language data, Persson examines three different ways of responding to polar questions that proffer an inference from prior talk for confirmation: repetition, voilà, and exact(ement). The analysis shows that these different types of responses take up different epistemic and agentive stances, such as, reclaiming authorship over the inference (with the repetition), attributing a degree of epistemic agency to the addressee (with voilà), and crediting the addressee with greater epistemic independence (with exact(ement)). The chapter thus highlights the complexity of confirmation as a type of responsive action and shows how language-specific resources are deployed to convey nuanced epistemic stances. Chapter 6 examines confirming responses in Korean conversation. Lee focuses on requests for confirmation in the form of ‘pseudo-tag questions’, a type of polar interrogative constructed by attaching a committal suffix (-ci) to the final verb of the turn. The analysis examines both unelaborated particle responses to such confirmation requests and elaborated conforming and nonconforming responses. Lee shows that particle responses accomplish nothing more than mere confirmation; in contrast, elaborate responses are used to ward off unintended hearings and problematize the legitimacy of the question. Overall, these findings support the argument that simple particle responses constitute an unmarked response option to requests for confirmation. In Chapter 7, Hennoste, Rääbis, Rumm, and Laanesoo examine two different affirmative response particles, jah and jaa, in responses to requests for confirmation in Estonian. They show that these two particles, while both enacting confirmation, accomplish somewhat different actions and have different sequential implications. The particle jah accomplishes an unproblematic confirmation, leading to sequence closure, while the particle jaa conveys some kind of misalignment between the interlocutors, engendering sequence expansion. These findings highlight the importance of examining question-answer sequences in relation to larger courses of action they participate in and expand our understanding of languagespecific response repertoires. Chapter 8 broadens our understanding of response possibilities further by examining a different kind of response practice available to bilingual speakers: shifting into another language. Drawing on data from Spanish-English bilingual speakers, C. W. Raymond analyzes the differences between language concordant and non-concordant (i.e. code-switched) responses to polar questions. The analysis shows that language concordance is an unmarked response option, while nonconcordant responses are used for cause to convey an agentive stance. Overall, these findings argue for a broader consideration of response repertoires.
Chapter 1. Introduction
Chapters 9 through 11 focus on extended responses to polar questions. Chapter 9 by Hayashi and Hayano draws on Japanese data to examine how responses to polar questions become extended beyond a simple confirmation. They analyze different kinds of clausal extensions and how they are used to modify the turn initial confirmation, for example to qualify it or to problematize the underlying assumptions of the question. The study explores the division of labor between different kinds of grammatical resources for extending a confirming response. In Chapter 10, Vepsäläinen, Sundqvist, Sorjonen, and Hakulinen analyze Finnish responses to interrogative polar questions that do not contain a separate element for expressing positive or negative polarity. The chapter examines a range of interactional contexts in which the polarity of the response might be conveyed indirectly or left for the recipient to infer. The analysis shows that such responses display the answers’ orientation to the action relevance of the question and are designed to further the progressivity of the ongoing project. In Chapter 11, Robinson and Heritage examine a particular interactional context in which a polar question does not make conditionally relevant a particle (yes) answer. The chapter analyzes English-language physician-patient interaction and focuses specifically on physicians’ solicitations of patients’ additional concerns (such as, Is there something else you’d like to address today?). They show that when such questions are designed so as to renew a prior ongoing questioning activity, patients normatively and unproblematically simply report an additional concern (if they have one), without including a yes particle in the response. This finding qualifies our understanding of type-conformity (as proposed by G. Raymond 2003) and underscores the importance of sequential context for analyzxsing question-answer sequences. In the final chapter of the volume, Chapter 12, Couper-Kuhlen, Thompson, and Fox inquire whether, in English, polar questions with the word any (such as, Are there any questions?) in fact prefer a negative response. Any is typically understood to be a negative polarity marker that implements a negative tilt so that a negative response is a preferred one. The chapter argues, however, that this is not always the case, and that the action implemented by the polar question and its sequential position exert stronger influence on the shape of the response.
8.
Concluding remarks
The studies collected in the volume enrich and diversify our understanding of polar questions and their responses. Taken together, they present a nuanced view of polar responses as a situated social action. Yet many questions remain.
29
30
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
First, what is the scope and meaning of ‘type conformity’? It seems almost certain that type conformity – i.e., the preference for particle responses to polar questions – is not a universal of human interactional conduct. Further, even where ‘type conforming’ responses constitute a numerical majority of responses to polar questions, their use and likelihood of occurrence is strongly shaped by several interactional and linguistic factors including: –
– –
the sequential context in which the question-answer sequence takes place, e.g., whether questions accomplish primary vs. secondary actions and the epistemic stances conveyed by the question (Bolden, this volume; Harjunpää and Ostermann, this volume; Weidner, this volume); the perceived objective of the question within a larger course of action (Pomerantz 2017; Robinson and Heritage, this volume; Vepsäläinen et al., this volume) in a particular social and institutional context; and grammatico-sequential relevancies raised by the question, e.g., the deployment of finite verb repeats in response to V-interrogatives in Finnish (discussed above) or the use of finite verb repeats in response to negatively formatted interrogatives in Russian (Bolden, this volume), Polish (Weidner, this volume), and other languages.
More cross-linguistic work is clearly mandated here. Second, the studies collected in this volume demonstrate the importance of going beyond the type conformity/non-conformity binary. Many languages, including English, deploy a wide range of particles to address different sequential and epistemic situations: these clearly need exploration on a language-bylanguage basis, and in comparative perspective. Future studies are needed to better understand a full diversity of response options, including different types of repetitions, different types of response particles and their prosodic variations, multilingual response practices, and embodied non-vocal resources. Relatedly, we need to specify the work of full vs. partial repeat responses: for example, are ‘larger’ repeats stronger than smaller ones, and if so in what sense? And do finite verb repeats (used in many languages, but not English) have a special status vis-àvis other kinds of responses? Third, for questions that implement information/confirmation requests (vs. other actions, such as requests for actions, invitation, etc.), how does question design relate to ‘preference’ for a particular type of response? That is, to what extent does a polar question prefer agreement, and how do aspects of question design contribute to the ‘strength’ of this preference (Couper-Kuhlen, Thompson, and Fox, this volume)? This issue is quite unresolved, even for English (see the discussion above), let alone the other languages of the world and needs to be investigated both in single languages and comparatively.
Chapter 1. Introduction
Fourth, how are polar responses shaped by actions implemented by polar questions requesting information/confirmation vs. using similar resources for implementing a deontic action (requesting, inviting, etc.)? More researched is called for to explicate constraints and relevancies different actions implemented via polar interrogatives impose on responses and different aspects of response design for managing these relevancies. It is our hope that the studies collected in this volume – both their findings and the questions they have raised – stimulate research on polar questions and their answers in many more typologically diverse languages.
References Alvanoudi, Angeliki. 2022. “Polar answers and epistemic stance in Greek conversation.” Pragmatics 32 (1): 1–27. Betz, Emma, and Arnulf Deppermann. 2018. “Indexing priority of position: Eben as response particle in German.” Research on Language and Social Interaction 51 (2): 171–193. Bolden, Galina B. 2008. “Reopening Russian conversations: The discourse particle -to and the negotiation of interpersonal accountability in closings.” Human Communication Research 34: 99–136. Bolden, Galina B. 2009a. “Beyond answering: Repeat-prefaced responses in conversation.” Communication Monographs 76 (2): 121–143. Bolden, Galina B. 2009b. “Implementing delayed actions.” In Conversation Analysis: Comparative perspectives, ed. by Jack Sidnell, In Studies in interactional sociolinguistics, 326–354. Cambridge: Cambridge University Press. Bolden, Galina B. 2016. “A simple da?: Affirming responses to polar questions in Russian conversation.” Journal of Pragmatics 100: 40–58. Bolden, Galina B. 2017. “Requests for here-and-now actions in Russian conversation.” In Imperative turns at talk: The design of directives in action, ed. by Marja-Leena Sorjonen, Elizabeth Couper-Kuhlen and Liisa Raevaara, 175–211. Amsterdam: John Benjamins. Bolden, Galina B. 2018. “Nu-prefaced responses in Russian conversation.” In Between turn and sequence: Turn-initial particles across languages, ed. by John Heritage and Marja-Leena Sorjonen, 23–58. Amsterdam: John Benjamins. Bolden, Galina B., Alexa Hepburn, and Jenny Mandelbaum. 2023. “The distinctive uses of right in British and American English interaction.” Journal of Pragmatics 205: 78–91. Bolinger, Dwight L. 1946. “Thoughts on ‘Yep’ and ’Nope.” American Speech 21 (2): 90–95. Bolinger, Dwight L. 1978. “Yes-no questions are not alternative questions.” In Questions, ed. by Henry Hiz. Dordrecht: Reidel. Brown, Penelope. 1998. “Conversational Structure and Language Acquisition: The Role of Repetition in Tzeltal.” Journal of Linguistic Anthropology 8 (2): 197–221. Brown, Penelope. 2010. “Questions and their responses in Tzeltal.” Journal of Pragmatics 42: 2627–2648.
31
32
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
Brown, Penelope, Mark A. Sicoli, and Olivier Le Guen. 2021. “Cross-speaker repetition and epistemic stance in Tzeltal, Yucatec, and Zapotec conversations.” Journal of Pragmatics 183: 256–272. Clayman, Steven E. 2002. “Sequence and solidarity.” In Advances in group processes: Group cohesion, trust, and solidarity, ed. by Edward J. Lawler and Shane R. Thye, 229–253. Amsterdam: Elsevier Science. Clayman, Steven E. 2010. “Address terms in the service of other actions: The case of news interview talk.” Discourse & Communication 4 (2): 161–183. Clayman, Steven E. 2013. “Agency in response: The role of prefatory address terms.” Journal of Pragmatics 57: 290–302. Couper-Kuhlen, Elizabeth. 2012. “Some truths and untruths about prosody in English question and answer sequences.” In Questions: Formal, functional and interactional perspectives, ed. by J. P. De Ruiter, 123–145. Cambridge: Cambridge University Press. Couper-Kuhlen, Elizabeth, and Margret Selting. 2018. Interactional Linguistics: Studying Language in Social Interaction. Cambridge: Cambridge University Press. Curl, Traci S., and Paul Drew. 2008. “Contingency and action: A comparison of two forms of requesting.” Research on Language and Social Interaction 41 (2): 129–153. Davidson, Judy. 1984. “Subsequent versions of invitations, offers, requests, and proposals dealing with potential or actual rejection.” In Structures of Social Action, ed. by J. Maxwell Atkinson and John Heritage, 102–128. Cambridge: Cambridge University Press. De Ruiter, J. P. (ed.) 2012. Questions: Formal, functional and interactional perspectives. Cambridge: Cambridge University Press. Deppermann, Arnulf. 2013. “Turn-design at turn-beginnings: Multimodal resources to deal with tasks of turn-construction in German.” Journal of Pragmatics 46 (1): 91–121. Deppermann, Arnulf, Lorenza Mondada, and Simona Pekarek Doehler. 2021. “Early responses: An introduction.” Discourse Processes 58 (4): 293–307. Deppermann, Arnulf, and Axel Schmidt. 2021. “Micro-sequential coordination in early responses.” Discourse Processes 58 (4): 372–396. Drake, Veronika. 2015. “Indexing uncertainty: The case of turn-final or.” Research on Language and Social Interaction 48 (3): 301–318. Dryer, Matthew S. 2013a. “Polar questions.” In World atlas structures, ed. by Matthew S. Dryer and Martin Haspelmath. Leipzig: Max Planck Institute for Evolutionary Anthropology. https://wals.info/chapter/116 Dryer, Matthew S. 2013b. “Position of polar question particles.” In World atlas structures, ed. by Matthew S. Dryer and Martin Haspelmath. Leipzig: Max Planck Institute for Evolutionary Anthropology. https://wals.info/chapter/116 Enfield, Nick J. 2010. “Questions and responses in Lao.” Journal of Pragmatics 42 (10): 2649–2665. Enfield, Nick J., Tanya Stivers, Penelope Brown, Christina Englert, Katarina Harjunpää, Makoto Hayashi, Trine Heinemann, Certie Hoymann, Tiina Keisanen, Mirka Rauniomaa, Chase Wesley Raymond, Federico Rossano, Kyun-Eun Yoon, Inge Zwitserlood, and Stephen C. Levinson. 2019. “Polar answers.” Journal of Linguistics 55 (2): 277–304.
Chapter 1. Introduction
Enfield, Nick J., Tanya Stivers, and Stephen C. Levinson. 2010. “Question-response sequences in conversation across ten languages: An introduction.” Journal of Pragmatics 42 (10): 2615–2619. Englert, Christina. 2010. “Questions and responses in Dutch conversations.” Journal of Pragmatics 42 (10): 2666–2684. Farkas, Donka F., and Kim B. Bruce. 2010. “On reacting to assertions and polar questions.” Journal of Semantics 27: 81–118. Farkas, Donka F., and Floris Roelofsen. 2019. “Polarity particles revisited.” Semantics and Pragmatics 12 (15): 1–25. Floyd, Simeon, Elisabeth Norcliffe, and Lila San Roque (eds.). 2018. Egophoricity. Typological studies in language (TSL). Philadelphia: John Benjamins Publishing Company. Freed, Alice F., and Carol M. Ehrlich (eds.). 2010. ”Why do you ask?”: The functions of questions in institutional discourse. New York: Oxford University Press. Geluykens, Ronald. 1988. “The myth of rising intonation in polar questions.” Journal of Pragmatics 12 (4): 467–485. Giannakidou, Anastasia. 2011. “Negative and positive polarity items: Variation, licensing, and compositionality.” In Semantics: An international handbook of natural language meaning, ed. by C. Maienborn, K von Heusinger and P. Portner, 1660–1712. Berlin: Mouton. Goffman, Erving. 1979. “Footing.” Semiotica 25 (1–2): 1–29. Golato, Andrea, and Zsuzsanna Fagyal. 2008. “Comparing single and double sayings of the German response token ja and the role of prosody: A conversation analytic perspective.” Research on Language and Social Interaction 41 (3): 241–270. Haakana, Markku, and Salla Kurhila. 2009. “Other-correction in everyday conversation: Some comparative aspects.” In Talk in Interaction: Comparative Dimensions, ed. by Minna Laakso Markku Haakana and Jan Lindström, 152–179. Helsinki: Finnish Literature Society (SKS). Haakana, Markku, and Laura Visapää. 2014. “Eiku – korjauksen partikkeli? [Eiku – the particle of repair?” Virittäjä 1: 41–71. Hakulinen, Auli. 2001. “Minimal and non-minimal answers to yes-no questions.” Pragmatics 11 (1): 1–15. Hakulinen, Auli, Leelo Keevallik Eriksson, and Jan Lindström. 2003. “Kuule, kule, hördu: Projicerande praktiker i finska, estniska och svenska [Kuule, kule, hördu: Projecting practices in Finnish, Estonian and Swedish].” In Grammatik och Samtal: Studier till minne av Mats Eriksson, ed. by Bengt Nordberg, Leelo Eriksson Keevallik, Kerstin Thelander and Mats Thelander, 199–218. Uppsala universitet. Hakulinen, Auli, and Eeva-Leena Seppänen. 1992. “Finnish kato: From verb to particle.” Journal of Pragmatics 18 (6): 527–549. Hamdani, Fakry, and Scott Barnes. 2018. “Polar questions in colloquial Indonesian: A pilot study.” Journal of Pragmatics 132: 1–20. Hayano, Kaoru. 2013. “Territories of knowledge in Japanese conversation.” PhD PhD dissertation, Radboud Universiteit. Hayashi, Makoto. 2010. “An overview of the question–response system in Japanese.” Journal of Pragmatics 42 (10): 2685–2702.
33
34
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
Hayashi, Makoto, and Kaoru Hayano. 2018. “A-prefaced responses to inquiry in Japanese.” In Between turn and sequence: Turn-initial particles across languages, ed. by John Heritage and Marja-Leena Sorjonen, 193–223. John Benjamins. Heinemann, Trine. 2005. “Where grammar and interaction meet: the preference for matched polarity in responsive turns in Danish.” In Syntax and Lexis in Conversation: Studies on the Use of Linguistic Resources in Talk-in-Interaction, ed. by Auli Hakilinen and Margret Selting, 375–402. Amsterdam: John Benjamins. Heinemann, Trine. 2006. “’Will you or can’t you?’: Displaying entitlement in interrogative requests.” Journal of Pragmatics 38 (7): 1081–1104. Heinemann, Trine. 2010. “The question–response system of Danish.” Journal of Pragmatics 42 (10): 2703–2725. Heritage, John. 1984. Garfinkel and ethnomethodology. New York: Polity Press. Heritage, John. 1988. “Explanations as accounts: A conversation analytic perspective.” In Analyzing everyday explanation: A casebook of methods, ed. by Charles Antaki, 127–144. Beverly Hills: Sage. Heritage, John. 2010. “Questioning in medicine.” In ”Why do you ask?”: The functions of questions in institutional discourse, ed. by Alice F. Freed and Carol M. Ehrlich, 42–68. New York: Oxford University Press. Heritage, John. 2012a. “The epistemic engine: Sequence organization and territories of knowledge.” Research on Language and Social Interaction 45 (1): 30–52. Heritage, John. 2012b. “Epistemics in action: Action formation and territories of knowledge.” Research on Language and Social Interaction 45 (1): 1–29. Heritage, John. 2015. “Well-prefaced turns in English conversation: A conversation analytic perspective.” Journal of Pragmatics 88: 88–104. Heritage, John. 2018. “Turn-initial particles in English: The cases of oh and well.” In Between Turn and Sequence: Turn-Initial Particles Across Languages, ed. by John Heritage and Marja-Leena Sorjonen, 149–184. Amsterdam: John Benjamins. Heritage, John, and Steven Clayman. 2010. Talk in action: Interactions, identities, and institutions. Malden, MA: Wiley-Blackwell. Heritage, John, and Chase Wesley Raymond. 2021. “Preference and polarity: Epistemic stance in question design.” Research on Language and Social Interaction 54 (1): 39–59. Heritage, John, and Geoffrey Raymond. 2005. “The terms of agreement: Indexing epistemic authority and subordination in talk-in-interaction.” Social Psychology Quarterly 68: 15–38. Heritage, John, and Geoffrey Raymond. 2012. “Navigating epistemic landscapes: Acquiescence, agency and resistance in responses to polar questions.” In Questions: Formal, functional and interactional perspectives, ed. by J. P. De Ruiter, 179–192. Cambridge: Cambridge University Press. Heritage, John, and Marja-Leena Sorjonen. 1994. “Constituting and maintaining activities across sequences: and-prefacing as a feature of question design.” Language in Society 23 (1): 1–29. Heritage, John, and Marja-Leena Sorjonen (eds.). 2018. Between turn and sequence: Turninitial particles across languages. Amsterdam: John Benjamins. Holmberg, Anders. 2015. The syntax of yes and no. Oxford: Oxford University Press.
Chapter 1. Introduction
Houtkoop-Steenstra, Hanneke. 1987. Establishing agreement: An analysis of proposalacceptance sequences. Dordrecht, Holland: Foris. Hoymann, Gertie. 2010. “Questions and responses in ╪Ākhoe Hai||om.” Journal of Pragmatics 42 (10): 2726–2740. Jones, Bob Morris. 1999. The Welsh answering system. New York: De Gruyter Mouton. Kamio, Akio. 1997. Territory of information. Pragmatics & Beyond New Series. 48. Philadelphia: John Benjamins. Keevallik, Leelo. 2010. “Minimal answers to yes/no questions in the service of sequence organization.” Discourse Studies 12 (3): 283–309. Kendrick, Kobin H., and Francisco Torreira. 2014. “The timing and construction of preference: A quantitative study.” Discourse Processes 52 (4): 255–289. Kim, Stephanie Hyeri. 2018. “Two types of trouble with questions: A comparative perspective on turn-initial particles in Korean.” In Between turn and sequence: Turn-initial particles across languages, ed. by John Heritage and Marja-Leena Sorjonen, 97–118. John Benjamins. König, Ekkehard, and Peter Siemund. 2007. “Speech act distinctions in grammar.” In Language Typology and Syntactic Description, ed. by Timothy Shopen, 276–324. Cambridge: Cambridge University Press. Labov, William. 1972. “The transformation of experience in narrative syntax.” In Language in the inner city: Studies in the Black English Vernacular, ed. by William Labov, 354–396. Philadelphia: University of Pennsylvania Press. Labov, William, and David Fanshel. 1977. Therapeutic discourse: Psychotherapy as conversation. New York: Academic Press. Laury, Ritva. 2018. “The Finnish verb repeat response: Its emergence and its nature as a formulaic expression.” Journal of Pragmatics 123: 139–150. Lee, Seung-Hee. 2015. “Two forms of affirmative responses to polar questions.” Discourse Processes 52 (1): 21–46. Lee, Seung-Hee. 2017. “Extended responding: Interaction and collaboration in the production and implementation of responses.” In Enabling Human Conduct: Studies of talk-ininteraction in honor of Emanuel A. Schegloff, ed. by Gene H. Lerner Geoffrey Raymond and John Heritage, 105–124. Amsterdam: John Benjamins Publishing. Lerner, Gene H. 1996. “Finding “face” in the preference structures of talk-in-interaction.” Social Psychology Quarterly 59 (4): 303–321. Levinson, Stephen C. 2010. “Questions and responses in Yélî Dnye, the Papuan language of Rossel Island.” Journal of Pragmatics 42 (10): 2741–2755. Levinson, Stephen C. 2012. “Interrogative intimations: On a possible social economics of interrogatives.” In Questions: Formal, functional and interactional perspectives, ed. by J. P. De Ruiter, 11 – 32. New York: Cambridge University Press. Lindström, Anna. 1997. “Designing social actions: Grammar, prosody, and interaction in Swedish conversation.” PhD Dissertation, University of California. Lindström, Anna. 2005. “Language as social action: A study of how senior citizens request assistance with practical tasks in the Swedish home help service.” In Syntax and lexis in conversation, ed. by Auli Hakulinen and Margret Selting, 209–230. Amsterdam: John Benjamins.
35
36
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
Lindström, Anna. 2017. “Accepting remote proposals.” In Enabling human conduct: Naturalistic studies of talk-in-interaction in honor of Emanuel A. Schegloff ed. by Geoffrey Raymond, Gene H. Lerner and John Heritage, 125–142. Amsterdam: John Benjamins. Mazeland, Harrie, and Leendert Plug. 2010. “Doing confirmation with ja/nee hoor: Sequential and prosodic characteristics of a Dutch discourse particle.” In Prosody in Interaction, ed. by Elisabeth Reber Dagmar Barth-Weingarten and Margret Selting, 161–188. Amsterdam: John Benjamins. Mushin, Ilana. 2018. “Diverging from ‘business as usual’: Turn-initial ngala in Garrwa conversation.” In Between Turn and Sequence: Turn-initial particles across languages, ed. by John Heritage and Marja-Leena Sorjonen, 119–154. Amsterdam: John Benjamins. Ogden, Richard, and Sara Routarinne. 2005. “The communicative functions of final rises in Finnish intonation.” Phonetica 62 (2–4): 160–175. Persson, Rasmus. 2020. “Taking isue with a question while answering it: Prefatory particles and multiple sayings of polar response tokens in French.” Research on Language and Social Interaction 53 (3): 380–403. Pomerantz, Anita. 1980. “Telling my side: “limited access” as a “fishing” device.” Sociological Inquiry 50 (3–4): 186–198. Pomerantz, Anita. 1984. “Agreeing and disagreeing with assessments: some features of preferred/dispreferred turn shapes.” In Structures of social action: Studies in conversation analysis, ed. by J. Maxwell Atkinson and John Heritage, 57–101. New York: Cambridge University Press. Pomerantz, Anita. 1988. “Offering a candidate answer: An information seeking strategy.” Communication Monographs 55: 360–373. Pomerantz, Anita. 2017. “Inferring the purpose of a prior query and responding accordingly.” In Enabling Human Conduct: Studies of talk-in-interaction in honor of Emanuel A. Schegloff, ed. by Gene H. Lerner Geoffrey Raymond and John Heritage, 61–76. Amsterdam: John Benjamins Publishing. Pope, Emily Norwood. 1976. Questions and answers in English. Edited by Jerrold J. Katz. Janua linguarum: Series practica 226. The Hague: Mouton. Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik. 1985. A Comprehensive grammar of the English language. London: Longman. Raevaara, Liisa. 1993. “Kysyminen toimintana. Kysymys-vastaus-vieruspareista arkikeskustelussa [Asking as an action. Question-answer-adjacent pairs in everyday conversation].” Lisensiaatintyö, Helsingin yliopiston suomen kielen laitos. Raevaara, Liisa. 2001. “Kysymisestä ja vastaamisesta institutionaalisena toimintana [On asking questions and answering as an institutional activity].” In Keskustelunanalyysin näkymiä, ed. by Mia Halonen and Sara Routarinne, 47–69. Raymond, Chase Wesley. 2015. “Questions and responses in Spanish monolingual and Spanish–English bilingual conversation.” Language & Communication 42: 50–68. Raymond, Chase Wesley. 2018. “Bueno-, pues-, and bueno-pues-prefacing in Spanish conversation.” In Between turn and sequence: Turn-initial particles across languages, ed. by John Heritage and Marja-Leena Sorjonen, 59–96. Amsterdam: John Benjamins.
Chapter 1. Introduction
Raymond, Chase Wesley, and John Heritage. 2021. “Probability and valence: Two preferences in the design of polar questions and their management.” Research on Language and Social Interaction 54 (1): 60–79. Raymond, Chase Wesley, Jeffrey D. Robinson, Barbara A. Fox, Sandra A. Thompson, and Kristella Montiegel. 2021. “Modulating action through minimization: Syntax in the service of offering and requesting.” Language in Society 50 (1): 53–91. Raymond, Geoffrey. 2003. “Grammar and social organization: Yes/No type interrogatives and the structure of responding.” American Sociological Review 68: 939–967. Raymond, Geoffrey. 2010a. “Grammar and social relations: Alternative forms of yes/no type initiating actions in health visitor interactions.” In Why do you ask?: The function of questions in institutional discourse, ed. by Alice F. Freed and Susan Ehrlich, 69–107. Oxford: Oxford University Press. Raymond, Geoffrey. 2010b. “Prosodic variation in responses: The case of type-conforming responses to yes/no interrogatives.” In Prosody in interaction, ed. by Dagmar Barth-Weingarten, Elisabeth Reber and Margret Selting. Amsterdam: John Benjamins. Raymond, Geoffrey. 2013. “At the intersection of turn and sequence organization: On the relevance of “slots” in type-conforming responses to polar interrogatives.” In Units of talk – Units of action, ed. by Beatrice Szczepek and Geoffrey Raymond, 169–206. Amsterdam: Benjamins. Roberts, Felicia, Piera Margutti, and Shoji Takano. 2011. “Judgments concerning the valence of inter-turn silence across speakers of American English, Italian, and Japanese.” Discourse Processes 48 (5): 331–354. Robinson, Jeffrey D. 2020a. “One type of polar, information-seeking question and its stance of probability: Implications for the preference for agreement.” Research on Language and Social Interaction 53 (4): 425–442. Robinson, Jeffrey D. 2020b. “Revisiting preference organization in context: A qualitative and quantitative examination of responses to information seeking.” Research on Language and Social Interaction 53 (2): 197–222. Roelofsen, Floris, and Donka F. Farkas. 2015. “Polarity particle responses as a window onto the interpretation of questions and assertions.” Language 91: 359 – 414. Rossano, Federico. 2010. “Questioning and responding in Italian.” Journal of Pragmatics 42 (10): 2756–2711. Rossi, Giovanni. in press. Systems of social action: The case of requesting in Italian. Oxford: Oxford University Press. Rossi, Giovanni, Mark Dingemanse, Simeon Floyd, Julija Baranova, Joe Blythe, Kobin H. Kendrick, Jörg Zinken, and N. J. Enfield. 2023. “Shared cross-cultural principles underlie human prosocial behavior at the smallest scale.” Scientific Reports 13 (1): 6057. Sacks, Harvey. 1987. “On the preferences for agreement and contiguity in sequences in conversation.” In Talk and social organization, ed. by Graham Button and John R E. Lee, In Intercommunication, 54–69. Philadelphia: Multilingual Matters. Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson. 1974. “A simplest systematics for the organization of turn-taking for conversation.” Language 50: 696–735.
37
38
Galina B. Bolden, John Heritage & Marja-Leena Sorjonen
Sadock, Jerrold M., and Arnold M. Zwicky. 1985. “Speech act distinction in syntax.” In Language typology and syntactic description: Clause structure, ed. by Timothy Shopen, 155–196. Cambridge: Cambridge University Press. Sandman, Erika. 2016. “A grammar of Wutun.” Doctoral dissertation, Faculty of Arts, Department of World Cultures, University of Helsinki. http://hdl.handle.net/10138 /168427 Sandman, Erika, and Karolina Grzech. 2022. “Egophoricity and evidentiality: Different categories, similar discourse functions: Insights on conversational data from the Tibetan Plateau and the Amazonian Foothills.” Interactional Linguistics 2 (1): 79–109. Schegloff, Emanuel A. 1996a. “Confirming allusions: toward an empirical account of action.” American Journal of Sociology 104 (1): 161–216. Schegloff, Emanuel A. 1996b. “Turn organization: one intersection of grammar and interaction.” In Interaction and grammar, ed. by Elinor Ochs, Emanuel A. Schegloff and Sandra A. Thompson, In Studies in interactional sociolinguistics, 52–133. Cambridge: Cambridge University Press. Schegloff, Emanuel A. 2007. Sequence organization in interaction: A primer in conversation analysis. Cambridge: Cambridge University Press. Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks. 1977. “The preference for selfcorrection in the organization of repair in conversation.” Language 53: 361–382. Searle, John R. 1969. Speech acts: An essay in the philosophy of language. Cambridge: Cambridge University Press. Sidnell, Jack. 2007. “‘Look’-prefaced turns in first and second position: Launching, interceding and redirecting action.” Discourse Studies 9 (3): 387–408. Sidnell, Jack, and Tanya Stivers (eds.). 2013. The handbook of Conversation Analysis. Oxford: Blackwell. Sinclair, John McHardy, and Malcolm Coulthard. 1975. Towards an Analysis of Discourse: The English Used by Teachers and Pupils. First Edition. London: Oxford Univ Pr. Sorjonen, Marja-Leena. 1996. “On repeats and responses in Finnish Conversations.” In Interaction and grammar, ed. by Elinor Ochs, Emanuel A. Schegloff and Sandra A. Thompson, In Studies in interactional sociolinguistics, 277–327. Cambridge: Cambridge University Press. Sorjonen, Marja-Leena. 2001a. Responding in conversation: A study of response particles in Finnish. Vol. 70 Pragmatics & beyond. Philadelphia: John Benjamins. Sorjonen, Marja-Leena. 2001b. “Simple answers to polar questions: The case of Finnish.” In Studies in interactional linguistics, ed. by Margret Selting and Elizabeth Couper-Kuhlen, In Studies in discourse and grammar, 405–431. Philadelphia: John Benjamins. Sorjonen, Marja-Leena. 2018. “Questions and answers.” In On their structural and interactional relationships across languages, ed. by Theodossia-Soula Pavlidou, In Studies in Spoken Discourse, 11–32. Thessaloniki: Institute of Modern Greek Studies Aristotle University Thessaloniki. Sorjonen, Marja-Leena, Elizabeth Couper-Kuhlen, and Liisa Raevaara (eds.). 2017. Imperative turns at talk: The design of directives in action. Amsterdam: John Benjamins. Steensig, Jakob, and Paul Drew. 2008. “Introduction: Questioning and affiliation/ disaffiliation in interaction.” Discourse Studies 10 (1): 5–15.
Chapter 1. Introduction
Steensig, Jakob, and Trine Heinemann. 2013. “When ‘yes’ is not enough – as an answer to a yes/no question.” In Units of talk – Units of action, ed. by Beatrice Szczepek and Geoffrey Raymond, 207–242. Amsterdam: Benjamins. Stefani, Elwys De. 2021. “Embodied Responses to Questions-in-Progress: Silent Nods as Affirmative Answers.” Discourse Processes 58 (4): 353–371. Stivers, Tanya. 2004. “”No no no” and other types of multiple sayings in social interaction.” Human Communication Research 30 (2): 260–293. Stivers, Tanya. 2010. “An overview of the question-response system in Americal English conversation.” Journal of Pragmatics 42 (10): 2772–2781. Stivers, Tanya. 2019. “How we manage social relationships through answers to questions: The case of interjections.” Discourse Processes 2 (4): 191–209. Stivers, Tanya. 2022. The book of answers: Alignment, autonomy, and affiliation in social interaction. Foundations of Human Interaction. New York: Oxford University Press. Stivers, Tanya, and Nick J. Enfield. 2010. “A coding scheme for question-response sequences in conversation.” Journal of Pragmatics 42 (10): 2620–2626. Stivers, Tanya, Nick J. Enfield, Penelope Brown, Christina Englert, Makoto Hayashi, Trine Heinemann, Certie Hoymann, Federico Rossano, Jan Peter De Ruiter, Kyun-Eun Yoon, and Stephen C. Levinson. 2009. “Universals and cultural variation in turn-taking in conversation.” PNAS 106 (26): 10587–10592. Stivers, Tanya, and Makoto Hayashi. 2010. “Transformative answers: One way to resist a question’s constraints.” Language in Society 39 (1): 1–39. Walker, Traci. 2014. “Form ≠ function: The independence of prosody and action.” Research on Language and Social Interaction 47 (1): 1–16. Walker, Traci, Paul Drew and John Local 2011. “Responding Indirectly.” Journal of Pragmatics 43 (9): 2434–2451. Weber, Elizabeth G. 1993. Varieties of questions in English conversation. John Benjamins. Weidner, Matylda. 2018. “Treating something as self-evident: No-prefaced turns in Polish.” In Between Turn and Sequence: Turn-initial particles across languages, ed. by John Heritage and Marja-Leena Sorjonen, 225–250. John Benjamins Publishing. Yoon, Kyung-Eun. 2010. “Questions and responses in Korean conversation.” Journal of Pragmatics 42 (10): 2782–2798. Zinken, Jörg. 2016. Requesting responsibility: The morality of grammar in Polish and English family interaction. New York: Oxford University Press. Zinken, Jörg, and Eva Ogiermann. 2013. “Responsibility and action: Invariants and diversity in requests for objects in British English and Polish interaction.” Research on Language and Social Interaction 46 (3): 256–276.
39
chapter 2
Repetitional responses to polar questions in Russian conversation Galina B. Bolden Rutgers University
This chapter examines repetitional responses to polar questions that implement requests for confirmation, information, or assistance in Russian conversation, contrasting them with particle responses, such as da (‘yes’) and net (‘no’). The analysis will show that Russian repetitional responses tend to be a marked response option deployed to disalign from some dimension of the interrogative or the action it implements. For example, repetitional responses may be deployed to reassert its speaker’s epistemic authority in the service of confirming a candidate understanding, to insist on the veracity of a statement following an enactment of disbelief or doubt, to disalign from the action implemented by an information request, or – in response to requests for service – to contest the stance that the provision of assistance might be problematic. The analysis suggests that different types of repetitional responses (such as lexical vs. longer repeats) may enact more or less disaligning stances vis-à-vis the initiating action. Overall, what exactly is being accomplished via a repetitional response is sensitive to a number of considerations, including the sequential and action environment of its deployment, the design of the repetitional response, and the repertoire of available response options. Keywords: Russian, polar questions, polar responses, repetitions, response design, question design, requests, repair, news-marks, conversation analysis
1.
Introduction
This chapter examines repetitional responses to polar questions that implement requests for confirmation, information, or assistance in Russian conversation, contrasting them with particle responses, such as da (‘yes’) and net (‘no’). Russian is considered one of the “verb-echo” languages (Holmberg 2015: 65) in which a polar question may be answered with a repeat of the focal element of the question, usually of the verb. Russian grammarians typically consider such echo responses https://doi.org/10.1075/slsi.35.02bol © 2023 John Benjamins Publishing Company
Chapter 2. Repetitional responses in Russian
as functionally equivalent to particle responses (Timberlake 2004). This chapter will show, however, that – not unlike some other languages – Russian repetitional responses are, for the most part, a marked response option deployed to resist some dimension of the polar question or of the action it implements. Prior conversation analytic research into question-answer sequences has demonstrated that questions (and polar questions in particular) set up a number of constraints on responses, and responses either embrace or resist these constraints (e.g., Raymond 2003; Heritage 2003; Heritage and Raymond 2012; Raymond 2010a; Sacks 1987; Bolden 2009a; Stivers and Hayashi 2010; Bolden 2016; Heritage 2010). Thus, a response to a polar question may be designed in such a way as to agree or disagree with the tilt of the question (preferred vs. dispreferred responses), conform or not to its topical and action agendas, and convey congruent or incongruent epistemic and evaluative stances vis-à-vis the question (Heritage and Clayman 2010; Raymond 2003; Heritage and Raymond 2012; Bolden 2016). Overall, resistant responses are typically characterized by departures from a most minimal form – an agreeing particle (e.g., Raymond 2003; Thompson, Fox, and Couper-Kuhlen 2015; Stivers 2019). Along these lines, cross-linguistic research has shown that, in comparison to response particles, repetitional responses tend to be a pragmatically marked response option (Enfield et al. 2019; Heritage and Raymond 2012; Bolden 2016; Enfield and Sidnell 2015; Enfield, Stivers, and Levinson 2010; Raymond 2003; Stivers 2019; Hayashi and Hayano 2018; Stivers 2010), though their deployment and prevalence vary across languages (Keevallik 2010; Brown 2010; Harjunpää & Ostermann, this volume; Sorjonen 2001b; Weidner, this volume). Raymond (2003) and Heritage & Raymond (2012) show that, in English, particle responses affirm the factual accuracy of the proposition in question and “accept the terms of the question unconditionally, exerting no agency with respect those terms,” thereby acquiescing to the question’s terms, stances, and presuppositions (Heritage and Raymond 2012: 183). In contrast, repetitional responses are deployed “for cause” (Raymond 2003). They may “assert the respondent’s epistemic and social entitlement” and “exert agency with respect to [the question’s] terms” and presuppositions (Heritage and Raymond 2012: 186). This chapter demonstrates that, broadly speaking, Russian follows this pattern in which repetitional responses are deployed to disalign from some aspect of the polar interrogative and/or the action implemented by it. The analysis will show that what exactly is being accomplished via a repetitional response is shaped by its sequential and action environment, aspects of its design, and the repertoire of available response options. Thus, repetitional responses may be deployed to reassert their speakers’ epistemic primacy and authority (for example, when confirming a candidate understanding), resist the stance of disbelief or doubt enacted
41
42
Galina B. Bolden
by a news-mark, disalign with the action implemented by an information request, or resist the stance that the provision of assistance might be problematic (when responding to requests for services). The chapter also explores the possibility that different forms of repetitional responses may accomplish slightly different actions and, specifically, that partial lexical repeats may be less pragmatically marked than longer repetitional responses and may, in some contexts, be fully conventionalized. This chapter is organized as follows. After a brief discussion of data and method, I introduce the basic forms Russian polar questions take and preview different types of repetitional responses. Then, I examine repetitional responses in five of the most common contexts (contrasting them with particle responses): in response to (1) negative polarity interrogatives, (2) understanding checks, (3) news-marks, (4) requests for information, and (5) requests for service.
2. Data and method The data for this chapter come from a diverse corpus of video and audio recorded conversations between native speakers of Russian (primarily between friends and family, but also doctor-patient and workplace interactions), recorded in the United States1 and in Russia.2 The analysis draws on a collection of over 500 polar questions. The collection includes polar questions designed in various ways (see the section below) that accomplish diverse social actions, including questions that seek information or confirmation, mark something as news, initiate repair, request assistance, etc. The collection contains both confirming and disconfirming types of responses. Seventy-five of the over 500 polar questions in the collection (i.e., approximately 15%) are responded to with a repetitional response.3 (There are additionally a few instances of composite responses that include repetition.) In contrast, over three hundred (333) responses (approximately 67%) contain a particle response (including various versions of da, net, and hm mm). The rest of the cases are responses of other kinds (such as those that neither confirm nor disconfirm the proposition). Note that I oversampled repetitional responses by specifi1. While many of the participants in US dataset are bilingual in English and Russian, only those who are dominant in Russian and who had lived in Russia for most of their lives were included in this study. 2. I would like to thank Julija Baranova and Aleksandr Shirokov for sharing some of their video recordings with me. 3. See the section on repetitional response forms for more detail.
Chapter 2. Repetitional responses in Russian
cally looking for them, and the actual proportion of repetitional responses in the dataset would be lower. The relative rarity of repeats when compared to particle responses suggests that, overall, it is a pragmatically marked response option, though the qualitative analysis presented in this chapter paints a more nuanced picture.
3.
An overview of Russian polar questions
Unlike a number of other languages (Enfield, Stivers, and Levinson 2010), Russian does not ordinarily rely on morpho-syntactic resources – such as question particles (but see below) or subject-auxiliary inversion – to form polar questions (e.g., Comrie 1984). Here I present the most widespread forms that polar questions take in my conversational data. Very commonly, polar questions are distinguished from statements by their intonation. Intonational interrogatives tend to convey a relatively steep epistemic gradient between the speaker and the addressee, with the questioner claiming a K- (i.e., unknowing) position vis-à-vis the recipient’s K+ (i.e., knowing) position (Heritage 2010). In Russian, the questioning intonation (typically, a pitch accent in a sharp rise-fall pattern) is carried by the word that conveys the focus of the question (Hirst and Di Cristo 1998; Svetozarova 1998; Comrie 1984; Meyer and Mleinek 2006). Depending on the structure of the utterance, the questioning intonation may or may not be at the end of the turn constructional unit (or TCU – see Sacks, Schegloff, and Jefferson 1974). Most commonly, the questioning intonation is carried by the predicate (Comrie 1984). In such cases, the entire proposition is questioned, as in Excerpt 1. (See the Appendix for a description of the transcription conventions that are unique to this chapter.) Excerpt 1. (GM3 0:50) 01
DUS:
palu,chili/no< my ˇo:chen’ [pozna prishli,/a yes we received but we very late came PRT Yes we did but we got home very late, and…
And in Excerpt 13, a lexical (and, in this case, full) repeat is followed by da (‘yes’): Excerpt 13. (PF10 1:25) 01
INN:
02
GRI:
03
Ana kagdata v Rasi,i zhil[a/ da]/ she sometime in Russia lived yes She used to live in Russia/ yes [Mh mm,]/ e-Ru¿skaja/ .h[hh Russian.F.SG {She’s} Russian?
Chapter 2. Repetitional responses in Russian
04
INN:
05
GRI:
[R^uskaja / [da/ Russian.F.SG yes Russian yes [°Agah/ panjatna/° understood I see
The analysis in the subsequent sections will suggest that in Russian it may be useful to make an analytic distinction between partial lexical repeats (especially of the verb or another focal element of the interrogative) and more elaborate (partial or full) repetitional responses implemented by phrasal or sentential TCUs. It appears that, in repeating a larger portion of the question, the speaker does more work to problematize something about the interrogative or the action it implements. On the other hand, partial lexical repeats (especially of the verb) may be less pragmatically marked than other repetitional responses (and in some contexts fully conventionalized). I am now going to turn to the analysis of environments in which repetitional responses occur and what they may accomplish interactionally, contrasting their usage with particle responses. I start with the most common and conventionalized environment where repetitional responses are deployed – in response to negative polarity interrogatives.
5.
Responding to negative polarity interrogatives
As noted above, in Russian, polar questions may be formatted negatively, so as to invite the addressee to agree (or disagree) with a negative assertion. In response to such interrogatives, the particle net (‘no’) may be used to agree with the negative proposition, which would constitute a response that aligns with the tilt of the question. For instance, in Excerpt 14, Rita’s question (line 1) suggests that Vitalij didn’t call, and Misha’s Ne:t (in line 2) affirms this (that Vitalij didn’t call). Excerpt 14. (RP13 0:20) 01
RIT:
Slush=tebe Vital’ka ne zvani?l/= listen you NAME not called Listen Vitalij didn’t call you?
02
MIS:
=Ne:t,/ no
03
RIT:
.hA/=patamushta vas ne byla do,ma/ PRT because you.PL not were home Oh/ because you weren’t home/
49
50
Galina B. Bolden
While the negative proposition can be affirmed with the particle net (as in Excerpt 14 above), to disconfirm the negative proposition – i.e., to assert the opposite – a repetitional response appears to be pragmatically required. For example, in Excerpt 15, Marina inquires whether Olga is presenting at a conference. Marina’s question (line 1) is designed as a negative polarity intonational interrogative and is, thus, tilted towards a no-type answer – i.e., that Olga will not be presenting at the conference. Excerpt 15. (JB9–1 12:30) 01
MAR:
02 03
Ty na kanferencii ne vystupa¿esh/ you.SG at conference not present.2SG You are not presenting at the conference? (.)
OLG:
04
|Vystupaju/| present.1SG {I} am | ((nod)) | S dak[ladam/ with report with a report
05
MAR:
06
olg:
07
MAR:
08
OLG:
[Za?vtr|a/ tomorrow tomorrow? |((nods twice))/(.) Da sht[o ty/ PRT what you Really [Ja eë vidu sama/ I it chair.1SG myself I’m chairing it
09
(0.7)
10
(0.3)/((Marina nods))
11
OLG:
i vystupaju/ and present.1SG and presenting
In line 3, Olga asserts that she is, in fact, presenting at the conference, in contrast to what Marina’s question design suggested. To assert this, Olga uses a repeat of the verb vystupaju (‘presenting’), modifying it for first person (replacing the second person singular suffix -esh in the question with the first person singular suffix -ju in the response). The repetitional response, produced as a prosodically complete lexical TCU and accompanied by a nod (line 3), is subsequently extended
Chapter 2. Repetitional responses in Russian
with an increment (line 4). Marina takes this up as news (lines 5 and 7), eliciting further elaboration from Olga (lines 8–9). Excerpt 16 is another example of a repeat used to affirm the opposite of what is being proposed via a negative polarity interrogative. In line 1, Greg asks whether the tape recording machine he is using to record this conversation (which Katya had given him) will show when the tape ends. The question is tilted towards a pessimistic (for the recording project) net (i.e., that there won’t be an indicator of when the tape ends), conveying a high degree of doubt in a positive outcome. Excerpt 16. (PF2 1:00) 01
GRE:
02
KAT:
03
GRE:
Bea&Joa leaned over smartphone-> +(0.3) +turns paper page-> s- s (0.4)
BEA: nar
página dois +aqui.+ page two here ->+.....+
06
NAR: -> é ↑dois? is {it} two
07
(0.4)
08
BEA: => é.
09 10
(0.4) NAR:
a: é verdade >aqui ach’ que tá (o contrário [o malah; the Mullah
05
BEA: => é:;
06
NAR:
07
BEA:
é. e as pessoas est- voltavam pra estudá com ele né. and people stu- returned to study with him right
In Excerpt 1, Nar requests confirmation of the page number with a turn containing the copula verb é (ser.3sg) (line 6). Bea provides confirmation by repeating the verb (line 8). In contrast, in Excerpt 2, line 4, Nar initiates repair of the pronominal person reference in line 1 by requesting confirmation of the person’s name with a phrasal turn, without any verb. Here as well, Bea confirms Nar’s understanding as correct with é (ser.3sg), but now using it as a particle, as it does not repeat anything from the inquiring turn. In Excerpt 3, a doctor and a pregnant woman are getting ready for a physical obstetric examination. Doc asks about the patient’s own observations of the baby’s movement as compared to the last visit (line 2 ainda, ‘still’ moving a lot), thus treating it as locally new information. Excerpt 3. other-than-é repeat, See (POSTO020806EEliana 00:01:54) 01
DOC:
02
DOC: -> tá se mexendo bastante a↑inda; be(estar).3SG REFL move.GER much still is {she/he} still moving a lot
03
PAT: => tá; be.3SG is
04
(18.0)
05
DOC:
°vamo vê como é que° tá esse nenê aqui_ let’s see how is this baby here (1.0) ((Pat walks towards examination table))
°(ele vai crescê mais ainda)° he is going to grow even more
Doc’s question contains the existential verb tá ‘is’ (estar, temporary existence). In line 3 Pat responds with a minimal repeat of the verb tá ‘is’ in the question. The
Chapter 3. Responding to polar questions in Brazilian Portuguese
response provides the requested information and indicates nothing exceptional, which Doc treats as sufficient and unproblematic by proceeding in the routine activity. In this study, we report our findings on the types of responses implemented by é both as a particle and as a repeat vis-à-vis other types of repeats in their sequential and action environments. We will argue that the use of the é-particle (and possibly the é-repeat) as a locally sufficient, positive answer that accepts the terms of the question and the assumptions encoded in it (cf. Raymond 2003) is specialized and limited to confirming information that the participants treat as already present and in some way established in the conversation. This happens in Excerpt 2 in the simple clarification of understanding of prior talk. By comparison, é-responses seem insufficient in contexts where participants orient to the confirmable matter as new – i.e., not having been previously addressed or sufficiently established – thereby inviting a more committed confirmation of this matter being true and locally relevant for the ongoing activity. Most importantly, é-responses do not affirm new information. Instead, speakers tend to provide new information or a more committed confirmation by using minimal repeats (as in Excerpt 3). In the following, we first examine é-particle and é-repeat responses, and then the repeat responses other than é. We begin from environments in which the response alternatives are used as freestanding polar responses, i.e., as a single element in the turn. These uses are then compared to responses further elaborated by the speakers. É-responses and repetitional responses tend to occur with distinct types of elaboration and action trajectories, which provide evidence of the participants’ orientations to the kinds of epistemic and action commitments that each response type conveys. Before proceeding to the analyses, we introduce the linguistic resources for asking and answering in BP, and present our data.
2.
Polar questions and responses in Brazilian Portuguese
2.1 Question design Polar questions in BP have no interrogative morpho-syntax. Instead, speakers use other turn design features, such as prosody and tags (né; não é), to compose turns that get treated as implementing a polar question. Polar questions may have an intonational marking with a rising final pitch or a pitch peak at an earlier point of the utterance (Moraes 1999), but not much is known about BP interrogative prosody in naturally occurring interactions. As in other languages, many declara-
79
80
Katariina Harjunpää & Ana Cristina Ostermann
tively formatted utterances invite confirmation on a pragmatic basis, mainly epistemic domains (Labov and Fanshel 1977; Pomerantz 1980; Couper-Kuhlen and Selting 2018: 181–182; Heritage 2012; Bolden 2016; Sorjonen 2001a: 58–72). The design of these turns in BP forms a cline from asserting to asking (Levinson 2012), on which requests for information and confirmation are not distinguished by syntactic or morphological criteria (nor consistently by prosody). An exception, however, is the deployment of tags (especially né; não é), which display the speaker’s assumption that the state of affairs prevails and thereby characteristically invite confirmation from the more knowledgeable participant rather than genuinely seek new information (e.g., Heritage and Raymond 2012).
2.2 Response variants Linguistic research on BP has traditionally held the view that speakers provide affirmative answers to polar questions mainly by repeating an element (e.g., verb, adverb, noun) from the question (Armstrong 2008; Urbano et al. 1993; Guimarães 2007; Santos 2003; Kato et al. 2023). Other response variants, in addition to é-responses, include the particle sim (‘yes’), the particle-like isso (dem2, lit. ‘that’), vocalizations such as mm-m and ãrrãm, and of course, embodied responses. Prior research on responses in BP is mostly based on the NURC1 sociolinguistic interviews archive. Another research branch focuses on European Portuguese (EP), a variety that differs in many ways from BP – in particular, in the (infrequent) use of é-responses. Repeats are argued to be the default, unmarked response to polar questions, in opposition to sim, both in BP (Urbano et al. 1993; Guimarães 2007; Armstrong 2008) and EP (Kato et al. 2023). The particle sim is said to convey weak commitment in EP (Kato et al. 2023: 430), and to be more frequently used in EP as compared to BP, a variety of Portuguese in which é plays a more central role “in the evanescence of sim answers” (ibid.: 28). According to a variationist study on everyday talk in BP (Rosemeyer & Schwenter 2019), repeats occur in responses to actions with a steeper epistemic gradient and show a stronger proof of alignment between interlocutors’ common ground than é-responses. In her conversation analytic study of police interviews in BP, Guimarães (2007) finds that sim is used “for cause,” showing problems with the question or the implicated next action, such as displaying that the information should already be known to the questioner. É, on the other hand, responds unproblematically to phrasal and some clausal questions and displays agreement with assessments and statements seeking agreement (Guimarães 2007: 109). In our data é is also used 1. Projeto da Norma Urbana Oral Culta (Educated Urban Speech Norm Project).
Chapter 3. Responding to polar questions in Brazilian Portuguese
for agreement, although we do not discuss assessment sequences here. As Armstrong (2008: 294) points out, in confirming propositions “activated” and “recoverable” in the discourse, such as phrasal candidate understandings and agreement displays, the é-response bears resemblance to nii in Finnish (Sorjonen 2001a). More recently, Enfield et al. (2019) have discussed the BP response alternatives in a quantitative study across 14 languages. The occurrence of é both as a particle and a verb repeat influences the analysis of the two main groups compared: repeats and “interjection” responses (i.e., particles, minimal vocalizations, and embodied responses). É was coded as a repeat when it had a possible repeat source in the question and otherwise as a particle. “Interjections” constituted only 63% of affirmative responses in BP in comparison to over 80% in 11 of the 14 languages examined, thus making repeats (minimal and full) relatively frequent in BP within the set of languages studied. However, after considering the type of data used in prior studies on BP and the ambiguous status of é-responses, Enfield et al. (2019) conclude that BP is not far from the overall response pattern found across the languages. Enfield et al. (2019) suggest – in contrast to earlier linguistic studies on BP – that “interjections” constitute an unmarked answer type across actions and languages (including BP) due to their higher frequency and because they assent to the questioner’s greater “sequential and thematic agency” (i.e., to the questioner setting constraints on the responsive action and thematizing the proposition, ibid.: 299). By comparison, the less frequent repeat responses are said to resist that agency by repeating the proposition, which can, in some languages, be competitive or emphasize shared access to knowledge. In this chapter we shed further light on polar responses in BP by suggesting that both é-responses and minimal repeats (i.e., other than é) have “home” environments in which they occur as structurally-fitted positive answers that align with the action and epistemic implications of the question. We follow Sorjonen’s terminology (2001a: 36; 2001b), in which confirmation refers to answers that treat the information as known to some extent to the questioner, while affirmation refers to answers that provide new information. While é-responses and repetitional responses appear divided between these main jobs (i.e., é-responses for confirmation and repetitional responses for affirmation), they are also used in each other’s environments in specific ways. For example, a repeat (rather than an é-response) may provide a “stronger” confirmation when it is relevant (see also Keevallik 2010; Sorjonen 2001a; b). We will also examine what features of polar questions invite a minimal or elaborated response. With respect to wh-questions, Thompson, Fox, and CouperKuhlen (2015) show that instead of structural type conformity, the preferred response types can be explained by a distinction between questions inviting specification (minimal response) or telling (maximal response). In polar questions,
81
82
Katariina Harjunpää & Ana Cristina Ostermann
interrogative design can invite an elaboration of the response beyond an initial polarity token (Raymond 2010b). In the absence of interrogative syntax in BP, our analyses draw on studies that show how expanded answers are invited in other ways. For instance, expansion can become relevant when speakers request confirmation by partially repeating a prior speaker’s statement and demonstrating that they did not know this information prior to its delivery, when they indicate a knowledge discrepancy, or when they request specification (Steensig and Heinemann 2013). Crucially, declarative questions that concern matters not yet addressed (vs. ones returning to prior talk) have been shown to elicit explanations, accounts, or topical talk (Seuren and Huiskes 2017). The current study will show that when BP speakers treat a confirmable or affirmable matter as relevant for topicalization or expansion, they use a verb repeat (other than é) rather than the é-particle. Although we explore these orientations by examining expanded responses, we suggest that the distinct response alternatives convey different orientations regardless of whether a response is expanded.
3.
The corpus
Our corpus includes two sets of data: (i) video recordings in homes and in a reading group (5 h 10 min), partly the same as in Enfield et al. (2019); and (ii) audio recordings of gynecological, obstetric, and fetal medicine consultations (2 hours), recorded in Southern and South-Eastern regions in Brazil. The analysis in this chapter is based on 91 cases of é-responses and different repetitional responses, which include minimal as well as non-minimal repeats (more than one element repeated). This data subset resulted from excluding from our initial collection of 196 instances all newsmarks, collaborative completions, deontic domain actions,2 and double questions. É-repeat comprises cases in which the question contains é as a copula verb. In addition, the category includes four cases in which the question involves é only as part of the tag né. Despite being a reduced, grammaticalized form of the negative tag question with the actual verb (não é neg be.3sg) (Figueiredo 2015; Ostermann and Almeida 2017), né constitutes a potential repeat source for é-responses.
2. Other question-formatted actions that overwhelmingly take repeats as responses in our data include invitations, requests (e.g., for permission), and offers: A: qué começá ↑lendo, ‘(do you) want to start reading’ B: quero (want.1sg). They are excluded from this study for their deontic rather than epistemic domain of action. Another typical use excluded from the study is é both as the initiating and responsive turn in a news receipt sequence (A: é? B: é).
Chapter 3. Responding to polar questions in Brazilian Portuguese
Table 1. Instances of each response alternative Response form
N
é-particle
31
é-repeat
19
minimal repeat (other than é)
26
non-minimal repeat
15
Total
91
We have excluded from repetitional responses those with modified initial polar elements (Stivers 2005; Stivers and Hayashi 2010), leaving in only those with necessary adaptations for speaker change (e.g., from 2sg to 1sg). This reduces somewhat artificially the role of repeat responses because non-minimal repeats are often somewhat modified due to dynamically progressing ongoing actions and activities.
4.
É-responses
In this section, we first show a basic use of the é-particle as a freestanding confirmation of matters already addressed in prior talk (4.1). Then we move on to showing how the é-particle in other types of environments is treated as not (sufficiently) committing to a positive answer and/or not fully aligning with the sequential implications of the action to which it responds. This is observed in the subsequent elaboration (4.2) or in how the speaker resumes prior action after é (4.3). We have found no major differences in the use of é-repeats in this regard (4.4).
4.1 É-particle as a freestanding response: Confirming already addressed matters The particle é occurs as a freestanding, sufficient, aligning response to a simple other-initiated repair that seeks to confirm a prior reference (as in Excerpt 2) or reconfirm a correction, as in the following excerpt. Môn’s question about the street number of Bea’s apartment – where they currently are – is expanded into a repair sequence.
83
84
Katariina Harjunpää & Ana Cristina Ostermann
Excerpt 4. é-particle, Seventy (Rio2017_Churrasco 01:33:40) 01
MÔN:
02
trezentos sessenta né? three hundred sixty right (0.6)
03
BEA:
04
MÔN: -> se↑tenta; seventy
05
BEA: => é.
06
tre↑zentos se↓tenta. three hundred seventy
(10.0)
Môn checks for the street number while writing Bea’s address in a text message to a guest (line 1). Bea provides a correction, the numeral 370 (line 3) instead of 360. Môn checks the corrected segment by producing a partial repeat setenta, ‘seventy’ (line 4), and Bea confirms it as correct with a freestanding é (line 5). After this, both participants return to their smartphones. The é-particle thus confirms an element that was present in the immediately prior turn, and it is followed by the participants’ return to the main ongoing activity (cf., Keevallik 2010 on particles in “non-primary actions” vs. repeats in “main actions,” and Enfield et al. 2019 on particles in “subsidiary sequences”). In Excerpt 5, the patient uses the é-particle to confirm an already present, shared understanding without involving repair. Excerpt 5. é-particle, A bit shorter (POSTO010806LMonique 00:01:11) 01
DOC:
02
PAT:
03
DOC:
>e a menstruaçã:o, vem todo mês, corretamente,= and the menstruation comes every month correctly
04
PAT:
= it comes every twenty two days in twenty two
05
>°então° va:mos tratá isso aí direitinho< so let’s treat this there properly.
tá:: já:. it’s horrible already
>↑sempre foi assim =↓tá (.) mais curti:nho um pouco_ PRT more short.DIM ART little alright a bit shorter
07
PAT: => °é:°
08
DOC:
>mas tudo bem. é natural [issomh:m ↑n’épor causa da< criança eu acho né, but {it} is because of the child I think né
03
(0.7)
04
DOC:
05
pode sê::. (0.3) >parece que tá pe↑sado< could be seems heavy [né,= doesn’t it
06
PAT:
[>olha aqui.< isso< (.) >ele< s’embola::; look here this he rolls up
07
DOC:
m-hm;
09
(3.6)
10
DOC: -> e o cigarro já parou? and ART cigarette already stop.PST and the cigarette have you already stopped
11
PAT: => ↑é::. ↑eu: nã- eu:: fumava assi:m be(ser).3SG 1SG NEG 1SG smoke.PST ADV é I don- I used to smoke
12
agora eu não fumo mais como eu fumava antes. now 1SG NEG smoke.1SG anymore like 1SG smoke.PST before now I don’t smoke anymore like I used to before
13
hhh (.) me dá: enjô::o essas coisas. it causes me nausea these things
14
(0.5)
15
DOC:
por que não aproveita pra pará_ why don’t you use the chance to quit
16
PAT:
é:: eu vou pará: si:m ma:s_ é I will quit yes but
Doc’s turn in line 10 displays his knowledge that Pat smokes and the expectation that quitting is desirable or even expected (‘have you already quit’), but the interrogative prosody with a final rising pitch indicates Doc does not treat this as assumed. The turn sets a preference for affirming this information as a new state of affairs, in contrast to the situation before. In line 11, however, Pat responds to Doc with a turn that starts with the particle é, produced with a high prosodic onset and vowel lengthening, which together indicate some hesitance, or even make it hearable as tilting towards negative polarity (Jefferson 1978). The rest of
Chapter 3. Responding to polar questions in Brazilian Portuguese
the response modifies the assumptions about her smoking and demonstrates only some progress toward the goal. This and Doc’s subsequent question (line 15) show that the matter is not actually affirmed: Pat has not (yet) quit smoking. The following excerpt shows a request for confirmation with a final rising pitch receiving an expanded é-particle response (line 9). The pregnant woman in a fetal medicine clinic checks whether everything else in her child’s situation – besides the need to have the fetus’ navel hernia examined by a surgeon – is ‘normal.’ The doctor responds with an é-particle followed by a lengthy explanation. Excerpt 7. é-particle, Normal (ACONGEN_aline_JEFERSON_07_01_14 00:45:01) 01
DOC:
02
03
tá? que daí: vai- vai se:r >aí< (.) alright? ‘cuz then it wil- will be bem importante tá:.= very important alright
PAT:
04
=mhm, (0.5)
05
DOC:
06
PAT:
tá .h e e:: alright .h and and
07
DOC:
°°(>daí é isso fora ↑i::sso::↑ (.) ↑(tudo) nor↑mal, PREP that all normal except for that (everything) normal
10
DOC: => ^é. a::; >como eu já falei< tem cuida- al↑guns é uh like I already said there are mea- some
11
cuidados que a gente vai ter que ter. .hh ã: .h measures that we will need to take uhm
12
ã: (.) >como eu lhe falei< ↑pode >por exemplo< uhm like I told you there can for example
13
(>um< problema) oftalmológico an ophtalmological problem
//6 lines omitted ((Doc explains possible problems)) 20
DOC:
como ele tem uma chance um ↑pouco maior que (-) as he has a bit higher chance than
87
88
Katariina Harjunpää & Ana Cristina Ostermann
21
ter do que outras crianças em ge↑ral? to have {it} than other children in general
22
PAT:
mhm,
23
DOC:
.h a gente fa:z o e↑xame: (.) pra ver ↑se te:m. we do the exam to see if {he} has
24
PAT:
25
DOC:
26
27
PAT:
28
DOC:
29
PAT:
30
DOC:
[mhm, [se a gente identificá >°alguma coisa a gente if we identify something we já tenta tratá°parteligaa cento< oitenta even to hundred, eighty
11
CAI:
12
AUR:
[↑ah °(então) não (deu nada)° ah so (it’s not a big deal) aí eu parei o car↑ro e na hora que (a gente so/then I stopped the car and at the moment we
13
abriu vi-) a< opened sa- the
14
CAI: -> boil.3SG {it/the thing} boiled
15
AUR: =>
16
AUR:
[é. é
[aí a< tampa ta↑va ali;] so/then the cap was there
Chapter 3. Responding to polar questions in Brazilian Portuguese
17
CAI:
18
[↑é; claro que a pa] é é of course since the p- é porque .hh parô de circular né aí ele ferve because {it} stopped circulating right so it boils
19
ime- (-) i↑mediatamente. imeimmediately
19
AUR:
20
CAI:
21
AUR:
(ngn) (.) [(-)
[mas ↑aí cê conseguiu voltar. but so you managed to return
>não então< aí Matias colocou água? well no then Matias put water ((continues telling))
Right after Aur has announced the news to preface her telling (lines 1, 3), Cai presents for the first time his candidate diagnosis of the problem with the car (line 6). Aur receives this only with a nod, and goes on to tell what is currently known about the damage. Later on (omitted segment), Cai asks whether the car actually stopped running. After disconfirming this, Aur engages in an extended telling of how they discovered something was wrong with the car. At a point when Aur is about to deliver the climax of the story (discovering that the radiator cap was off ), Cai self-selects to offer, for the third time, the conclusion that the radiator ‘boiled’ (line 14), with a confident voice (lengthening, step down in pitch on the second syllable). Aur confirms this minimally with an é-particle (line 15), but then, in a separate prosodic unit (line 16), resumes her telling that had been cut off by Cai’s turn, while Cai explicates his reasoning in overlap (lines 17–19). Aur does not take up Cai’s diagnosis in the ongoing activity, thus contrasting with Cai’s insistence on it. The é speaker may be seen as resisting to validate the candidate diagnosis at this point of her telling, in which the diagnosis short-cuts the telling by anticipating its conclusion. The case illustrates the use of the é-particle response when the questioner has displayed a “confident” understanding (Heritage 1984: 320–323) of an aspect related to, and possibly inferred from, the prior talk but not made explicit in it. The questioner’s independent conversational contribution is based on their own prior or currently formed understanding, and it may consist of “ancillary matters” (Jefferson 1984) in that inquiring about them shifts the discussion away from where the previous speaker was going (e.g., Antaki 2012). The é-particle seems, on the surface, a sequentially adequate confirmation, yet it conveys something less than what the speaker could do to align with the implications for the next action, thus indicating that the action responded to is not relevant for advancing the ongoing activity.
91
92
Katariina Harjunpää & Ana Cristina Ostermann
In sum, the é-particle appears to be in its home environment in confirming matters that both participants treat as already present in the interaction. When it occurs in environments where the questioner indicates the need to establish something new here and now – inviting the more knowledgeable, responding speaker to commit to a previously unaddressed matter by validating its content and/or its relevance for the activity underway – the é-particle response comes across as not fully confirming it and/or as treating it as ancillary to the ongoing activity. It does less than what the question would have enabled in this response slot.
4.4 É-repeat After having presented patterns in the use of the é-particle, we now turn to uses of é-repeats. Similarly to the é-particle, é-repeats appear as fully confirming, unelaborated responses in subsidiary sequences when the confirmable matter has already been dealt with in prior talk, as in the following segment, Excerpt 9, line 14. The excerpt also contains an é-repeat (line 7) elaborated by a more substantial repetition. During an obstetric consultation, a pregnant patient asks about sensations in her abdomen (lines 1–5). Excerpt 9. é-repeat, Hardening (POSTO030506Bjucilda 00:00:52) 01
PAT:
e o >qué é que eu< ia pefalá: é norma↑:l esse and what I was going to as- say is {it} normal
02
negócio de:: (1.2) de vez em quando dá uns this thing of sometimes having some
03
endurecimento na barri:ga_ (0.3) forte dói, (0.4) hardening in the belly strong hurting
04
.hh aí pára (.) daí eu vou pra contá no no relógio then {it} stops then I count in the clock
05
assim passa e não dá ma:is (né); so {it} passes and doesn’t start again (right)
06
(1.4)
07
DOC:
^é::. ↑é norma::l↑ queri:da:_ tu não pode tê isso é {it} is normal dear you can’t have that
08
aí: (.) é: seguido uma atrás da outra que não passa uh uh one after another that doesn’t pass
09
né: aí seria um trabalho de parto prematuro mas:= right then it would be entering premature labor but
10
PAT:
=mhã,
Chapter 3. Responding to polar questions in Brazilian Portuguese
11
DOC:
12
eventualme::nte assi:m tu tê: you having {it} in that occasional manner ↑isso não ↑tem pro[ble:ma.] that’s not a problem
13
PAT: ->
14
DOC: => é↓:a.
15
(9.0)
[ >é] norma:l< né; {it} is normal né
Doc elaborates the first confirming é-repeat (line 7) with é normal, delineated from the elaboration by forming its own intonation contour. In the elaboration, Doc repeats more of the question (é normal ‘is it normal’), possibly due to the distance between the question and the beginning of Pat’s turn (line 1), which makes the first é in Doc’s response less recognizable as a repeat from line 1. The elaboration by a more substantial repeat then serves to strengthen the initial é in a position where é alone might not sustain confirmation. (The initial é thus comes to resemble the insufficient é-particle responses discussed in Sections 4.2 and 4.3.) The repeat is followed by Doc’s specification that verifies Pat’s sensations to be, in fact, normal and unproblematic in her situation. Our focus sequence occurs after ‘normality’ has already been established. Pat requests confirmation again (é normal né, line 13) in a summarizing turn that draws directly from Doc’s prior talk, without bringing any independent new contribution or suggesting expandability of the information (cf. Seuren and Huiskes 2017). Despite the slight overlap of the Pat’s é on line 13, both é tokens (lines 13, 14) are clearly hearable, allowing the identification of the latter one as a repeat. The é-repeat thus confirms an already established and shared matter, and closes the discussion. In certain contexts, however, é-repeats can treat a confirmable matter as not quite accurate or as ancillary regarding the ongoing activity, similarly to what was shown for é-particles. This is illustrated by Excerpt 1 (reproduced from the Introduction), where the é-repeat is used to confirm the repair initiation. Excerpt 1. reproduced 01
NAR:
02 nar 03 04
?:
página::, page >>Bea&Joa leaned over smartphone
+(0.3) +turns paper page-> s- s (0.4)
93
94
Katariina Harjunpää & Ana Cristina Ostermann
05
BEA: nar
página dois +aqui.+ page two here ->+.....+
06
NAR: -> é ↑dois? is {it} two
07
(0.4)
08
BEA: => é.
09 10
(0.4) NAR:
a: é verdade >aqui ach’ que tá (o contrário tem f↑i:lho dona↑ melissa. have.2/3SG child HON ((name)) do {you} have children mrs Melissa
06
PAT: => te::nho; have.1SG {I} have
07
(3.7)
08
DOC:
09
PAT:
10
11
DOC:
12
PAT:
partos nor↑ma:is; natural deliveries i::sso. that that’s right (2.4) quantos fi:lhos_ how many children do:is_ two
95
96
Katariina Harjunpää & Ana Cristina Ostermann
Questions regarding previous pregnancies typically occur in an earlier phase, during history taking, but in this consultation, the topic has not yet been raised. Since the information is relevant for the investigation of Naboth cysts, it is elicited later, during the physical exam of Pat’s cervix. The question–answer sequence (lines 5–12) consists of a series of questions in which each new response allows the doctor to move forward to the next (routine) question about children and childbirth (cf. Heritage 2010). In line 6, Pat provides a response to Doc’s question with the verb repeat tenho, deictically adjusted from the 2nd/3rd-person singular form in the question to the 1st-person singular form. The repeat provides information from which a next question relevantly follows, whereas isso, ‘that(’s right)’ in line 9 confirms Doc’s assumption about normal deliveries in a way that emphasizes he has formulated the correct conclusion and that this matter does not need to be opened up by further questions (see also line 33 in Excerpt 7). By producing only minimal responses, Pat orients to the institutional activity of Doc proceeding in the list of questions, and by continuing in the list of questions, Doc treats the responses as uneventful. In everyday conversation, a question like ‘do you have children’ would normally invite more talk. Minimal repeats in our data typically show an orientation to the progression of some main activity, from lists of questions to multimodal activities (see also Excerpt 3). It appears that the use of repeats in responding to requests for new information (i.e., questions with a steep epistemic gradient, see Heritage 2012) is reflected in the use of repeats for confirmation in contexts that make relevant a response in which the speaker epistemically commits to, validates, and/or expands on the confirmable matter. Such contexts were already examined in Section 4.2, where the “weak” é-particle was used as an insufficient response, and now we turn to cases that receive a “strongly” committing, expanded repetitional response (other than é).
5.2 Repeat expanded: Committing and specifying Minimal as well as non-minimal repeats in BP occur as aligning responses to requests for confirmation that deal with matters which have not yet been (sufficiently) established in the current discussion or do not match the questioner’s earlier knowledge, thus occasioning more committed and substantial confirmation (e.g., Steensig and Heinemann 2013). (Repeats have also been shown to validate an interlocutor’s contribution made from a relatively knowing stance in the case of candidate understanding and collaborative completion, Schegloff 1996; Oloff 2014). Excerpt 11 shows how information that departs from the prior conversation is confirmed as true, elaborated, and integrated in the conversation. The partici-
Chapter 3. Responding to polar questions in Brazilian Portuguese
pants are reminiscing about their joint experiences as exchange students in Germany. In line 7, Môn announces her trouble in remembering at which party the events they are talking about took place. Excerpt 11. minimal repeat, Festa (Rio2017_Churrasco 00:06:03) 01
FER:
e o Amos +também assi::m, and Amos also like this +dance movements w arms->
02
MÔN:
é [eu lembro dessa festa. foi festa do f-= é I remember that party it was the party of e-
03
FER:
04
MÔN: => cê lembra no fi↑nal.< you remember in the end
05
(0.4)
06
HEL:
07
MÔN:
08
09
[é::+ o Amos é aquele (deus) holandês; ((to BEA)) é Amos is that dutch (god) ->+
não [me lembro disso. I don’t remember that [>(mas) eu tava co-< eu tava confun°dindo com° (but) I was coI was confusing {it} with -> carnaval.=↑teve um carna↑val tam[bé:m lá, carnival be/exist.PST ART carnival also ADV carnival {there} was a carnival too there
FER: =>
[↑teve que be/exist.PST REL {there} was as
10
a gente foi pro bar irlandês; ART we go.PST PREP.ART bar Irish we went to the Irish bar
11
(0.5)
12
FER:
13
o irish pub. (1.4) ((Mon looking away, touching her chin))
14
MÔN:
15
FER:
é: acho que é desse dia que eu tô lem[brando. é I think that it’s that day I’m recalling
16
com aquela menina da Austrália. with that girl from Australia
[com aquele< with that
mas eu tô tendo to be there with Ana and .h but I’m having
04
impulsos< de volta. Beata foi quem me[:: an impulse to return Beata was the one who me
05
NAR:->
06
[e and ainda e↑xiste o grupo de lei↑tura; still exist.3SG ART group PREP reading does the reading group still exist
07
BEA:
08
MAR:=>
09
NAR:
e[xiste_ exists [existe né, .h só que é num horário m:uito complexo; exist right {it} is just at a very complicated time qual q[ue é. which is {it}
Chapter 3. Responding to polar questions in Brazilian Portuguese
10
MAR:
[umas seis da tarde:; about six in the afternoon
In lines 1–4, Mar tells that she ended up not attending the meetings anymore after having a child, but that she is now interested in joining again. Although this implies the group is still active, Nar’s turn in lines 5–6 treats the matter of whether the very same reading group still exists as not established in the talk so far. It is first confirmed by Bea, a current member of the group, by repeating the verb existe, ‘exist.’ The main recipient, Mar, does the same almost simultaneously and then elaborates the response, beginning with só que ‘it’s just,’ to reveal a potential obstacle to attending the meetings: it is difficult to reach the location at that time (due to the traffic jams in the city). Thus, instead of treating Nar’s checking of the group’s existence as irrelevant or ill-fitted (in contrast to, e.g., the é-response in Excerpt 8), the repeat responses by the two speakers commit to the fact of the group’s existence and accept topicalizing it. Moreover, Mar‘s elaboration of the concrete circumstances of accessing the group relevantly anticipates an underlying line of action related to Nar’s interest in the group. In later talk, Nar is invited (and accepts) to attend a Sufi meeting with them; Mar’s elaboration has provided relevant information in this regard. As illustrated by Excerpts 11 and 12, the initial repeat-formatted confirmation holds even when it is elaborated by further details. The specification can anticipate (or treat as problematic, cf. Heritage and Raymond 2012) a line of action that the questioner could be seen as taking in asking the question, yet without compromising the confirmation as such (Sacks 1992: 415). By structurally building the elaboration on the repeat, the speaker can display commitment to the proposition introduced by the questioner while launching further talk that makes the information coherent and compatible with regard to locally relevant circumstances. By contrast, the é-particle does not seem fit for these types of orientations, as shown in Section 4.2, and may reject the assumptions involved in the prior action in a more serious (and insidious) manner than repeats. In sum, the repeat displays commitment to the information epistemically (as true) and in terms of action (e.g., telling more), whereas é-responses (é-particles and é-repeats) tend to align weakly with the accuracy and local relevance of the matter or even avoid straightforward positive answering. In the next analytic section, we discuss further the role of repeats in BP in light of a final case.
5.3 A further look at what repetitional responses convey The analyses presented thus far suggest that BP speakers use minimal repetitional polar responses to commit to the accuracy of the information while possibly elaborating on it. In this section, we discuss repeats in an environment in which
99
100
Katariina Harjunpää & Ana Cristina Ostermann
the participants insist on their differing views and invoke issues of responsibility. With this case, we wish to further relate the findings of this chapter to the crosslinguistic discussion of agentive and epistemic displays in repetitional responses. In Excerpt 13, a doctor and patient have a different understanding of what exams the patient was supposed to have taken. Pat has brought the exam reports for Doc to see, and in line 1, Doc claims the absence of the obstetric ultrasound exam report. Excerpt 13. minimal repeat, Ultrasound (POSTO030506BJucilda 00:00:17) 01
DOC:
02
(>°c’dêawhere is ↑foi pedi:do ↓e:co, be(ser).3SG.PST ask.PTPC scan was {a} scan requested
06 07
(0.3) DOC:=> f↓o::i, be(ser).3SG.PST was
08
(1.9)
09
DOC:
10
PAT:
°foi [pedido a e:co°.] be.3SG.PST ask.PTPC ART scan the scan was requested [eu vo::u fazê:: ]: sema:- >essa semana I will do {it} we- this week
11
agora que tem marcado a uma< em alto largo (.) now that there’s an {appointment} in Alto Largo
12
.hh >que eu achei que não tinha pedido eco< as I thought that a scan had not been requested
13
DOC:
ti::nha, have.3SG.PST {it} had
Pat’s open repair initiation in line 3 displays a problem of hearing or understanding Doc’s request for the exam report. After Doc reissues the request (line 4), Pat questions the grounds for making the request by displaying she was not aware that an ultrasound exam had been requested, and thus that she was expected to bring
Chapter 3. Responding to polar questions in Brazilian Portuguese
its report (line 5). By doing this, she treats the matter as not established in prior talk, and also deflects the blame for the missing report away from herself.4 Pat’s turn is prosodically marked, showing skepticism. What does the minimal repeat in line 7 convey as a response to this action? Especially in English, repeat responses have been regarded as a marked response format to various actions implemented by polar questions, involving an inherent form of resistance to the questioner’s agentive or epistemic rights concerning the matter asked about or the fact of having uttered it (Stivers 2005; Enfield et al. 2019; Enfield and Sidnell 2015), and as leverage for disagreement (Heritage and Raymond 2012). An argument along those lines would perhaps suggest that Doc is using the repeat to push back against Pat’s action (e.g., her right to question whether the report was legitimately expected) by means of independently asserting that the exam had been ordered. However, in our analysis of BP we have not found empirical grounds for saying that a minimal repeat, in itself, does any extra work beyond its main use of affirming/confirming previously unestablished information – this applies also for the epistemic and deontic tensions in Excerpt 13. When Doc responds with a minimal repeat at line 7, he aligns with Pat’s orientation to the earlier request as unexpected. By responding with a minimal repeat, Doc treats the matter as having become newsworthy and in need of reinforcement (cf. Sorjonen 2001b: 416). That is, the use of the minimal repeat here coheres with the uses seen in the earlier sections – orienting to the novelty, tellability, or veracity of a piece of information as projected by the question. Instead, other features in the responsive action contest the question: the turn’s emphatic prosodic design through stress and lengthening (f↓o::i, line 7),5 which conveys some discordance, and Doc’s lack of elaboration. Pat’s question displays a discrepancy between her earlier knowledge and what Doc treats as given (i.e., that the exam had been referred). To adequately respond to such a question, the responder needs to not only confirm but also address the knowledge discrepancy (Bolden 2016: 47; Selting 1996; Steensig and Heinemann 2013: 222). Doc, however, does not provide any account or explanation to bridge the gap in their views. By instead continuing (after a silence of 1.9 s) with a full repeat (line 9), Doc implies
4. This case (line 5) shares the common feature of repair initiations receiving repeat confirmations in our data: the repair is prolonged prior to the sequence with the repeat. Confirmations with minimal repeats mark the joint arrival at a final solution, or, when elaborated, provide further information concerning the complex trouble source. 5. For the prosody of minimal answers conveying resistance to an initiative action, e.g., disregarding its action implications, or its evaluative or epistemic stance, see Bolden (2016) and Raymond (2010a).
101
102
Katariina Harjunpää & Ana Cristina Ostermann
there is nothing to explain, thereby “sticking to his guns.” Notice also that the syntax of the question and its response (lines 5 and 9) is the same apart from Doc adding the definite article on ‘scan’, thereby changing to a form that treats the exam report as assumed. His action differs from the cooperative, specifying elaborations after repetitional responses seen in the earlier excerpts. Yet, similarly to the earlier analyses, the minimal repeat itself orients to the confirmable matter as a piece of information that needs to be “created” (Sorjonen 2001a: 53–56) here and now. An é-particle response in such an environment could rather bypass the questioner’s attempt to bring up the matter.
6.
Discussion
This study has shown that the core use of the é-particle as a freestanding, sufficient response is in providing confirmation of something in the prior talk, for instance, clarification of reference. In our data, the é-particle does not occur alone as an affirmative response to requests for information. In this environment, it can occur as an initial response token anticipating that the speaker does not fully commit to those facts in the subsequent talk. In a similar vein, when the é-particle is used in responses to requests for confirmation that, instead of simply dealing with something in the prior talk, put forward something new (e.g., matters departing from the prior line of talk, ‘confident’ candidate understandings), the unfolding of the interaction characteristically shows that the é speaker does not actually commit to confirming or to handling the issue as relevant. The observed uses of é-repeats are similar to é-particles in that they also display sufficient and closing-implicative confirmation when the question returns to an already established or obvious understanding, and when the question makes relevant something more, they provide weak confirmation. In these other environments, repetitional responses (other than é) are employed for more committed affirming/confirming. The findings are summarized in Table 2: Table 2. Environments of the response variants Response type
Home environment
é-responses (particles and repeats)
Requests for confirmation
Requests for information (rare) and requests for confirmation orienting to novelty: Weak commitment
minimal repeats
Requests for information
Requests for confirmation orienting to novelty: Relevantly strong commitment
Other environments
Chapter 3. Responding to polar questions in Brazilian Portuguese
As shown in Table 2, minimal repeats (of elements other than é) do the work of affirming new information in our BP data.6 Moreover, when responding to requests for confirmation that treat the matter as newly addressed, repeats can provide confirmation that validates the confirmable matter and accepts incorporating it in the local action or ongoing activity. With the grammatical resources of the repetitional response, speakers take on the responsibility to establish that piece of information for the interaction and make a strong enough “morphosyntactic commitment” (see Thompson, Fox, and Couper-Kuhlen 2015: 155, 316 on assessments) to it, thus fulfilling what the first action asks for. This warrants providing specification or explanation, topicalizing the matter, and otherwise treating it as relevant to the ongoing action or activity. Our findings indicate that the use of minimal repeats for confirmation in BP is primarily linked to their functioning elsewhere as answers to requests for new information. In other words, some requests for confirmation are closer to requests for information – not with regard to the questioner’s knowledge status, as they can display a high degree of knowledge and confident understanding, but to the type of response expected (e.g., the possibility to topicalize and expand the information). These orientations are present in the responder’s choice between response alternatives: é-responses and repeats orient to varyingly grounded local expectations for the knowledgeable party to merely “check the box” vs. validate and elaborate on what is asked. Our analysis of BP thus shows a somewhat different picture in relation to studies that have discussed repeats as conveying epistemic or agentive displays that challenge the constraints imposed by coming ‘second’ in a sequence (in addition to Enfield et al. 2019, for English, see, e.g., Raymond 2003; Heritage and Raymond 2012; Stivers 2005; Enfield and Sidnell 2015; for Estonian, see Keevallik 2010: 297; and for Russian, see Bolden 2016, this volume). One of the main arguments has been that when responding speakers utter the proposition that was contained in the question, they are actually asserting it rather than simply assenting to it. Yet, in this regard, there are crucial differences between languages. Languages such as Brazilian Portuguese and Finnish (among others) allow answering by recycling the prior turn very minimally, typically by repeating only a copula or an auxiliary verb (‘is’, ‘was’, ‘can’). These minimal repeats are sequentially and syntactically fitted to the responsive position (Laury 2018; Kato et al. 2023), leave most of the source structure (e.g., word order) in the question intact (see 6. The comparative study by Enfield et al. (2019: 292) also reports that repetitions in all the examined languages are relatively most frequent as responses to non-declaratively formatted requests for new information, although no overall sequential, action, or morpho-syntactic context was identified where repetition-type answers were more likely than interjections.
103
104
Katariina Harjunpää & Ana Cristina Ostermann
Hakulinen and Sorjonen 2009; Sorjonen and Hakulinen 2009 for assessments in Finnish), and do not reassert much from the proposition contained in the question. They can be regarded as formally more dependent on the initiating action than, for example, an ‘it is’ response in English (to is it x?). These issues complicate drawing cross-linguistic conclusions about asserting vs. assenting to a proposition. Brazilian Portuguese and Finnish (and some other languages) offer other types of resources for speakers to adjust their responses, including the minimality–maximality continuum of the repeat (Hakulinen 2001), the combination and (word) order of responsive elements in the turn, and prosody. BP and Finnish are pro-drop languages, which also makes subject (and object) inclusion/exclusion a resource available for epistemic displays (Armstrong 2008; Hakulinen and Sorjonen 2009). Despite minimally re-asserting the matter asked about, repeats do seem to provide “more agentive” (cf. Heritage and Raymond 2012; Enfield et al. 2019) responses in the sense that the speaker acts as the one who can affirm and elaborate on the accuracy of the proposition. Similar to findings on English, repeats respond to actions that invite or “mandate” a “greater degree of commitment to a future course of action”, for which a particle response could “imply insufficient agency and commitment to a course of action being assented to.” (Heritage and Raymond 2012: 187). However, whereas the “more agentive” repetitional responses in English have been found to claim “more epistemic rights over the information required than the original polar question conceded” (ibid., 185), in responses to polar questions in BP, repeats seem to claim just the appropriate amount of rights for continuing the action in a way that the question warrants. Thus, BP repeats appear more related to the aspect of commitment: the speakers show commitment to taking the expected knowledgeable role and to engaging with the course of action. Here our findings align with Thompson, Fox, and Couper-Kuhlen (2015: 310), who found no evidence that minimal or extended clausal repeats as agreeing responses to assessments in English show epistemic primacy, and concluded that “agency and symmetry of commitment” are stronger factors in the choice of grammatical format. In sum, the speakers’ selection of a repeat over an é in a polar response does not seem to be a marked epistemic display but one that satisfies the norms of responding, which differ between sequence types and sequential environments (Thompson, Fox, and Couper-Kuhlen 2015: 272). Naturally, differences between requesting information/confirmation and other actions discussed in the literature on repeats also bear on the conclusions to be made about repetitional responses. In requests for information/confirmation the recipient’s greater agentive/epistemic role is sequentially implicated (Heritage 2012), which would seem to reduce the need to specifically claim epistemic or
Chapter 3. Responding to polar questions in Brazilian Portuguese
agentive rights, in comparison to, e.g., assessments (and despite the general “lead” in thematizing something in the first position, Enfield et al. 2019). Finally, this study has aimed at refining our understanding of the uses of é-responses and repeat responses in Brazilian Portuguese, proposing not to treat repeats as a marked response alternative across contexts but as having a division of labor with é and other response formats according to the type of action and sequence. Further investigation of naturally occurring BP interactions is necessary regarding the grammatical, sequential, and prosodic composition of polar questions, how it ties to what constitutes an unproblematic answer to those actions (Thompson, Fox, and Couper-Kuhlen 2015: 284), and the answerers’ resources for additional epistemic work. Regarding é as a particle vs. repeat, further investigations could reveal differences in terms of sequential contexts, repeat sources, as well as prosodic variants of é.
Acknowledgements We would like to thank the editors and reviewers of this volume for their most valuable critical and constructive feedback and suggestions, and to CNPq (Process 307527/2022-0) for funding the second author. All remaining issues are of course our own responsibility.
References Antaki, Charles. 2012. “Affiliative and Disaffiliative Candidate Understandings.” Discourse Studies 14 (5): 531–547. Armstrong, Meghan E. 2008. “Pragmatic Restrictions on Affirmative Response Choice in Brazilian Portuguese.” In Selected Proceedings of the 10th Hispanic Linguistics Symposium, ed. by Joyce B. de Garavito and Elena Valenzuela, 288–299. Somerville, MA: Cascadilla Proceedings Project. Bolden, Galina. 2016. “A simple da?: Affirming Responses to Polar Questions in Russian Conversation.” Journal of Pragmatics 100: 40–58. Couper-Kuhlen, Elizabeth, and Margret Selting. 2018. Interactional linguistics. Studying language in social interaction. Cambridge: Cambridge University Press. Enfield, Nick J., and Jack Sidnell. 2015. “Language Structure and Social Agency: Confirming Polar Questions in Conversation.” Linguistics Vanguard 1 (1): 131–143. Enfield, Nick J., Tanya Stivers, Penelope Brown, Christina Englert, Katariina Harjunpää, Makoto Hayashi, Trine Heinemann, Gertie Hoymann, Tiina Keisanen, Tiina, Mirka Rauniomaa, Chase W. Raymond, Federico Rossano, Kyung-Eun Yoon, and Stephen C. Levinson. 2019. “Polar Answers.” Journal of Linguistics 55 (2): 277–304.
105
106
Katariina Harjunpää & Ana Cristina Ostermann
Figueiredo, Giacomo. 2015. “Uma Descrição Sistêmico-Funcional dos Marcadores Discursivos Avaliativos em Português Brasileiro: a Gramática das Partículas Modais.” Alfa: Revista de Linguística 59 (2): 281–307. Guimarães, Estefania. 2007. Talking About Violence: Women Reporting Abuse in Brazil. Ph.D. thesis, University of York. Hakulinen, Auli. 2001. “Minimal and non-minimal answers to yes-no questions.” Pragmatics 11 (1): 1–15. Hakulinen, Auli, and Marja-Leena Sorjonen. 2009. “Designing Utterances for Action: Verb Repeat Responses to Assessments.” In Talk in interaction. Comparative dimensions, ed. by Markku Haakana, Minna Laakso, and Jan Lindström, 124–151. Helsinki: Finnish Literature Society. Heritage, John. 1984. “A Change-of-State Token and Aspects of Its Sequential Placement.” In Structures of social action: Studies in Conversation Analysis, ed. by J. Maxwell Atkinson and John Heritage, 299–345. Cambridge: Cambridge University Press. Heritage, John. 2010. “Questioning in Medicine.” In “Why do you ask?”: The function of questions in institutional discourse, ed. by Alice Freed and Susan Ehrlich, 42–68. New York: Oxford University Press. Heritage, John. 2012. “Epistemics in Action: Action Formation and Territories of Knowledge.” Research on Language and Social Interaction 45: 1–29. Heritage, John, and Geoffrey Raymond. 2012. “Navigating Epistemic Landscapes: Acquiescence, Agency and Resistance in Responses to Polar Questions.” In Questions. Formal, Functional and Interactional Perspectives, ed. by Jan P. de Ruiter, 179–192. Cambridge: Cambridge University Press. Jefferson, Gail. 1978. “What’s in a “Nyem”?”, Sociology 12 (1): 135–139. Jefferson, Gail. 1984. “On Stepwise Transition from Talk about a Trouble to Inappropriately Next-Positioned Matters.” In Structures of social action: Studies in conversation analysis, ed. by J. Maxwell Atkinson, and John Heritage, 191–222. Cambridge: Cambridge University Press. Kato, Mary A., Martins, Ana Maria, and Jairo Nunes. 2023. The Syntax of Portuguese. Cambridge: Cambridge University Press. Keevallik, Leelo. 2010. “Minimal Answers to yes/no Questions in the Service of Sequence Organization.” Discourse Studies 12 (3): 283–309. Labov, William, and David Fanshel. 1977. Therapeutic Discourse: Psychotherapy as Conversation. New York: Academic Press. Laury, Ritva. 2018. “The Finnish Verb Repeat Response: Its Emergence and Its Nature as a Formulaic Expression.” Journal of Pragmatics 123: 139–150. Levinson, Stephen. 2012. “Interrogative Intimations: On a Possible Social Economics of Interrogatives.” In Questions: Formal, functional and interactional perspectives, ed. by Jan P. De Ruiter, 11–32. New York: Cambridge University Press. Lindström, Anna. 2017. “Accepting Remote Proposals.” In Enabling Human Conduct: Studies of talk-in-interaction in honor of Emanuel A. Schegloff, ed. by Geoffrey Raymond, Gene H. Lerner and John Heritage, 125–142. Amsterdam: John Benjamins Publishing Company.
Chapter 3. Responding to polar questions in Brazilian Portuguese
Moraes, João Antônio de. 1999. “Intonation in Brazilian Portuguese.” In Intonation Systems. A Survey of Twenty Languages, ed. by Daniel Hirst and Albert Di Cristo, 179–194. Cambridge: Cambridge University Press. Oloff, Florence. 2014. “L’évaluation des complétions collaboratives: analyse séquentielle et multimodale de tours de parole co-construits.” SHS Web of Conferences 8, 4e Congrès Mondial de Linguistique Française, 2125–2145. Ostermann, Ana Cristina, and Camila de Almeida Teixeira. 2017. “Small Things that Matter a Lot’: the Deontic Nature of ‘né.” Presentation at 18th AILA Congress. Rio de Janeiro. Pomerantz, Anita. 1980. “Telling My Side: “Limited Access” as a “Fishing” Device.” In Language and social interaction. Sociological Inquiry 50 (3/4): 186–198. Pomerantz, Anita. 1984. “Agreeing and Disagreeing with Assessments: Some Features of Preferred/Dispreferred Turn Shapes.” In Structures of Social Action: Studies in Conversation Analysis, ed. by J. Maxwell Atkinson and John Heritage, 57–101. Cambridge: Cambridge University Press. Raymond, Geoffrey. 2003. “Grammar and Social Organization: Yes/No Interrogatives and the Structure of Responding.” American Sociological Review 68 (6): 939–967. Raymond, Geoffrey. 2010a. “Prosodic variation in responses: the case of type-conforming responses to yes/no interrogatives.” In Prosody in Interaction, ed. by Dagmar Barth-Weingarten, Elisabeth Reber & Margrit Selting, 109–130. Amsterdam: John Benjamins. Raymond, Geoffrey. 2010b. “Grammar and Social Relations. Alternative Forms of Yes/No–Type Initiating Actions in Health Visitor Interactions.” In “Why do you ask?”: The function of questions in institutional discourse, ed. by Alice Freed and Susan Ehrlich, 87–107. New York: Oxford University Press. Raymond, Geoffrey. 2013. “At the Intersection of Turn and Sequence Organization. On the Relevance of “Slots” in Type-Conforming Responses to Polar Interrogatives.” In Units of Talk – Units of Action, ed. by Beatrice Szczepek Reed and Geoffrey Raymond, 69–206. Amsterdam: John Benjamins. Rosemeyer, Malte, and Scott A. Schwenter. 2019. “Echoic and non-echoic confirming affirmative responses in spoken Brazilian Portuguese.” Journal of Pragmatics 141, 80–101. Sacks, Harvey. 1992. Lectures on conversation (vols. 1 & 2). Oxford: Basil Blackwell. Santos, Ana Lúcia. 2003. “The Acquisition of Answers to yes/no Questions in European Portuguese: Syntactic, Discourse and Pragmatic Factors.” Journal of Portuguese Linguistics 2: 61–91. Schegloff, Emanuel A. 1996. “Confirming allusions: Toward an empirical account of action.” American Journal of Sociology 102 (1): 161–216. Selting, Margret. 1996. “Prosody as an Activity-Type Distinctive Cue in Conversation: the Case of So-called “Astonished” Questions in Repair Initiation.” In Prosody in Conversation: Interactional Studies, ed. by Elizabeth Couper-Kuhlen and Margret Selting, 231–270. Cambridge: Cambridge University Press. Seuren, Lucas M., and Mike Huiskes. 2017. “Confirmation or Elaboration: What Do Yes/No Declaratives Want?” Research on Language and Social Interaction 50 (2): 188–205. Sorjonen, Marja-Leena. 2001a. Responding in Conversation: A Study of Response Particles in Finnish. Amsterdam: John Benjamins.
107
108
Katariina Harjunpää & Ana Cristina Ostermann
Sorjonen, Marja-Leena. 2001b. “Simple answers to polar questions: The case of Finnish.” In Studies in interactional linguistics, ed. by Margret Selting and Elizabeth Couper-Kuhlen, 405–429. Amsterdam: John Benjamins. Sorjonen, Marja-Leena, and Auli Hakulinen. 2009. “Alternative responses to assessments.” In Conversation analysis: Comparative perspectives, ed. by Jack Sidnell, 281–303. Cambridge: Cambridge University Press. Steensig, Jakob, and Trine Heinemann. 2013. “When ‘yes’ is not enough as an answer to a yes/no question.” In Units of Talk – Units of Action, ed. by Beatrice Szczepek Reed and Geoffrey Raymond, 207–241. Amsterdam: John Benjamins. Stivers, Tanya. 2005. “Modified repeats. One Method for Asserting Primary Rights From Second Position.” Research on Language and Social Interaction 38 (2): 131–158. Stivers, Tanya. 2019. “How We Manage Social Relationships Through Answers to Questions: The Case of Interjections.” Discourse Processes 56 (3): 191–209. Stivers, Tanya, and Makoto Hayashi. 2010. “Transformative answers: One way to resist a question’s constraints.” Language in Society 39 (1): 1–25. Thompson, Sandra A., Barbara A. Fox, and Elizabeth Couper-Kuhlen. 2015. Grammar in Everyday Talk: Building Responsive Actions. Cambridge: Cambridge University Press. Urbano, Hudinilson, Leonor Fávero, Maria Lúcia Andrade, and Zilda Aquino. 1993. “Perguntas e respostas na conversação (‘Questions and responses in conversation’).” In Gramática do português falado. Vol. III, ed. by Atalíbio Teixeira de Castilho, 75–97. Campinas: Editora da Unicamp.
chapter 4
Responses to polar questions in Polish Matylda Weidner
Kazimierz Wielki University
Using data from naturally occuring Polish conversations this chapter demonstrates how speakers of Polish answer polar questions. It offers a detailed conversation analytic examination of two response formats, particularly in their unexpanded form: (i) particle responses, which are used for confirming the terms and the action agenda of the question and which display the participants’ understanding of the status of the question as subsidiary to a larger course of action and (ii) repetitional responses, which are understood to do more than simple confirmation (e.g., to fullfil requests). By identifying the interactional circumstances for the choice of a particular format of a response, the chapter highlights the importance of social action as key to building and understanding polar question-answer sequences in Polish. Keywords: conversation analysis, Polish, polar questions, responsive action, particles, repeats, social action format, action formation, action recognition, epistemics
1.
Introduction
This chapter offers a conversation analytic investigation of responses to polar questions in Polish and thus contributes to already existing extensive research on questions and answers across typologically diverse languages. Although research on these prototypical adjacency pair types (Schegloff 2007: 13–14) within conversation analysis and interactional linguistics has been abundant (e.g., CouperKuhlen 2012; Couper-Kuhlen and Selting 2018), none of the prior cross-linguistic surveys of questions and answers included a Slavic language (cf., Enfield, Stivers, and Levinson 2010; Enfield et al. 2019). Using conversation analysis, this chapter investigates the use of particle and repetitional responses, taking into consideration their place in the unfolding sequence of action and their relation relative to the polar questions that these two forms are responsive to. More generally, the
https://doi.org/10.1075/slsi.35.04wei © 2023 John Benjamins Publishing Company
110
Matylda Weidner
analyses presented here are aimed at introducing Polish into the expanding forum of comparative cross-linguistic studies of social interaction. The conversation analytic method is well established in research on questionanswer sequences across languages and settings (e.g., Bolden 2016; Boyd and Heritage 2006; Enfield et al. 2019; Hakulinen 2001; Heinemann 2005; Heritage 2002, 2010; Houtkoop-Steenstra and Antaki 1997; Keevallik 2010; Raymond 2003; Raymond and Heritage 2006; Sorjonen and Heritage 1991; Sorjonen 2001a, 2001b). Prior research demonstrated that questions, being the firsts in an adjacency pair, introduce a number of constraints on what responses, or the seconds, should relevantly engage with. These constraints are related to topical and action agendas, presuppositions, epistemic gradients and preference for a particular type of response (e.g., Bolden 2009; Heritage 2010; Heritage and Clayman 2010; Heritage and Raymond 2012; Raymond 2003, 2010; Stivers, Mondada, and Steensig 2011). Responses, in turn, can be produced in various formats so that as Schegloff (2007: 252) observes, “doing a relevant SPP is the prime way a recipient of a prior turn can show their understanding of what the prior turn was doing and what it made relevant to be done next, and thereby grounds that turn’s efficacy as an action”. Drawing on naturally-occurring data, I examine the linguistic practices that speakers of Polish use to design their responses to polar questions so that they are fitted (or not) to particular social action formats implemented by those questions. Specifically, this study looks at how particle and repeat responses address (or redress) the terms and agendas of polar questions, taking into account the particular sequential contexts in which these second pair parts are produced. As Heritage and Raymond (2012: 192) put it, “answer design is a ‘second window’ into the landscape of rights and obligations that are projected by question design.” In the following sections, I will first describe the current state of knowledge regarding responses to polar questions in Polish and subsequently offer a conversation analytic examination of particle responses and repeat responses to polar questions. My investigation demonstrates that particle responses are used predominantly for responding to questions that seek confirmation in a range of sequential contexts (e.g., repair initiations and upshot formulations). Repeat responses analyzed thus far are used as seconds to more focal informationseeking questions or in contexts where more than a minimal confirmation is required.
2.
Questions and responses in Polish: Prior research
One of the general characteristics of Polish is its morphological richness. In addition, Polish is often classified as a “scrambling” language (Witkoś 2008), with a
Chapter 4. Responses to polar questions in Polish
relatively free word order, a tendency to drop personal pronouns in the subject position and the possibility to leave out the copula verb. All of this is reflected in both question and response design. There is no interrogative syntax in Polish, so that the syntax of polar questions and declaratively formatted utterances looks alike, e.g. Idziesz do szkoły? go.prs.2sg to school.gen You are going to school? vs. Idziesz do szkoły. go.prs.2sg to school.gen You are going to school. Some polar questions can be additionally marked with a question particle czy in sentence-initial position, e.g.: Czy idziesz do szkoły? prt go.prs.2sg to school.gen CZY you are going to school? In Polish, czy functions both as a conjunction (used, for instance, in alternative questions or in reported speech) and a question particle (used only in polar questions). In the former case, it is translated as ‘if, whether, or’, whereas in the latter case it is untranslatable. Even though the question particle is often regarded as a formal marker of Polish polar questions (Dryer 2013), it is, in fact, optional. What is more, the prosodic patterns of polar questions in Polish vary, rising intonation contour being far from the only possibility (see Couper-Kuhlen 2012 for a discussion of final intonation contours in English relative to the actions accomplished by particular questions as well as the stance embodied in these actions). Thus, the interpretation of a particular utterance as either an assertion or a declaratively formatted question in Polish relies heavily on the epistemic positioning of the participants (Heritage 2013; Keevallik 2010; Labov and Fanshel 1977: 100–01; Sorjonen 2001a, 2001b). Responses to polar questions in Polish have so far only been described within the framework of traditional theorizing on non-naturally occurring questions and anwers within the philosophy of grammar. Ajdukiewicz (1974 [1965]) suggested that polar questions have only two available direct answers: tak (‘yes’) and nie (‘no’). At that time, direct answers were conceptualized as “these sentences which everybody who understands the question ought to be able to recognize as the simplest, most natural, admissible answers to the question” (Kubiński 1980: 12). However, no actual conversational data was provided to illustrate these
111
112
Matylda Weidner
conceptualizations, which is one of the biggest shortcomings of the currently available studies on Polish questions and responses. This chapter seeks to redress this shortcoming by identifying the interactional circumstances for the participants’ choice of a particular format of a response.
3.
Data and method
I draw my collection of cases from a varied corpus of telephone conversations and co-present interactions. For this chapter, I examined over 14 hours of interaction in which I identified 447 responses to polar questions. In selecting instances for the collection, whenever necessary, language-specific grammatical features, morphological richness or pro-drop quality of Polish (which make ellipsis structurally allowable), were taken into consideration. For instance, when questions addressed in a second person singular (masz ‘have.2SG’) or a third person singular form (ma ‘have.3SG’)1 are answered in the first person (mam ‘have.1SG), such responses are considered repeats. Table 1 provides an overview of responses to polar questions identified in my corpus (447 responses). Adding up to 260 instances, more than half of the responses in the collection (58%) have tak ‘yes’, nie ‘no’, or the particle no ‘PRT’ as the first item. In my data, they are used a bit more frequently as turn-initial elements (135 instances) than stand-alone items (125 instances). I will refer to these as particle responses throughout the chapter. They can be further subdivided into unmarked particle responses tak ‘yes’ (and other versions thereof, e.g. ta ‘yep’) and nie ‘no’, which confirm or affirm a proposition (see Enfield et al. 2019 for a most recent discussion), and the particle response no,2 which is a marked alternative with a stancerelated function (Weidner 2018). Similar to the set of languages covered in the study by Enfield et al. (2019), repetitional responses constitute a less frequent response option in my data, with 64 instances (14%) of full, partial and modified repeats (Bolden, this volume; Keevallik 2010; Sorjonen 1996, 2001a, 2001b; Stivers 2005) and 40 additional instances where a repeat is followed by either a particle or something else (9%).3 The answers that are the focus of this study were limited
1. This third person singular verb form often co-occurs with a subject in the third person singular formal form, i.e. Pan, Pani ‘sir, madam’. 2. Notwithstanding the enticing similarity in spelling, this untranslatable particle in Polish should not be confused with the English negative ‘no’. 3. Responses identified as “other” include cases such as the one below, where two audio engineers (AE1 and AE2) are talking about which microphones to use for the recording session. The
Chapter 4. Responses to polar questions in Polish
Table 1. Responses to polar questions in Polish Response type
Number
% of total
Tak
66
15%
Nie
31
7%
No
28
6%
Tak + X
51
12%
Nie + X
46
10%
No + X
28
6%
Tak/nie/no + repeat
10
2%
X + tak/nie/no
27
6%
Repeat
64
14%
Repeat + tak/nie/no
3
1%
Repeat + X
37
8%
Other
56
13%
Total
447
100%
* Percentage totals subject to rounding effects
to stand alone particle responses and repeats, though I will provide brief illustrations of some other formats.
4.
Responses to polar questions in Polish: Analysis
The following sections offer a conversation analytic investigation into the difference between unmarked particle responses (that is, tak and nie), marked no (which is a stance-indexing particle) and repetitional responses. As the analysis will show, even though the response slot in the adjacency pair provides for the possibility of using any of the alternatives, these response formats are not functionally equivalent options. I begin with unexpanded particle responses (i.e., tak,
question produced by AE1 receives a response that is neither a particle nor a repeat, but which operates on the semantics of the word nie działa ‘not working’: 01
AE1:
a ten czarny zupe::łnie nie działa:? the black {one} doesn’t work at all?
02
AE2:
martwy. dead.
113
114
Matylda Weidner
nie and no), documenting that the most frequent sequential contexts of their occurrence are social actions requiring locally-relevant confirmations.
4.1 Particle responses: tak, nie and no Table 2 reviews the arrangement of particle responses among 287 questionanswer pairs that had tak, nie, no as part of their design, demonstrating that 125 of these are single particle responses. Table 2. Distribution of particle response formats Particle responses
Distribution of response types
Tak
23% (n = 66)
Nie
11% (n = 31)
No
10% (n = 28)
Tak/nie/no + X
44% (n = 125)
Tak/nie/no + repeat
3% (n = 10)
X + tak/nie/no
9% (n = 27)
Total
100% (n = 287)
Particle responses in my data are used in sequences where polar questions request confirmation or initiate repair (cf. Keevallik 2010; Sorjonen 2001a, 2001b; Stivers 2010). In these environments particle responses provide straightforward answers that conform to the terms of the question and fit its action agenda. Overwhelmingly, in these sequential contexts, tak and nie directly address the local here-and-now projects that the questions implement (i.e., obtaining confirmation or resolving problems with hearing or understanding). This pattern is illustrated in two excerpts below. The first example is a fragment of an interaction between two audio engineers in a music studio. One audio engineer is looking down into his laptop (C=computer guy), while the other has his headphones on (he wears the headphones in such a way that one ear is fully covered, while the other ear is only partially covered) and is processing music on the second computer and the mixing table (H=headphones guy). The audio engineers are debating purchasing an additional license for music software, which turns out to be quite costly. In lines 1 and 2, C informs H that the license is not merely an extension to the software that they already bought and are working with, but that it is a different software altogether. In response to C’s informing, H now takes the headphones off and turns to C and, with a polar question tesz:: y płatny::? ‘also paid’ (line 3), H requests confirmation
Chapter 4. Responses to polar questions in Polish
that the new software requires an additional payment. In this context, H’s question is hearable as addressing the local project of clarifying that this ‘different program’ is not free. Excerpt 1. (Music studio_Program; C = computer guy, H = headphones guy) 01
C:
c 02
#no::. hmfhh (.) on nie jest tu najtań-=bo to= PRT he not be.2SG here cheap cause it NO4 it is not the cheapest cause it #sits behind H and looks down into his laptop-------=jest >inny program w ogóle< słu:chaj.# be.2SG different program in general listen.2SG.IMP is a different program altogether listen. --------------------------------------------># ((H takes headphones off and turns to C))
03
H:
tesz:: y płatny::? also paid? (0.1)#(0.9) c #scrolls on the touch pad--->>
04
C:
→ #tak. yes. #head down looking at the screen->
c 05
(0.2)
06
H:
ile? how much?
07
C:
stówę. a hundred.
08
(0.4)
09
H:
10
C:
11
H: h
12
C:
dolków? dollar.PL.DIM dollars? → ta. yeah. *fj:: * phew *shakes his head* nazy:wa się go, (0.2) go button, call.3SG REFL it.ACC {it’s} called go (0.2) go button, --->>
4. This no is not a disagreement token, but a particle in Polish, which has been intentionally left untranslated but rendered in capitals in the English idiomatic translation line.
115
116
Matylda Weidner
C responds to H’s question with a positive particle tak. (line 4), thus confirming the candidate proposition in question. Following an exchange between H and C about the exact price of the software (lines 6–7), H inquires about the currency in which the price is given. H’s question in line 9 puts forth a candidate understanding that the currency is US dollars (H uses a colloquial diminutive form dolków?) and it is designed for confirmation. C confirms the accuracy of H’s candidate understanding using an abbreviated colloquial version of the positive particle (ta., line 10). Here, particle responses are used to address a series of questions that are contributory to the primary course of action underway, i.e., the decision to buy the program or not (see Keevallik 2010 for a related discussion of minimal responses in Estonian). Excerpt 2 shows a fragment from a history-taking phase of a neurological consultation, during which the doctor, through a series of polar questions, aims to establish the patient’s circumstance at the time of his ischemic attack. Here, two follow-up questions that offer an inferential understanding of the patients’s history receive two straightforwad answers, a single denial nie in line 2 and a single confirmation tak in line 4. Excerpt 2. (Neurology; D = doctor, P = patient) pan5 był sam wtenczas w domu:? sir be.3SG.PST alone then PREP home.LOC sir {youV} were home alone at that time?
01
D:
02
P:
03
D:
04
P:
05
D:
uhm, .hh to ja bym- y:: żona była? ta:k? czy ktoś:then I be.1SG.COND wife be.3SG.PST yes or somebody uhm, .hh then I would {your} wife was {there}? yes? or somebody
06
P:
to akurat nie jest żo:na. it actually not be.3SG wife it actually isn’t {my} wife.
07
D:
niE:, no,
08
→ nie. no. ktoś by:ł, somebody be.3SG.PST {there} was somebody, → tak. yes.
(0.3)
5. Pan is a masculine gender honorific term in Polish (a V-form).
Chapter 4. Responses to polar questions in Polish
09
D:
10
n^o: to nie ma znacze:nia akurat ż(h)adnego. y::::= PRT it not have.3SG meaning.GEN actually none NO it doesn’t matter at all actually. y:::: =gdyby ona (.) przy:szła do mnie i jeszcze mi opowiedziała. if she come.3SG.PST to me.GEN and still me.DAT tell.3SG.PST if she {would} (.) come to me and also tell me.
Both particle responses (lines 2 and 4) acquiesce to the terms of the questions, by offering unexpanded answers in the sequential context of routine questioning during the history-taking phase. The patient’s nie. (line 2) negates the candidate non-optimized6 proposition put forth by the question (Heritage and Clayman 2010: 143–145; Stivers 2007: 54–57) that he was home alone at the time of his ischemic attack, while the subsequent tak. (line 4) confirms the doctor’s logical inference that there was somebody home during the incident. These answers, in their unelaborated form, align with the terms and the action agenda of the questions, whose local aim is to establish whether somebody accompanied the patient during his ischemic attack, and move the sequence forward. Moreover, as evidenced in the subsequent talk, they are supportive of the larger course of action underway. As it turns out, the doctor’s initial question is retrospectively hearable as a pre-request (Schegloff 1980, 2007), where the subsequent expansion paves the way for the doctor to formulate the request proper (line 10), i.e., for the co-present person to come and describe the episode. As is also shown in Table 2, 44% of polar questions (125/287) received a particle (tak/nie/no) + X response. These responses have tak/nie/no in turn-initial position, followed by an expansion (which is other than a repeat) that may be either through-produced or constitute a successive TCU. For illustration of this format, consider Excerpt 3, in which a wife (W) and a husband (H) have just finished their breakfast and the husband is making coffee for both of them. The wife’s question at line 1 seeks confirmation that the coffee is ready, perhaps implementing a pre-request. Excerpt 3. (PECII_PL_Breakfast) 01
W:
02
H:
już tą kawkę mamy:? already this coffee.DIM have.1PL {we} already have this coffee?
→ ta:k, jeszcze tylko cukier. yes still only sugar yes, except for sugar.
The husband first confirms with a turn-initial particle tak and then expands his answer to specify that sugar has not been added yet (line 2). 6. A non-optimized question is one that nominates a less desirable outcome (Heritage 2010).
117
118
Matylda Weidner
What we have seen so far is that in Polish, stand-alone tak, which is a positive response, and nie, which is a negative response, align with the question’s terms and (dis-) affirm or (dis-) confirm the propositional content of the question. They are no-problem particle responses that “accept the terms of the question unconditionally, exerting no agency with respect to those terms, and thus acquiescing in them” (Heritage and Raymond 2012: 183). Looking back at Table 2, it is noticeable that another particle often found in responses to polar questions in Polish is no, a marked particle response with a stance-related function. Even though the particle no is often used as a positive response particle, it furnishes a marked response that is different from the positive particle tak ‘yes’ (see Stivers 2018 for her discussion of marked interjections in English). As discussed in Weidner (2018), the primary function of no is to treat the content of the questioner’s turn as already known or self-evident, based on (i) common knowledge and common epistemic access; (ii) the fact that the issue has already been mentioned on some prior occasion and should thus already be known to the participants; or (iii) when the questioner assumes a more knowledgeable position (K+ position, Heritage 2012b), in which case no is used to push back on that assumption. No also has a second function, namely to invoke the speaker’s “my-side” perspective¨ (Heritage 2015; Pomerantz 1980; Weidner 2018: 228), which highlights the participants’ epistemic status in relation to their epistemic access or rights to some particular information. Excerpt 4, which comes from the same kitchen interaction as Excerpt 3, illustrates the stance-related function of no. Here, the wife is chopping sweet pepper on the kitchen counter and the husband is standing next to her, frying bacon for breakfast on the stove. While chopping, the wife recalls an incident of hurting her finger (line 1), indicating the point of the injury (line 3) after a brief pause. The husband takes notice of the finger and subsequently formulates a question that attends to the mentioned incident (line 4). Excerpt 4. (PECII_PL_Breakfast; W = wife, H = husband) 01
W:
02 03
(0.4) W:
h 04
s:ię:: wal:nę:łam w ten palec mocno. REFL hit.1SG.PST in this finger hard.ADV {I} hit myself hard on this finger.
H:
t^ut*aj. wiesz?* here know.2SG here you know? *looks at wife’s finger* uh: kiedy- w:- wcz^oraj. when yesterday.
Chapter 4. Responses to polar questions in Polish
05
W:
→ no, PRT
NO 06
H:
07
W:
m[hm, [co tak latałam. what yes fly.1SG.PST.IPFV when {I was} in a hurry.
The question at line 4 is initially designed as a content (‘when’) question in search for information, but its design changes mid-TCU to culminate as a polar question that seeks confirmation of the candidate proposition that it was yesterday. The wife’s no (line 5) attests to the propositional content of the husband’s question as obvious and emphasizes her own perspective (“my side”) (Weidner 2018). Prior turns put forth epistemic assumptions about the wife’s and the husband’s access to knowledge, where the wife’s t^utaj. wiesz? ‘here. you know’ (line 3) prompts the husband to acknowledge the incident. The husband seeks to specify the time of the incident, wcz^oraj ‘yesterday’, offering a knowledgeable guess. With a no response, the wife qualifies the matter as self-evident and asserts her own epistemic primacy regarding this particular piece of information (Weidner 2018). Having seen an example of a no response, it is now relevant to compare the particle tak ‘yes’ and the particle no. Excerpt 5, taken from a neurological consultation, illustrates the difference between the two. The patient’s question, starting at line 1, draws on the preceding diagnostic discussion and requests information from the doctor concerning the necessity to repeat a computer tomography scan (CT scan). The patient’s description of the test is quite vague (takie (.) zdjęcie_ ‘this (.) photo_’, lines 1–4). The doctor’s subsequent turn (line 5) requests confirmation of her candidate understanding of the reference as denoting a computer tomography scan. Excerpt 5. (Neurology; D = doctor, P = patient) 01
P:
02
03
z tym, z tym że pani7 doktor,= with this with this that madam doctor but but madam {youV} doctor
=bo ostatnio mia’em te:ż [takie] takie= cause lately have.1SG.PST also such such cause last time {I} also had this this D:
[ta:k¿] yes
7. Pani is a feminine gender honorific term in Polish (a V-form). which is not directly translatable into English. In this chapter they are indicated with sir and madam and the marking (youV ) in parenthesis.
119
120
Matylda Weidner
04
P:
=(.) zdjęcie_ czy to by było konieczne? photo.ACC PRT it COND be.3SG.PST necessary photo CZY it would be necessary?
05
D:
ale które zdjęcie, komputerowe::? but which photo computer.ADJ but which photo, {the} computer {tomography}?
06
P:
07
D:
08
P:
→ °°tak.°° yes. głow↑y:? head.GEN {of the} head? → no:_ PRT
NO 09
D:
nie:, bo tu ni:c złego nie ma. no becasue here nothing bad not have.3SG no, because {there} is nothing wrong here.
The patient’s °°Tak.°° at line 6 is a particle response, which provides an answer to the just prior understanding check. The doctor’s subsequent turn (line 7) further explicates the referent and provides a candidate specification of the photo (line 4). The doctor knows that the patient had an ischemic attack and has already had a spine X-ray, thus a CT scan of the head is the obvious logical choice in this context. The patient answers (line 8) with a stand-alone particle response No: that communicates that this is indeed what he meant. In this sequential context of two consecutive understanding checks, tak is a neutral-stance affirmative response, while no orients to the-known-in-common character of a piece of information. It conveys the patient’s stance towards the doctor’s second understanding check as seeking confirmation of a proposition that, from his point of view, should already be self-evident, given that the doctor has already signaled some recognition of the referent (line 5). Hence, in response to polar questions, no provides confirmation with an additional epistemic layer that invokes the speaker’s “my side” perspective and points towards the obviousness of the proposition being confirmed (cf. Weidner 2018). Notably, the actual answer to the patient’s question (line 9) is given once the referent is obvious to both participants, and comes in a more elaborated form. Thus, while tak and no were particle responses to auxiliary confirmation-seeking questions, the nie + elaboration answer format at line 9 addresses the initial agenda of the patient’s information-seeking question (from lines 1–4). In sum, even though the examples differ with respect to the epistemic aspects, the particle responses analyzed so far share some commonalities: they provide direct answers that fit the action agenda of the question and display the partici-
Chapter 4. Responses to polar questions in Polish
pants’ understanding of the status of the question as subsidiary to a larger course of action (e.g., a request proper or a negotiation of the legitimacy of buying a new expensive software).
4.2 Repetitional responses The overview of response types shown in Table 1 demonstrates that particle responses outnumber repetitional responses in the collection analyzed so far. Moreover, there appears to be a significant difference between the two with respect to their function and placement relative to the larger sequence of action underway, which I will explore in this section after illustrating the formats that repetitional responses take in the data, i.e., full, partial and modified repeats (Extracts 6–9). Table 3 below re-presents the distribution of repetitional responses Table 3. Distribution of repetitional response formats Repetitional responses Repeat
Distribution of response types 62% (n = 64)
Repeat + tak/nie/no Repeat + X
3% (n = 3) 35% (n = 37)
Total
100% (n = 104)
In Excerpt 6, father (F) and son (S) are talking about their past experiences of commuting to work. Here, the father recounts the story of his previous company car and the number of kilometers he drove with it. The turn of interest is at line 6, where the father expands on his prior story by pursuing a response (Pomerantz 1984; Stivers and Rossano 2010) from the son with a question dużo? ‘a lot?’ – a question that solicits agreement with the proffered assessment. The son’s response, dużo. ‘a lot.’, is a (full) lexical repeat of the question. Excerpt 6. (PECII_PL_Car drive; F = father, S = son) 01
F:
02
03
tym służbowym autem zrobiłem= this.INST company.INST car.INST do.1SG.PST I did ninety thousand kilometers with
=dziewięćdziesiąt tysię:cy kilometrów, nie? ninety thousand kilometer.PL no this company car, no? S:
no=no. PRT PRT
NO=NO.
121
122
Matylda Weidner
04
F:
05
przez trzy lata. during three years. (0.2)
06
F:
→ dużo? a lot?
07
S:
→ dużo. a lot.
While line 7 in Excerpt 6 above showed an instance of a lexical repeat, line 10 in Excerpt 7 below illustrates a full sentential repeat (for an extended analysis of this fragment see Excerpt 15 below). Excerpt 7. (PECII_PL_Car drive; B = brother, S = sister) 09
B:
→ a8 (.) p-piątek masz wolny::? CONJ friday have.2SG free Friday {you} have free?
10
S:
→ piątek mam wolny. friday have.1SG free Friday {I} have free.
A partial (lexical) repeat is featured in Excerpt 8, which comes from a car drive. The talk between a boyfriend (B) and a girlfriend (G) concerns the video recording that is being made inside the car. At line 1 and 2, the boyfriend asks his girlfriend to make a call to the data collector to inquire whether they are allowed to stop at a gas station while the camera is running. The question contains an impersonal modal auxiliary verb można ‘can’ (Zinken and Ogiermann 2011), which frames it as asking for permission to stop. The girlfriend’s response, produced in overlap with the second TCU, grants permission with a partial repeat that selects only the impersonal modal verb from the prior question, altering it prosodically. Excerpt 8. (PECII_PL_Car drive) 01
02
B:
→ >dowiedz się czy< można9 się zatrzy:mać,= inquire.2SG.IMP REFL PRT can REFL stop.INF inquire if {one} can stop =[bo bym se zatankował.] cause be.COND RP fuel.3SG.PST cause {I} would refuel.
8. In Polish a is a conjunction of contrast, which projects that the following action will somehow depart from the prior. In this function a does not occur turn-initially in declarative sentences in Polish. Since it is often wrongly translated as ‘and’ and thus confused with a conjunction of addition i ‘and’, I have decided to leave it untranslated in this chapter. 9. Można and trzeba (in Excerpt 12 below) are modal auxiliary verbs in Polish that cannot be combined with a grammatical subject, i.e. marked for person (see Zinken and Ogiermann 2011, for a detailed discussion of this impersonal form).
Chapter 4. Responses to polar questions in Polish
03
G:
→
[ can {one} can.
]
An instance of a modified (partial) repeat (cf. Stivers 2005) is shown in Excerpt 9, which features three participants inside a car, one sitting in the back (BK) and two in the front (DR and PS). The talk is about a mutual friend (Paulo) who shaved his head bald. In his repetitional response in line (adjusted for person widzieliśmy ‘we saw’), the driver shifts the stress on the verb, from a marked antepenultimate in line 1 to an unmarked penultimate stress in line 2.10 Excerpt 9. (PECII_PL_Car drive; BK = passenger in the back seat, DR = driver, PS = passenger in the front seat) 01
BK:
widzie:liście w ogóle Paulo:? see.2PL.PST in general NAME {you} saw Paulo at all?
02
DR:
widzieli:śmy. see.1PL.PST {we} saw
03
PS:
mhmh. [no. PRT
mhmh. [NO 04
05
DR:
] ja= I ] I
[wczoraj.] yesterday =strzelałam że przegrał zakład. shoot.1SG.PST.IPFV that lose.3SG.PST bet said that {he} lost {a} bet.
I now turn to examining the action contexts in which repetitional responses are typically used and what they do. Generally, taking into consideration their place in the unfolding sequence of action, repetitional responses provide unambiguous answers to polar questions that require more than straightforward confirmations. For instance, when a question is interpreted as primarily seeking information about matters that are in the recipient’s domain of knowledge or when a question is understood to be the focal action within a sequence, a repetitional response is more likely to occur (see Excerpts 12 and 13 below). Moreover, in contrast to particles, repeats can be used to push back on some constraints embodied in the design of the question, contest claims put forth by the questioner (for exam10. The most common lexical stress pattern in Polish is on the penultimate syllable. Words of foreign origin (e.g. matematyka ‘mathematics’, prezydent ‘president’) allow for both penultimate and antepenultimate alternative. Here, in Excerpt 9, at line 1, the antepenultimate stress may be marking the question as conveying an unexpected stance.
123
124
Matylda Weidner
ple with respect to epistemic access) or respond to “double-barreled” actions (cf. Excerpt 10 below). Let us look at a number of specific instances of repetitional responses. Excerpt 10 is a fragment taken from the recording studio data corpus. Here, a repetitional response orients to the initiating action as both an informationseeking question and a request, that is, a “double-barreled” action (Schegloff 2007). The short segment presented here has only two actively interacting participants, but in fact there are four persons co-present in the studio (two of them behind a glass shield and available via microphones). Singer one (S1) and singer two (S2) stand behind the microphones, with their headphones on and ready to sing. In line 1 and 2, S1 produces a polar question that requests information about the potential availability of a Kleenex among the co-present parties, at the same time inspecting his own pockets in search for one. Excerpt 10. (Recording studio_Kleenex; S1 = Singer one, S2 = Singer two) 01
S1:
s1 02
s1 s2
03
S2:
s1 s2 04
S1:
#ktoś ma chusteczkę do nosa tuta::j?= # somebody have.3SG kleenex to nose here somebody has {a} kleenex here? #looks down inspects his pockets searching a kleenex→# *#=może:?# maybe maybe? *#looks up# *S2 searches his pockets* (0.2)
→ #mam.* (0.2) * have.1SG {I} have. #turns to S2 *hands kleenex to S1* °°o:: dzięki.°° °°O11 thanks.°°
The question is addressed to any of the co-present parties. It includes the nonrecognitional indefinite pronoun ktoś ‘somebody’, a grammatically fitted thirdperson verb ma (have.3SG), and is downgraded with the turn-final może:? ‘maybe’ (Heritage and Raymond 2005; Heritage 2012b), which modulates its imposition (Ogiermann 2009) and epistemic stance. The embodied conduct by S1, who is looking down and searching his pockets for a Kleenex, supports the 11. O is often deployed as an indication of a noticing in Polish. In some contexts it can also be used analogously to ‘oh’, i.e. as a change-of-state token (Heritage 1984). Even though Swan (2002: 420) proposes that the Polish “little word” “O” be translated as There! You see!, I decided to leave it untranslated.
Chapter 4. Responses to polar questions in Polish
impersonal design of his turn. The question is received by the second singer (S2) with a repetitional answer (line 3), produced as an appropriately inflected firstperson verb repeat (mam. ‘I have’). The answer is immediately followed by an embodied action of handing the Kleenex to S1 (line 3). Prior conversation analytic research has shown that the same formats, for instance questions, can be used as vehicles for accomplishing a range of actions, such as requests for service (e.g., Schegloff 1988, 2007: 76–81). Here, S2’s ‘more agentive’ repetitional response might be “motivated” (Heritage and Raymond 2012; Raymond 2003) by the second singer’s understanding of the action as a request addressed to multiple recipients, and where S2 self-selects to fulfill it (Keevallik 2010). The evidence for this being a “double-barreled” action (Schegloff 2007) is retrievable from the participants’ own conduct. S1 receives the verbal repeat and the action of handing the Kleenex with a noticing token o and a verbalized dzięki ‘thank you’ (line 3). This expression of gratitude acknowledges the just provided help as volunteered (Zinken, Rossi, and Reddy 2020) and retrospectively orients to the initial question as a request. Looking back at the analysis of particle responses in the prior section, which were shown to be supportive of some larger course of action (e.g., the doctor’s request in Excerpt 2), a positive particle tak could have been heard as treating the initiating action as a pre-request (see Levinson 1983: 356–64 for a discussion of indirect speech acts and pre-requests). As a consequence, a simple tak response could put S1 into a potentially sensitive position of having to make an explicit request to S2 (see also Bolden, this volume; Rossi 2015 for a related investigation of hai x ‘do you have x’ sequences in Italian). A repetitional response can also be used to address a mismatch between expectations put forth by the design of the question and the answer. In this case, the mismatch is related to both the presuppositions conveyed by the polar question and its grammatical polarity. Excerpt 11 shows a fragment of an interaction in a music studio, which also functions as a practice room for a vocal ensemble. There are three participants (two audio engineers and one non-regular visitor to the studio, who is not involved with the ensemble) in the studio, but only two of them (audio engineer, AE and the visitor, V ) are involved in the sequence of interest here. Having noticed that the audio engineers are dismantling pretty much all the equipment in the studio (line 1), V draws an inference that the ensemble does not need to use the equipment currently, which means that they will not not be doing a concert in the immediate future. Excerpt 11. (Music studio_Equipment; V = visitor, AE = audio engineer) 01
V:
ale to znaczy że nie:: nic (.) nie macie, żadnego= but it mean.3SG that not nothing not have.2PL none but it means {you} don’t have anything, {you} don’t have
125
126
Matylda Weidner
02
03
=teraz:y:: bałaganu? koncertu, nic? now mess.GEN concert.GEN nothing any concert now, nothing? AE:
ae 04 05
→ *m↓amy, * have.1PL {we} have *head nod* (0.2)
AE:
p↑o kole:i.=m:h↑y:hyhyhyhyhy. after turn one after another m:h↑y:hyhyhyhyhy
This inference, which is packaged as a polar question and interspersed with negative elements (lines 1–2), conveys a very steep tilt towards a no-response that would align with this preference12 (Clayman and Heritage 2002; Heritage 2010; Heritage and Raymond 2021). Here, there are five successive negative elements in the design of the turn (nie::/nic/nie/żadnego/nic?), which magnify the question’s bias toward a response that would confirm the proposition that the ensemble does not have any concerts. In line 3, the audio engineer (AE) produces a repetitional response that goes against the preference and the polarity of the question (see also Bolden, this volume, for a resonating example of an “echo repeat” used to disconfirm a negatively formatted interrogative in Russian). The answer is a repeat of the verb in an appropriately inflected form (the adjustment is from second person plural macie ‘you have’ to first person plural mamy ‘we have’) that lacks negation and conveys the contrary of what the question presupposed. Crucially, it should be noted that this question could have also been responded to with a particle response, tak or nie, though both particles would have been ambiguous as to the scope of their action. In this sequential context, the verb repeat offers an unambiguous disconfirmation of the proposition conveyed by the question and simultaneously asserts the opposite. The verbal action done by the repeat is further reinforced by a synchronized embodied action by the audio engineer (he nods during the production of m↓amy,). That this is a potentially unexpected response is evidenced in subsequent talk (line 5). The stream of laughter, produced by the audio engineer mid-turn, works as an “offense remedy” (Jefferson, Sacks, and Schegloff 1977), orienting to his prior answer as possibly departing from expectations set by the question and manipulating its potentially incongruent character (where the expectation put forth by the design of the preceding question might have been set towards a confirmation of no upcoming concerts). 12. Polish is a negative concord language and accepts constructions with multiple negations, which reinforce the negative tenor of the sentence (Swan 2002; Zagorska Brooks 1975).
Chapter 4. Responses to polar questions in Polish
Repetitional responses in my data are also used to respond to informationseeking polar questions. Excerpt 12 is an illustration of an interaction between a mother and two sons, where, similarly to the example discussed above, a repeat is used to provide an unambiguous answer. The younger son (Jurek) and mom, who is cleaning the table after breakfast, are in the dining room. The older son (Filip) is behind mom, in the living room and without visual access to mom. Just prior to this fragment, Jurek draws a wetwipe from a Kleenex box in front of mom to wipe his nose. The question at line 1 is addressed to Filip, next to whom there is a box with regular (dry) tissues. Excerpt 12. (PECII_PL_Brkfst_20160306_cam1; M = mom, F = Filip, J = Jurek) 01
M:
j m 02
#*t^am nie ma już chusteczek Fili::::p? * there not have.2SG no more tissue.PL NAME {there} are no more tissues there Filip? *wipes his nose with wetwipes in front of mom----->* #cleans the table and looks at Jurek--------------> (0.2)
03
M:
04
F:
05
M:
06
m
tych suchy::::ch?# these dry.PL.GEN these dry {ones}? → not have.2SG {there} are not. dobra. to trzeba wymienić, bo jak będziecie wycierać= good it one has to change.INF cause when be.3PL.FUT wipe.INF alright. then one has to exchange {them}, because if {you} =tymi mokrymi to będziecie mieć podrażnioną skórę. these wet.PL it be.3PL.FUT have.INF irritated skin.ACC keep using {the} wet {ones you} will have irritated skin. --------------------------------------------------->>
Line 1 is a B-event statement (Labov and Fanshel 1977: 100–101), produced as a negatively formatted question that requests information about something that mom has no access to and Filip does (she is turned with her back to the Kleenex box). Her access is inferential. Mom usually refills the Kleenex boxes in the house and so this question expresses a proposition that is counter to mom’s expectations. The design of the question, with a lexical element już (glossed as ‘no more’)13 and a prolonged articulation of the son’s name, conveys a surprised stance that
13. In the English translation line this lexical element is rendered as ‘more’, where in fact it should have been ‘no more’. However, due to the fact that this would create a grammatically anomalous sentence in English, i.e., ‘There are not no more towels there Filip?’, the second negation was omitted.
127
128
Matylda Weidner
there are no more tissues in the box. Mom’s initial question (at line 1) receives no response. Only after mom specifies the referent as the dry tissues (as opposed to the wet ones) does Filip produce an answer (line 4). It is a repetitional response, which uses the impersonal subjectless existential verb phrase nie ma ‘are not’ (Lipińska-Grzegorek 1981) from mom’s question and is produced at an audibly slower pace. Given the specifics of this participation framework, with Filip having direct visual access to the Kleenex box, the repeat, produced at a slower pace, foregrounds his direct access to the information regarding the material object in question and his rights over the requested piece of information. Mom, who does not turn around to check the validity of this answer, relies entirely on Filip’s independent judgement. The repetitional response is treated as informative enough and neither mom nor Filip elaborate on this answer. Rather, mom’s dobra ‘alright’ closes the sequence (Beach 1993; Betz et al. 2021) and initiates another course of action, in which she formulates an objective necessity to refill the Kleenex box (see Zinken and Ogiermann 2011). Excerpt 13, which comes from the history-taking phase of a neurological consultation, provides evidence for the observation that repetitional responses occur in contexts where the stakes of the question are higher (cf. Heritage and Raymond 2012). The following example is different from the prior one, in that from the initial repetitional response, the patient proceeds into sequence expansion and a continuation of the topic (see Sorjonen 2001a, 2001b for Finnish). The patient in this excerpt is a man in his late forties. He was admitted to the hospital because of a problem with his lumbar and cervical spine and troublesome dizziness. This fragment comes from the history-taking phase of the consultation and follows a series of questions that the doctor produces in order to gather information from the patient about the symptoms related to his spine problems (e.g., Boyd and Heritage 2006). The doctor’s question (lines 1–2) inquires about a matter that has not been discussed extensively in this consultation, but which connects the pain in cervical spine and dizziness with another candidate symptom, headaches. The doctor’s A proszę mi powiedzieć ‘Please tell me’ marks a shift to a new topical agenda (Weidner 2015) and the ensuing question specifies the terms of the shift. Excerpt 13. (Neurology_Spine; D = doctor, P = patient) 01
D:
a
CONJ
02
03
proszę mi powiedzieć=czy= please.1SG me.DAT tell.INF PRT please tell me if
=bóle gło:wy są. pain.PL head.GEN be.3PL {there} are headaches. P:
→ są. be.3PL {there} are.
Chapter 4. Responses to polar questions in Polish
04
(0.3)
05
D:
06
P:
07
często pana głowa bo:li, often sir.ACC head hurt.3SG {youV} often get headaches, → hhhhh. (.) czę:sto i: praktycznie= hhhhh. often and practically =no: y: y::: stały środek (.) pyralgina. PRT permanent.ADJ means NAME NO pyralginum {is my} regular remedy
The doctor’s question is designed as a request for information about matters that are not shared (Heritage 2010, 2012a). The patient’s answer (line 3) is a repeat of the copula verb są (literally translated as ‘are’) from the doctor’s turn. Here, both a particle and repeat response would address the question’s terms and provide an unambiguous response. However, in contrast to a particle tak ‘yes’, the repeat not only aligns with the question’s terms, but it also upgrades the importance of and, potentially, the consequentiality of the piece of information that it provides, thus treating the question as focal for the ongoing activity. Hence, this repetitional response conveys an implication of the continuing relevance of talk as crucial for the current activity of history taking. Following a moment of silence, the doctor orients to this relevance and asks a second question (line 5), which elaborates on the patient’s answer and seeks confirmation of the proposition that the headaches are frequent. The patient’s response does more than just confirm (lines 6–7). It starts with an audible sigh, followed by a moment of silence and only then does the patient produce the first full word. The word is a repeat of the temporal adverb często ‘often’,14 which is then succeeded by an additive conjunction i:. This initial design of the patient’s responsive turn is hearably projecting more talk to the turnin-progress. Moreover, the design features of the turn may also be contributing to an interpretation of this turn as a potential complainable. Formulating his answer from his own perspective (Pomerantz 1980), the patient extends it, volunteering details from his everyday life, which he treats as potentially relevant for further medical treatment (Stivers and Heritage 2001). This example also illustrates a tendency, documented elsewhere, for repeats to function as vehicles from which to launch additional units of talk (Heritage and Raymond 2012; Stivers 2005). Similar to the prior one, Excerpt 14 illustrates the use of a repetitional answer to a question about matters clearly in the recipient’s domain of knowledge. The excerpt comes from an audio-recorded interaction between a doctor and a patient in an emergency room. The patient came to the emergency unit because of a knee 14. This response is an illustration of the [repeat + X] format from Table 1 and 3.
129
130
Matylda Weidner
injury. As part of the admission routine for walk-in patients, he must have already filled in a form to indicate the reason for his visit prior to this interaction. In lines 1–3, the attending doctor, who is probably reading from the patient’s file, asks a question about the placement of the injury, as part of medical history taking. The question is composed of two units, where the first TCU (line 1) puts up a proposition that a leg is hurting, while the incremental expansion (line 3) selects the right leg as the locus of the pain and offers this selection up for confirmation. Excerpt 14. (Emergency room_Knee; D = doctor, P = patient) 01
D:
02 03
(0.2) D:
04 05
P:
→ prawa. right {the} right (one) (0.8)
D:
08 09
prawa. right {the} right (one). (.)
06 07
i bo:li no:ga::, and hurt.3SG leg and {the} leg hurts,
jaki ma pan zawód? what is your profession? (0.3)
P:
stolarz. carpenter.
The first TCU (line 1) is syntactically possibly complete after the word noga ‘leg’ (at the end of the turn) and it may be a potential space for the patient to provide the information about which leg the pain concerns (note also the lenghtening of the word no:ga:: ‘leg’). Yet, he does not. Following a momentary silence (during which the doctor might be reading the patient’s chart), the doctor extends his turn with an increment produced from a knowledgeable position (K+) (Heritage 2012b; Schegloff 1996), which names the right leg and presents it as a candidate proposition. The patient responds, with a verbatim repeat prawa ‘right’. This repetitional response does more than tak would do. While tak would simply accept the propositional content of the doctor’s question and its accuracy, the repeat prawa offers an upgraded confirmation about which leg hurts. Even though the doctor may already have had access to the piece of information in question (e.g., from the patient’s admission file), the patient now draws on his own experience of pain
Chapter 4. Responses to polar questions in Polish
to reassert this proposition. In contrast to the previous excerpt, here neither the doctor nor the patient elaborate on this answer, rather, the doctor proceeds with further questioning (line 7). As shown earlier (in Excerpt 7), repetitional responses can also come in the form of full sentential repeats. These answers involve an almost verbatim repetition of the question with appropriately inflected forms related to person marking. Excerpt 15, which depicts three family members during a car drive, is an illustration of such a repeat. The fragment begins with the brother commenting on the beautiful weather outside, a topic that is picked up by all three family members (line 1–3). The sequence in focus is “touched off ” (Jefferson 1978) by this topic and takes place between a sister (S) and a brother (B); its theme concerns the sister’s plans for the Corpus Christi weekend. Corpus Christi, which is a national holiday in Poland, is celebrated on Thursdays and many people bridge it with the weekend by taking the Friday off as well. That year, Corpus Christi was on the 11th of June. In lines 3–8, the sister states that she might go to Skorzow (a family cottage at the lake) on Friday, after Corpus Christi. Excerpt 15. (PECII_PL_Car drive; 3 family members, S = sister, B = brother, F = father) 01
B:
ładna pogoda nie? beautiful weather no?
02
F:
bar[dzo ładna. very beautiful
03
S:
04
S:
05
[ się zrobiło=no j↑a:: właśnie nie:= beautiful.ADV REFL become.PST PRT I just not it got beautiful=NO I actually don’t know =wiem czy jadę do tego Skorzowa czy nie. know.1SG PRT go.1SG to this NAME PRT not if {I’m} going to this Skorzow or not. (1.2)
06
S:
a15 jeśli ja:dę, no to dopiero (.) y: w:: piątek. nie, CONJ if go.1SG PRT it only in friday not if {I} go, NO then only on Friday. right,
07
F:
((sigh))
08
S:
tam wte:dy, (.) po tej
there then after this.F eleventh after this eleventh
15. As stated before, a is a contrastive conjunction in Polish, which does not have a proper translation in English.
131
132
Matylda Weidner
09
B:
10
S:
11 12
a (.) p-piątek masz wolny::? CONJ friday have.2SG free Friday {you} have free? → piątek mam wolny. friday have.1SG free Friday {I} have free. (2.5)
F:
nowy fiat, nie? new fiat, no?
Her announcement is spread over a number of turns, each of which unpacks the details of the specific date she has in mind. By mentioning the possibility of going away on Friday (line 6–8), the sister can be understood as implying that she is not going to work on this particular day (after the eleventh). The brother’s subsequent question (line 9) is addressed at precisely this understanding of his sister’s turns-so-far. Formulating the question as piątek masz wolny::? ‘Friday you have free’ implies that the employer has given the sister a day off, not that she has unilaterally decided to take the Friday off. Moreover, the design of his question, with a turn-initial contrastive conjunction a, a stretch of the last vowel and a turn-final rising intonation contour may further convey that this is a surprising or unexpected inference. The answer, delivered as a full sentential repeat (line 10), offers an upgraded confirmation. It also displays a stance towards the implications of the question and indicates that S encountered trouble with the action-implications of her brother’s prior turn. On one hand, by confirming her brother’s inference, this repeat claims the speaker’s greater epistemic authority over the proposition conveyed by the question and upgrades her primary rights to this inference. On the other hand, it orients to the mismatch of expectations implicated in the initiating action. While for the brother the interpretation gathered from his sister’s turn may be tentative or even unexpected, for the sister, with her direct knowledge of her own work schedule, this is an obvious conclusion. The repeat exerts agency over the expectations put forth by the question and foregrounds the sister’s unequivocal primacy over the proposition in a way in which an alternative response, tak, would not do (tak would simply acquiesce to the proposition of the question). The sequence is not elaborated any further and after a longer silence, a new topic is initiated by the father (line 12). In sum, repeats in my data are used to respond to polar questions which are understood to do more than request unproblematic confirmation (e.g. implement requests, offer epistemically agentive inferences) or initiate repair within larger courses of action. Depending on the dynamics of the sequence of talk in which the question-answer pairs occur, repeats are used to provide information to ques-
Chapter 4. Responses to polar questions in Polish
tions that are understood as focal and not subsidiary within an ongoing activity (see Keevallik 2010 on her discussion of “primary actions”), to volunteer to fulfill a request implemented via a polar question, or to push back on some constraints embodied in the design of a question. In contexts where the question problematizies matters that the recipient knows more about, or matters that go against the expectations put forth by the questioner, repeats deal with these asymmetries in a marked fashion. They claim ownership of and rights to some particular piece of knowledge or reassert agency over the question’s terms (e.g., by underscoring the rights to the question’s propositional content or by adjusting the focus of the question) and agenda. Moreover, as illustrated above, repetitional responses also deal with addressing incongruences between expectations put forth by the design of the question and the answer.
5.
Concluding remarks
This chapter has been written with the aim of paving the way for a further investigation into the architecture of responses and questions in Polish, focusing in particular on particle and repetitional answers to polar questions. The findings presented in this chapter are based on detailed sequential analyses of actual interactions and thus differ from what has been proposed so far on the basis of noninteractional data. For instance, whereas prior descriptions of responses to polar questions in Polish have postulated that the relevant choice is between two direct answers, an affirmative tak ‘yes’ and a negative nie ‘no’, this chapter shows that interactants have a range of options to choose from and that the choices they make regarding a particular answer-type are not equivalent. Initial sections of this chapter provided illustrations of what Ajdukiewicz (1974 [1965]) referred to as “direct answers” in his theorizing, that is, particle responses that acquiesce to the question terms (with a contrastive example illustrating the difference between tak and no). With examples taken from naturallyoccurring interaction, I demonstrated that participants in Polish talk-in-interaction use particle responses to address polar questions that request confirmation or initiate repair (cf. Keevallik 2010; Sorjonen 2001b, 2001a; Stivers 2010). Overwhelmingly, in these sequential contexts, tak and nie provide straightforward answers that conform to the terms of the question and fit its action agenda. Subsequently, the chapter investigated repetitional responses, starting with an illustration of their most common formats in the data and then offering a detailed analysis of a number of cases. Repeats were shown to be used predominantly as unambiguous responses to requests for information or to push back on some constraints embodied in the design of a question. When questions
133
134
Matylda Weidner
problematizied matters that the recipient knew more about or matters that went against the expectations problematized by the questioner, repeats were selected to claim ownership of and rights to some particular piece of knowledge or to reassert agency over the question’s terms and agenda. Repetitional answers are also to be found in sequential contexts where the initiating turn is a “double barreled” action. This chapter has also shown that the selection of one answer type over another is sensitive to the status of the question in the progressing sequence of action, which influences the construction of a particular response (Schegloff 2007: 77–78). This is an empirically emergent observation and, as this chapter illustrates, the evaluation and the relevance of a given response type seems to be tightly tied to the specific action implemented by a particular question in its context. Hence, while particle responses are used predominantly to address confirmation-seeking questions and orient to the question as contributory to some larger course of action, repetitional responses deal with larger agendas and are used to address (information-seeking) questions interpreted as focal for the activity in progress. Finally, I acknowledge the fact that some issues have only been touched upon; the particulars of question design, the prosodic features of the response, the precise timing of the response relative to the question, all have only been signaled in passim. Moreover, the findings presented in this chapter are mostly qualitative, thus furnishing an exciting space for future quantitative elaborations (e.g., the relationship between question and answer type). These are all promising avenues of prospective research. Some consolation comes from the fact that one has to start with something, thus this chapter discussed the answering apparatus people use in responding to polar questions in Polish, guided by Sacks’s (1987 [1973]: 67) teaching that “you cannot find what they’re trying to do until you find the kinds of things they work with”.
Acknowledgment I would like to thank John Heritage, Marja-Leena Sorjonen and Galina Bolden for their invaluable guidance, patience and support.
Chapter 4. Responses to polar questions in Polish
References Ajdukiewicz, Kazimierz. 1974 [1965]. Pragmatic logic [Logika pragmatyczna]. Dordrecht: Reidel. Beach, Wayne. 1993. “Transitional Regularities for Casual “okay” Usages.” Journal of Pragmatics 19: 325–352. Betz, Emma, Arnulf Deppermann, Lorenza Mondada, and Marja-Leena Sorjonen, eds. 2021. OKAY across Languages: Toward a Comparative Approach to Its Use in Talk-in-Interaction. John Benjamins. Bolden, Galina. 2009. “Beyond answering: repeat-prefaced responses in conversation.” Communication Monographs 76 (2): 121–143. Bolden, Galina. 2016. “A simple da?: Affirming responses to polar questions in Russian conversation.” Journal of Pragmatics 100: 40–58. Boyd, Elizabeth, and John Heritage. 2006. “Taking the History: Questioning During Comprehensive History Taking.” In Communication in Medical Care: Interactions between Primary Care Physicians and Patients, edited by John Heritage and Douglas Maynard. 151–184. Cambridge, England: Cambridge University Press. Clayman, Steven E., and John Heritage. 2002. “Questioning Presidents: Journalistic deference and adversarialness in the press conferences of Eisenhower and Reagan.” Journal of Communication 52 (4): 749–775. Couper-Kuhlen, Elizabeth. 2012. “Some truths and untruths about final intonation in conversational questions.” In Questions. Formal, Functional and Interactional Perspectives, edited by Jan P. De Ruiter. 123–145. Language, culture & cognition, 12; New York: Cambridge University Press. Couper-Kuhlen, Elizabeth and Margret Selting. 2018. Interactional Linguistics: Studying Language in Social Interaction. Cambridge: Cambridge University Press. Dryer, Matthew S. 2013. “Position of polar question particles.” In The World Atlas of Language Structures Online, edited by Martin Haspelmath, Matthew Dryer, David Gil, and Bernard Comrie. Leipzig: Max Planck Institute for Evolutionary Anthropology. Enfield, Nick J., Tanya Stivers, and Stephen Levinson. 2010. “Question-response sequences in conversation across ten languages: An introduction.” Journal of Pragmatics 42: 2615–2619. Enfield, N. J., Tanya Stivers, Penelope Brown, Christina Englert, Katariina Harjunpää, Makoto Hayashi, Trine Heinemann, et al. 2019. “Polar Answers.” Journal of Linguistics 55 (2): 277–304. Hakulinen, Auli. 2001. “Minimal and non-minimal answers to yes-no questions.” Pragmatics 11 (1): 1–15. Heinemann, Trine. 2005. “Where grammar and interaction meet: The preference for matched polarity in responsive turns in Danish.” In Syntax and Lexis in Conversation: Studies on the use of linguistic resources in talk-in-interaction, edited by Auli Hakulinen and Margaret Selting. 375–402. Amsterdam: John Benjamins. Heritage, John. 1984. “A change-of-state token and aspects of its sequential placement.” In Structures of social action, edited by J. Maxwell Atkinson and John Heritage. 299–345. Cambridge: Cambridge University Press.
135
136
Matylda Weidner
Heritage, John. 2002. “Ad hoc inquiries: two preferences in the design of ‘routine’ questions in an open context.” In Standardization and Tacit Knowledge: Interaction and Practice in the Survey Interview, edited by Douglas Maynard, Hanneke Houtkoop-Steenstra, Nora Cate Schaeffer and Johannes van der Zouwen. 313–333. New York: Wiley Interscience. Heritage, John. 2010. “Questioning in medicine.” In ”Why Do you Ask?”: The Function of Questions in Institutional Discourse, edited by Alice F. Freed and Susan Ehrlich. 42–68. New York: Oxford University Press. Heritage, John. 2012a. “Epistemics in Action: Action Formation and Territories of Knowledge.” Research on Language & Social Interaction 45: 1–25. Heritage, John. 2012b. “The Epistemic Engine: Action Formation, Sequence Organization and Territories of Knowledge.” Research on Language and Social Interaction 45: 25–50. Heritage, John. 2013. “Epistemics in conversation.” In Handbook of Conversation Analysis, edited by Jack Sidnell and Tanya Stivers. 370–394. Boston: Wiley-Blackwell. Heritage, John. 2015. “Well-prefaced turns in English conversation: A conversation analytic perspective.” Journal of Pragmatics 88: 88–104. Heritage, John, and Geoffrey Raymond. 2005. “The terms of agreement: Indexing epistemic authority and subordination in assessment sequences.” Social Psychology Quarterly 68 (1): 15–38. Heritage, John, and Steven Clayman. 2010. Talk in Action: Interactions, Identities and Institutions. Oxford: Blackwell-Wiley. Heritage, John, and Geoffrey Raymond. 2012. “Navigating Epistemic Landscapes: Acquiescence, Agency and Resistance in Responses to Polar Questions.” In Questions edited by J-P. De Ruiter. 179–192. Cambridge: Cambridge University Press. Heritage, John, and Chase Wesley Raymond. 2021. “Preference and Polarity: Epistemic Stance in Question Design.” Research on Language and Social Interaction, 39–59. Houtkoop-Steenstra, Hanneke, and Charles Antaki. 1997. “Creating happy people by asking yes-no questions.” Research on Language and Social Interaction 30 (4): 285–313. Jefferson, Gail. 1978. “Sequential aspects of storytelling in conversation.” In Studies in the organization of coversational interaction, edited by Jim Schenkein. 219–248. New York: Academic Press. Jefferson, Gail, Harvery Sacks, and Emanuel A. Schegloff. 1977. “Preliminary Notes on The Sequential Organization of Laughter.” Pragmatics Microfiche. Keevallik, Leelo. 2010. “Minimal answers to yes/no questions in the service of sequence organization.” Discourse Studies 12 (3): 283–309. Kubiński, Tadeusz. 1980. An Outline of the Logical Theory of Questions. Berlin: AkademieVerlag. Labov, William, and David Fanshel. 1977. Therapeutic Discourse: Psychotherapy as Conversation. New York: Academic Press. Levinson, Stephen. 1983. Pragmatics. Cambridge: Cambridge University Press. Lipińska-Grzegorek, Maria. 1981. “Existential sentences in english and polish.” In Theoretical Issues in Contrastive Linguistics, edited by Jacek Fisiak. 347–364. Current Issues in Linguistic Theory, 12: John Benjamins Publishing Company.
Chapter 4. Responses to polar questions in Polish
Ogiermann, Ewa. 2009. “Politeness and in-directness across cultures: A comparison of English, German, Polish and Russian requests.” Journal of Politeness Research 5: 189–216. Pomerantz, Anita. 1980. “Telling my side: ‘Limited access’ as a ‘fishing device’.” Sociological Inquiry 50: 186–198. Pomerantz, Anita. 1984. “Pursuing a Response.” In Structures of Social Action, edited by J. Maxwell Atkinson and John Heritage. 152–164. Cambridge: Cambridge University Press. Raymond, Geoffrey. 2003. “Grammar and social organization: Yes/No interrogatives and the structure of responding.” American Sociological Review 68: 939–967. Raymond, Geoffrey. 2010. “Grammar and social relations: Alternative forms of Yes/No-type initiating actions in health visitor interactions.” In ”Why Do You Ask?”: The Function of Questions in Institutional Discourse, edited by Alice F. Freed and Susan Ehrlich. 87–107. New York: Oxford University Press. Raymond, Geoffrey, and John Heritage. 2006. “The epistemics of social relations: Owning grandchildren.” Language in Society 35: 677–705. Rossi, Giovanni. 2015. “Responding to pre-requests: The organisation of hai x ‘do you have x’ sequences in Italian.” Journal of Pragmatics 82: 5–22. Sacks, Harvey. 1987 [1973]. “On the Preferences for Agreement and Contiguity in Sequences in Conversation.” In Talk and Social Organisation, edited by Graham Button and John R. E. Lee. 54–69. Clevedon, England: Multilingual Matters. Schegloff, Emanuel A. 1980. “Preliminaries to Preliminaries: ‘Can I Ask You a Question’.” Sociological Inquiry 50: 104–152. Schegloff, Emanuel A. 1988. “Presequences and Indirection: Applying Speech Act Theory to Ordinary Conversation.” Journal of Pragmatics 12: 55–62. Schegloff, Emanuel A. 1996. “Turn organization: One intersection of grammar and interaction.” In Interaction and grammar, edited by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson. 52–133. Cambridge: Cambridge University Press. Schegloff, Emanuel A. 2007. Sequence organization in Interaction: A Primer in Conversation Analysis Volume 1. Cambridge, England: Cambridge University Press. Sorjonen, Marja-Leena. 1996. “On repeats and responses in Finnish conversations.” In Interaction and Grammar, edited by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson. 277–327. Cambridge: Cambridge University Press. Sorjonen, Marja-Leena. 2001a. “Simple answers to polar questions: The case of Finnish.” In Studies in Interactional Linguistics edited by Margret Selting and Elizabeth Couper-Kuhlen. 405–431. Amsterdam: Benjamins. Sorjonen, Marja-Leena. 2001b. Responding in Conversation: A study of response particles in Finnish. Pragmatics & Beyond; Amsterdam: John Benjamins. Sorjonen, Marja-Leena, and John Heritage. 1991. “And-prefacing as a feature of question design.” In Leikkauspiste [Intersections: Essays in Honor of Auli Hakulinen], edited by L. Laitinen, P. Nuolijärvi, and M. Saari. 59–74. Helsinki: Finnish Literature Society. Stivers, Tanya. 2005. “Modified Repeats: One Method for Asserting Primary Rights from Second Position.” Research on language and social interaction 38 (2): 131–158. Stivers, Tanya. 2007. Prescribing Under Pressure: Parent-Physician Conversations and Antibiotics. New York City: Oxford University Press.
137
138
Matylda Weidner
Stivers, Tanya. 2010. “An overview of the question-response system in American English conversation.” Journal of Pragmatics 42: 2772–2781. Stivers, Tanya. 2018. “How We Manage Social Relationships Through Answers to Questions: The Case of Interjections.” Discourse Processes 56 (3): 191–209. Stivers, Tanya and John Heritage. 2001. “Breaking the sequential mold: Answering ‘more than the question’ during comprehensive history taking.” Text 21 (1/2): 151–185. Stivers, Tanya and Federico Rossano. 2010. “Mobilizing Response.” Research on Language and Social Interaction 43 (1): 3–31. Stivers, Tanya, Lorenza Mondada, and Jakob Steensig. 2011. The morality of knowledge in conversation. Cambridge: Cambridge University Press. Swan, Oscar. 2002. A grammar of contemporary Polish. Bloomington, Indiana: Slavica Publishers. Weidner, Matylda. 2015. “Telling somebody what to tell : “Proszę mi powiedzieć” in Polish doctor-patient interaction.” Journal of Pragmatics 78: 70–83. Weidner, Matylda. 2018. “Treating something as self-evident: No-prefaced turns in Polish.” In Between Turn and Sequence. Turn-initial particles across languages, edited by John Heritage and Marja-Leena Sorjonen, 225–250. Amsterdam: John Benjamins. Witkoś, Jacek. 2008. “On the correlation between A-type scrambling and lack or weak crossover effects.” Studia Anglica Posnaniensia 44. Zagorska Brooks, Maria. 1975. Polish Reference Grammar. The Hague: Mouton & Co. Zinken, Joerg, and Ewa Ogiermann. 2011. “How to propose an action as objectively necessary: the case of Polish Trzeba X (“one needs to x”).” Research on Language and Social Interaction 44 (3): 263–287. Zinken, Jörg, Giovanni Rossi, and Vasudevi Reddy. 2020. “Doing more than expected: Thanking recognizes another’s agency in providing assistance.” In Mobilizing others: Grammar and lexis within larger activities, edited by Carmen Taleghani-Nikazm, Emma Betz and Andrea Golato, 253–278. Amsterdam: John Benjamins.
chapter 5
Three practices for confirming inferences in French talk-in-interaction Repetition, voilà, and exact(ement) Rasmus Persson Uppsala University
Three practices for doing confirmation in French are investigated (repetition, the particle voilà and the adverb/adjective exact(ement)) as responses to inferences, i.e., requests for confirmation that display some understanding of prior talk. While broadly restricted to the context of inferences, at a level of finer action-sequential granularity the paper looks in turn at responses to four different types of inference-making confirmationseeking first pair parts. The analyses show how each confirmation practice does slightly different work depending on the type of initiating action, while some commonalities also hold for each confirmation practice across inference-types. Through repetitional confirmation, confirmers claim authorship over what is inferred. With voilà-confirmations, confirmers attribute some epistemic agency to the interlocutor with respect to the forming of the confirmable understanding. By confirming with exact(ement), confirmers treat the inference-producer as having reached the proffered understanding even more independently. The findings reveal both more context-free and more context-sensitive aspects of the work done by the response forms. Keywords: French, confirmations, response tokens, epistemics, inferences, collaborative completions, other-continuations, candidate understandings, formulations
1.
Introduction
This chapter explores three practices for doing confirming second pair parts (SPPs) in French talk-in-interaction. Rather than consider responses to all types of polar questions, this paper will focus on confirming responses to inferences in a wide sense, that is, first pair parts (FPPs) that make explicit some form of https://doi.org/10.1075/slsi.35.05per © 2023 John Benjamins Publishing Company
140
Rasmus Persson
understanding of coparticipants’ prior talk(-so-far) and invite confirmation of that understanding. In particular, the paper homes in on three response types: (1) repetitional confirmation, (2) the particle voilà, and (3) the adverb/adjective exact(ement) ‘exact(ly)’. These three response types are marked forms of confirmation, compared with affirmative response particles such as oui/ouais (‘yes/ yeah’), which constitute the unmarked response forms for all types of confirmables considered in this study, doing mere and minimal confirmation. Participants have been shown to orient to the different ways in which inferences can arise (or rather be treated as having arisen), including issues such as whether some understanding is based on an inferential process, or essentially on the words in the just prior talk (Sorjonen 2018). Similarly, inferences may be understood as intersubjective (i.e., explicating what was implied and putatively ‘intended’), or unilateral, that is, concluded from prior talk but not necessarily ‘intended’ (Deppermann 2018). Permeating these phenomena are also issues of epistemics, such as the certainty with which the inference is proffered (as indexed through turn design). While Sorjonen (2018) and Deppermann (2018) show how such issues may inform the design of the inference (specifically, the turn preface used), this chapter focuses on how participants may orient to epistemic issues through the turn design when confirming an inference. Because inferences demonstrate some level of understanding of the interlocutor’s prior talk, they are inherently requests for confirmation involving a relatively shallow epistemic gradient (Heritage and Raymond 2012), as opposed to requests for information, designed to convey no pre-existing understanding of the matter in question. (However, not all requests for confirmation are inferences; see below.) As these inferences request confirmation of the explicated understanding, they constitute polar questions in a functional sense, despite lacking the prototypical formal attributes of questions: none of the types of inferences considered here has ordinary interrogative morphosyntax, and none of them relies on any particular prosodic or phonetic format for establishing the conditional relevance of confirmation (or alternatively, rejection). This focus on inferences entails the exclusion of a certain number of polar questions, such as requests for information (i.e. questions claiming no previous knowledge, including many questions with the interrogative particle est-ce que). In addition, some types of requests for confirmation can be argued not to involve any inferential work, e.g. questioning repetitions, which reproduce prior talk verbatim. Also excluded are contingent questions or ‘follow-up questions’ (Heritage and Sorjonen 1994), which – although they do build on prior talk – are geared more towards sequence progression than displaying an understanding of the prior turn. The analysis is based on a core collection of 185 instances, drawn from approximately 22 hours of data. The instances are drawn from several different
Chapter 5. Three practices for confirming inferences in French
sources of (European) French talk-in-interaction, including both video recordings of face-to-face interaction and audio recordings of telephone conversations. Both everyday conversation and institutional interaction (from several different settings) are represented. Most of the recordings are publicly available, either through the CLAPI database,1 or through the Parole Publique project (the UBS and OTG corpora; Nicolas et al. 2002). The three response types investigated in this study all accomplish samepolarity (i.e., confirming) responses primarily occupied with confirming the proffered inference, thus asserting the confirmer’s ultimate epistemic authority over the matter. However, the three response practices differ, first of all, with respect to how the confirmer treats the confirmable as having been arrived at. In short, (1) repetitional confirmation attributes the authorship over the inference to confirmers themselves (hereafter “Self ”), (2) voilà treats the inference as a combined product of what Self had previously said and the inference-producer’s (“Other’s”) inferential work, and (3) exact(ement) serves to treat Other as having more independently reached the understanding that is being confirmed. Furthermore, the three also differ in terms of how definitive (or alternatively, how provisional) the confirmation is taken to be. Repetitions and exact(ement), each in their own way, generally confirm the proffered understanding rather definitively on the terms proposed and strongly propose sequence closure. On the other hand, voilà (while still an unambiguous confirmation that may often stand as sequence closure) appears to leave more room for further refinement of the proposed wording, thus doing a more provisional sort of confirmation. The remainder of the chapter is organised as follows. In Section 2, some background is provided on confirming responses, with particular focus on the three target practices. In Section 3, the work of each confirmation type is considered in four types of inference environments. Section 4 summarises the findings, draws together some commonalities across inference-types, and considers the implications of the findings.
2.
Background: Response types in French and other languages
In this section, a survey is provided of work with relevance for the three response types: (1) repetitional confirmation, (2) voilà, and (3) exact(ement). Response systems have been investigated in various typological studies, and languages are sometimes said to use one (or possibly several) of the following systems: (1) a yes/no system, (2) an agree/disagree system, or (3) an echo (repetition) answer 1. http://clapi.ish-lyon.cnrs.fr/
141
142
Rasmus Persson
system (König and Siemund 2007: 320–22), of which the first two can be said to use interjections. In fact, many if not all languages may draw on several of these systems, with resources for responding that include both response particles (interjections) and repetitions (Enfield and Sidnell 2015; Enfield et al. 2019). French is no exception: various particles as well as repetition are used for doing confirmations. However, given that overt subjects are required in French, verb-only repetitions do not occur in French (unlike in, e.g., Finnish, Estonian and Russian – see Sorjonen 2001; Keevallik 2010; Bolden this volume – where verb-only repetitions occur with different functions than full clausal repeats). Furthermore, French does not have pro-repeats (subject pronoun + copula/auxiliary) like those found in, for example, English (Stivers 2005: 138–43). On the other hand, both clausal and phrasal repetitions are used as confirmations in French, including in confirmations of inferences. Repetitional confirmations in English have been shown to implement confirmations that assert greater agency than more acquiescent response particles, e.g., yes, given that repetitions have greater indexical independence than particles (Heritage and Raymond 2012; Enfield and Sidnell 2015). When responding to explications of allusions, repetitional confirmation may also indicate that the confirmer had indeed essentially conveyed as much in what preceded (Schegloff 1996). In the environment of collaborative completions, confirmations are regularly accomplished through repetition, and previous studies have shown the use of repetitions for integrating others’ contributions into an emerging (extended) turn (Lerner 2004; Oloff 2014). In particular, repetitions are often used in French for confirming other-completions in word search environments (Mondada 1999: 30). The particle voilà occurs in numerous turn and sequence environments (Mondada 2018), and this review will focus on confirmation-related ones. Mondada (2018) and Oloff (2016) examine cases where voilà does confirmation of collaborative completions and other-continuations, and find that voilà and its combinations with other turn-components (including repeats) convey specific epistemic stances, beyond the binary value of confirmation over disconfirmation. Voilà has also been found to occur as a confirming response to candidate understandings and formulations (Bruxelles and Traverso 2006: 82ff; Persson 2014: 84–87) and to other closure-implicative conclusions (Oloff 2016: 254–57). Furthermore, voilà occurs as a “volunteered” (re )confirming response to registering repeats, but not as a response to repair-initiating repeats, which make confirmation conditionally relevant (Persson 2015: 599f ). Taken together, earlier work on confirmation uses of the particle voilà suggests the following. Although voilà allows the confirmer to claim ultimate authority over what is being confirmed, it occurs where – and indexes the very fact that – the FPP speaker has some independent access to the matter at hand. That is, the confirmable derives from knowl-
Chapter 5. Three practices for confirming inferences in French
edge or interactional history to which the addressee also has (some) access. It is therefore unsurprising that while voilà works as a confirmation token after various types of inferences, voilà does not seem to occur as a confirming response to requests for information, i.e., questions that encode no pre-existing knowledge (Delahaie 2009), unlike, e.g., oui/ouais and repetitions. Finally, it is worth recalling that previous research has found that voilà often functions as a confirming response in collaborative list environments (Mondada 2018: 384; Bruxelles and Traverso 2006: 85f ). There is little previous research on the interactional work of tokens like exactement used as confirmations, but Li (2007) shows that exactly is used as a confirming response to formulations in English. Stivers (2019) analyses some other English adverbs (of course, certainly, absolutely, definitely, totally and right) as ‘upgraded’ interjections, which propose a problem with the asking of the question, while ‘downgraded’ interjections (probably, possibly and maybe) propose a problem with answering it. However, French exactement – and possibly English exactly – may not be possible responses to polar questions generally, but rather have more circumscribed responsive uses, for instance as a confirming response to inferences or assertions of second-hand knowledge, or as an agreeing response in other environments. In particular, exact(ement) does not appear to occur as a response to requests for information. In the present dataset, the adverbial form exactement appears to be used interchangeably with the (less frequent) adjectival form exact.2 Both occur as stand-alone tokens implementing confirmation, without reference to any preceding predicate. Since the work of exact is seemingly indistinguishable from that of exactement, the two will not get separate treatments in what follows.
3.
Confirming inferences
3.1 Types of inferences The focus in this paper is on confirming responses to inferences, i.e., FPPs that explicate and offer up for confirmation some form of understanding of coparticipants’ prior talk(-so-far). Before considering the work of the three focal (marked) response practices, in this subsection we map out these FPPs, distinguishing four
2. This may or may not illustrate a more general pattern, but note that the adjective probable ‘probable’ occurs as a stand-alone token in some response environments where the adverb probablement ‘probably’ could conceivably also have been produced (cf. Stivers 2019 on probably).
143
144
Rasmus Persson
types of inferences, and showing instances where they are responded to with (unmarked) affirmative particles oui/ouais. The general sequential structure can be schematically represented as follows: 1
Self:
Prior talk
2 FPP Other (inference-producer):
Inference
3 SPP
Confirmation (e.g. oui/ouais; repetition; voilà; exact(ement))
Self (confirmer):
The actions that are grouped together here under the blanket term ‘inferences’ take different (more specific) forms, and may be formally tied to prior talk to various degrees. In order of decreasing dependence on prior talk, on the one end are collaborative completions, including word search solutions (Lerner 1991, 1996, 2004), where Self ’s syntactically (or action-wise) incomplete turn gets completed by another speaker (Other). This initiates a collaborative turn sequence, with the conditionally relevant response from the original speaker being confirmation (preferred) or rejection/amendment (dispreferred) (Lerner 2004). A minimal illustration is given in Excerpt 1, where the talk revolves around a photography competition that Judith has entered. (Throughout this chapter, inferences are marked with a single arrow, and confirmations with a double arrow and grey highlighting.) Excerpt 1. (CLAPI.glasgow_22:29) 01
JUD:
02 03
c’est sur une journée en- uniquement it=is in one day only is it a single day (.)
PAT:
.t (0.3) ouais ouais hh (.) °ouais ouais c’est une yeah yeah yeah yeah it=is one .t yeah yeah °yeah yeah it’s a single
04
journée° (.) mais ça commence à neuf heures on day but it starts at nine hours IMPRS day but it starts at nine and we
05
a
fini à dix-neuf heures finish.PPCP at nineteen hours finished at seven PFV
06
07
JUD:
et après le: c’est quoi le: and then the.SG it=is what the.SG and then the: what’s the: (0.5)
Chapter 5. Three practices for confirming inferences in French
08
PAT: → le: (.) les prix the.SG the.PL prizes the: the prizes
09
(.)
10
JUD: ⇒ ouais yeah yeah
11
PAT:
.h (.) le premier prix c’est u:n reflex numérique the first prize it=is a reflex digital .h the first prize is a digital reflex camera
In line 8, following a halt in the progressivity of Judith’s turn, Patricia displays an understanding of the turn-so-far by proffering a completion of Judith’s content question (line 6), albeit repairing the singular article to plural. Judith confirms the completion in line 10 with an unmarked ouais ‘yeah’, and in line 11 Patricia begins a multi-unit answer to Judith’s content question. Slightly more independent are other-continuations such as appendor questions (Sidnell 2012; Sacks 1995; see also Kendrick 2015: 175 on continuing candidate understandings), which structurally build on, and recomplete, a coparticipant’s turn that was already possibly complete. Again, this has the effect of soliciting confirmation from Self (Sidnell 2012). For example, see Excerpt 2, where Justine tells her two friends Arnaud and Albine about a habit of her boyfriend’s (Séb). Excerpt 2. (CLAPI.rupture_23:51) 01
JUS:
02
mais Séb c’qu’il fait (0.6) souvent il avance but Séb what=he does often he forwards but what Séb does often he’ll fast-forward le film tu sais (.) *fwuit (0.3) fwuit (0.3) fwuit the film you know ONOM ONOM ONOM the film y’know (.) swish (0.3) swish (0.3) swish *stepwise swiping w hand----->
03
ARN:
QUOI what WHAT
04
JUS:
fwuit (0.3) fwuit* jusqu’à la fin ONOM ONOM until=PREP the end swish (0.3) swish all the way to the end -->*
05
ARN: → pour voir c’qui s’passe aprè[s ] for see:INF what happens after to see what happens later
06
JUS: ⇒
[OU]I yes YES
145
146
Rasmus Persson
07
ALB:
non no no way
08
JUS:
AA:H ITJ
AARGH:
Arnaud’s other-continuation in line 5 builds on and recompletes Justine’s turn (lines 1–2 and 4), thus seeking confirmation. Justine provides this in line 6 with oui ‘yes’, before Albine produces a token of ritualised disbelief (line 7) and Justine takes an affective stance towards Séb’s reported behaviour (line 8). Like collaborative completions, other-continuations are one way to display understanding and intersubjectivity by virtue of the joint production of a common linguistic structure, even though the other-continuation is subject to confirmation in the next turn3 (Deppermann 2015). The next step in terms of independence is replacing candidate understandings (Kendrick 2015), whereby, e.g., a referring expression or a lexical item from Self ’s prior talk gets replaced by an alternative – and sometimes more specific – corresponding expression, to be confirmed by Self. The following extract from a tourist office encounter illustrates this type of inference. Excerpt 3. (OTG.1SB0027_0:00) 01
CLI:
bonjour je suis à la recherche d’une carte des hello I am on the search for=a chart of.the hello I’m on the lookout for a chart of the
02
transports sur Grenoble transports in Grenoble ((public)) transports in Grenoble
03
(0.7)
04
EMP: → un plan a map a map
04
(0.4)
05
CLI: ⇒ ouais yeah yeah
06
(0.6)
3. The exception is ‘same-directionality’ completions and continuations (vs. ‘reversedirectionality’). In same-directionality completions, both the original turn and the completion/ continuation are addressed to the same participant, as in co-tellings of a story to a third party, and in such cases, no confirmation is relevant.
Chapter 5. Three practices for confirming inferences in French
07
EMP:
voilà ((hands over document)) there_you_are there you are
08
CLI:
merci thank_you thank you
The employee proffers the indefinite noun phrase un plan ‘a map’ as a candidate understanding, replacing the client’s une carte ‘a chart; a map’, with reference to the public transport’s network map. The client minimally confirms this candidate understanding with ouais ‘yeah’ in line 5, before the employee responds to the request (line 7). Finally, formulations (Heritage and Watson 1979, 1980; Heritage 1985) are a practice by which Other explicates an understanding of Self ’s earlier talk (by summarising, glossing or developing the gist or an upshot of it), usually with weaker formal connections to the wording of prior talk. Like the other inferencetypes, formulations make conditionally relevant a confirming (or disconfirming) response (Heritage and Watson 1979, 1980; Heritage 1985). See for instance Excerpt 4, where Judith and Patricia talk about a friend of Patricia’s, who Judith also knows. Excerpt 4. (CLAPI.glasgow_35:51) 01
JUD:
02 03
et elle habite où maintenant and she lives where now and where does she live now (0.5)
PAT:
!t (0.3) elle a trouvé un studio:- (.) ben she PFV find.PPCP a studio ((flat)) well !t (0.3) she found a studio flat:(.) well
04
il est pas loin d’Perrach:e il e:st genre à: it is NEG far from=Perrache it is like PREP it’s not far from Perrache ((station)) it’s: like:
05
trois minutes de ºl’école quoi donc euhº (0.8) three minutes from the=school like so uh three minutes from ºschool like so uhº (0.8)
06
ouais c’est tout p’tit c’est eu:h- (0.6) yeah it=is ITJ small it=is uh yeah it’s really small it’s uh:(0.6)
07
genre sous les toits like under the roofs like under the eaves
08
(.)
147
148
Rasmus Persson
09
JUD:
10 11
mmhm mmhm mmhm (0.4)
PAT:
mais ’fin moi j’trouve ça tout mignon tu vois but I_mean 1SG I=find that ITJ cute you see but I mean I think that’s really cute y’know
12
’fi:n (0.4) c’est pas genre super coquet I_mean it=is NEG like super pretty I mean: (0.4) it’s not like super pretty
13
super machin: ºmais euhº super thingy but uh or super whatever: ºbut uhº
14
(0.2)
15
JUD: → elle habite seule °du_coup° she lives alone then she lives alone °then°
16
PAT: ⇒ ouais yeah yeah
17
(0.5)
18
JUD:
c’est plus petit qu’ici it=is more small than=here is it smaller than this place
In line 15, Judith formulates an upshot of Patricia’s earlier talk, drawing on the remarks about the size of the apartment to infer that the friend lives alone. This formulation is also marked as such through du coup ‘then’. Patricia confirms with ouais ‘yeah’ (line 16) and Judith then asks a question about the size of the apartment (line 18). In sum, all these inference-type turns are FPPs that project a confirming SPP. However, there is also a sense in which these inferences are not exclusively initiating – they are also responsive to the talk of which they proffer an understanding. As pointed out by Lerner (2004, regarding collaborative turn sequences), this corresponds to launching a retro-sequence, where the sequence beginning is located retrospectively (Schegloff 2007). The inference is presented as subsequent to prior talk, but without having been made relevant by that prior. The adjacency pair structure is thus found in the [inference]–[confirmation] part of the sequence.
Chapter 5. Three practices for confirming inferences in French
3.2 Digression: Marked and unmarked confirmations As mentioned above, this paper deals with three marked types of confirmation, in contrast to unmarked confirmations (see Excerpts 1 through 4). For French polar questions in general (including inferences), the unmarked confirming response forms are the affirmative particles oui/ouais ‘yes; yeah’ and the negative particle non ‘no’, for positively and negatively framed FPPs, respectively.4 The practices analysed here are three of the most common marked types of confirmations produced in response to inferences, specifically.5 Before moving on to the main analysis, a digression on ‘markedness’ is in order, so as to clarify the relationship between marked and unmarked options. The descriptors ‘marked/unmarked’ here refer not so much to sheer frequency, but evoke the argument that unmarked response forms merely and minimally confirm the FPP, exert no agency with respect to the terms of the FPP, and acquiesce in those terms (cf. Raymond 2003 on “conforming” responses; Heritage and Raymond 2012). Unmarked options also tend to promote sequential progressivity (over sequence expansion), as in Excerpts 1–4, although marked confirmations also regularly achieve sequence closure (Heritage and Raymond 2012). Furthermore, unmarked options do no additional work, in terms of epistemic or agentive stance; in particular, I argue that they demonstrate no analysis on the part of the confirmer of how an inference was arrived at. This is supported by the circumscribed use of marked forms of confirmation. The fact that voilà and exact(ement) are not used for responding to information requests (see Section 2) suggests that voilà/exact(ement) are specialised confirmation practices, and serve not to do generic confirmation, but to deal specifically with the issues involved in confirming provisional understandings, thus interactionally ratifying them. On the other hand, unmarked oui/ouais are used for information-requests and confirmation-requests alike. Furthermore, while the three marked confirmation practices are fitted to epistemically different scenarios (i.e., different ways in which inferences are occasioned),6 oui/ouais appears to be used across the board within the collection of inferences. In addition, the unmarked oui/ouais are also frequently combined with all three marked
4. While this study considers confirmation exclusively, it can be noted that typical options for disconfirmation are non ‘no’ and si ‘yes’ for positively and negatively framed FPPs, respectively. 5. Other marked confirmation-types can also be found, although not quite as frequently, at least in the examined data (e.g. tout à fait ‘absolutely’; c’est ça ‘that’s it’, cf. Excerpt 13). 6. Although, of course, these marked practices are indexical, and therefore they may also at times be used in (analysably) strategic ways, to make claims about epistemic agency that are seemingly incongruous with the scenario as analysable from other indications in the context.
149
150
Rasmus Persson
response practices, as seen in some of the excerpts, again suggesting that the unmarked options are less specialised.7 In Sections 3.3 through 3.6, we will look successively at each of the four inference-types (constituting similar but distinct sequential environments) and the work that the three types of confirmation do in each case. We will consider both the contextual particularisation of the response practices for different types of inferential FPPs, and the commonalities which appear to hold across all types of inferences.
3.3 Collaborative completions Collaborative completions of a coparticipant’s utterance (e.g. Lerner 1991, 1996, 2004) vary along a continuum in terms of how strongly projectable they are, i.e., how much Other can rely on Self ’s turn-so-far in order to infer a completion that may be proffered. As we shall see, the choice among the three focal confirmation practices is sensitive to this variation. 3.3.1 Collaborative completion + repetition Excerpt 5 is a first case of a collaborative completion, from a conversation where Judith is recounting her recent trip to certain parts of Scotland, where Patricia has been previously. Excerpt 5. (CLAPI.glasgow_5:32) 01
JUD:
02
03
super cher le ferry en plus no but it=is super pricey the ferry in addition super pricey the ferry too t’as vu c’e:st [euh you=PFV see.PPCP it=is uh you saw it’s: uh
PAT:
]
[ouais] pour la voiture eu:h yeah for the car uh yeah for the car uh:
7. Further work is required for ascertaining how such [marked+unmarked] combinations are used differently from the marked option, e.g. how the work of voilà ouais differs from voilà. It should also be said that combinations of repetition/voilà/exact(ement) do occur with some regularity. These combined forms appear difficult to account for with a single coherent analysis; separate accounts may be needed respectively for through-produced ‘packages’ (e.g. voilà + repetition; see Mondada 2018: 385; Bruxelles and Traverso 2006: 84), and for contingently expanded series of confirmations, successively produced in response to an emerging FPP (e.g. in more or less overlap with the confirmation turn). This latter type of combination cannot be considered here due to space restrictions.
Chapter 5. Three practices for confirming inferences in French
04
05
(.) ça coûtait combien [>j’m’rappelle plusI can’t remember< JUD:
[mm et PRT and mm and
06
07
tu payes par les per- eu:h [en plus] you pay per DEF.PL (per-) uh in addition you pay per the peruh: in addition PAT: →
[par per]sonne per person per person
08
→ [ouais ] yeah yeah
09
JUD: ⇒ [par per]sonne per person per person
10 11
]
(2.1) JUD:
et donc après ben qu’est_ce_qu’on a fait euhm and so after PRT what=Q =we PFV do.PPCP uhm and so after that well what did we do uhm
Judith runs into trouble a few words into line 6, where she begins a hearably ungrammatical structure by including the definite plural article les before cutting off what is recognisably the beginning of the word personne(s) ‘person(s)’ and producing a lengthened eu:h. Patricia proffers a grammatically correct completion (without the article les) as par personne in line 6, and Judith confirms the completion with a repeat in line 9. Judith’s confirmation comes in overlap with Patricia’s turn-final ouais (line 8). This tag-like ouais may strengthen the bias towards a confirming response (however, as can be seen in Excerpts 6 and 9, this appended ouais does not pre-empt a confirmation, unlike what Lerner 2004: 233f suggests for similarly placed yes/yeah). A gap follows before Judith resumes her extended telling in line 11. As in other cases of collaborative completions confirmed through repetition, the completion is highly projectable at the point where Other (here, Patricia) comes in: not only has an ungrammatical version of it been begun and aborted (line 6), but the previous mention of a vehicle fare (line 3) also means that the issue of a passenger fare is one relevant topical development. Patricia’s collaboration appears to consist in helping out when the turn has stalled, by uttering something that was clearly coming. In this group of cases, the high projectability of the completion is a feature to which the selection of response type is fitted: a repetition here comes off as confirming Other’s completion as what Self was (claimably) about to say, or starting to say, thus claiming authorship over it.
151
152
Rasmus Persson
3.3.2 Collaborative completion + voilà In Excerpt 6, drawn from everyday conversation, Jean’s collaborative completion is produced in overlap with Julie’s own (slightly different) completion. Excerpt 6. (CLAPI.chat_15:05) 01
JUL:
NON mais parce_que c’est intéressant en fait y no but because it=is interesting in fact there no well ’cause it’s interesting actually there
02
a vraiment les ONT des animaux et are really DEF people who have INDF animals and are really people who have animals and
03
pis y en a d’autres en fait qui: then there of.it are others in fact who then there are others actually who:
04
[qui n’(en) ont pas quoi] who NEG=of.it have NEG PRT who don’t have (any) PRT
05
JEA: → [>qui n’ont jamais< eu: ] who NEG=PFV never have.PPCP >who’ve never< had ((any))
06
JUL: ⇒ [voilà
]
PRT
VOILÀ 07
JEA:
[et qui e:]t qui n’auront jamais ouais and who and who NEG=have:FUT never yeah and who and: who never will yeah
08
JUL:
ouais yeah yeah
09 10
(0.4) JEA:
c’est vrai qu’moi j’aimerais bien en avoir it=is true COMP=1SG I=like:COND well of.it have:INF I must say that I’d like to have
11
un ouais one yeah one yeah
12
(.)
13
JEA:
’fin t’sais un p’tit chien en fait d’appartement PRT you=know a small dog in fact ATT=apartment I mean y’know a small dog actually an apartment one
Here, Julie is beginning to build a contrast between people who do have animals (note the accent on the verb ONT ‘have’, line 2), and ‘the others’. At the point
Chapter 5. Three practices for confirming inferences in French
where Jean comes in, after the progressivity-halting sound-stretch on the relative pronoun qui in line 3, there is consequently a strong projection that the second category (‘others’) will be specified by a verb phrase that contains some negated version of ‘having animals’. However, the two versions end up slightly different: Julie’s completion is in present tense, and Jean’s completion is in present perfect tense, with the negation n(e) … jamais ‘never’. After having produced her own completion in line 4, Julie confirms Jean’s completion(-so-far) with voilà (line 6). However, in overlap with voilà, Jean begins to extend the specification with a future tense component (line 7), and a turn-final ouais. It is clear in this case, and other cases with voilà, that some version of the turn completion is strongly projected, although not a single, precise and definite wording of that completion; here this is obvious from the actual production of two alternative completions. Julie’s selection of voilà as the specific form of confirmation is responsive to this very sequential feature, and confirms the completion as one adequate option, among several possible ones. Another kind of demonstration of the specific work done by voilà is given in Excerpt 7. The client (CLI), who is ordering some catering, will confer with her manager about the available budget, before calling the caterer (CAT) back. Excerpt 7. (CORAL.FTELPV26_0:38) 01
CLI:
et si j’vous ai pas rappelé d’ici ce and if I=you.DAT have NEG call_back:PPCP by this and if I’ve not called back by tonight
02
soir c’est que: c’est qu’ evening it=is COMP it=is COMP it’s ’cause: it’s ’cause
03
c’est pas possibl[e it=is NEG possible it’s not possible
04
CAT: →
05
CLI: ⇒ voilà (.) [hein]
]
[c’est] qu’ c’est pas bon it=is COMP it=is NEG good it’s ’cause it’s no good
PRT
PRT
VOILÀ (.) PRT 06
CAT:
07
CAT:
08
CLI:
[(*) ] c’est qu’ça s’ra pa:s d’accord it=is COMP=it be:FUT NEG okay (*) it’s ’cause it’s not happening okay .hh [donc ] euh ça s’passe à l’hôpital /…/ so uh it happens at DEF=hospital .hh so uh it takes place at the hospital /…/ [°d’accord°] okay °okay°
153
154
Rasmus Persson
After having produced an if-component and begun the then-component, CLI halts the turn progressivity with a sound stretch and a recycling (line 2). However, CAT does not come in until the very end of line 3. Thus, rather than an anticipatory completion, CAT produces an alternative completion (line 4), subsequent to CLI’s own. In line 5, CLI confirms CAT’s alternative completion with voilà, orienting to the multiple possible and equally viable ways to formulate the then-component. This is further oriented to in the next turn (line 6), where CAT produces yet another version of it, followed by an acceptance token (d’accord ‘okay’). By continuing to speak after this (line 7), CAT thus treats CLI’s prior voilà as acknowledging that CAT has apparently “got it” close enough by now. Generally, voilà confirmations serve to treat a proffered completion as demonstrating a good enough understanding of the talk-so-far, while not identical with the talk that was about to be produced (see Mondada 2018: 383; Oloff 2016: 255, 258 for further instances of voilà confirming a ‘competing’ but compatible completion; cf. also Bruxelles and Traverso 2006: 82f ). In some cases where Self ’s own turn-completion is overlapped by Other’s collaborative completion, Self may overtly display that both completions are adequate ways to put it, by first confirming the collaborative completion with voilà, and then repeating their own completion (see Oloff 2016: 258). The availability of several viable completions (compare with Section 3.3.1) means that the completion-speaker has somewhat more epistemic agency in picking one, and voilà allows treating the completionspeaker as having found one wording that acceptably conveys “the gist” of what was about to be said. 3.3.3 Collaborative completion + exact(ement) The collection contains only one case of exact(ement) as a confirmation of a collaborative completion, and it has the adjectival form exact (the same case is analysed in Oloff 2014: 2136–37). Nonetheless, what can be noticed here is that the proffered completion is considerably more difficult to project from the talk (and this difficulty may also account for the relative rarity of such cases). Excerpt 8. (CLAPI.kiwi_4:00) 01
ÉLI:
le
jeudi j’ai juste eu:h psycho j’ai thursday I=have only uh psychology I=have on thursdays I’ve only got uh: psychology I’ve DEF
02
03
qu’ une heure et demie just one hour and half just got an hour and a half BÉA:
ouais yeah yeah
Chapter 5. Three practices for confirming inferences in French
04 05
(0.4) ÉLI:
e:t le vendredi normalement j’ai une fois sur and DEF friday usually I=have one time PREP and: on fridays usually once every two weeks
06
07
08
09 10
11
deux tu sais int- eu::hm::: ºcomm[entº] two you know (int-) uhm how I have y’know int- uh::m:: what’s it called BÉA: →
[ ana]°lyse analysis conversation
→ conversationnelle° conversational analysis (0.3) ÉLI: ⇒ exact .hhh MAI:S euh le truc c’est que: EXACT but uh the thingie it=is COMP EXACT .hhh but: uh the thing is: du_coup même si euh je l’ai pas /…/ then even if uh I it.ACC=have NEG then even if uh I don’t /…/
As Élise is detailing her weekly class schedule to two friends, she cuts off in line 6 after int-, and initiates a word search with a stretched eu::hm:: and comment ‘what’s it called’. Béatrice comes in with a collaborative completion (lines 7–8), that Élise confirms with exact (line 10). Had a repetitional confirmation been deployed here, it would arguably open up for the interpretation that Élise had already mentioned this class in the recent past. But in fact, there are no indications in the preceding talk (or at any point in this conversation) providing any cues as to where Élise’s turn is going – in fact, what Élise has begun to say is something different (possibly the beginning of interactions, but it is unclear how accessible this is to the participants). Thus, Béatrice goes beyond what she could reasonably be expected to infer from the talk so far, drawing substantially on her own knowledge and figuring out the appropriate turn completion on her own, despite Élise’s nearly saying something else (and despite the fact that Élise should expectably know her own schedule better than Béatrice). With this in mind, it appears that exact (and possibly exactement) is used when the subsequently accepted completion is not projectable from the talk, and the authorship of the completion is attributable to Other. Also note that Béatrice’s completion is allowed to stand as final, as Élise herself does not produce any alternative or specified version – the proffered completion is not only “good enough” (cf. 3.3.2) but “exactly” right.
155
156
Rasmus Persson
In summary, repetitions confirm the collaborative completion proffered by Other as the completion (that is, just what Self was projectably about to say), whereas voilà confirms the completion as being one adequate completion, selected by Other, among several possible projectable completions (and accordingly, the wording may sometimes be further worked on by either participant, after voilà). With exact(ement), Self acknowledges even more epistemic independence on the Other’s part: Other’s completion is treated as actually being a more adequate one than what Self was projectably beginning (or “trying”) to say. Note that none of these epistemic stances are conveyed through unmarked response forms.
3.4 Other-continuations In appendor questions and other types of reverse-directionality othercontinuations structurally fitted to Self ’s talk, Other displays an understanding of the host turn by proffering for confirmation an expansion of it, at a point of possible completion (Sidnell 2012). The type of subsequent confirmation may index how Other’s continuation stands in relation to the host turn, from Self ’s perspective, ranging from a disambiguating expansion that spells out something taken for granted, to a highly ‘proactive’ development whereby Other displays what they have figured out. 3.4.1 Other-continuation + repetition In Excerpt 9, the host of a radio phone-in show (Éric) checks his understanding that the caller Fred was inspected specifically for whether he owns a TV, at a point where the reason for the mentioned inspection may not yet be clear (as explained later, Fred has declared not owning one, making him exempt from paying a TV license). Excerpt 9. (MOUV.21–12_32:12) 01
FRE:
je fais partie de:s deux pourcent d:es de I make part of:the two percent of:the of I’m one of the: two percent of the:
02
français >j’ai appris ça récemment< qui french I=PFV learn:PPCP that recently who french people >I learnt that recently< who
03
n’ont pas la télé: ça fait trois ans que NEG=have NEG DEF TV it makes three years COMP don’t have a TV: for three years
04
je n’ai pas la télé I NEG=have NEG DEF TV I haven’t had a TV
Chapter 5. Three practices for confirming inferences in French
05
(.)
05
ÉRI:
06
FRE:
a::[h toi non_plu:s ] PRT you neither a::h you neither: [.hh j’ai déjà-] u:n petit message aux I=have (already-) a little message to:the .hh I’ve alreadya: little message to the
07
auditeurs [ne t]richez pas= listeners NEG cheat:IMP NEG audience do not cheat=
08
ÉRI:
09
FRE:
[ouais] yeah yeah =parce_que j’ai déjà été contrôlé trois fois because I=PFV already AUX check:PASS three times =because I’ve already been inspected three times
10
(0.4)
11
ÉRI:→
.hh pour le:: la télé [là ouais] for the.M the.F TV PRT yeah .hh for the:: the TV PRT yeah
12
FRE:⇒
13
⇒
[pour eu:h] pour la for uh for the.F for uh: for the télévisi[on hein j’ai déjà été contrôlé donc]= television PRT I=PFV already AUX check:PASS so television PRT I’ve already been inspected so=
14
ÉRI:
15
FRE:
[ouai:s .hh mais euh tiens .h euh-] yeah but uh you_know_what uh yeah: .hh but uh you know what .h uh=euh si vous déclarez sur votre déclaration uh if you.PL declare on your.PL declaration =uh if you declare on your tax return:
16
d’impô:t eu:h en disant eu:h je vais ATT=tax uh PREP say:GER uh I FUT uh: saying uh: I’m going to
17
eu[:h uh uh:
18
ÉRI:
]
[ouais ouais] (là) on est contrôlé ouais yeah yeah PRT IMPRS AUX check:PASS yeah yeah yeah (then) you get inspected yeah
157
158
Rasmus Persson
Éric’s inference (line 11) is built by picking up a lexical item from the prior talk (télé, line 3 and 4) and producing a prepositional phrase that is readily hearable as grammatically tied to the verb phrase in line 9, more specifically forming an argument to the main verb contrôlé ‘inspected; checked’. Although by then, Fred has begun his confirming turn, Éric’s turn continues past the word télé, with the particle là, which may index an anaphoric link or mark Éric’s turn as inferential, and a response particle ouais (which is tag-like insofar as it is an appended turncomponent that appears to strengthen the expectability of a confirming response by overtly anticipating one). Fred confirms with a repetition (lines 12–13), albeit using the full form télévision ‘television’ rather than télé ‘TV’. By repeating the very words of Éric’s disambiguating other-continuation, Fred confirms it as the understanding (and not merely an acceptable one). Similarly to confirmations of allusions (Schegloff 1996; Sorjonen 2001: 76–80), repetitional confirmation here serves to treat the continuation as previously conveyed, and as already apparent from Fred’s prior talk, although unsaid. This is also evidenced by the fact that Fred next goes on to repeat the core elements of the host turn as well (line 13: j’ai déjà été contrôlé ‘I’ve already been inspected’), orienting to the possibility that Éric may have missed some of what was explicitly said (and not only what was contextually available, through the previous mention of television ownership): by going on to restate what was explicitly said, Fred is displaying that he is not saying anything new, suggesting that the confirmed understanding was also previously conveyed. Generally, repetitional confirmation of other-continuations serves to confirm the proffered continuation as a given, since it was implicitly conveyed through the host turn as it occurs in its sequential context. 3.4.2 Other-continuation + voilà Another set of cases, exemplified by Excerpt 10, have other-continuations that are readily understood as possible – but not “given” – ways of explicating what the host turn was designed to convey. Excerpt 10 is drawn from a service encounter in a bakery, where the recording itself is addressed. Excerpt 10. (CLAPI.boulangerie1_5:16) 01
CUS:
>vous êtes filmés aujourd’hui< you.PL AUX film:PASS today >are you being filmed today
> // (( 32 seconds omitted. When SAL returns, CUS is )) // (( looking at a notice on the counter, concerning )) // (( the ongoing recording. )) 05
SAL:
06
c’est pour voir j’crois dans le:s différents it=is for see:INF I=think in DEF different it’s for seeing I believe in: different pays comment ça s’passe countries how it happens countries how things are done
07
CUS: → °ouais [(da:]ns) les° échanges yeah in DEF exchanges °yeah (in:)° exchanges
08
SAL:
[hein] PRT
PRT 09
SAL: ⇒ ouai+s voilà yeah PRT yeah VOILÀ cus +gaze towards baker-->
10
CUS:
cus cus
>‡ça l+’oblige à s’tenir‡ à_carreau< that 3SG.ACC=forces to REFL=hold:INF in_check >that forces him to lie low< ‡gesture towards baker--------‡ -->+gaze towards salesclerk-->>
The salesclerk’s (SAL’s) turn at lines 5–6 is potentially complete, so the customer’s prepositional phrase da:ns les échanges (line 7) is hearable as an othercontinuation of SAL’s turn, although prefaced by a quiet ouais. The confirmation is done with ouais + voilà (line 9), where voilà treats the customer’s contribution as one confirmation-worthy explication of the highly indexical line 6. Les échanges is one possible way of referring to the recorded and studied activity and setting, but not a particularly privileged one (note that the lexical item les échanges selected here was not used in prior talk). For instance, with other customers, SAL refers to the object of study as le(s) commerce(s) ‘trade; shops’. Les échanges also refers primarily to the commercial, transactional activity, but may alternatively be heard as referring to interaction more generally. Thus, voilà displays an orientation to the other-continuation as one appropriate way to explicate how the host turn should be understood, selected by Other among several.
159
160
Rasmus Persson
3.4.3 Other-continuation + exact(ement) Excerpt 11 comes from the radio show hosted by Éric, and Pierre is a guest reviewing the Israeli comedy TV series Avoda Aravit. Excerpt 11. (MOUV.12–04_1:15:00) 01
PIE:
euh on va voir une chose intéressante qui uh IMPRS FUT see:INF a thing interesting which uh we see one interesting thing which
02
est une RElation d’amour entre une (0.3) euh une is a relation ATT=love between a.F uh a.F is a romantic relationship between a (0.3) uh an
03
arabe (0.5) e:t (.) un juif c’que- c’qui arab and a.M jew.M (REL.ACC) REL.NOM arab (FEM) (0.5) and: (.) a jew (MASC) who- which
04
s’voit pas beaucoup à la télévision euh PASS=see NEG much on DEF television uh isn’t seen very much on uh israeli
05
israélienne .hhh et on va voir des p’tites israeli and IMPRS FUT see:INF INDF little television .hhh and we see little
06
séquences par exemple LUI il va vouloir mett’ sequences for instance 3SG he FUT want:INF put:INF scenes for instance he ((the main character)) wants
07
ses enfants dans une école israélienne juive (.) his children in a school israeli jewish to put his children in a jewish israeli school (.)
08
pa[rce_que-] because because-
09
ÉRI: → [ah
pour] les intégrer [voilà ouais] for them.ACC integrate:INF PRT yeah oh to integrate them I see yeah PRT
10
PIE: ⇒
[ben exacte]ment PRT EXACTEMENT well EXACTEMENT
11
.hh et donc euh >lala maîtresse va leur and so uh (the-) the teacher FUT them.DAT .hh and so uh >the- the teacher says to
12
faire bon j’vous préviens tout_de_suite< à la say PRT I=2PL.ACC inform right_now in the them well let me tell you right now< in the
13
cour de récré: le jeu PRÉféré des p’tits yard ATT break the game favourite of.DEF little school yard the favourite game of israeli
Chapter 5. Three practices for confirming inferences in French
14
israéliens C’EST .hhh LE soldat et israeli it=is DEF soldier and kids is .hhh playing at soldiers and
15
qu’est_ce_qu’on bute quand on est what=Q =IMPRS kill when IMPRS is what do you kill when you’re an
16
[soldat israélien] soldier israeli israeli soldier
17
ÉRI:
18
[°hh ha hh°
]
(0.3)
19
ÉRI:
[(
20
PIE:
[on
)]
bute des palestiniens] et donc voilà le kill INDF palestinians and so that_is the you kill palestinians and so that’s the IMPRS
21
genre de de séquences qu’on va voir sort of of sequences REL=IMPRS FUT see:INF sort of of scenes that we’ll see
After one TCU in the multi-unit telling is recognisably complete at the end of line 7, Éric produces an other-continuation (line 9) where the prefatory changeof-state token ah ‘oh’ marks the understanding as arrived at just-now. Note that the other-continuation is begun in overlap with Pierre’s turn (line 8) just as Pierre is apparently about to provide an account for the reported action (parce que‘because-’), and Éric anticipates this by providing a possible motive: integrating the children. After the other-continuation, Pierre comes in (line 10) and confirms with exactement prefaced by ben (roughly ‘well’), before Pierre resumes the telling (line 11). While the ben-preface could be an orientation to the preemptiveness and sequential ill-fittedness of Éric’s contribution, in the midst of a telling not yet nearing possible completion, exactement acknowledges that Éric figured this out before it could expectably have been gleaned from the talk. That Éric’s continuation did indeed anticipate the development of Pierre’s turn is evidenced by the fact that Pierre’s (just about recognisable) beginning of an account (line 8) never gets resumed – it is Éric’s account (line 9) that is allowed to stand as final. In sum, through repetition, Self can confirm an other-continuation as the way that Self ’s prior talk should necessarily be understood, whereas voilà confirms the continuation as a reasonable one among a range of possibilities, and exactement confirms the other-continuation as entirely accurate while acknowledging that it anticipated something that was not yet projectable from the talk-so-far.
161
162
Rasmus Persson
3.5 Replacing candidate understandings This section outlines the specifics of the three response types when occurring in the environment of replacing candidate understandings (Kendrick 2015), whereby Other proffers some turn component replacing a part of Self ’s turn, leaving Self to confirm whether it is an adequate representation of what Self meant. In what follows, it is argued that the selection of each confirmation type indexes a specific relation between the replaced and the replacement. 3.5.1 Replacing candidate understanding + repetition In Excerpt 12, a student (STU) is calling her university, and UNI takes the call. Here, STU is about to give the reason-for-the-call. Excerpt 12. (UBS.251_0:01) 01
STU:
oui bonjour j’appelle pour prendre un rendez_vous yes hello I=call for take:INF an appointment yes hello I’m calling to make an appointment
02
03
pour mon inscription s’il_vous_[plaît ] for my enrolment please for my enrolment please UNI: →
04
05
→ R:Éinscription [c’est ça] reenrolment it=is that REenrolment is that it STU: ⇒
06 07
[pour v]ot’ for your for your
[oui réin]scription yes reenrolment yes reenrolment (.)
UNI:
.hhh eu:h oui: patientez s’il_vous_plaît uh yes wait:IMP please .hhh uh: yes: please hold
In line 3–4, UNI proffers a phrase containing a replacing candidate understanding (pour vot’ R:Éinscription), overtly marked as provisional by the ‘checking’ expression c’est ça ‘is that it’. UNI produces it with a salient accent on the first syllable (R:É-), which constitutes the prefix that sets the reference apart from the student’s reference. The institutional import of this is that the administrative procedure is different for those who are enrolling for a first time, and those who are continuing their studies. In overlap, starting immediately after the candidate understanding, STU confirms with oui followed by a repeat of the replacing word. In confirming the alternative reference specifically by reproducing it herself – as
Chapter 5. Three practices for confirming inferences in French
if doing the reference again but properly – STU does not merely confirm, but also manifestly corrects her own, prior reference, conveying that it was indeed what she “meant to say” by what she actually said. Here, and also more generally, repetitions in response to replacing candidate understandings take on the job of accepting correction (cf. Jefferson 1972: 302, 1987). 3.5.2 Replacing candidate understanding + voilà In Excerpt 13, a candidate understanding is proffered that replaces one term (DESS) for another (master 2) that is in line with the post-Bologna Process8 nomenclature. Excerpt 13. (UBS.469_0:03) 01
STU:
02
con[cer]na:nt le: D.E.S.S. de droit des aff:ai:res regarding the DESS ATT law ATT business regarding: the: DESS ((degree)) in business law:
03
UNI:
04
STU:
05 06
.hh j’aimerais un renseignement s’il_vous_plaît I=like:COND INDF information please .hh I’d like some information please
[oui] yes yes .hhhhhh (0.3)
UNI:
07
eu:h oui >j’vais vous passer l’secrétariat< .hhh uh yes I=FUT 2.DAT pass:INF the=secretariat uh: yes >I’ll connect you to the secretaries< .hhh → le: D.E.S.S. c’est-à-dire le master deux maintenant the DESS it=is=to=say the master two now the: DESS that is to say the master’s year two now
08
(0.4)
09
STU: ⇒ voilà c’est [ça ] PRT it=is that VOILÀ that’s it
10
UNI:
11
STU:
[c’est] ça (.) °d’accord° it=is that okay that’s it (.) okay °mm° PRT
°mm°
8. A standardisation of qualifications in European higher education.
163
164
Rasmus Persson
In line 6, UNI begins to respond to STU’s request, but then in line 7, UNI first repeats the item le: D.E.S.S. before producing a lexical paraphrase marker (c’est-àdire ‘that is to say’) and the reformulation (le master deux) followed by the adverb maintenant ‘now’. The paraphrase marker and the time reference frame the candidate understanding as a synonymous but more current term, without treating le D.E.S.S. as erroneous. STU confirms (line 9) with voilà followed by another marked confirming expression, the formulaic c’est ça which, here in response position, treats the prior turn as checking an understanding. This is registered by UNI with a repeated c’est ça and d’accord ‘okay’ in third position (line 10). Whereas a repetitional confirmation from STU would have been hearable as acknowledging correction here, voilà serves to treat the replacement term as a paraphrase and only that, not rejecting the original reference but rather acknowledging the more current term as another valid option. Excerpt 14 is another case in point, from a guided tour of a manor house. The guide (GUI) has asked the visitors (VI1, VI2, VI3 and VI4) to make guesses about the purpose of some copper pots on display, and here she gives the answer. Excerpt 14. (CLAPI.manoir2_15:20) 01
GUI:
gui 02
*c’est madame* qui a raison ça n’est pas pour les it=is madam who has reason it NEG=is NEG for the it’s the lady ((here)) who is right it’s not for *points VI3--* MARrons c’est pour RÉchauffer le lit chestnuts it=is for warm:INF the bed chestnuts it’s for warming the bed
03
VI2:
04
VI1: →
05
ben [oui: PRT yes of course
]
[(pour) mett’ les ce]ndres for put:INF the cinders (for) putting the cinders ((in)) (0.5)
06
VI1:
[(°et puis °)] and then (°and then °)
07
GUI: ⇒ [voilà PRT
VOILÀ
des] cendres ou des braises INDF.PL cinders or INDF.PL embers cinders or embers
In line 4, the visitor VI1 offers an interpretation of GUI’s ‘correct’ answer, using the same grammatical format (the preposition pour ‘for’ followed by an infinitive phrase). This candidate understanding explicates how the warming might work, and is not taken as a competing answer to the original question – it starts from
Chapter 5. Three practices for confirming inferences in French
and paraphrases GUI’s answer, rather than rejecting it. GUI confirms with voilà (line 7), and also goes on to further work on the wording by adding an alternative word (braises ‘embers; coals’) to VI1’s suggestion, explicitly marking both VI1’s suggestion and her own as viable alternatives with ou ‘or’. Like in other uses of voilà, the confirmable is treated as one adequate and confirmation-worthy option among several, but not a final and definitive one. 3.5.3 Replacing candidate understanding + exact(ement) In Excerpt 15, from a university setting, STU has called about a so-called DAEU degree, and just prior to the excerpt, has been referred to another person at the university. Here, the call-taker UNI refers to both the relevant organisational unit (line 3) and the person (line 4), and it is the former that becomes the target of the candidate understanding (line 5). Excerpt 15. (UBS.496_1:04) 01
STU:
02
03
ah d’accord c’est elle qui saura plus me PRT okay it=is she who know:FUT more me.DAT oh okay she’s the one who’ll be better placed to dire d’accord say:INF okay advise me okay
UNI:
04
oui (.) c’est ça c’est l’bureau qui s’en yes it=is that it=is the=office REL REFL=of.it yes (.) that’s it it’s the office in charge of occupe c’est vot’ secrétaire takes_care it=is your secretary that it’s your secretary
05
STU: → c’est à l’ADEFOPE [ quoi it=is at DEF=ADEFOPE PRT it’s at the ADEFOPE PRT
06
UNI: ⇒
07
08
]
[c’est la] secrétaire- exactement it=is the secretary- EXACTEMENT it’s the secretary- EXACTEMENT c’est la secrétaire du D.A.E.U. littéraire it=is the secretary of.the DAEU literary it’s the secretary for the DAEU literary degree
STU:
d’accord mais j’pensais savoir si quelqu’un /…/ okay but I=thought know.INF if someone okay but I thought I’d find out if someone /…/
Unlike UNI’s non-recognitional reference (l’bureau qui s’en occupe ‘the office in charge of that’, lines 3–4), STU’s replacing candidate understanding (line 5) involves a recognitional (the acronym l’ADEFOPE), which is also somewhat jar-
165
166
Rasmus Persson
gonistic (specific to this university, and presumably not known to every potential student). The turn ends with the final particle quoi, which regularly occurs at the end of reformulations, among other things. At line 6, UNI begins extending the person reference, before she cuts off and confirms STU’s candidate understanding with exactement. UNI then immediately resumes with the person reference and adds more descriptive detail to it, further accounting for the referral. The lexical semantics of exactement seems to be mobilised to acknowledge not only that the reference is adequate, but that it is optimal. Although the institutional representative herself, UNI, used a less specific circumlocution, there are no corrective overtones (cf. Section 3.5.1), since the original reference is not quite rejected – simply, a non-recognitional was used where a recognitional turned out to be possible (Sacks and Schegloff 1979). Furthermore, exactement appears to be selected as an appropriate option for this particular epistemic scenario, where STU used the more proper reference form, reflecting more knowledgeability than was foreseen by UNI’s initial reference. In fact, throughout the collection and across all inference-types, confirmations with exact(ement) are regularly deployed in cases where the inference-producer has displayed greater independent epistemic access than the talk so far had made expectable (see e.g. Excerpts 8 and 11). In sum, confirmation by repetition of the proffered candidate understanding serves to recognise the replacement as correct, and as what Self “meant to say”, or even appropriately should have said (thus, Self ’s repetitional SPP may orient to Other’s FPP as to some degree corrective, even though Self retains the opportunity and authority to confirm or reject the candidate alternative reference as the intended meaning). Voilà confirmations, on the other hand, do not acknowledge correction but treat Other’s alternative reference as simply being another way of putting it, and underscore that Self ’s original form was (also) appropriate for its local purpose. Exact(ement) appears to acknowledge Other’s alternative reference as an optimal one – and as testifying to more knowledgeability than could have been expected.
3.6 Formulations Offering a formulation of prior talk may amount to summarising or glossing it, or explicating its gist or upshot (Heritage and Watson 1979, 1980; Heritage 1985). Unlike the other types of inferences (3.3–3.5), these are overwhelmingly done in clausal or sentential declarative form. Self ’s choice of confirmation practice is sensitive to Other’s epistemic agency, i.e., how far beyond the words Other’s formulation goes in terms of inferential work (cf. Thompson, Fox, and Couper-Kuhlen 2015: 128–34).
Chapter 5. Three practices for confirming inferences in French
3.6.1 Formulation + repetition As could be expected based on prior work on English (see, e.g., Schegloff 1996), repetitional confirmation in this context indexes that the formulated understanding spells out something that prior talk had conveyed inexplicitly; put differently, repetition serves to ‘confirm an allusion’. Excerpt 16 shows a case where a student calls to ask about when exam results will be available (lines 1–2), and the call-taker answers with an approximate time (line 4). This answer is the target of several post-expansion sequence initiations: first a repetition that occasions confirmation (line 5–6), then another turn that defers acceptance of the answer by questioning its credibility on the basis of previous experiences (lines 7–9). This is responded to with an account and a reiteration of the answer (lines 10–13). Finally, there is the focal formulation (line 15) that elicits confirmation (line 17), before the sequence is closed and the pre-closure of the call is initiated (lines 18–20). Excerpt 16. (UBS.119_0:06) 01
STU:
et j’voudrais savoir eu:h à_partir_de quelle and I=want.COND know.INF uh from which and I’d like to know at what time
02
heure s’ront affichés les résultats hour be.FUT posted the results the results will be posted
03
(.)
04
UNI:
.hh eu:h VE:RS dix_sept heures hh uh around seventeen hours .hh uh: arou:nd five o’clock hh
05
STU:
dix_sept heures seventeen hours five o’clock
06
UNI:
oui yes yes
07
STU:
.hhhh d’accord >p’ce_que l’année dernière on okay ’cause DEF=year last IMPRS .hhhh okay ’cause last year they’d
08
nous avait dit dix_sept heures et m-< us.DAT had say.PPCP seventeen hours and told us five o’clock and (n-)
09
vers midi c’était déjà: affiché around noon it=was already posted around noon they were already: posted
167
168
Rasmus Persson
10
UNI:
ben c’est possible aussi hein mais les PRT it=is possible too PRT but the well that’s possible too y’know but the
11
secrétaires prennent toujours une marge (au cas où) secretaries take always a margin in case REL secretaries always keep a margin (in case)
12
y aura un souci: eu:::h surimposé there be:FUT a problem uh added_on there’s a problem: that uh::: comes up
13
.hhh donc je dis bien V:ERS dix_sept heures hein so I say indeed around seventeen hours PRT .hhh so I do say AROUND five o’clock y’know
14
(0.2)
15
STU: → d’accord ça peut être un peu avant okay it can be:INF a bit before okay it might be a bit earlier
16
(0.4)
17
UNI: ⇒ !t (0.4) ça peut être un peu ava[nt] it can be:INF a bit before !t (0.4) it might be a bit earlier
18
STU:
19
UNI:
[ d]’accord hh okay okay hh °voilà° PRT
°right° 20
STU:
bon ben j’vous remercie PRT PRT I=you.ACC thank alright well thank you
After the student (lines 7–9) mentions last year’s results coming around noon, and not at five as foreseen, UNI (in lines 10–13) underscores the approximate nature of the time reference; note the accent on V:ERS ‘around’, which was not included in STU’s repetition in line 5. STU’s formulation in line 15 involves rather little inferential work, doing little more than articulating the implication of the accented word in relation to the prior talk about a time margin. After some delay (a 0.4 s pause, a talk-projecting click, and another 0.4 s pause), UNI confirms with repetition in line 17, which gives a sense that UNI is answering patiently, but treating the inference as spelling out what should already be clear. In other cases in the collection, the confirmer continues their turn after the repetition with expressions such as c’est évident ‘that’s obvious’, lexically conveying the redundancy of the formulation turn.
Chapter 5. Three practices for confirming inferences in French
3.6.2 Formulation + voilà In Excerpt 17, from a radio phone-in show, the (French) caller Baptiste (BAP) is spending a year in Finland as part of a student exchange program. After a multiunit response about which class Baptiste is in (lines 3–4 and 6), the host Éric proffers a formulation (lines 8–9) of the gist of Baptiste’s answer, thus initiating a post-expansion sequence. Excerpt 17. (MOUV.11–04_50:15) 01
ÉRI:
02
03
t’es en quelle classe ’fin tu: tu fais quoi you=are in which class I_mean you you do what which class are you in I mean what: what are you comme études as studies studying
BAP:
04
eu:h en franc:e l’année dernière j’étais e:n uh in france the=year last I=was in uh: in france: last year I was in the: penultimate premièr:e euh scientifique .hh first uh scientific year of the science stream ((high school))
05
ÉRI:
d’accord okay okay
06
BAP:
et donc quand je rentre: je vais e:n terminale hhhh and so when I go_back I go in final and so when I go back: I’ll be in: final year hhhh
07 08
(.) ÉRI: → d’accord donc t’es à l’é- mm: okay so you=are PREP the=(e-) okay so you’re in the e- mm:
09
→ dans l’équivalent de la première là in the=equivalent of the first PRT in the equivalent of the penultimate year now
10 11
(0.3) BAP: ⇒ voilà hhh PRT
VOILÀ hhh 12
13
ÉRI:
mais- mais je- j’savais même pas qu’ça (but-) but (I-) I=knew even NEG COMP=that but- but I- I didn’t even know that this existait ce genre de programme pour le::s existed this sort of program for the sort of ((exchange)) program existed for::
169
170
Rasmus Persson
14
pour les lycé:ens c’e:st for DEF high_school_students it=is for high school students it’s:
The multi-unitness of Baptiste’s response suggests that there is no straightforward answer to Éric’s question (strictly speaking, he is currently not in any class in the French system, but doing an additional year). However, Éric’s formulation (line 8) nonetheless sums it up in terms of an equivalence, which is one sort of operation that requires some inferential work (going some way beyond the words said). The formulation gets confirmed with voilà (line 11), before the host moves on (lines 12–14). While Éric’s formulated understanding may not quite have been what Baptiste designed his turn to convey, the particle voilà allows Baptiste to confirm it as a warranted and acceptable approximation of the gist of what he said. Generally, voilà serves to confirm the formulation while recognising it to be a somewhat more epistemically agentive inference, that reformulates prior talk in way that gets “close enough” to the essence of Self ’s talk (unlike confirmations with repetition and exact(ement), both of which would treat the formulation as “spot on”, albeit on different grounds). 3.6.3 Formulation + exact(ement) In Excerpt 18, the university call-taker is explaining the enrolment procedure to a prospective student’s mother (CAL). The understanding negotiated here will turn out to be the separation of two issues: when to call and make an appointment, and when to make it for (UNI’s turn at line 3 is ambiguous). Excerpt 18. (UBS.436_3:57) 01
UNI:
she calls for make:INF an appointment
02
CAL:
oui voilà yes PRT yes right
03
UNI:
donc entre le sept et le dix_sept septembre so between the seven and the seventeen september so between the 7th and the 17th september
// (( 46 seconds of talk not directly related )) // (( to the time for the appointment )) 04
05
CAL:
et faut prendre le rendez_vous ava:nt and is_necessary take:INF the appointment before and one must make the appointment befo:re le sept the seven the 7th
Chapter 5. Three practices for confirming inferences in French
06
UNI:
(1.3) ben oui[: ] PRT yes (1.3) well yes:
07
CAL:
[mm] PRT
mm 08 09
(0.4) UNI:
10 11
(.) UNI:
12
13
oui yes yes
il faut prendre- enfi:n .h it is.necessary take:INF I_mean .h one must make- I mean: .h en prin[cip]e quoi in principle PRT in principle y’know
CAL:
[mm ] PRT
mm 14
CAL:
[oui parce_que: ] yes because yes because:
15
UNI:
[à- à_moins_qu’e]lle veuille venir que le: un- unless=she want:SBJV come:INF only the un- unless she doesn’t want to come until the:
16
que le qui:nze ou le:: >j’s’pas comment only the fifteen or the I=know=NEG how until the 15th or the:: >I don’t know how
17
ça tombe moi< that falls 1SG that plays out
sept et so_that=it AUX.SBJV set:PASS between the seven and so that it’s scheduled: between the >7th and
171
172
Rasmus Persson
22
→ le dix_sept< the seventeen the 17th
ama ne::- ku: yueysussi-la kulay-ss-ci ne. probably you DEM USC-DECL say-PST-COMM you Probably you::- uh: you said you’re at USC ci.
08
LIA:=> ney: PRT
Yes: 09 10
11
(0.3) HAN:
ne ama han tal hwu-ccum-ey: you probably one month after-about-LOC You probably after about a month: nay chinkwu waiphu ka-l ke-ta keki my friend wife go-FUT thing-DECL there my friend’s wife will go there
In producing the pseudo-tag question marked with -ci (line 7), Han invokes a past event when Lia had said she’s at USC and indicates his epistemic access. Han seeks to (re-)confirm his prior knowledge of the information he had already been told about in the past, thereby taking up a knowing stance. In so doing, he also seeks to secure the background that is preliminary to the upcoming action.
Chapter 6. Complexities of responding
At line 8, Lia produces a type-conforming ney: ‘yes:’ alone in response. She simply confirms the matter at hand, fully acquiescing to the terms and constraints of the question (Lee 2015; see Raymond 2003). Lia treats the question as seeking confirmation and merely confirms the questioner’s prior knowledge, acquiescing to the knowing stance taken up by the questioner. In so doing, Lia also orients to the preliminary nature of the current sequence, not developing the sequence further but inviting progression to the upcoming projected action. At lines 10–11, Han moves to launch the projected announcement. Similarly, in Excerpt 4 below the respondent produces an affirmative particle alone in response to a pseudo-tag question addressing the information that has been told in prior talk. Jin and Yoo have been talking for a while on the phone. In the segment below, Jin launches a telling about a person they both know, Ms. Miso Pak. She had previously said that Ms. Pak became very pretty (see lines 1–4) but has recently realized that she is not (see lines 27–28). Jin’s telling concerns how and in what aspects she has realized Ms. Pak is not pretty. At lines 1–4, Jin starts by invoking a past event when she had told Yoo that Ms. Pak became very pretty, which serves as a background of the telling. Yoo produces ung ‘yes’ a few times (lines 5–6), establishing recognizability of the past event and inviting Jin to continue. Before moving on to the telling itself, Jin produces a wh-question about the time of the past event (line 7). Yoo responds by conveying its recency, rather than providing a specific time sought by the question (line 9; see Fox and Thompson 2010). Excerpt 4. (Two women) 01
JIN:
kuntey iss-canh-a cepen-ey by.the.way exist-NCHAL-IE last.time-LOC By the way you know last time
02
way: ku wuli ceki: nwukwu-nya pak.miso-ssi why DEM we DEM who-Q ((full name))-HT uh we well: who is it Ms. Miso Pak
03
nay-ka .hh nemwu ippecye-ss-ta-ko I-NOM very become.pretty-PST-DECL-COMP I .hh {said she} became very pretty
04
nay-ka kulay-ss-canh-a [cinanpen-ey po-kwu-se:= I-NOM say-PST-NCHAL-IE last.time-LOC see-CONN-after [] {I said} after I saw {her} last time=
05
YOO:
[ung PRT
[] Yes 06
YOO:
=ung ung PRT
PRT
=Yes yes
187
188
Seung-Hee Lee
07
JIN:
08
kukey encey-ni? DEM.NOM when-Q When was that? (0.2)
09
YOO:
mwe e- elma.an.tway-ss-e what not.long-PST-IE Well it was not long ago
10
JIN:=> elma.an.tway-ss-ci, not.long-PST-COMM It was not long ago ci,
11
YOO:=> ung. PRT
Yes. ((meaning ‘no’ in English)) 12
JIN:
a kulenikka ceki ku: sangka an-eyse pwa-kacikwu ITJ so DEM DEM store inside-LOC see-because Oh so well uh: because I saw her inside a store
13
hwan:hay poye-ss-kwuna=eckucey shiny look-PST-FR a.few.days.ago she looked shi:ny=a few days ago
14
nay-ka tto way .hh wuli thimjang-nim ttalaI-NOM again why we team.manager-HT follow when I again uh .hh followed our team manager-
15
ne an o-n nal you NEG come-ATTR day the day you didn’t come
16
YOO:
ung ung PRT
PRT
Yes yes //10 lines of transcript of JIN’s telling (13 seconds) are omitted// 27
28
JIN:
kulayse pak.miso-ssi manna-ss-nuntey .hh so ((full name))-HT meet-PST-CIRCUM So I met Ms. Miso Pak and .hh pakk-eyse po-nikka ani-te-la-kwu: outside-LOC see-CAUS not-RETROS-DECL-CONN when I see her outdoors she’s not {pretty}:
At line 10, Jin produces a pseudo-tag question, repeating and seeking confirmation of the prior response. In constructing the question, Jin fully repeats Yoo’s response (line 9) but with modification of its sentence-ending suffix. She uses the committal -ci ending (line 10) in replacement of the informal ending -e in the prior response (line 9), thus formulating a pseudo-tag question. Jin thereby indicates the information as already available, taking up a knowing stance. She regis-
Chapter 6. Complexities of responding
ters, and seeks to confirm, the information that has just been directly stated in the prior turn. In response, Yoo confirms the question with an affirmative particle alone at line 11. (Note that Korean is an agree-disagree language, as described earlier, and it is the affirmative particle, ung, that confirms the question – i.e., that ‘it was not long ago’ – in this context, although it means ‘no’ in English because the question has a negative element.) In producing the affirmative particle alone (line 11), Yoo simply confirms the question by accepting the terms and constraints of the question as well as the questioner’s knowing stance indexed in the question. In so doing, Yoo also invites Jin to move on to the larger activity in progress. In the following turns, Jin moves on to the telling, first dealing with her previous, mistaken assessment about Ms. Pak – that Ms. Pak looked shiny inside a store (lines 12–13) – and then elaborating on a more recent event when she has met Ms. Pak and realized she is not pretty (lines 13–28). In sum, respondents produce affirmative particles alone in response to pseudo-tag questions seeking confirmation of “old” information that had already been told in the past or in the interaction. Pseudo-tag questions in the dataset are largely addressed to the state of affairs to which questioners have pre-existing access by reference to interactional history, prior interaction, and so on. When responding to such questions with unelaborated yes particles alone, respondents treat the question as seeking confirmation of the information that is already available, and thus as indexing the questioner’s knowing stance. Unelaborated typeconforming affirmative particles that occur solo in the turn merely confirm the proposition formulated by the question. They do so by accepting not only the terms and constraints set by the question but also the questioner’s knowing stance indexed in the question.
5.
Elaborated type-conforming responses
We have shown that confirming responses to pseudo-tag questions tend to be unelaborated. Because pseudo-tag questions are largely addressed to the information that is already available and index questioners’ knowing stance, unelaborated, minimal type-conforming yes particles providing a mere confirmation can be made specifically relevant (see Heritage and Raymond 2012; Lee 2015). Thus, when a type-conforming response is constructed with more than a simple affirmative particle in response to pseudo-tag questions, it can be done “for cause” (cf. Raymond 2003). Such elaborated type-conforming responses in which affirmative particles occur with additional components in the turn are uncommon in the dataset (Table 1). In elaborated responses, each element of a
189
190
Seung-Hee Lee
type-conforming affirmative particle and elaboration seems to engage in different interactional work. An affirmative yes particle confirms and accepts the question’s candidate proposition; and elaboration tends to address the action implemented by the question, while departing from the terms and epistemic positioning set by the question. Through the use of elaborated type-conforming responses, respondents – while confirming – work to avoid negative or unwanted consequences that an affirmative particle alone can bring in particular sequential and interactional contexts. For example, consider Excerpt 5 from an earlier part of the same telephone conversation as Excerpt 3. As described above, Lia is a graduate student in the US and Han is her friend in Korea. Apparently they are not in a relationship to make international calls to talk to one another. When getting a call from Lia, Han produces a wh-question that displays a presupposition that Lia has come to Korea (line 1). At line 2 Lia rejects the presupposition in response, implying that she is making an international call. Given this response, Han first claims a change of state in his understanding (e ‘oh’; see Heritage 1984), and then produces a polar question addressed to Lia’s whereabouts by displaying his revised understanding (line 3). Lia produces a simple, confirming yey:: ‘yes::’ response (line 4), which can raise a question or surprise about her international call. At line 6 Han displays a surprise through the intoned production of a response cry o::: ‘wow:::’ (Goffman 1978). Excerpt 5. (Stressed out) 01
HAN:
encey wa-ss-e. when come-PST-IE When did you come.
02
LIA:
.hhh hankwuk ani-ey-yo(h). heh heh Korea no-COP-POL .hhh I’m not in Korea(h). heh heh
03
HAN:
e
mikwuk-i-ya? America-COP-IE Oh you are in the States? ITJ
04
LIA:
.hh yey:: PRT
.hh Yes:: 05 06
(0.5) HAN:
o:::= ITJ
Wow:::= 07
LIA:=> =nolla-sye-ss-cyo:. be.surprised-HON-PST-COMM.POL You are surprised: ci.
Chapter 6. Complexities of responding
08 09
(0.4) HAN:=> e
toykey nollay-ss-e. very be.surprised-PST-IE Yes I’m very surprised. PRT
10
LIA:
11
HAN:
hah hah hah .hh kunyang hanpen [haypwa-ss-e-yo= just once do-PST-IE-POL [] Hah hah hah .hh I just called= [chakha-ta= nice-DECL [] You’re nice=
12
LIA:
=sayngkakna-se think-because =Because I thought of you
At line 7 Lia produces a pseudo-tag question, which is addressed to Han’s surprise that has been conveyed in his response cry (line 6) as well as by implication in his prior, mistaken assumptions about Lia’s call and her whereabouts (lines 1–4). Lia thus seeks confirmation of Han’s surprise that is already made available in the interaction. This can make a simple confirmation relevant in the next turn. At the same time, however, Lia’s question proposes a first assessment about Han’s state of mind. While downgrading the first-position assessment by using the form of a pseudo-tag question (see Heritage and Raymond 2005; Raymond and Heritage 2006), Lia brings up and articulates Han’s surprise in her own words, something that has been conveyed but not explicitly stated so far. Lia thus claims thematic agency over Han’s surprise, which Han has primary rights to assert (Enfield et al. 2019; Heritage and Raymond 2005; Raymond and Heritage 2006). In addition, by explicitly verbalizing Han’s surprise, which had only been indirectly conveyed in prior talk, Lia also joins with Han in treating the call as surprising. She proposes an affiliative, “empathic” understanding through the question (see Heritage 2011). Lia’s pseudo-tag question thus puts Han into a position in which he can risk being heard as merely reactive, as making no independent claim about his own surprise nor addressing Lia’s affiliative understanding – when producing a mere confirmation as made relevant by the form of the question (see Heritage and Raymond 2005; Raymond and Heritage 2006). At line 9, Han constructs an elaborated type-conforming response in managing these contingencies. He not only confirms the question with an affirmative particle (e ‘yes’), but also constructs an upgraded second assessment. By asserting his upgraded surprise, Han not only indicates the independence of his assessment but also provides an agreement with the first assessment (see Heritage and Raymond 2005; Hayano 2011; Raymond and Heritage 2006). Han thus avoids being heard as merely going along with the questioner, as possibly conveying that
191
192
Seung-Hee Lee
he is not really surprised about, and as a result may not appreciate, Lia’s international call. He achieves a stronger affiliation with Lia, who has verbalized and recognized his surprise, potentially implying an appreciation of the international call. Note that Han implies an appreciation by producing a compliment at line 11. Thus, through the production of the elaborated type-conforming response, Han confirms the question and avoids possibly unwanted interactional understandings, while departing from the relevance of a mere confirmation set by the form of the question. In Excerpt 6 below, the respondent also deals with complex contingencies in producing an elaborated type-conforming response. At the beginning of the call, Min asks whether Lee is at home (line 1) and, in her next turn, produces an announcement that she is done with some activity (line 4). It is not articulated in the call what that activity is, but apparently both parties have a shared understanding about what it refers to. In response to the announcement (line 4), Lee produces an acknowledgement, e: ‘yes:’, and follows with a pseudo-tag question proposing an assessment of the activity (line 5). Excerpt 6. (Home) 01
MIN:
a
enni cip-i-sey-yo? sister home-COP-HON-POL Oh sister you’re home? ITJ
02 03
(0.2) LEE:
ung: PRT
Yes: 04
MIN:
05
LEE:=> e: nemwu himtul-ci::. PRT too hard-COMM Yes: it’s too hard:: ci.
06
MIN:=> yey hh cang:nan ani-yey-yo hh .hh PRT joke no-COP-HON Yes hh it’s no jo:ke hh .hh
07
a: .hh a(hh) cikum ta kkuthna-ss-eyo. ITJ ITJ now all finish-PST-POL Oh: .hh oh(hh) I’m just all done.
08
=> son-i mak pelpelpelpel ttellye-yo cikum(h)= hand-NOM severely brrr shake-POL now my hand is now(h) shaking badly brrrr= LEE:
=cincca? really =Really?
At line 5, in first producing e: ‘yes:’ Lee can acknowledge an inference from Min’s announcement, treating the prior turn (line 4) as displaying that the activity has
Chapter 6. Complexities of responding
been hard work. (Note that Min produces her announcement with in-breaths and outbreaths at line 4.) Lee then proposes this understanding explicitly in the form of a pseudo-tag question, indexing a knowing stance (line 5). She treats the information as already conveyed in the prior turn, thus as in need of a mere confirmation. While seeking confirmation by marking the question with the committal -ci ending on the surface, Lee can nonetheless invite its recipient to not merely confirm the question. On the one hand, Lee’s question proposes a first assessment. While downgrading her assessment through the use of a pseudo-tag question (see Heritage and Raymond 2005; Raymond and Heritage 2006), Lee proposes, and articulates in her own words, a first assessment about the activity Min has just completed. This can make Min – who has primary rights to evaluate the activity – assert her own assessment in response (see Heritage and Raymond 2005; Raymond and Heritage 2006). On the other hand, Lee’s question occurs in response to an announcement (line 4) that projects an action of elaborating (Jefferson 1981; Maynard 2003). In producing the question, Lee displays an understanding about, and articulates explicitly, what Min has implicitly conveyed in her announcement. Lee thus proposes an empathic evaluation in affiliation with Min’s announcement (see Heritage 2011), inviting Min to move on to the projected sequential action of elaborating on the announcement. All these aspects of the question can pose complex contingencies for the recipient to manage. The recipient may simply confirm the question addressed to the information she has already conveyed. However, the recipient not only has initiated an announcement sequence that projects an elaboration of the announcement (Jefferson 1981; Maynard 2003), but also has primary rights to evaluate the activity in question with first-hand experience (Heritage and Raymond 2005; Raymond and Heritage 2006). The use of an affirmative particle alone in the response turn can thus be ill-fitted in terms of the sequential trajectory initiated by the announcement, as well as in terms of the recipient’s epistemic rights. At lines 6–7, Min addresses these contingencies by constructing an elaborated type-conforming response. She first confirms the question with an affirmative particle, yey ‘yes’, as made relevant by the form of the question. She then follows with elaboration providing a second assessment. In so doing, Min not only uses a different, but stronger evaluative expression, cang:nan ani-yey-yo ‘it’s no jo:ke’ (line 6), but also describes the physical condition she is directly experiencing at the moment (line 7). Min thus evidences her immediate, first-hand access to the activity she has just completed, in comparison with Lee’s assessment of the activity in general. In addition, Min also elaborates on her announcement (line 4) in providing these additional components. She accomplishes the sequentially relevant action of elaborating, which is set in motion by her announcement (Jefferson
193
194
Seung-Hee Lee
1981; Maynard 2003). By constructing elaboration in addition to the affirmative particle, Min works to defend her epistemic rights, achieve the projected sequential action, and confirm the question. Min manages multiple contingencies of responding within the boundaries of a ‘yes’ response. Thus, in Excerpts 5–6 above, respondents construct elaborated typeconforming responses in dealing with pseudo-tag questions proposing first assessments. Respondents first produce an affirmative particle, confirming the question. They then provide elaboration that is addressed to the action performed by the question as well as to their epistemic rights to assert their evaluation, although departing from the expectation of a mere confirmation encoded in the form of the question. In Excerpt 7 below, the respondent seems to have a different purpose in producing an elaborated type-conforming response. Kim and Han are graduate students at different universities in the US. Kim called Han after a long period of no contact. It turns out that Kim graduated from the graduate school and moved to New York (see lines 1–6), but Han had not been updated on this. Right after the conversation opening, at line 1 Han produces a wh-question about Kim’s whereabouts, but repairs that in favor of a polar question. He first uses an indexical keki ‘there’ and then specifies that by providing a particular location, Indiana, where Kim used to live (line 1). At line 3 Kim provides a disconfirming response. He produces a negative ‘no’ particle and names a substitute location. He further elaborates on this by telling that he has finished at line 6. In so doing, Kim does not specify what he has finished, and could be conveying either that he has finished a semester and is visiting New York, or that he has finished, that is, has graduated from the school and moved to New York. Excerpt 7. (Bar exam) 01
HAN:
02 03
(0.5) KIM:
04 05
eti-ya, keki-ya? intiayna-ya? where-IE there-IE Indiana-IE Where are you, you’re there? You’re in Indiana?
ani na- cikum nyuyok-i-ya PRT I now New.York-COP-IE No I’m- now in New York (.)
HAN:
e::::= ITJ
Oh::::= 06
KIM:
=e::.=kkuthna-ko: nyuyok-ey wa-ss-ta PRT finish-CONN New.York-LOC come-PST-DECL =Yes:.=I’ve finished and: came to New York
Chapter 6. Complexities of responding
07 08
(0.8) HAN:=> a- hakkyo ta- ani- panghak- a ceki ceki- (0.3) ITJ school all PRT vacation ITJ DEM DEM Oh- the school all- no- vacation- oh uh uh- (0.3)
09
=> keki panghak-i-ci? = cikum there vacation-COP-COMM now it’s a vacation there ci?=now:.
10 11
(0.2) KIM:=> e:
a na colephay-ss-e. I graduate-PST-IE Yes: well I graduated. PRT ITJ
12 13
(1.2) HAN:
a
cincca-ya? real-IE Oh really? ITJ
At lines 8–9, Han addresses this ambiguity and proposes his candidate understanding in the form of a pseudo-tag question. He seems to first start by proposing that Kim has finished the school (hakkyo ta- ‘the school all-’; line 8). However, Han repairs that in favor of a pseudo-tag question asking for confirmation that it’s a vacation (line 9). This proposes an understanding that Kim is only visiting New York because he has finished his semester and is now on vacation. In producing this question, Han does not orient to the prior turn (line 6) as providing an update on a totally new life event such as graduation. Instead, Han proposes an understanding that displays his pre-existing knowledge about Kim’s life, treating the matter as already available. Han thereby tries to (re-)establish shared knowledge with Kim, which he failed to do in prior talk (see lines 1–6). In response to the pseudo-tag question, Kim first produces an affirmative particle, confirming that it’s a vacation, but then follows with elaboration that in effect departs from the question (line 11). He produces an announcement that he graduated, which revises the question’s mistaken, underlying assumption that Kim is still in school (line 9). Kim thereby avoids an inappropriate implication the yes particle alone can bring, that Kim is on vacation and is only visiting New York. The elaboration (line 11) thus departs from the question, displaying its inappropriateness. This shows that there is a division of labor in Kim’s elaborated response. The affirmative particle confirms the question at its face value, treating the proposition on the surface as true and correct; and the elaboration departs from the question and its underlying assumption, resisting an irrelevant understanding that is implied by the question. Through this division of labor, Kim can also acknowledge Han’s efforts to establish common ground by confirming shared
195
196
Seung-Hee Lee
knowledge, while constructing the elaboration as a departure from the question (see Clark 1996; Enfield 2006). In sum, respondents deal with various contingencies in producing elaborated type-conforming responses. Their use of an affirmative particle alone, while confirming the question, can bring unwanted or ill-fitted consequences in which respondents merely go along with and do not really affiliate with the questioner (Excerpt 5), fail to claim epistemic rights and to develop a projected action (Excerpt 6), or lead to misunderstandings about their circumstances and disaffiliate with the questioner who is trying to establish common ground (Excerpt 7). Thus, respondents produce elaboration so as to address complex contingencies involved in providing confirmation. Through the production of elaborated responses, respondents engage in multiple interactional work in which they confirm the question and in effect affiliate with the questioner, while departing from the relevance of a mere confirmation. This suggests a division of labor in elaborated type-conforming responses, with a yes particle and elaboration each addressed to different contingencies of responding (see Stivers 2011).
6.
Nonconforming responses
In only a small number of cases, respondents provide confirming responses to pseudo-tag questions without the use of type-conforming affirmative particles (16.3%; Table 1). They construct nonconforming responses (Raymond 2003), treating the question as problematic in various ways. Consider Excerpt 8, for example, which illustrates two instances of nonconforming responses to pseudo-tag questions. Cho and Jay are undergraduate students at the same department. Jay is preparing for Chartered Financial Analyst (CFA) exam and planning to enroll in a preparation course for the Level 1 CFA exam. He has been making rounds of calls to find people who would be interested in studying for the course together. When he called Cho and asked about this, Cho showed no interest. Given Cho’s rejection, Jay asked whether Cho knows people who are preparing for the Level 1 exam. In response, at lines 1–2 Cho tells that those preparing for the CFA exam are now studying for Level 2. At lines 3–4, Jay acknowledges that lots of people took the Level 1 exam last year, conveying an agreement with Cho’s position. At lines 6–8, Cho again asserts that those around are all preparing for Level 2 (see lines 1–2). This can imply the inappropriateness of Jay’s agenda concerning the Level 1 exam.
Chapter 6. Complexities of responding
Excerpt 8. (Speak louder) 01
CHO:
02
cwunpiha-nun ay-tul-un ta prepare-ATTR kid-PL-TOP all {Well} those preparing are all
03
cikum i-cha kongpwuha-nuntey mwe=ay-tul ta.= now two-level study-CIRCUM what kid-PL all. now studying for Level Two=all of them.= JAY:
=kukka ay-tul: il-cha manhi pwa-ss-canh-a:. so kid-PL one-level a.lot see-PST-NCHAL-IE =So lots of them took Level One:.
04
caknyen-ey. last.year-LOC last year.
05
(0.5)
06
CHO:
kulenikka: i-cha: ta iso two-level all two So: Level Two: all Two-
07
cwupyen-ey: iss-nun ay-tul-un ta around-LOC exist-ATTR kid-PL-TOP all those around: are all
08
i-cha cwunpihay = cikum two-level prepare:IE now preparing for Level Two=now
09
(0.2)
10
JAY:1=> cwupyen-ey iss-nun ay-tul taaround-LOC exist-ATTR kid-PL all Those around are all-
11
1=> cwupyen-ey iss-nun ay-tul-ila-nun key around-LOC exist-ATTR kid-PL-DECL-ATTR thing.NOM by those around
12
1=> wuli kkwa chinkwu-tul yaykiha-nun ke-ci:. our department friend-PL talk-ATTR thing-COMM you mean our friends in the department: ci.
13 14
15
16
(0.8) CHO:1=> kkwa chinkwu-tul mwe department friend-PL what Friends in the department {and} uh 1=> kunyang mwe nay-ka a-nun talun chinkwu-tul just what I-NOM know-ATTR different friend-PL just uh other friends I know JAY:2=> ta i-cha ha-ci? all two-level do-COMM All are doing Level Two ci?
197
198
Seung-Hee Lee
17
(0.2)
18
CHO:2=> ku-ci ta i-cha [ha-ci be.so-COMM all two-level do-COMM [] Of course all are doing Level Two
19
JAY:
[a: kulenikka ITJ so [] Oh: so
20
ppalli il-cha hay-ya-ci.= fast one-level do-NECESS-COMM I have to do Level One fast.=
21
=aissi il-cha an ha-ko ITJ one-level NEG do-CONN =Gee without doing Level One
22
i-cha ha-l swu.eps-canh-a two-level do-ATTR cannot-NCHAL-IE I cannot do Level Two
At lines 10–12 (1=>), Jay starts by repeating the prior turn (cwupyen-ey iss-nun aytul ta- ‘those around are all-’; line 10), which can be geared to registering Cho’s prior assertion. However, Jay abandons that, and produces a pseudo-tag question (lines 11–12) addressed to the reference of cwupyen-ey iss-nun ay-tul ‘those around’ in Cho’s prior turn (lines 7–8). By proposing wuli kka chinkwu-tul ‘our friends in the department’ as a candidate reference in particular (line 12), Jay narrows down the range of people preparing for Level 2, as well as those Cho has relationships with. Jay can thus implicate a possibility for finding people preparing for Level 1 beyond friends in the department, further pursuing his agenda. In proposing this candidate understanding, Jay requests confirmation by using the form of a pseudo-tag question (lines 11–12). In response, Cho constructs a nonconforming response (lines 14–15; 1=>). He first partially repeats the term used in the question, kkwa chinkwu-tul ‘friends in the department’, without producing wuli ‘our’ (line 14; cf. line 12), which can potentially increase the range of friends in the department beyond their mutual friends. Then Cho continues by including ‘other friends’ from Cho’s other kinds of relationships, further broadening the range of people preparing for Level 2 (line 15). By expanding the scope of people Cho had referred to in his prior turn (lines 6–8), Cho resists the implication of the pseudo-tag question. Cho’s nonconforming response (lines 14–15), although (partially) confirming the question, departs from the question’s terms and constraints as well as the underlying agenda pursued by the question. Following these several turns at talk establishing that most people are preparing for Level 2 (lines 1–15), Jay produces another pseudo-tag question that is again
Chapter 6. Complexities of responding
addressed to this issue (line 16; 2=>). He seeks to re-confirm the information that had already been directly asserted a few times across prior turns. Although requiring a mere confirmation and indexing his knowing stance by using the form of a pseudo-tag question, Jay still treats the matter as in need of confirmation. In response to the pseudo-tag question, Cho again constructs a nonconforming response (line 18; 2=>). This time Cho produces a marked confirmation, ku-ci ‘of course’, and a full repetition of the question, ta icha ha-ci ‘all are doing Level 2’. These two forms of confirmation engage in different interactional work. First, in producing the marked confirmation, Cho orients to the self-evident nature of the matter at hand, which had already been explicitly stated and established in prior talk. He treats the issue as no longer a questionable matter, problematizing the legitimacy of the question, as the questioner has already been provided with the information in the interaction (see Stivers 2011, 2019). In producing the repetitional response following the marked confirmation (line 18), Cho asserts the proposition formulated by the question. He independently asserts the information, rather than constructing the turn as a mere confirming, reactive answer to the question. Cho thus claims his thematic agency and epistemic rights over the proposition (see Enfield et al. 2019; Heritage and Raymond 2012; Stivers 2005). By producing these two nonconforming elements, Cho provides confirmation but treats the question as problematic in two different aspects. At lines 19–22, Jay in the end abandons his agenda and shifts the attention to his own need to take Level 1 exam. Finally, in Excerpt 9 below the respondent seems to be more agentive in constructing a nonconforming response. Kim and Jin have been talking about cosmetics and makeup. Kim has been considered an expert and Jin has been complaining that she is not good at makeup. Kim has elaborated on her own makeup, providing advice, and now turns to ask about how Jin does her makeup. At line 1, Kim starts by asking what kind of foundation Jin uses. Jin first describes a particular brand she uses for a makeup base (lines 2 and 4), which is typically applied before foundation, and then moves on to talk about foundation (line 6). At line 7, however, Kim goes back to the makeup base, seeking to confirm the particular product Jin uses. Jin confirms by repeating ‘the white one’ (line 8), and Jin and Kim share the merits of that particular makeup base (6 lines of transcript omitted). At line 16, Jin goes back to talk about foundation, the discussion of which has been halted (see line 6), but again Kim cuts in with pseudo-tag questions about the makeup base Jin uses (lines 17–19). Excerpt 9. (Busy) 01
KIM:
ne:- ceki hwawunteyisyen mwe-l palu-nya?, you DEM foundation what-ACC apply-Q You:- uh what do you use for foundation?
199
200
Seung-Hee Lee
02
JIN:
yosay-nun na-to t- tiol kke: these.days-TOP I-also Dior thing These days I also {use} D- Dior:
03
KIM:
e PRT
Yes 04
JIN:
ku: meyikuep.peyisu palu-kwu, DEM makeup.base apply-CONN Uh: makeup base and,
05
KIM:
e PRT
Yes 06
JIN:
hwawunteyisyen-un= foundation-TOP For foundation=
07
KIM:
=meyikuep.peyisu hayan ke? makeup.base white thing =Makeup base {you mean} the white one?
08
JIN:
hayan ke. white thing The white one.
09
KIM:
e
e
PRT PRT
Yes yes //6 lines of transcript (9 seconds) are omitted: the parties share the merits of applying the particular makeup base product// 16
JIN:
kulikonase meyikuep- a ceki hwawunteyisyen-un, and.then makeup- ITJ DEM foundation-TOP And then makeup- oh uh for foundation,
17
KIM:=> a
ne kuke-n ani-ci? you DEM-TOP no-COMM Oh you don’t mean that ci? ITJ
18
=> ku wenlay: tiol-eyse nao-te-n kuke malko DEM originally Dior-LOC come.out-PST-ATTR DEM not {You mean} not the one originally: from Dior
19
=> malha-nun ke-ci, say-ATTR thing-COMM {but} mean {the other one} ci,
20
JIN:=> hayan ttwukkeng talli-n ke:. mac-[e. white lid be.hung-ATTR thing right-IE [] The one with a white lid:. That’s right.
21
KIM:
[hayanwhite [] White-
Chapter 6. Complexities of responding
22
JIN:=> hayan ttwukkeng talli-n meyikuep.peyisu.= white lid be.hung-ATTR makeup.base The makeup base with a white lid.=
23
=> =thyupu hyeng-ulo toy-n ke= tube form-INST be-ATTR thing =The one in a tube=
24
KIM:
25
JIN:
=thyupu hyeng. hayan [ke. tube form white thing [] =In a tube. The white one. [e
e. kuke malha-nta-nikka say-DECL-CAUS [] Yes yes. That’s what I’m talking about PRT PRT DEM
In producing pseudo-tag questions (line 17–19), Kim again seeks to confirm the particular product Jin uses for a makeup base. Although Kim and Jin have already established a shared understanding about the product (lines 7–15), Kim here again requests confirmation about the makeup base Jin refers to and treats the issue as unresolved (lines 17–19). This time Kim uses different, but unspecific expressions in describing the product (cf. line 7). In the first pseudo-tag question (line 17), Kim uses an indexical kuke ‘that’ and refers to the product Jin uses as not ‘that’. Then in the following pseudo-tag question (lines 18–19), Kim specifies the prior indexical kuke ‘that’ as ‘the one originally: from Dior’, thus describing the product Jin uses as ‘not the one originally: from Dior’ (line 18); but she does not further specify the other product she is referring to in the question (line 19). (Note that Kim does not articulate but only implies ‘the other one’ in Korean.) While providing these unspecific expressions, Kim tries to confirm the particular product Jin uses out of several products from the same cosmetics brand. At line 20, Jin constructs nonconforming responses. She first articulates a new description, ‘the one with a white lid:’, replacing the term provided in the question. Given the questioner’s use of unspecific expressions in referring to the product, Jin provides an alternate, more specific description that can better establish the referent. She thus modifies and resets the terms of the question, constructing a transformative response (Stivers and Hayashi 2010). Jin then produces mac-e ‘that’s right’ (line 20), asserting confirmation that it is the one that they have been talking about with the merits discussed. Given Kim’s continued efforts to confirm the product, Jin asserts correctness and thereby suggests that the question may be unnecessary because the information has already been provided (see Stivers 2019). When Kim starts to repeat Jin’s expression in overlap (line 21), Jin again produces her description by specifying ke ‘the one’ (line 20) as meyikhuep.peyisu ‘makeup base’ (line 22) and follows with another description, ‘the one in a tube’ (line 23). She thus provides more details about the product, again modifying the
201
202
Seung-Hee Lee
terms of the question, so as to better establish the referent. Jin’s transformative responses (lines 20 and 22–23) are thus oriented to, and can accomplish, the agenda pursued by the question, although departing from the terms of the question (see Stivers and Hayashi 2010). In sum, respondents tend to treat a question as problematic in constructing nonconforming responses. They may treat the matter at hand as irrelevant by virtue of prior talk, as already established in the interaction, or as not a matter under question any more. They thus depart from the terms and constraints of the question in various ways. This may not be necessarily disaffiliative, however, because nonconforming responses may serve to better accomplish the agenda pursued by the question (see Excerpt 9). In constructing nonconforming responses without the use of affirmative particles, respondents may obliquely confirm the proposition while engaging in other interactional work addressed to various aspects of the question, such as its framing, relevance, legitimacy, agenda, etc.
7.
Confirming responses, question formats, and epistemic gradients
So far we have examined three forms of confirming responses to pseudo-tag questions: unelaborated type-conforming responses, elaborated type-conforming responses, and nonconforming responses. In these responses, speakers tend to treat the question as addressing the information that is already available, and thus as indexing the questioner’s knowing stance. In most cases, respondents acquiesce to a knowing stance of the questioner by producing an unelaborated affirmative particle alone in response. Prior research has shown that respondents similarly orient to declarative or unmarked forms of questions as indexing questioners’ knowing stance in Korean conversation (Lee 2015). They tend to produce an affirmative particle alone in response to declarative questions with informal endings. In contrast, in response to interrogative forms of questions with overt interrogative endings, respondents tend to construct elaboration in addition to an affirmative particle (Lee 2015). They treat such interrogative polar questions as indexing questioners’ unknowing stance. Building on this research (Lee 2015), we can suggest that in Korean conversation, both tag and declarative forms of polar questions can be treated as encoding questioners’ more knowing stance, in contrast to interrogative forms of questions, which tend to be treated as indexing questioners’ unknowing stance. In the dataset under analysis, pseudo-tag and declarative forms of questions seem to indicate different degrees of epistemic gradient between questioner and respondent (see Heritage 2010; Heritage and Raymond 2012). Pseudo-tag questions with the committal -ci ending seem to encode a stronger epistemic com-
Chapter 6. Complexities of responding
mitment to the candidate answer, conveying a flatter epistemic gradient between questioner and respondent than declarative questions with informal endings. This can be grounded by cases in which a questioner replaces the form of a pseudo-tag question with a declarative question, specifically when facing a potential problem projected by the respondent. Excerpt 10 is a case in point. Cho is giving Lee a callback, after having missed a call from Lee. In dealing with the work of identification/recognition in the telephone opening, Cho produces a ‘second’ summons and invites recognition by the recipient (line 2; Lee 2006). Lee displays recognition by producing her answer in a non-polite form (line 4). At line 5, Cho launches a question addressing whether Lee has just called. He constructs the question in the form of a pseudo-tag question with the committal -ci ending, seeking confirmation. Note that by using the verb hay-ss ‘did’, Cho does not specify what Lee did, although this is understandable as meaning ‘called’ in this context, especially with the use of pangkum- ‘just-’ that marks recency (line 5). Excerpt 10. (Working out) 01
LEE:
yeposeyyo? hello Hello?
02
CHO:
yeposeyyo? hello Hello?
03 04
(0.3) LEE:
ung. hh PRT
Yes. hh 05
CHO:=> ung:.=pangkum- (0.3) hay-ss-ess-ci, PRT just do-PST-PPCP-COMM Yes:.=You just- (0.3) called ((literally ‘did’)) ci,
06 07
-> (0.3) LEE:-> ung? PRT
Huh? 08
CHO:=> mwe cenhwahay-ss-ess-e? what call-PST-PPCP-IE Well you called?
09 10
(0.6) LEE:
e.= kuntey:: kukey isangha-te-la-kwu PRT and DEM.NOM strange-RETROS-DECL-CONN Yes.=And:: that was strange
203
204
Seung-Hee Lee
At line 6, there ensues a 0.3-second gap, which can project a potential problem or disagreement with the question (Sacks 1987). Indeed, in the next turn (line 7) Lee does not respond to the question but initiates a repair (Schegloff, Jefferson, and Sacks 1977). She produces one of the open-class repair initiators available in Korean, ung? ‘huh?’, which does not identify the trouble source and can indicate or project some problem or disagreement with the prior turn (see Drew 1997). Faced with a potential problem projected by the delay (line 6) and the repair initiation (line 7), Cho repairs his prior question at line 8. He first produces a turn-initial mwe ‘well’, which can downgrade the likelihood of a hypothesis proposed in the upcoming question. Then he produces a revised question. Cho not only specifies the verb, hay-ss ‘did’, in the prior question (line 5) as cenhwahayss ‘called’ (line 8). He also modifies the form of question by replacing the prior pseudo-tag question (line 5) with a declarative question (line 8). That is, Cho uses the informal -e ending in the revised question (line 8), in replacement of the committal -ci ending in the prior question (line 5). This repair of the question format suggests that the declarative, unmarked form of question with the informal ending (line 8) may embody a weaker epistemic commitment to a candidate answer than the pseudo-tag question. Given a potential problem or disagreement projected by the respondent, the questioner may downgrade his epistemic commitment in constructing a revised question. Note that Cho also deploys a rising intonation in producing the declarative form of question (line 8), which can also display his downgraded epistemic commitment, unlike in the production of the pseudo-tag question (line 5). At line 10, Lee finally produces a confirming response to the revised question and launches a telling, displaying that the trouble has been resolved with Cho’s repair of the question. Thus, questioners may construct a declarative form of question in replacement of the initial, pseudo-tag question, when a potential problem or disagreement with the question is projected. This revision of the question format suggests that participants may orient to pseudo-tag and declarative questions differently, treating pseudo-tag questions as indicating a stronger degree of epistemic commitment to the proposition than declarative questions. While both forms of questions tend to be treated as indexing questioners’ knowing stance, pseudo-tag questions seem to be treated as conveying a flatter epistemic gradient between questioner and respondent than declarative, unmarked forms of questions in Korean conversation (see Heritage 2010; Heritage and Raymond 2012). Note that this is distinctive from the epistemic gradient between questioner and respondent encoded in different forms of polar questions in English. Among the most common forms of polar questions including yes/no interrogative, declarative, and tag questions, declarative questions are treated as encoding the smallest
Chapter 6. Complexities of responding
information gap between questioner and respondent in English conversation (Heritage 2010; Heritage and Raymond 2012). Declarative questions in general assert a candidate answer and embody a “nearly equal epistemic footing” with their recipient (Heritage and Raymond 2012: 181), in contrast to yes/no interrogative questions, which indicate a steeper epistemic gradient between questioner and respondent (Heritage 2010; Heritage and Raymond 2012; Raymond 2010; Stivers 2010). Declarative questions are also treated as proposing a stronger commitment to a candidate answer than tag questions, while both forms encode questioners’ knowing stance (Heritage 2010; Heritage and Raymond 2012). This is unlike declarative forms of polar questions in Korean conversation, which tend to be treated as indexing a lesser degree of epistemic commitment than pseudo-tag questions. The difference between Korean and English in this regard may relate to distinctive grammatical resources for constructing polar questions in each language. In English, tag questions are constructed in the form of an interrogative with subject-auxiliary inversion that is appended at the end of a candidate proposition, whereas declarative questions assert a candidate proposition. Questioners may thus embody a stronger epistemic claim in constructing the proposition in declarative form, not using the interrogative syntax at all even in a tag form (see Heritage 2010; Raymond 2010). In contrast, in Korean the syntax of pseudo-tag questions as well as that of declarative questions does not mark any specific type of sentence. Both pseudotag and declarative forms can be used across various sentence types, including declaratives, interrogatives, and imperatives (Lee 1994, 1999). The grammatical difference between the two forms, at least on the surface, lies in the use of different sentence-ending suffixes: a committal -ci ending in pseudo-tag questions and an informal -e/a ending in declarative questions. The committal suffix -ci explicitly expresses the speaker’s commitment to, or belief in, the stated proposition (Lee 1999), whereas the informal -e/a ending lacks any such specific epistemic marking (see Kim 2011). Thus, when questioners choose to use -ci instead of -e/a in formulating the sentence ending of an utterance, they can overtly mark their epistemic commitment to the stated proposition. Pseudo-tag questions can thus embody a stronger epistemic claim with a flatter epistemic gradient between questioner and respondent compared to declarative questions, in which questioners do not use any morpho-syntactic marking that can overtly express such epistemic claims. Korean questioners may thus rely on the use of particular sentence-ending suffixes in indicating their epistemic stance, as distinct from English questioners, who have syntactic resources for constructing different forms of question. This suggests that questioners’ epistemic stance can be indexed and understood in dif-
205
206
Seung-Hee Lee
ferent ways in each language, depending on the language-specific grammatical resources available for question formation.
8. Conclusion This paper examined confirming responses to pseudo-tag questions in Korean conversation. It focused on three forms of confirmation: unelaborated typeconforming responses, elaborated type-conforming responses, and nonconforming responses. In the dataset, respondents tend to treat the question as addressing a state of affairs that is already available by reference to prior talk in the interaction, interactional history, shared knowledge, and so on. They thus orient to pseudo-tag questions as making a mere confirmation specifically relevant in the next turn, indexing a knowing stance of the questioner. In most cases, respondents provide a simple confirmation with an affirmative particle alone, acquiescing to the terms and constraints of the question as well as to the questioner’s knowing stance as indexed in the question. However, when producing elaborated type-conforming responses, respondents do not merely confirm but engage in multiple forms of interactional work. While confirming the proposition with affirmative particles, respondents also provide elaboration, which departs from the relevance of a mere confirmation. They may thus address the sequential action of the question, defend their epistemic rights, and affiliate with the questioner, avoiding unwanted understandings that a simple confirmation can bring. In contrast, in constructing nonconforming responses, respondents depart from the question’s terms and constraints altogether. They tend to treat the question as problematic. These three forms of confirming responses can embody different degrees of acquiescence or agency. Unelaborated type-conforming responses may be fully acquiescent. They provide a mere confirmation, which is made specifically relevant by the form of the pseudo-tag question, encoding a questioner’s epistemic commitment to the proposition. Elaborated type-conforming responses seem to be only partially acquiescent. While confirming the question with an affirmative particle, respondents follow by providing elaboration, thus departing from the relevance of a mere confirmation. They seem to embody a step-by-step move from acquiescence to agency, although this may not be necessarily disaffiliative. Nonconforming responses may be more agentive. They assert the question’s proposition or its modified version, treating the question itself as irrelevant and problematic. Building on prior research (Heritage 2010; Heritage and Raymond 2012; Lee 2015; Raymond 2010), the findings of this paper also suggest that unelaborated
Chapter 6. Complexities of responding
and elaborated type-conforming responses can have different interactional import. While both forms are type-conforming, elaborated responses can engage in interactional work other than providing confirmation. They can deal with multiple contingencies involved in responding, such as the sequential action of the question, relative epistemic rights over the proposition, and affiliation with the questioner, which may not be (appropriately) addressed with an affirmative particle alone. In dealing with these contingencies, elaborated responses may depart from the question. This suggests not only that type-conforming responses may not accept all aspects of the question but also that there is a division of labor in each element of elaborated type-conforming responses (see Stivers 2011). Even when constructing type-conforming responses, respondents may consider various aspects of the question to accept or resist within the ‘yes’ response. Thus, the respondent’s choice of a particular form of response represents a solution to the multiple contingencies involved in responding. While confirming a question, respondents can engage in the complex interactional work of managing various issues raised by the question, including relevance of the question, underlying agendas pursued by the question, sequential activities in progress, relative epistemic rights between participants, affiliation with the questioner, particular sequential and interactional contexts, etc. The design of responses to polar questions is thus a result of respondents’ complex considerations about the question with its language-specific resources in particular contexts.
References Bolden, Galina. 2016. “A Simple Da?: Affirming Responses to Polar Questions in Russian Conversation.” Journal of Pragmatics 100: 40–58. Bolinger, Dwight. 1978. “Yes-No Questions Are Not Alternative Questions.” In Questions, ed. by Henry Hiz, 87–105. Dordrecht, Netherlands: Reidel. Clark, Herbert H. 1996. Using Language. Cambridge: Cambridge University Press. Drew, Paul. 1997. “‘Open’ Class Repair Initiators in Response to Sequential Sources of Trouble in Conversation.” Journal of Pragmatics 28: 69–101. Enfield, N. J. 2006. “Social Consequences of Common Ground.” In Roots of Human Sociality: Culture, Cognition, and Interaction, ed. by N. J. Enfield, and Stephen C. Levinson, 399–430. Oxford: Berg. Enfield, N. J., Tanya Stivers, Penelope Brown, Christina Englert, Kaatariina Harjunpää, Makoto Hayashi, Trine Heinemann, Gertie Hoymann, Tina Keisanen, Mirka Rauniomaa, Chase W. Raymond, Federico Rossano, Kyung-Eun Yoon, Inge Zwitserlood, and Stephen C. Levinson. 2019. “Polar Answers.” Journal of Linguistics 55: 277–304. Fox, Barbara, and Sandra A. Thompson. 2010. “Responses to Wh-Questions in English Conversation”. Research on Language and Social Interaction 43: 133–156. Goffman, Erving. 1978. “Response Cries.” Language 54: 787–815.
207
208
Seung-Hee Lee
Hayano, Kaoru. 2011. “Claiming Epistemic Primacy: Yo-Marked Assessments in Japanese.” In The Morality of Knowledge in Conversation, ed. by Tanya Stivers, Lorenza Mondada, and Jakob Steensig, 58–81. Cambridge: Cambridge University Press. Hepburn, Alexa, and Galina Bolden. 2017. Transcribing for Social Research. Thousand Oaks, CA: Sage. Heritage, John. 1984. “A Change-of-State Token and Aspects of its Sequential Placement.” In Structures of Social Action: Studies in Conversation Analysis, ed. by J. Maxwell Atkinson, and John Heritage, 299–345. Cambridge: Cambridge University Press. Heritage, John. 1998. “Oh-Prefaced Responses to Inquiry.” Language in Society 27: 291–334. Heritage, John. 2010. “Questioning in Medicine.” In “Why Do You Ask?”: The Function of Questions in Institutional Discourse, ed. by Alice F. Freed, and Susan Ehrlich, 42–68. Oxford: Oxford University Press. Heritage, John. 2011. “Territories of Knowledge, Territories of Experience: Empathic Moments in Interaction.” In The Morality of Knowledge in Conversation, ed. by Tanya Stivers, Lorenza Mondada, and Jakob Steensig, 159–183. Cambridge: Cambridge University Press. Heritage, John, and Geoffrey Raymond. 2005. “The Terms of Agreement: Indexing Epistemic Authority and Subordination in Talk-in-Interaction.” Social Psychology Quarterly 68: 15–38. Heritage, John, and Geoffrey Raymond. 2012. “Navigating Epistemic Landscapes: Acquiescence, Agency and Resistance in Responses to Polar Questions.” In Questions: Formal, Functional and Interactional Perspectives, ed. by Jan P. de Ruiter, 179–192. Cambridge: Cambridge University Press. Jefferson, Gail. 1981. The Abominable “Ne?”: A Working Paper Exploring the Phenomenon of Post-Response Pursuit of Response. Department of Sociology, University of Manchester. Jefferson, Gail. 2004. “Glossary of Transcript Symbols with an Introduction.” In Conversation Analysis: Studies from the First Generation, ed. by Gene H. Lerner, 13–31. Amsterdam: John Benjamins. Kim, Haeyeon. 1999. “The Form and Function of Questions in Korean Conversation.” Discourse and Cognition 6: 211–247. Kim, Mary Shin. 2011. “Negotiating Epistemic Rights to Information in Korean Conversation: An Examination of the Korean Evidential Marker -tamye.” Discourse Studies 13: 435–459. Kim, Stephanie Hyeri. 2015. “Resisting the Terms of Polar Questions through Ani (‘No’)-Prefacing in Korean Conversation.” Discourse Processes 52: 311–334. Lee, Hyo Sang. 1994. “Discourse-Pragmatic Functions of Sentence-Type Suffixes in Korean.” In Theoretical Issues in Korean Linguistics, ed. by Young-Key Kim-Renaud, 517–539. Stanford, CA: CSLI. Lee, Hyo Sang. 1999. “A Discourse-Pragmatic Analysis of the Committal -ci in Korean: A Synthetic Approach to the Form-Meaning Relation.” Journal of Pragmatics 31: 243–275. Lee, Seung-Hee. 2006. “Second Summonings in Korean Telephone Conversation Openings.” Language in Society 35: 261–283. Lee, Seung-Hee. 2013. “Response Design in Conversation.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell and Tanya Stivers, 415–432. Oxford: Wiley-Blackwell. Lee, Seung-Hee. 2015. “Two Forms of Affirmative Responses to Polar Questions.” Discourse Processes 52: 21–46.
Chapter 6. Complexities of responding
Lee, Seung-Hee. 2017. “Acquiescence and Resistance in Disconfirming Responses to Polar Questions.” Discourse Processes 54: 124–142. Maynard, Douglas W. 2003. Bad News, Good News: Conversational Order in Everyday Talk and Clinical Settings. Chicago: University of Chicago Press. Pomerantz, Anita. 1988. “Offering a Candidate Answer: An Information Seeking Strategy.” Communication Monographs 55: 360–373. Raymond, Geoffrey. 2003. “Grammar and Social Organization: Yes/No Interrogatives and the Structure of Responding.” American Sociological Review 68: 939–967. Raymond, Geoffrey. 2010. “Grammar and Social Relations: Alternative Forms of Yes/No-Type Initiating Actions in Health Visitor Interactions.” In “Why Do You Ask?”: The Function of Questions in Institutional Discourse, ed. by Alice F. Freed, and Susan Ehrlich, 87–107. Oxford: Oxford University Press. Raymond, Geoffrey, and John Heritage. 2006. “The Epistemics of Social Relations: Owning Grandchildren.” Language in Society 35: 677–705. Sacks, Harvey. 1987. “On the Preferences for Agreement and Contiguity in Sequences in Conversation.” In Talk and Social Organization, ed. by Graham Button, and John R. E. Lee, 54–69. Clevedon: Multilingual Matters. Sadock, Jerrold, and Arnold Zwicky. 1985. “Speech Act Distinctions in Syntax.” In Language Typology and Syntactic Description, Volume 1, Clause Structure, ed. by Timothy Shopen, 155–196. Cambridge: Cambridge University Press. Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks. 1977. “The Preference for SelfCorrection in the Organization of Repair in Conversation.” Language 53: 361–382. Sohn, Ho-Min. 1999. The Korean Language. Cambridge: Cambridge University Press. Sorjonen, Marja-Leena. 2001. “Simple Answers to Polar Questions: The Case of Finnish.” In Studies in Interactional Linguistics, ed. by Margret Selting, and Elizabeth Couper-Kuhlen, 405–431. Amsterdam: John Benjamins. Stivers, Tanya. 2005. “Modified Repeats: One Method for Asserting Primary Rights from Second Position.” Research on Language and Social Interaction 38: 131–158. Stivers, Tanya. 2010. “An Overview of Question-Response System in American English Conversation.” Journal of Pragmatics 42: 2772–2781. Stivers, Tanya. 2011. “Morality and Question Design: “Of Course” as Contesting a Presupposition of Askability.” In The Morality of Knowledge in Conversation, ed. by Tanya Stivers, Lorenza Mondada, and Jakob Steensig, 82–106. Cambridge: Cambridge University Press. Stivers, Tanya. 2019. “How We Manage Social Relationships through Answers to Questions: The Case of Interjections.” Discourse Processes 56: 191–209. Stivers, Tanya, and Makoto Hayashi. 2010. “Transformative Answers: One Way to Resist a Question’s Constraints.” Language in Society 39: 1–25. Suh, Cheong-Soo. 2006. Korean Grammar. Seoul: Hanyang University Press. Yoon, Kyung-Eun. 2010. “Questions and Responses in Korean Conversation.” Journal of Pragmatics 42: 2782–2798.
209
chapter 7
The division of labor between the particles jah and jaa ‘yes’ as responses to requests for confirmation in Estonian Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo University of Tartu
This chapter examines the use of the Estonian particles jah and jaa ‘yes’ as responses to positively formulated requests for confirmation in ordinary interaction. We will show what actions these particles perform, in which sequential contexts they are used, and whether they can be interpreted as equivalent or not. Our analysis reveals that although both jah and jaa confirm the accuracy of the proposition presented in the question, the particles are not equivalent. The most important difference between jah and jaa lies in how the sequence continues after the response. After jah, the local sequence is closed, while after jaa, the sequence is expanded by the questioner. By using jaa the answerer indicates that the sequence is open for expansion and allows the interlocutor to choose how to continue. In addition, the context of jaa is more limited and partially different from the context of jah. In the case of jaa, the interactional stances of the interactants are systematically incongruent. Specifically, the questioners express their non-neutral interactional stance in the expansion of the sequence, while the answers respond neutrally or express a different interactional stance. Overall, jah could be classified as an ‘unmarked’ response particle which only confirms the question’s proposition, while jaa is pragmatically a more ‘marked’ particle used for ‘special purposes.’ Keywords: polar question, confirmation, response particles, Estonian, ordinary conversation
1.
Introduction
In this paper, we investigate the division of labor between the particles jah and jaa ‘yes’ as responses to positively formulated requests for confirmation in Estonian
https://doi.org/10.1075/slsi.35.07hen © 2023 John Benjamins Publishing Company
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
ordinary interaction. Particle responses to polar questions have been an important subject in recent conversation analytic and interactional linguistic studies. For the purposes of this chapter, it is important to highlight four aspects of that research. First, studies on different languages (e.g., on English and Korean) have shown that the choice of response type and the trajectory of the talk after the response are related to the questioner’s epistemic stance (e.g., Heritage & Raymond 2012; Heritage 2012; Lee 2015). Questions that request information and have a steep epistemic gradient between the speaker and the hearer tend to get longer, more elaborate responses, which project an expansion of the question-answer sequence. By contrast, questions with a flatter epistemic gradient that request confirmation of some information tend to be answered with a simple particle, which is followed by sequence closure. Second, requests for confirmation perform a range of different social actions. One central action of requests for confirmation is other-initiated repair (Schegloff, Jefferson, and Sacks 1977), which deals with problems of hearing or understanding and takes the form of a candidate understanding or a repeat. Besides initiating repair, requests for confirmation are sometimes related to the unexpectedness of what the previous speaker has said. Rossi (2020: 512) uses the term “interactional stance” for such cases, distinguishing two types of actions: displaying surprise (e.g., Selting 1996; Wilkinson & Kitzinger 2006) and questioning “the truthfulness, acceptability or appropriateness of what has been said” (Rossi 2020: 513). While some authors exclude unexpectedness from repair, others interpret it as initiating repair to deal with the problem of the acceptability of preceding talk (e.g., Svennevig 2004). In addition, in some cases, the concept of ‘newsmark’ has been employed. With a newsmark the speaker treats an informing as news, which presents an extraordinary or remarkable fact (Heritage 1984: 339; see also Sorjonen 2001a: 223,300). Newsmarks can be formulated as repeats, but more typical devices for newsmarks are expressions of “ritualized disbelief ” (really, seriously, etc.) (Heritage 1984: 239–244). Different actions carry different implications for the trajectory of the turn and/or sequence. Repair initiations for problems of hearing or understanding typically lead to closure of the departure from the main line of talk (Kendrick 2015: 166; Rossi 2020: 512). By contrast, newsmarks and questions that express unexpectedness tend to lead to sequence elaboration (Heritage 1984: 339; Rossi 2020: 512). Third, it has been argued that responses can be divided into ‘unmarked’ and ‘marked’ variants, expressed by different particles. Particles like yes, yeah or yep have been described as pragmatically ‘unmarked’ minimal responses, which only confirm the question’s proposition (factual accuracy) and align with the epistemic stance conveyed by it. By contrast, particles like of course, probably, sure, etc. are
211
212
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
treated as ‘marked’ response variants reserved for special purposes (e.g., Stivers 2018). Fourth, there is very little research concerning how ‘unmarked’ response particles operate in interaction. For the purposes of our chapter, it is important to highlight two studies which show that different particles or different prosodic variants of a response particle perform different tasks in interaction (see also Raymond 2010, 2013). Sorjonen (1996, 2001a: 45–92, 2001b) shows that Finnish particles joo and nii(n) are not equivalent. Their use is related to the format of the question and to the action they implement. Sorjonen shows that joo can be used as a response to verb-interrogatives (e.g., voitteko, ‘can you’, where -ko is a question clitic) which question the entire proposition. In such cases, however, the interrogative encodes an assumption that the information is already shared to some extent by the participants, and joo provides confirmation.1 If the question is formulated as a non-V-interrogative (declarative, non-clausal utterance or focused interrogative), then nii(n) is the primary response, although joo is also possible. While joo is possible as a response to a non-V-interrogative, Sorjonen argues that “it differs from niin in one respect: it treats what the prior speaker took up in her or his utterance as a departure from some main line of talk and thus as closure relevant.” (Sorjonen 2001b: 417; see also Sorjonen 1996, 2001b: 419,421). In another study, Bolden (2016) analyzes the Russian response particle da ‘yes’. Bolden distinguishes two prosodic variants of da and shows that these variants perform different social actions. Prosodically ‘relatively unmarked’ da occurs as a response to other-initiations of repair (candidate hearings or understandings) and confirms the accuracy of the proposition (Bolden 2016: 44). Besides this, da is used as a response to newsmarks that express the unexpectedness of the information (surprise, astonishment, etc.), which she terms “evaluative stance vis-à-vis the proposition being confirmed” (Bolden 2016: 46). Bolden shows that by modulating the prosody of the particle da recipients can convey their congruent or incongruent evaluative stance vis-à-vis the question (Bolden 2016: 46–49). Bolden does not address how sequences continue after different variants of da. This chapter contributes to this literature by analyzing Estonian particles jah and jaa as responses to the requests for confirmation. We examine sequential and action environments for the two response particles, their sequence closure implicativeness, and prosodic realizations. We explore the following questions. Is the sequence always closed after the particle response or is the closure related to sequential context? Is closure affected by the social action performed by the question, the choice of the response particle or the prosody of the particle? What 1. Sorjonen argues that as a response to a V-interrogative that requests information, a repeat of the finite verb of the interrogative marks the response as new information.
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
actions do the particles perform; do they only confirm the proposition or do they also convey different evaluative stances vis-à-vis the question?
2.
Confirmation-seeking polar questions and their responses in Estonian: An overview
The primary difference between polar question formats in Estonian lies in the distinction between requests for information and requests for confirmation (Hennoste 2012; Hennoste, Rääbis, and Laanesoo 2017). In this section, we give an overview of the formats for requests for confirmation and their responses in ordinary spoken Estonian. Table 1. Formats of positively formulated requests for confirmation in ordinary spoken Estonian (constructed examples, question mark marks a question) * Question format
Example
Translation
“Plain” declarative
Sa oled abielus? Abielus?
‘You are married?’ ‘Married?’
Declarative + jah
Sa oled abielus jah? Abielus jah?
‘You are married JAH?’ ‘Married, JAH?’
Declarative + eks/ eksole/ onju
Sa oled abielus eks/eksole/ onju?
‘You are married right?’ ‘Married, right?’
Declarative + vä
Sa oled abielus vä?
‘You are married VÄ?’ ‘Married VÄ?’
* The word ‘declarative’ is used for clausal and non-clausal utterances. The formats presented here differ from formats of requests for information. Requests for information are typically formulated in Estonian with the utterance-initial particle kas, and also with inversion. We do not deal with them in this paper.
The three primary positively formulated question formats for requests for confirmation in Estonian conversation are “plain” declarative utterances, utterances with the final particle jah, and utterances with the final particle vä.2 Tags 2. The question particle vä comes from the conjunction või ‘or’ and has been used as a question particle at least since the end of the 19th century. The particle has several prosodic variants (vä, või, ve, võ, väh, veh), among which vä and või are the most commonly used. Või can also be used as a conjunction, other variants are employed only as question particles. Hereinafter we use the generalizing form vä. The question particle jah comes from the response particle jah ‘yes’ and it is used as a question marker dating back at least to the end of the 19th century (Hennoste, Rääbis, and Rumm 2019).
213
214
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
like eks, eksole, onju ‘right’ are only rarely used (Erelt & Metslang 2017: 519–521; Keevallik 2010: 286; Hennoste, Rääbis, and Laanesoo 2017). The three main formats are not equivalent. By using vä, questioners downgrade their certainty about the proposition and often mark or highlight their disbelief. Jah-questions (and also question tags) express the accuracy of the propositional content of the question and imply strong expectations for confirmation (Hennoste, Rääbis, and Rumm 2019). Minimal positive answers to polar questions are divided into four groups in Estonian (also Keevallik 2009): – – – –
particles jah, jaa ‘yes’, mhmh ‘uh-huh’; ‘marked’ particles like ikka ‘certainly, surely’, muidugi ‘of course, surely’, võibolla ‘possibly, probably, maybe’, etc.; repetition responses; combinations of particles and repeats (jah + repeat, jah + muidugi, repeat + ikka, etc.).
The two primary groups of responses are the particles jah, jaa, mhmh, and repetition responses. Keevallik (2009, 2010) has argued that, in Estonian, particle responses are used to respond to what she called ‘non-primary actions’ (e.g. presequence initiations, repairs, checking questions, etc.), whereas repetitions are minimal responses to a question that is either an initiation of a sequence or is used as part of the ‘primary’ action (cf. also Sorjonen 2001b; Hakulinen et al. 2004: 1147–1148). By responding in a particular way, the interlocutor conjointly indicates whether the question has been interpreted as ‘primary’ or ‘non-primary’ action. According to our analysis, the minimal particle responses jah, jaa and mhmh are typical responses to requests for confirmation and they are used both in the ‘non-primary’ as well as in the ‘primary’ actions of the talk (Hennoste in press). The prior studies of particle responses in Estonian have not investigated differences in the use of jah and jaa (Hennoste 2000a: 1785–1793; Kasterpalu 2005; Keevallik 2009, 2010). In this chapter, we will focus on the particles jah and jaa as responses forming a turn of their own. Our aim is to find out what actions the particles perform, in which kinds of sequential contexts they are used, and whether they can be interpreted as equivalent or not.
3. Data and method Our data come from the Corpus of Spoken Estonian of the University of Tartu (SEKK; Hennoste et al. 2008). The corpus consists of 5012 recordings (821 hours)
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
and about 2.560,800 transcribed words (2023). We randomly selected 75 everyday face-to-face and telephone conversations (106,500 words) from the corpus. The participants in these conversations are friends, family members, schoolmates, or relatives. The database contains 395 question-answer sequences with responses to a positively formulated polar question. From these 395, we found 72 cases of jah and 54 cases of jaa as minimal responses, which constitute our collection. We analyze particle responses using the methodology of Conversation Analysis and Interactional Linguistics.
4.
Analysis
In this section, we analyze the use of two response particles in our collection. First, we describe the use of jah (Section 4.1) and then turn to jaa (4.2).
4.1 Jah: Confirming the accuracy of the proposition and closing the sequence In our data, jah is employed as a response to variously formulated requests for confirmation in diverse types of sequences. We will illustrate six usages of jah here. Excerpt 1 shows a typical case where jah is used as a response to an initiation of repair for a problem of understanding and projects closure of the sequence. The excerpt comes from a face-to-face conversation in which mother (MOT) and son (SON) are talking about basketball teams. The son has watched a basketball game between Estonia and Hungary on TV, which they have talked about earlier in the conversation (data not shown in the excerpt). After discussing other topics (during 32 turns), the mother returns to basketball with a request for information concerning the race of the players (line 1). Excerpt 1. (Basketball) 01
MOT:
kas siin musti mehi Q here black:PAR:PL man:PAR:PL are there black men
02
ka mängib=või. also play:3SG Q also playing here
03
(.)
04
SON:
kus Eestis= vä. where Estonia:INE Q where in Estonia
05
MOT:=> jah_ JAH
215
216
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
06
SON:
üks on: jah. one be:3SG yes there is one yes
In her question, the mother uses the adverbial siin ‘here’ (line 1) which causes a problem of understanding for the son. The son initiates a repair with the content question kus ‘where’ but proceeds immediately to a request for confirmation Eestis vä ‘in Estonia’ (line 4). The mother confirms the candidate answer with jah pronounced with a flat intonation (line 5). In the next turn (line 6), the son closes the repair sequence by answering the mother’s previous question, which shows that he treats jah as closing implicative (line 6). In Excerpt 2, jah is also used as a confirming response to a repair initiation, which this time deals with a problem of hearing. Here, two friends, Robert and Timo, are talking via mobile phone. Both of them had been at the beach that day but did not see each other there. The excerpt starts with Robert’s question about Timo’s whereabouts and Timo responds that he was just hanging out there (lines 1–4). Excerpt 2. (Beach) 01
ROB:
02 03
(0.7) TIM:
04
ma nisama passisin seal ringi I no.reason hang:IPFV:1SG there around I was just hanging around there (kuradi istusin.) devil:GEN sit:IPFV:1SG bloody sat {there}
05
ROB:
06
TIM:
07 08
kus= sa olid ise. where you.SG be:IPFV:2SG yourself where were you yourself
kus[kohas;] where.place:INE where [ääre]= peal. edge:GEN on.top on the edge (0.3)
ROB:
ah? PRT
huh 09 10
(0.3) TIM:
ÄÄRE PEAL. edge:GEN on.top on the edge
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
11
ROB:
12
TIM:=> °jah_° JAH
13
ROB:
14
ääre peal= vä. edge:GEN on.top Q on the edge
aa, ma=olin seal vetelpääste PRT I be:IPFV:1SG there lifeguard:GEN oh I was there at the lifeguard putka= jures. liiva= (bäl_) booth:GEN at sand:GEN on tower on the sand
In line 5, Robert initiates repair, asking again about the location. Timo’s response ääre peal ‘on the edge’ (line 6) is partly overlapping with Robert’s repair initiation, which causes a problem of hearing. This is seen in line 8, where Robert responds to Timo’s turn with an open-class repair initiator ah ‘huh’ (Drew 1997; see also Enfield et al. 2013; Hennoste in press). Timo responds by repeating his previous turn in a louder voice (line 10). Subsequently, Robert continues with a hearing check by repeating Timo’s answer. This turn is confirmed by Timo with jah (line 12). Robert treats jah as implying closure of the sequence by responding with a sequence-closing third using the change-of-state particle aa ‘oh’ (Heritage 1984; Schegloff 2007; Koivisto 2014; Kasterpalu & Hennoste 2016). In our next excerpt, jah is used as a response to a request for confirmation where the questioner has expressed a negative stance towards the unexpected information he received. In Excerpt 3, Uno (UNO) has called to a landline phone to talk to his sister, but his nephew Rein (REI; a 15-year-old boy) has answered the phone and said that his parents are not at home (see also Hennoste, Rääbis, and Rumm 2019: 63–64). The excerpt begins with Uno asking when Rein’s parents will come home so he could talk to his sister (line 1). Excerpt 3. (Coming home) 01
UNO:
02
et= ee (.) millal nad tulevad. PRT uhm when they come:3PL so uhm (.) when will they come (0.5)
03
REI:
nooh; südaöö paiku vast.= PRT midnight.GEN about probably well, at about midnight probably
04
UNO:
=südaöö paiku alles jah. midnight.GEN about not.until yes only about midnight right
05
REI:=> jah. JAH
217
218
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
06
07
UNO:
aga omme olete kodus muidu PRT tomorrow be:2PL home:INE PRT {you PL} are at home tomorrow põhimõtlselt=vä.3 in principle Q in principle
Rein starts his answer (line 3) to Uno’s question with the particle nooh, implying a possibly dispreferred answer to come (Hennoste in press; also Keevallik 2016). The answer provides a point of time, südaöö ‘midnight’ indicating a late arrival. Uno responds to the information with a question, südaöö paiku alles jah ‘only about midnight right’, perhaps conveying disappointment at the late arrival, to which Rein responds with jah ‘yes’, pronounced with a falling intonation (lines 4–5). Similarly to the previous excerpts, in this excerpt the sequence is closed by the questioner after the response jah (lines 6–7). Sequence closure after jah suggests that both interactants interpret jah as sequence-closing implicative. However, in contrast with the first two cases, here the questioner Uno has received unexpected information, which apparently conflicts with his expectation that he could talk to his sister, given that midnight is generally not a suitable time for a phone call (see also lines 6–7, where Uno asks about tomorrow’s plans). Such sequences tend to be expanded (Rossi 2020: 512). However, here expansion of the sequence does not take place. The reason for lack of expansion is related to the format of the question. Uno uses a question particle jah ‘right’, which often occurs in sequences in which the questioner’s prior knowledge or opinion has been refuted. With jah, the questioners indicate that they accept the new knowledge, but need the confirmation (Hennoste, Rääbis, and Rumm 2019; Hennoste in press). Thus, the question format hints that the confirmation is sufficient. Excerpt 4 also illustrates the usage of jah in a sequence in which the interactional stance of the question is not neutral. In this case, the questioner displays surprise at the information received in the prior turn. The excerpt comes from a landline phone call between friends Maarika (MAA) and Eve (EVE) who plan to travel together from a village to the city where they study. The excerpt begins with Maarika’s (the caller) turn tead=et mul on sulle üks palve ‘(you) know I have a request for you’ (lines 1–2), which comes immediately after the greetings and projects a request.
3. The subject NP is not an obligatory constituent in a clause in Estonian. The subject is always marked by the person suffix on the verb (here the plural second person suffix -te).
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
Excerpt 4. (Beet) 01
MAA:
02
03
tead= et mul on sulle know:2SG that I:ADE be:3SG you.SG:ALL {you} know I have üks palve. one request a request for you
EVE:
noh_ PRT
04
MAA:
et kas sul on peeti kodus. PRT Q you.SG:ADE be:3SG beet:PAR home:INE do you have beets at home
05
EVE:
beet:PAR beets
06
MAA:=> jah_ JAH
07
EVE:
ma arvan küll et peaks olema. I think:1SG PRT that should be:INF I think that there should be
08
MAA:
a
kas sa saaksid mulle mõned Q you.SG can:COND:2SG I:ALL some:PL could you take some for me, PRT
09
võtta= et ma omme=b ommikul paneks take:INF that I tomorrow morning:ADE put:COND tomorrow morning I would put them
10
ise enda kotti. myself my bag:ILL into my bag
Maarika’s first turn is a preliminary to preliminary (pre-pre) in Schegloff ’s terms (Schegloff 2007: 44–47), projecting that Maarika intends to make a request. Eve responds with the particle noh (line 3), which works as a go-ahead and urges the interlocutor to continue (Keevallik 2016: 216–218). Maarika proceeds to a question kas sul on peeti kodus ‘do you have beets at home’ (line 4). Instead of answering the question, Eve repeats the word peeti ‘beets’ at a slower pace and with a prolongation of the long vowel ee (line 5). The prolongation of the internal vowels of the word conveys unexpectedness in Estonian (Hennoste 2000b: 2038), here conveying surprise. In turn, Maarika responds with jah pronounced with a flat intonation (line 6) and Eve closes the sequence by responding to Maarika’s preceding question ma arvan küll et peaks olema (‘I think that there should be’, line 7).
219
220
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
Prosodically modified repeats as questions form a scale from repair to nonrepair (Thompson, Fox, and Couper-Kuhlen 2015: 60–64). If they are interpreted as newsmarks or as means that display unexpectedness (including surprise), they typically make sequence expansion relevant (Heritage 1984: 339–344; Sorjonen 1996: 302; Rossi 2020: 512). Here we can see that the expansion does not follow, the confirmation was enough for a questioner and she did not ask for additional information. Previous studies have shown that in this case the sequence type is crucial for interpretation. If the repeat cuts off an adjacency pair, then the repeats are more likely to be treated as a repair initiation and the sequence is closed after the response (Kurhila & Lilja 2017: 230). Also, it has been pointed out that additional information is necessary when a problem of understanding occurs. Excerpt 4 shows that Maarika interprets the question as a marker of surprise, not a request for additional information, and Eve’s progression of the sequence confirms this interpretation. Additionally, the question arises as to why Maarika used closure rather than the opportunity to expand this sequence or elaborate this topic immediately. The answer may lie in the fact that this is a delicate topic for her, as subsequent talk reveals that she needed beets for solving her digestive problems (not shown in the excerpt). Requests for confirmation in Excerpts 1–4 occurred in departures from the main line of talk. In the following Excerpt 5, particle jah is used as a response to a main line question. In this excerpt, two friends Uve (UVE) and Marge (MAR) are talking over a landline phone about Uve’s driving test. Marge asks questions about different parts of the exam and the excerpt starts with a third ja-prefaced (‘and’ prefaced, Heritage & Sorjonen 1994) question on a list (lines 1–2). Important here is the fact that Marge has attended and taken the exam in the same driving school earlier, thus she has some knowledge about the test. Excerpt 5. (Driving test) 01
MAR:
ja palju=te platsi peal sõitsite and much you.PL square:GEN on drive:IPFV:2PL and how much did you drive on the square
02
ühe= ringi onju. one:GEN round:GEN Q one round right
03
(0.5)
04
UVE:=> jah. JAH
05
MAR:
aa
kuidas test oli. how test be:IPFV:3SG oh how was the test PRT
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
Marge initiates her turn with a content question, which is then reformulated as a candidate answer question (Rumm 2019: 152–156). The question tag onju ‘right’ marks a flat epistemic gradient, indicating that the questioner is relatively sure about the accuracy of candidate answer. Uve responds with jah ‘yes’ with a falling intonation (line 4) and Marge closes the local sequence with a sequence-closing third aa ‘oh’ (Kasterpalu & Hennoste 2016) (line 5). This case shows that Marge treats jah as implying sequence closure and jah closes the sequence on the main line of conversation. Excerpt 6 illustrates a sequential context that is different from the previous cases. Here the excerpt shows how jah not only projects the closure of the local sequence, but also works to close the topic and a larger part of the conversation. Here, jah is used to respond to a request for confirmation in the main line of the talk at the beginning of a landline phone call between the mother (MOT) and the son (SON). The mother calls to ask if she should come to babysit. Excerpt 6. (Babysit) 01 02
((summons))
SON:
03
jah_ JAH (0.6)
04
MOT:
terviduss? hey
05
SON:
tere_ hello
06
(0.4)
07
MOT:
.hhhh noh; kuida elate. PRT how live:2PL well, how are you
08
SON:
ästi; fine
09 10
11 12
13
(0.3) MOT:
vihmaselt. rainy.ADV rainy (0.5)
SON:=> jah. JAH (0.5)
221
222
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
14
MOT:
.hhhhhh e kas te mind omme tahate. uhm Q you.PL I:PAR tomorrow want:2PL do you want me tomorrow
Unlike the previous examples, this is a situation where the interlocutors are oriented differently towards the course of the interaction. From the beginning of the call the son responds in ways which indicate that he is not interested in a longer conversation. For instance, the most common ways of responding to the phone in Estonian are jaa ‘yes‘ and hallo ‘hello‘. Here the son answers the phone with jah ‘yes’, which can be heard as implying that the interlocutor is uninterested in talking (Rääbis 2009: 44–49). Next, he does not reciprocate the familiar tone used by the mother in her colloquial, familiar greeting terviduss ‘hey’ (literally: ‘greetings’), but responds with the neutral greeting tere ‘hello’ (line 5). Then, when the mother initiates a new sequence with the topic proffer kuida elate ’how are you’ (line 7, Schegloff 2007), the son responds only minimally ästi ‘fine’ (line 8). Note that in Estonian, how-are-you questions do not expect a plain routine response (like ‘fine’) but seek a longer telling (i.e., a story, a report) as a response (Rumm 2019; Rääbis 2009; cf. Thompson, Fox, and Couper-Kuhlen 2015). Following this withholding response from the son, the mother makes another attempt to engage son in the conversation by proffering the topic of the weather via a request for confirmation vihmaselt ‘rainy’ (line 10). Following a 0.5 second gap (line 11), which may project a dispreferred answer (Kendrick & Torreira 2014), the son responds with jah without any elaboration (line 12). After another gap (line 13), the mother treats this response as a sequence closer, moving on to the new topic, her reason for calling (line 14). Thus, the mother can be understood as seeking an expanded conversation about the son’s life, but the son refuses by withholding elaborate responses. The son is oriented to the main reason for the call, i.e., baby-sitting, as evidenced by the further development of the talk, in which the son is considerably more responsive (not shown in the excerpt). Here jah closes the sequence and also functions to close the topic and the larger part of the conversation. To summarize, Excerpts 1–6 highlighted the use of the particle jah as a response to different types of requests for confirmation in different sequential contexts. We showed that jah confirms the proposition presented in the request for confirmation, and that it is followed by a closure of the local sequence. In most of our examples, jah projects the closure of a local sequence only (Excerpts 1–5). However, sometimes jah can also close a wider topic (Excerpt 6). In all cases, the speakers of jah accept the closing of the sequence, which shows that they also interpret their answer as sufficient and closure-relevant. Questions that are followed by jah-responses may be neutral (Excerpts 1, 2, 5), but they may also display different types of interactional stances, such as unexpectedness
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
(Excerpts 3, 4, 6). While unexpectedness generally leads to sequential expansion, in our data the jah-response always leads to the closure of the sequence. As we have seen, jah as minimal response is used in repair sequences (Excerpts 1, 2). In addition, it occurs as a response to requests for confirmation that are departures from the main line of the talk, but go beyond initiating repair (Excerpts 3, 4), as well as requests for confirmation that are in the main line of the talk (Excerpts 5, 6). The intonation of jah is in accordance with its sequence closing role. Jah is almost never uttered with a rising intonation, which often projects turn or sequence continuation (Keevallik 2003b; Asu 2006; Hennoste et al. 2009). There are only three cases (4%) in our data where jah has a slightly rising intonation. Mostly, jah is pronounced with a level (Excerpts 1, 2, 4) or falling intonation (Excerpts 3, 5, 6). Level intonation accounts for 65% (47 cases in our data) and falling intonation 31% (22 cases) of jah cases in our data.
4.2 Jaa: Confirming the accuracy of the proposition and implicating sequence expansion We will now turn to the particle jaa. In our data, the use of jaa differs from jah in two aspects. Firstly, in jaa-cases the request for confirmation to which jaa responds always conveys unexpectedness towards the information that the jaa speaker previously provided. The questioners topicalize the information, express their surprise, or question the truthfulness, accuracy, or appropriateness of what has been said. Secondly, with jaa the speaker indicates that the sequence is open for non-minimal post-expansion (Schegloff 2007: 148–161), and the interlocutor can choose how to continue. Excerpt 7 shows a typical jaa-response sequence, in which the question expresses the interlocutor’s surprise and disbelief towards the received information. In this excerpt, Mai (MAI) has called to her mother Linda (LIN). The talk is about Linda’s son and Mai’s brother Aare, who works in Finland. The excerpt begins when Linda makes a topic shift by stating that Aare and his son, who lives in Estonia, call each other almost every day (lines 1–4). The call was recorded in 2003 when calling abroad was expensive. Excerpt 7. (Calling to Finland) 01
02
LIN:
ja= nüüd (.) ja=nüüd sis helistab=õõ and now and now PRT call:3SG uhm and now and now {he} calls issile Soome ja issi helistab daddy:ALL Finland.ILL and daddy call:3SG to daddy to Finland and daddy calls
223
224
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
03
talle= ja (0.5) käib telefonidega pea he:ALL and go:3SG phone:COMP:PL almost to him and (0.5) they talk on the phone almost
04
iga päev jutt. every day talk every day
05
MAI:
ausalt= vä. honestly Q honestly
06
LIN:=> jaa? JAA
07
MAI:
(.) ee KAS e-v-v VENNA SAAB Q uhm Q uhm brother get:3SG does brother get to AGA KAS PRT
08
09
10
sis mm ODAVAMALT= vä. then uhm cheap:COMPA.ADV Q {call} cheaper then LIN:
Aare tegi no neil on vist NAME.M do:IPFV:3SG PRT they:ADE be:3SG probably Aare made well they probably have präägu sooduspakkumine. now discount.offer a discount offer now
Mai responds to the given information with the expression of ritualized disbelief ausalt=vä. ‘honestly’, which displays surprise and topicalizes the information received (line 5). Although this kind of question makes relevant a longer elaborate response, the interlocutor responds only with a minimal jaa (line 6), pronounced with rising intonation. Subsequently, Mai expands the sequence by asking for additional information about the matter of whether her brother can make the call more cheaply (lines 7–8), which Linda provides (lines 9–10). Here we can see that the question ausalt=vä. ‘honestly’ does not explicitly convey what exactly causes the speaker’s surprise and disbelief. The response jaa leaves the sequence open for expansion and allows the interlocutor to specify what caused the surprise. In addition, the excerpt shows that interlocutors’ stances towards the information are incongruent, which is made explicit in the expansion of the sequence. Mai expresses surprise and doubt in her question using the particles aga ‘but’, siis ‘then’ and a question particle vä (lines 7–8), but Linda’s response is neutral (lines 9–10). For Linda, the information about calling every day was not surprising or extraordinary, as she has information about a discount. In our next excerpt, jaa is also used as confirmation for a question that treats received information as unexpected. Differently from the previous excerpt, the
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
questioner expresses surprise, does not question the truthfulness of the information, and the request is formulated as a repeat. Excerpt 8 comes from a face-to-face conversation in which Siret (SIR) and her husband Rait (RAI) are talking about various everyday topics. The excerpt begins when Siret announces that they have received a letter and asks whether Rait wants to see it. After a short joking dialogue (lines 1–6), Siret brings the letter from another room. Excerpt 8. (Statistical Office) 01
SIR:
02
03
meile tuli üks kiri £tahad ma we:ALL come:IPFV:3SG one letter want:2SG I we got one letter do you want me näitan= vä.£ show:1SG Q to show {you}
RAI:
mqmm_ PRT
no 04
SIR:
.hhh ei= taha. NEG want {you} don’t
05
RAI:
hmq £no [näita= sis.]£ PRT PRT show:IMP:2SG then well show me then
06
SIR:
[heh]hehe ((laughter))
07
((SIR is going to other room to bring the letter))
08
(5.5)
09
SIR:
10 11
(
) ((says something from a distance))
(1.2) SIR:
12
statistikaametilt.=hh statistics.office:ABL from the Statistical Office (0.7)
13
RAI:
@statistikaamet.@ statistics.office Statistical Office
14
SIR:=> jaa_ JAA
15
RAI:
misasja see taab saada. what.thing:PAR it want.3SG get what does it want to get
225
226
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
16 17
(0.3) SIR:
18
se=on seal orantsi paberi peal on it be:3SG there orange:GEN paper:GEN on be:3SG it’s written there on the orange paper which kirjutad= mis: seletus °(mis tuleb teha.)° write:PPCP what account what must:3SG make:INF account is to be made
When Siret comes back to the room, she announces that the sender of the letter is the Statistical Office (line 11). In response, Rait repeats the previous turn @statistikaamet.@ ‘Statistical Office’ after a gap (line 13), albeit in the nominative case (the basic, dictionary form) rather than the case that Siret was using. In Estonian, the stress in compound words falls on the first word only. Rait’s intonation and the stressing of both units in the compound word (statistika and amet) show that this is not a neutral request for confirmation. Siret confirms with jaa (line 14), which is pronounced with level intonation. In the following turn, Rait expands the sequence by a question misasja see taab saada ‘what does it want to get’ (line 15). The interrogative used here is not a neutral request for information. Firstly, the colloquial question word misasja is typically used to express doubt or a negative stance. Secondly, the pronoun see ‘it’, which refers back to the statistikaamet, is emphasized, indicating that for Rait specifically the sender of the letter is surprising and unexpected. Here again we can see that the interlocutors’ interactional stances are incongruent. In her response (lines 17–18), Siret replies that the letter contains a paper that explains what kind of report is to be made. The response shows that Siret orients to Rait’s interrogative as a neutral what-question and that she does not display the surprise and negative stance that Rain expressed in his question. We conclude that, unlike the previous excerpt, the questioner does not question the truthfulness of the information given by the partner here, because Siret has already seen and read the letter (this is seen in lines 17–18). In the next excerpt, jaa is used as a response to the expression of ritualized disbelief. In this case, there is a discrepancy in the participants’ knowledge, which leads the questioner to doubt the truthfulness of the information received. Excerpt 9 comes from the same phone call conversation as Excerpt 5 and from the same series of questions about Uve’s exam. Excerpt 9. (Driving test) 01
MAR:
see oli aind koolieksam onju. it be:IPFV:3SG only school.exam Q it was just a school exam right
02
UVE:
jah. JAH
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
03
MAR:
04
mingeid vaatlejaid ei olnd; some:PAR:PL observer:PAR:PL NEG be:PPCP there were no observers aind Jüri ja see_ only NAME.M and this only Jüri and this
05
UVE:
ei (.) see politseist oli ka. NEG this police:ELA be:IPFV:3SG also no (.) the one from the police was there too
06
MAR:
oli ka= vä. be:IPFV:3SG PRT Q was {he} really
07
UVE:=> jaa. JAA
08
MAR:
tegelt ka= vä. really PRT Q really
09
UVE:
too kontrollis (.) meid. that check:IPFV:3SG we:PAR he checked up on us
10
MAR:
aa;
meil küll minu=meelest ei olnud. we:ADE PRT I:GEN mind:ELA NEG be:PPCP oh I don’t think we had PRT
After the question about the examination (line 1), Marge formulates a follow-up question mingeid vaatlejaid ei olnd; aind Jüri ja see ‘there were no observers only Jüri and this’ (lines 3–4). Uve disconfirms Marge’s assumption (line 4) and provides the correct information (line 5). Marge responds to that with an expression of ritualized disbelief (oli ka=vä; word by word ‘was also Q’, line 6). Marge’s newsmark conveys her disbelief, as the information contradicts her knowledge which she had expressed in her previous turn (in lines 3–4). At the same time, it treats the prior turn as newsworthy and asks for additional information. Uve only confirms her previous information with jaa, here uttered with falling intonation. Subsequently Marge expands the sequence with another expression of ritualized disbelief (tegelt ka=vä; word by word: ‘really also Q’, line 8). This time Uve responds by providing information about what the police did in the examination (line 9) and Marge closes the sequence (line 10). Similarly to Excerpts 7 and 8, the interactants here have incongruent interactional stances towards the information. For Marge the information about more observers in the exam is new and remarkable, which is represented by expressions of ritualized disbelief. For Uve, who was present at the scene, the presence of the
227
228
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
police representative was unremarkable and she responds to Marge’s last question in a neutral fashion. In the following excerpt, jaa is used as a response to the question in a sequence where a contradiction is found between the objective information of one participant and the subjective opinion of another participant. Excerpt 10 comes from a face-to-face conversation. Here, two couples have tasted raki, brought by Heivi (HEI) and Arvo (ARV ) from Greece. They have called it viinamarja samakas (bathtub vodka made of grapes), although the raki they actually drink is bought from a store. The excerpt begins when all participants have tasted the raki. We will concentrate on the dialogue between Merili (MER) and Heivi. Merili begins her turn in overlap with Kalju (KAL) and gives an assessment that the raki is not very strong (lines 2–3). This is a subjective opinion which is based on the fact that she has just tasted it. Excerpt 10. (Bathtub vodka) 01
KAL:
02
MER:
03
samakas pole ku[nagi paha. bathtub.vodka be.NEG never bad bathtub vodka is never bad
[ja= ta=pole nii] and it be.NEG so and it is not so [kange ka.] strong also strong too
04
ARV:
[hehe] ((laughter))
05
HEI:
nelikümmend? forty forty
06
]
(0.5)
07
MER:
on= vä;= be:3SG Q is it
08
HEI:=> =jaa? JAA
09
MER:
10
ARV:
11
HEI:
ma küll aru=ei=[saa.] I PRT understand.NEG I don’t get it [mt] Tch! .hhhhh ((köhib)) ((coughs))
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
12
MER:
[ma=i= tea_] I NEG know I don’t know
13
HEI:
[heh ] hehe £.hhhhh täiesti nelikümend.£= ((laughter)) totally forty totally forty
Heivi reacts to Merili’s opinion, saying that the strength of the raki is 40% (a typical strength of vodka in Estonia) (line 5). Merili responds with an expression of ritualized disbelief (on=vä ‘is it’, line 7), which Heivi confirms with a jaa with rising intonation (line 8). In her next turns, Merili expands the sequence. She begins by making an assertion based on her tasting experience ma küll aru=ei=saa ‘I don’t get it’ (line 9), thereby challenging Heivi’s information. Next (line 12), Merili continues by expressing doubt and avoiding an explicit disagreement with ma=i=tea ‘I don’t know’ (Keevallik 2003a: 93, 2011). In her response, Heivi laughs and upgrades her previous information about the strength of the drink with the adverb täiesti ‘totally’ (line 13). By that turn, the sequence is closed. Similarly to the previous excerpts, the response jaa does not close the sequence. In this excerpt, the doubt about the truthfulness of the information is not expressed by a new question, like in Excerpts 7–9, but the questioner uses assessments to express her disbelief (lines 9 and 12). Here again, we can see interactants’ different stances. Merili expands the sequence by expressing doubt, while at the same time Heivi responds to the expression of doubt with new and stronger confirmation, produced with laughter (line 13). Our last excerpt, Excerpt 11, also illustrates a case where the request for confirmation preceding jaa displays a stance that the received information is unexpected, treating it as surprising and also inappropriate. The excerpt comes from the beginning of a mobile phone call between friends Anu (ANU) and Ivika (IVI). At the start of the excerpt, Ivika initiates a pre-sequence (lines 4 and 7), testing the grounds for making a subsequent action. Following the second question (line 7), Anu asks directly about the reason why Ivika needs this information (line 9). Excerpt 11. (Orkut) 01
ANU:
jaa=hhh JAA
02
IVI:
tšau_ hey
03 04
(0.5) IVI:
oled oled kodus= vä. be:2SG be:2SG home:INE Q are you at home
229
230
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
05
(1.0)
06
ANU:
ei= ole;=hhh NEG be no
07
IVI:
a millal sa koju jõuad. but when you.SG home:ILL arrive:2SG but when will you get home
08 09
(1.3) ANU:
10 11
ma=i= tea=hhh miks=e miks sa tahad. I NEG know why why you.SG want:2SG I don’t know, why do you want (0.4)
IVI:
mul on= ee (0.9) mul on vaja I:ADE be:3SG uhm I:ADE be:3SG need I I need
12
et sa mu Orkutisse läheks. that you.SG I:GEN Orkut:ILL go:COND you to log in to my Orkut
13
(1.6)
14
ANU:
15
IVI:=> jaa_ JAA
16 17
et ma sinu Orkutisse läheks. that I you.SG:GEN Orkut:ILL go:COND that I would log in to your Orkut
(0.9) ANU:
okei,=hhh kule ma:::::: (1.2) okay listen I okay listen I
18
ma teen seda öösel. I do:1SG that at night I’ll do that at night
19
mul on natuke tegemist ka. I:ADE be:3SG some doing:PAR too I have some things to do
20
(0.5)
21
IVI:
no tee= sis öössel. PRT do:IMP:2SG then at night well, do it at night then
As a response, Ivika makes a request asking Anu to log in to her personal social media account (lines 11–12). A restart and hesitation (pause filler ee and
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
0.9-second pause) indicate that the formulating of this utterance is problematic for her. The problem is in the choice of the verb. Ivika emphasizes the word vaja ‘need’, which shows that the granting of the request is important. The request is followed by a 1.6-second gap (line 13), which indicates that there might be something problematic concerning the preceding turn for Anu. After the pause, Anu repeats the request, modifying it slightly (line 14). The particle et at the beginning of a turn ascribes the following content to the previous speaker, and thus makes confirmation relevant (Keevallik 2008: 129–130; Strandson 2001: 406). However, the question also indicates that Ivika’s request was unexpected and problematic. Anu uses the stressed long pronoun form sinu, used to highlight the contrastivity, instead of the short and unstressed variant su ‘your’. By doing so, she marks this part of the request as problematic for her. Also, the nearly full repeat of the previous turn expresses, among other things, that the social action of the previous turn was problematic for the speaker (Robinson & Kevoe-Feldman 2010). Ivika answers the question with jaa (line 15) without a gap, which indicates that for her there is not any problem concerning granting the request. The continuation of this sequence differs from the previous excerpts as the questioner does not expand the sequence. Instead, the response jaa is followed by a long gap (line 16) which indicates that Anu may be waiting for more information or that a dispreferred response might follow. In her next turn, Anu agrees to fulfill the request (lines 17–18). However, the design of the turn highlights that acceptance is problematic for her. This is shown by okei ‘okay’ followed by an exhalation, the particle kule ‘listen’, which often projects dispreferred and nonminimal second-pair parts (Keevallik 2003a: 69), prolongation of the pronoun ma and also by a long pause after ma. Anu might have moral problems with granting the request, since it is somewhat unusual to log in to another person’s personal social media account. This excerpt again demonstrates how an interlocutor treats jaa as a response that makes expansion relevant and the format of Anu’s agreement shows incongruence between the stances of the participants. Excerpts 7–11 demonstrated the use of the particle jaa as a minimal response and highlighted the differences in the usage of jaa compared to jah. In the case of jaa, information that elicits a request for confirmation was treated as unexpected, whether surprising or contradictory to the questioners’ previous knowledge, attitudes, interests, etc. The questioners do not close the local sequence after jaa, but instead expand it by asking for additional information (Excerpts 7, 8), by using assessment to express disbelief (Excerpt 10) or to strengthen doubt (Excerpt 9). At the same time, in all of the cases interlocutors’ interactional stances were incongruent in our data. The questioners expressed their non-neutral inter-
231
232
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
actional stance in the expansion of the sequence, while the partners responded neutrally (Excerpts 7–9) or expressed a different interactional stance (Excerpt 10). There is also a difference in the intonation of jaa compared to jah. Jaa is pronounced with slightly rising, flat or falling intonation (rising pitch contour in our data 39% (21 cases), falling 33% (18 cases) and flat 28% (15 cases)). Compared to jah, we see much more frequent usage of rising intonation, which often projects continuation in Estonian. At the same time, the sequence is continued regardless of the intonation of jaa. This reveals that the crucial element for the subsequent trajectory of the talk is the overall choice of the particle (jah or jaa), not the prosody of the particle.
5.
Conclusions and discussion
In this chapter, we have described the use of the Estonian particles jah and jaa as responses to positively formulated requests for confirmation in ordinary interaction. We examined what actions the particles perform, in what kinds of sequential contexts they appear, and whether they are functionally equivalent. Although both jah and jaa confirm the accuracy of the proposition presented in the question, our analysis reveals that there are systematic differences in their usage. The most important difference is found in how the sequence continues after the response. After jah, the local sequence is closed and the closing is accepted by both participants, while after jaa, the sequence is expanded. Both particles could be pronounced with falling, level and rising intonation. There is a difference in the frequency of rising intonation, typically projecting turn or sequence continuation, between the two particles. Jah as a closer of the sequence is almost never uttered with rising intonation. At the same time our analysis reveals that the intonation does not influence the continuation of the sequence. Regardless of the intonation, the sequence is closed after jah and expanded after jaa. We can conclude that the crucial element for the subsequent trajectory of the talk is the choice of the particle itself, not the prosody of the particle. The main context in which jah appears is a repair sequence, where jah occurs as a response to a problem of understanding (Excerpt 1) and to a hearing check (Excerpt 2). In addition, jah is used as a response to questions that express unexpectedness of the received information and go beyond initiating repair (Excerpts 3, 4). Jah is also used in the main line of the talk (Excerpts 5, 6). In all these contexts, jah projects closure of the local sequence. Schematized prototypical sequence of jah:
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
1. Q: request for confirmation 2. A: jah 3. Q: sequence closing third / new (sub)topic
The context of jaa is more limited and partially different from the context of jah. Jaa is used when the question expresses the unexpectedness of the provided information, displaying surprise or questioning the acceptability of the preceding talk (Excerpts 7–11). The sequence is continued after jaa by the questioner who asks for clarification, additional information, or adds some comment. The jaa producer also accepts the continuation of the sequence. Thus, jaa makes continuation of the sequence relevant instead of projecting sequence completion. Schematized prototypical sequence of jaa: 1. Q: request for confirmation + unexpectedness 2. A: jaa 3. Q: additional question / a comment
Actions that were responded to with jah and jaa can be divided into repair initiations and actions that go beyond repair. We showed that jah closed the sequence even if the unexpectedness expressed in the question would have required the expansion of the sequence. In some cases, the form of the request indicates that the questioner only orients to the confirmation (Excerpt 3), while in other cases the question form leaves the possibility of interpreting the problem behind the question differently (Excerpt 4). On the other hand, the speaker of jaa does not add information immediately after jaa in the same turn, although the questions require an expansion of the sequence. We explain this first and foremost by the form of the questions. The most typical question format answered by jaa is a repeat (mostly word or phrase) or an expression of ritualized disbelief. These question formats indicate that there was unexpectedness for the questioner, but do not specifically state what the unexpectedness consisted of. Thus, by using jaa, the answerers indicate that they are giving the questioners an opportunity to point out for themselves what caused the unexpectedness. In addition, in the case of jaa, the interactional stances of the interactants are systematically incongruent. In all cases, the questioners express their non-neutral interactional stance in the expansion of the sequence, indicating doubt (Excerpt 7, 9, 10) or negative attitude (Excerpt 8). However, the partner responds mostly neutrally (Excerpts 7–9) or expresses a different interactional (evaluative) stance (Excerpt 10). Finally, we will examine the usage of Estonian response particles and minimal particle responses from the perspective of pragmatic typology. Some languages have multiple unmarked particles (like Finnish joo and nii) or phonetic variants of one particle (like English yes, yep and yeah). Moreover, the same particle can
233
234
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
be pronounced with different types of intonation (e.g., Russian da). Responses formulated with different particles or different phonetic variants thereof are not equivalent. Firstly, different variants are used in response to differently formulated questions, like Finnish joo and nii (Sorjonen 1996, 2001a, 2001b). Secondly, different variants can indicate congruence or incongruence in the interactional stances between participants, like prosodic variants of da in Russian (Bolden 2016). In this chapter, we did not present all possible question formats which we have in our data. However, our analysis reveals that the selection of the response particle is not related directly to the format of the request for confirmation in Estonian. For example, both jah and jaa are used as responses to questions formulated by repeats (Excerpts 4 and 8). We cannot say that jaa itself indicates the incongruence. However, our analysis reveals that the choice of the particle is connected to the relevance of an incongruent interactional stance in Estonian. By using jaa the speaker indicates that the incongruence is relevant and gives the partner the possibility to expand the sequence to resolve the incongruence. In sum, our analysis highlights the point that, in different languages, different resources (e.g., particles and prosody) may be used differentially (cf. also ahah, ahaa and aa; Kasterpalu & Hennoste 2016). In addition, different questions and responses carry implications for the trajectory of the turn and/or sequence. Repair initiations for problems of hearing or understanding typically lead to sequence closure, while questions that display unexpectedness of the information given in the previous turn lead to sequence elaboration (Rossi 2020: 512). At the same time, different particles or their variants have different effects on turn or sequence continuation. The speakers guide the progression of the sequence by the response particle they choose. It has been argued that the particle yep instead of yes or yeah can project turn completion (Bolinger 1946; Heritage & Sorjonen 1994: 25; Raymond 2000: 43). In the case of Finnish non-V-interrogatives, joo suggests the closure of the departure, while nii makes continuation relevant (Sorjonen 2001b: 417, 421). In this regard, Estonian is somewhat like Finnish. Estonian jah (cf. Finnish joo) projects closing of the local sequence, while jaa (cf. Finnish nii) projects sequence continuation. There are languages in which multiple response particles and their phonetic variants are used. Some studies have argued that responses can be divided into ‘unmarked’ and ‘marked’ variants. ‘Unmarked’ responses only confirm the question’s proposition (factual accuracy) and align with the epistemic stance conveyed by it (e.g., Enfield et al. 2019: 287; Stivers 2018). In this typology, only jah could be classified as an ‘unmarked’ response particle. Jaa is a pragmatically ‘marked’ particle used for ‘special purposes’.
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
Funding The study was supported by the European Regional Development Fund (Centre of Excellence in Estonian Studies) and Estonian Research Council grant (PRG341 “Pragmatics overwrites grammar: subjectivity and intersubjectivity in different registers and genres of Estonian”) and by the Estonian Ministry of Education and Research (EKKM14–308 “A complete description of Spoken Estonian”, EKKM14–310 “Supplementation and equilibration of the University of Tartu corpus of spoken Estonian”, EKKM14–340 “Reference grammar of Estonian”).
References Asu, Eva Liina. 2006. “Rising Intonation in Estonian: An Analysis of Map Task Dialogues and Spontaneous Conversations.” In Fonetiikan Päivät 2006. The Phonetics Symposium 2006, ed. by Reijo Aulanko, Leena Wahlberg, and Martti Vainio, 1–8. Helsinki: Helsinki University. Bolden, Galina. 2016. “A Simple da?: Affirming Responses to Polar Questions in Russian Conversation.” Journal of Pragmatics 100: 40–58. Bolinger, Dwight L. 1946. “Thoughts on ‘Yep’ and ‘Nope’.” American Speech 21 (2): 90–95. Drew, Paul. 1997. “‘Open’ Class Repair Initiators in Response to Sequential Sources of Trouble in Conversation.” Journal of Pragmatics 28: 69–101. Enfield, N. J., Mark Dingemanse, Julija Baranova, Joe Blythe, Penelope Brown, Tyko Dirksmeyer, Paul Drew, Simeon Floyd, Sonja Gipper, Rósa S. Gísladóttir, Gertie Hoymann, Kobin H. Kendrick, Stephen C. Levinson, Lilla Magyari, Elizabeth Manrique, Giovanni Rossi, Lila San Roque, and Francisco Torreira. 2013. “Huh? What? – A First Survey in Twenty-One Languages.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 343–380. Cambridge: Cambridge University Press. Enfield, N. J., Tanya Stivers, Penelope Brown, Christina Englert, Katariina Harjunpää, Makoto Hayashi, Trine Heinemann, Gertie Hoymann, Tiina Keisanen, Mirka Rauniomaa, Chase Wesley Raymond, Federico Rossano, Kyung-Eun Yoon, Inge Zwitserlood, and Stephen C. Levinson. 2019. “Polar Answers.” Journal of Linguistics 55 (2): 277–304. Erelt, Mati, and Helle Metslang (eds). 2017. Eesti keele süntaks. Tartu: Tartu Ülikooli Kirjastus. Hakulinen, Auli, Maria Vilkuna, Riitta Korhonen, Vesa Koivisto, Tarja-Riitta Heinonen, and Irja Alho. 2004. Iso suomen kielioppi. Helsinki: Suomalaisen Kirjallisuuden Seura. Hennoste, Tiit. 2000a. “Sissejuhatus suulisesse eesti keelde IV. Suulise kõne erisõnavara 3. Partiklid.” Akadeemia 8: 1773–1806. Hennoste, Tiit. 2000b. “Sissejuhatus suulisesse eesti keelde V. Mõned mitteverbaalsed nähtused suulises kõnes.” Akadeemia 9: 2011–2038. Hennoste, Tiit. 2012. “Küsimuse vorm, episteemiline staatus ja episteemiline hoiak.” Keel ja Kirjandus 8/9: 674–695. Hennoste, Tiit. (in press). “Suuline keel.” In Eesti grammatika, ed. by Helle Metslang. Tartu: Tartu Ülikooli Kirjastus.
235
236
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
Hennoste, Tiit, Olga Gerassimenko, Riina Kasterpalu, Mare Koit, Andriela Rääbis, and Krista Strandson. 2008. “From Human Communication to Intelligent User Interfaces: Corpora of Spoken Estonian.” In Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC’08). Marrakech, Morocco. European Language Resources Association (ELRA). http://www.lrec-conf.org/proceedings /lrec2008/pdf/518_paper.pdf Hennoste, Tiit, Olga Gerassimenko, Riina Kasterpalu, Mare Koit, Andriela Rääbis, and Krista Strandson. 2009. “Towards an Intelligent User Interface: Strategies of Giving and Receiving Phone Numbers.” In Text, Speech and Dialogue. TSD 2009. Lecture Notes in Computer Science, vol 5729, ed. by Václav Matoušek, and Pavel Mautner, 347–354. Berlin/Heidelberg: Springer. Hennoste, Tiit, Andriela Rääbis, and Kirsi Laanesoo. 2017. “Polar Questions, Social Actions and Epistemic Stance.” STUF – Language Typology and Universals 70 (3): 523–544. Hennoste, Tiit, Andriela Rääbis, and Andra Rumm. 2019. “Estonian Declarative Questions: Their Usage and Comparison with vä- and jah-questions.” Journal of Pragmatics 153: 46–68. Heritage, John. 1984. “A Change-of-State Token and Aspects of Its Sequential Placement.” In Structures of Social Action: Studies in Conversation Analysis, ed. by J. Maxwell Atkinson, and John Heritage, 299–345. Cambridge: Cambridge University Press. Heritage, John. 2012. “Epistemics in Action: Action Formation and Territories of Knowledge.” Research on Language and Social Interaction 45 (1): 1–29. Heritage, John, and Geoffrey Raymond. 2012. “Navigating Epistemic Landscapes: Acquiescence, Agency and Resistance in Responses to Polar Questions.” In Questions: Formal, Functional and Interactional Perspectives, ed. by Jan P. de Ruiter, 179–192. Cambridge: Cambridge University Press. Heritage, John, and Marja-Leena Sorjonen. 1994. “Constituting and Maintaining Activities across Sequences: And-Prefacing as a Feature of Question Design.” Language in Society 23 (1): 1–29. Kasterpalu, Riina. 2005. “Partiklid jah, jaa ning jajaa naaberpaari järelliikmena eestikeelsetes müügiläbirääkimistes.” Keel ja Kirjandus 11: 873–890, 12: 996–1000. Kasterpalu, Riina, and Tiit Hennoste. 2016. “Estonian aa: A Multifunctional Change-of-State Token.” Journal of Pragmatics 104: 148–162. Keevallik, Leelo. 2003a. From Interaction to Grammar. Estonian Finite Verb Forms in Conversation. Acta Universitatis Upsaliensis. Studia Uralica Upsaliensia 34. Uppsala. Keevallik, Leelo. 2003b. “Terminally Rising Pitch Contours of Response Tokens in Estonian.” Crossroads of Language, Interaction and Culture, vol. 5: 49–65. Keevallik, Leelo. 2008. “Conjunction and Sequenced Actions: The Estonian Complementizer and Evidential Particle et.” In Crosslinguistic Studies of Clause Combining: The Multifunctionality of Conjunctions, ed. by Ritva Laury, 125–152. Amsterdam: John Benjamins. Keevallik, Leelo. 2009. “Üldküsimuse lihtvastuse funktsioonid.” Keel ja Kirjandus 1: 33–53. Keevallik, Leelo. 2010. “Minimal Answers to Yes/No Questions in the Service of Sequence Organization.” Discourse Studies 12 (3): 283–309.
Chapter 7. The division of labor between the particles jah and jaa ‘yes’
Keevallik, Leelo. 2011. “The Terms of Not Knowing.” In The Morality of Knowledge, ed. by Tanya Stivers, Lorenza Mondada, and Jakob Steensig, 184–206. Cambridge: Cambridge University Press. Keevallik, Leelo. 2016. “Estonian no(o)(h) in Turns and Sequences: Families of Function.” In NU and NÅ: A Family of Discourse Markers Across the Languages of Europe and Beyond, ed. by Peter Auer, and Yael Maschler, 213–242. Berlin/Boston: De Gruyter. Kendrick, Kobin H. 2015. “Other-Initiated Repair in English.” Open Linguistics 1 (1): 164–190. Kendrick, Kobin H., and Francisco Torreira. 2014. “The Timing and Construction of Preference: A Quantitative Study.” Discourse processes 52 (4): 255–289. Koivisto, Aino. 2014. “Displaying Now-Understanding: The Finnish Change-of-State Token aa.” Discourse Processes 52 (2): 111–148. Kurhila, Salla, and Niina Lilja. 2017. “Toisto ja korjauksen rajat.” Virittäjä 121 (2): 213–243. Lee, Seung-Hee. 2015. “Two Forms of Affirmative Responses to Polar Questions.” Discourse Processes 52 (1): 21–46. Rääbis, Andriela. 2009. Eesti telefonivestluste sissejuhatus: struktuur ja suhtlusfunktsioonid. Dissertationes linguisticae Universitatis Tartuensis 13. Tartu: Tartu Ülikooli Kirjastus. Raymond, Geoffrey. 2000. The Structure of Responding. Unpublished PhD thesis, University of California, Los Angeles, CA. Raymond, Geoffrey. 2010. “Prosodic Variation in Responses: The Case of Type-Conforming Responses to Yes/No Interrogatives.” In Prosody in Interaction, ed. by Dagmar Barth-Weingarten, Elisabeth Reber, and Margret Selting, 109–130. Amsterdam: John Benjamins. Raymond, Geoffrey. 2013. “At the Intersection of Turn and Sequence Organization: On the Relevance of “Slots” in Type-Conforming Responses to Polar Interrogatives.” In Units of Talk – Units of Action, ed. by Beatrice Szczepek Reed, and Geoffrey Raymond, 169–206. Amsterdam: John Benjamins. Robinson, Jeffrey D., and Heidi Kevoe-Feldman. 2010. “Using Full Repeats to Initiate Repair on Others’ Questions.” Research on Language and Social Interaction 43 (3): 232–259. Rossi, Giovanni. 2020. “Other-Repetition in Conversation across Languages: Bringing Prosody into Pragmatic Typology.” Language in Society 49: 495–520. Rumm, Andra. 2019. Avatud küsimused ja nende vastused eesti suulises argivestluses. Dissertationes linguisticae Universitatis Tartuensis 36. Tartu: Tartu Ülikooli Kirjastus. Schegloff, Emanuel A. 2007. Sequence Organization in Interaction. A Primer in Conversation Analysis. Volume 1. New York: Cambridge University Press. Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks. 1977. “The Preference for SelfCorrection in the Organization of Repair in Conversation.” Language 53: 361–382. Selting, Margret. 1996. “Prosody as an Activity-Type Distinctive Cue in Conversation: The Case of So-Called ‘A stonished’ Questions in Repair Initiation.” In Prosody in Conversation: Interactional Studies (Studies in Interactional Sociolinguistics 12), ed. by Elizabeth Couper-Kuhlen, and Margret Selting, 231–270. Cambridge: Cambridge University Press. Sorjonen, Marja-Leena. 1996. “On Repeats and Responses in Finnish Conversations.” In Interaction and Grammar (Studies in Interactional Sociolinguistics 13), ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 277–327. Cambridge: Cambridge University Press.
237
238
Tiit Hennoste, Andriela Rääbis, Andra Rumm & Kirsi Laanesoo
Sorjonen, Marja-Leena. 2001a. Responding in Conversation: A Study of Response Particles in Finnish. Amsterdam: John Benjamins. Sorjonen, Marja-Leena. 2001b. “Simple Answers to Polar Questions: The Case of Finnish.” In Studies in Interactional Linguistics, ed. by Margret Selting, and Elizabeth Couper-Kuhlen, 405–431. Amsterdam: John Benjamins. Stivers, Tanya. 2018. “How We Manage Social Relationships through Answers to Questions: The Case of Interjections.” Discourse Processes 56 (3): 191–209. Strandson, Krista. 2001. “Kuidas vestluskaaslane parandusprotsessi algatab?” Keel ja Kirjandus 6: 394–409. Svennevig, Jan. 2004. “Other-Repetition as Display of Hearing, Understanding and Emotional Stance.” Discourse Studies 6 (4): 489–516. Thompson, Sandra A., Barbara A. Fox, and Elizabeth Couper-Kuhlen. 2015. Grammar in Everyday Talk: Building Responsive Actions (Studies in Interactional Sociolinguistics 31). Cambridge: Cambridge University Press. Wilkinson, Sue, and Celia Kitzinger. 2006. “Surprise as an Interactional Achievement: Reaction Tokens in Conversation.” Social Psychology Quarterly 69 (2): 150–182.
chapter 8
Code-switching, agency, and the answer possibility space of Spanish-English bilinguals Chase Wesley Raymond
University of Colorado Boulder
The present study addresses the implicit monolingual bias in our understanding of polar question-answer sequences by examining them in bilingual conversation. Drawing upon a corpus of naturally-occurring conversational data amongst native Spanish-English bilinguals in the southwestern United States, I target and explore the ‘fit’ between polar questions and answers in this community of practice. In this report, I focus specifically on the dimension of language (non-)concordance between particle answers and the questions they address. Evidence is offered that speakers produce response particles that are concordant with the language of the question as the pragmatically unmarked answer format, whereas response particles that are non-concordant with the language of the question (i.e., ‘code-switched’) are marked, produced agentively and for cause, routinely indexing an emergent stance that is at variance with the terms established by the question. Some possible avenues for future comparative work on the expression of agency in question-answer sequences are explored in the Conclusion. Keywords: code-switching, bilingualism, agency, polar questions, answers, conversation analysis, English, Spanish (in the U.S.), heritage speakers
1.
Introduction
Speakers make selections from an array of options when asking questions in conversation. These include, amongst other things, the lexical composition of the turn and its morphosyntactic and prosodic packaging, the timing and sequential placement of its delivery, as well as any multimodal conduct related to its production and interpretation in context. Such elements of question design are best conceptualized as resources that questioners can configure in specific ways, at https://doi.org/10.1075/slsi.35.08ray © 2023 John Benjamins Publishing Company
240
Chase Wesley Raymond
specific moments, to do specific interactional work. That is, in addition to creating a sequential slot for an answer, questions simultaneously set topical and action agendas, embody presuppositions, convey an epistemic stance, and can incorporate preferences of various sorts – all through the particulars of their construction (for overviews, see Bolden, Heritage & Sorjonen, this volume; Hayano 2013; Heritage 2010; G. Raymond 2003; Sacks 1987 [1973]). In these and other ways, questions establish a complex and somewhat constraining environment for the very answers they make conditionally relevant. Consider the case of epistemic stance in the design of polar (yes/no) questions. If, as Bolinger (1978: 104) observed, “a question advances a hypothesis for confirmation, in any degree, not just in terms of polar opposites,” epistemic stance can be conceived of as the nature of the questioner’s commitment to the likelihood of that hypothesis. A question in English like Are you going to the party on Friday? – posed with interrogative morphosyntax and rising intonation – indexes a steep epistemic gradient, casting the questioner as relatively unknowledgeable about whether the recipient is or is not going to the party, and thereby requesting that information (Heritage 2012; Stivers 2010). In contrast, with a design like So you’re not going to the party on Friday. – posed with a so-preface (Bolden 2009), declarative morphosyntax and falling intonation (Heritage 2012; Stivers 2010), and incorporating negative polarity (Heritage & C. W. Raymond 2021) – the speaker indexes a far greater commitment to the proposition that the recipient’s attendance at the party is unlikely, and thereby is heard to be requesting confirmation of that hypothesis (see also C. W. Raymond et al. 2021). Borrowing a phrase from elsewhere in the research on social interaction, it might be said that these (and other) different practices of question design initiate, from the interactants’ perspectives, “alternative, but nonequivalent” courses of action through their construction (Atkinson & Heritage 1984: 53). Polar questions are one half of a canonical adjacency pair structure, with answers (upon their provision) providing the second-pair part of the sequence (Schegloff 2007). Questions thus open up what Stivers (2022) terms an “answer possibility space”, which recipients can then fill with some form of response. Importantly, in the same way as questioners select from amongst an array of resources to design their questions, so too do answerers construct their responsive turns in order to enact distinct stances through their answers. Thus, while Yeah and Of course are both semantically affirmative answer formats, and therefore both can be used to affirm the proposition embodied in the question to which they respond, Of course is produced and oriented to as additionally taking issue with the ‘askability’ of the question in a way that Yeah does not (G. Raymond 2003; Stivers 2010, 2011, 2022). In other words, notwithstanding sequentially ‘going second’ in terms of speaking, answerers are not left at the mercy of the
Chapter 8. Answer possibility space of Spanish-English bilinguals
“terms” (Heritage & G. Raymond 2012; G. Raymond 2003) set by the design of the questioning turn; although they can acquiesce to these terms, they may also agentively resist them in myriad ways through the format of their response (see, e.g., on English, Heritage & G. Raymond 2012; C. W. Raymond 2017: 31–33; C. W. Raymond et al. 2021; G. Raymond 2003; Stivers 2022; Stivers & Hayashi 2010; Thompson, Fox & Couper-Kuhlen 2015: 16–49). While conversation-analytic research on polar question-answer sequences has recently experienced a surge in cross-linguistic and comparative inquiry (e.g., Stivers, Enfield, and Levinson 2010; Enfield, et al. 2019), these investigations have embodied a primary focus on monolingual speakers, or at least on speakers interacting in an exclusively or primarily monolingual “mode” (Grosjean 1998) or “medium” (Gafaranga 2005). The present study aims to address this implicit monolingual bias (cf. Auer 2007) in our understanding of polar question-answer sequences by examining them in bilingual conversation. Drawing upon a corpus of naturally ocurring conversational data amongst native Spanish-English bilinguals in the southwestern United States, I target and explore the ‘fit’ between the design of polar questions and the design of particle answers in this bilingual communitity of practice, focusing specifically on the dimension of language (non-)concordance1 between the answer and the question it addresses. The chapter is structured as follows: I begin by briefly describing my data and methods, before offering a distributional overview of the bilingual answer possibility space of speakers in this community of practice. This initial quantification supports the notion that language choice itself is one of the terms that a polar question sets up through its design, and lays the groundwork for the more detailed analysis of individual exemplars that follows, where this hypothesis is further pursued. In the examination of cases, I review language-concordant particles first, followed by language-non-concordant ones, considering instances of both confirmation and disconfirmation. Evidence will be offered that speakers produce response particles that are concordant with the language of the question as the pragmatically unmarked answer format, whereas response particles that are nonconcordant with the language of the question (i.e., ‘code-switched’) are marked and produced agentively and for cause, routinely indexing an emergent stance that is at variance with the terms established by the question. Some avenues for future comparative work are then explored in the Conclusion.
1. I use the terms ‘concordance’ and ‘non-concordance’ to refer specifically to the relationship between the language of a first action and its responsive action (see C. W. Raymond 2020, where I formerly used ‘discordance’). Other terminology – e.g., ‘code-switching’, ‘language (dis)accommodation’ – does not account for sequence organization in this way.
241
242
Chase Wesley Raymond
2. Data & methods This study draws upon a corpus of naturally-occurring conversation (~200 hours) amongst native bilingual speakers of Spanish and English in the southwestern United States. These individuals are predominantly heritage speakers of Spanish who were born and/or raised in the U.S., and who thus acquired both languages during childhood and actively use both languages in adulthood with friends, family members and coworkers. The interactants in the data examined here are such individuals engaged in everyday conversation with one another. Centering the analysis on this sort of participant and context is, in and of itself, a crucial difference from the object of study in many previous investigations involving dimensions of ‘bilingual’ conversation – particularly those in institutional environments where participants are often not familiar with each other’s linguistic competencies and preferences (e.g., Hänggi 2022; Heller 1982; Gafaranga & Torras 2002; Mondada 2018; C. W. Raymond 2020; see also Cashman 2005; Wei 1994). For the aims of this study, the potentially confounding issues of intercultural communication and second-language pragmatics, as well as norms associated with particular institutional contexts, are set aside, as I focus solely on interactions involving native bilinguals, known to one another, engaged in ordinary conversation.2 Data were transcribed according to the conventions outlined by Jefferson (2004). Analysis was conducted according to the theory and methods of conversation analysis, which aims to uncover the practices, and normative organizations of practice, through which participants constitute and make sense of action in interaction (for overviews, see Clift 2016; Heritage 1984b; C. W. Raymond & Olguín 2022; Sidnell & Stivers 2013). By centering our inquiry on polar question-answer sequences, we hold constant our sequential focus, thereby establishing what Schegloff (1993: 106) calls an “environment of relevant possible occurrence” for different answer types. This approach in turn allows for both the detailed examination of the ‘fit’ between question and answer design in singular cases, as well as systematic comparison across collections of exemplars that are similar and dissimilar (see Clayman & Gill 2004; Clift & C. W. Raymond 2018; Schegloff 1996).
2. A range of additional epistemological issues should be carefully considered when examining bilingual (and bidialectal) conversational data. For further discussion, see Auer (1984, 1995), Gafaranga (2005), C. W. Raymond (2018, 2020), and Wei (1998, 2002).
Chapter 8. Answer possibility space of Spanish-English bilinguals
3.
The bilingual answer possibility space
When presented with a polar question, what does the repertoire of possibilities that answerers in this bilingual community of practice have at their disposal look like, and in what local questioning contexts are particular options from that repertoire recurrently produced?
3.1 Language concordance vs. non-concordance An initial impression of the bilingual answer possibility space can be gleaned from the descriptive statistics reported in Table 1. This quantitative overview draws upon a collection of 486 polar question-answer sequences from the abovementioned corpus of native Spanish-English bilinguals, assembled in accordance with the procedures outlined in Stivers and Enfield (2010) (see C. W. Raymond 2015).3 Table 1. Overview of the answer possibility space of Spanish-English bilinguals Response particles
427 (88%)
Language-concordant
359 (74%)
Language-non-concordant
68 (14%)
Other answer formats
59 (12%)
Repetitions
43 (8.8%)
Marked Particles (e.g., ‘of course’)
16 (3.3%)
Total
486 (100%)
As shown in the table, the vast majority of answers (88%) take the form of response particles,4 with other answer formats, like repetitions (8.8%) and marked particles (3.3%), being produced far less frequently. Moreover, if we probe the response particle category further, we see that these are overwhelmingly concordant with the language of the question to which they respond, with only a minority being language-non-concordant (i.e., code-switched). These descriptive statistics offer an initial piece of evidence suggesting that answerers attend to 3. Given the present focus on particle answers, non-answer responses (e.g., I don’t know), and transformative answers (Stivers & Hayashi 2010), are excluded from these figures; but see C. W. Raymond (2015). Note that in this earlier work, following Stivers and Enfield (2010), I referred to particle answers as ‘unmarked interjections’. 4. Here I refer to answers comprising of particles only, as opposed to those cases where particles are produced as turn-initial components of larger, clausal response formats.
243
244
Chase Wesley Raymond
the language of the question to which they are responding when formulating an answer, as opposed to selecting randomly or agnostically between the languages at their disposal.5 As an initial exemplar, consider Excerpt (1), in which a caller to a radio program6 produces three affirmative answers in a row, each one being concordant with the language of the question to which it responds – first English (yes), then Spanish (sí), then English (yes) again: Excerpt 1. (Call-In-12_9.40.29; C. W. Raymond 2020: 414–415) 01
HST:
Myrna:.=where are you. =así le dijo like-that DAT.3SG say.PFV.3SG =that’s how she ((Grace)) said it
16
17
a su c[ara? to her face to her ((supervisor’s)) face? RAQ:=>
17.1 raq:=>
[Yea:h. [((single nod))
5. Some other marked answer formats (repetitions, claro que sí ‘of course’) are considered in C. W. Raymond (2015: 60–64). And for an examination of así es (‘that’s it’) as an answer type in the context of sports broadcasting discourse, see Cashman and C. W. Raymond (2014) and C. W. Raymond and Cashman (2021). 6. This institutional example – from the same speech community but in radio talk – is used here purely to show three language-concordant question-answer sequences in rapid succession. The remainder of the examples presented will be from mundane conversational data, as described in Section 2.
Chapter 8. Answer possibility space of Spanish-English bilinguals
Excerpt 3. (F15.RA.10:10:00) 16
BRI: -> oh but- (.) so then, (.) so then they
17
-> can’t use ‘em on the weekend then.
18
(.)
19
BRI:
20
SEL: =>
er- (.) I gue[ss unless they [ha[Sí↓:↑: .h
[↑sí:,
This relative distribution of language concordance vs. non-concordance in polar answer design suggests that, for speakers in this bilingual community of practice, language choice is one of the terms established by a questioning turn. In other words, polar questions set up a local sequential expectation that the language of the question will be preserved in the answering turn, and answers depart from this norm ‘for cause’ (C. W. Raymond 2015, 2020). To explore this hypothesis further, in what follows I compare the use of language-concordant response particles with the use of langauge-non-concordant ones.
3.2 Language-concordant particles As previously noted, by far the most prevalent answer format in this bilingual community of practice is a language-concordant response particle – e.g., yeah/ no in response to an English question, sí/no in response to a Spanish question (Table 8.1). Such answers greatly outnumber all other answer formats, at nearly three-quarters of all answers provided (74%). The interactional environments in which answerers produce languageconcordant particles reveal them to be the pragmatically unmarked format for answering a polar question (C. W. Raymond 2015, 2020). Speakers use this answer format to deliver answers without “introduc[ing] any pragmatic turbulence” (Enfield et al. 2019: 291) into the local sequence, accepting the terms of the question as it was designed, “exerting no agency with respect to those terms, and thus acquiescing in them” (Heritage & G. Raymond 2012: 183). In this way, bilingual speakers display an orientation to language choice as a relevant feature of turn design – both in the construction of questions, as well as answers. Sacks’s (1987 [1973]) early observations about question design – namely that questioners work to formulate questions such that answerers can accept the terms as they are put forth in the questioning turn (see also Heritage & C. W. Raymond 2021) – are supported by the prevalence of language-concordant particle answers in the dataset. Consider Extract (4), in which Ana has been telling Joe about her visit to a shopping center this past weekend. Following this telling, Joe comments that it has been a while since he has been to the center, but that he used to go on the
245
246
Chase Wesley Raymond
weekends, with which Ana immediately affiliates (lines 1–5). Following a silence in which both participants are chewing their food (line 6), Ana asks andaba(h) con tu familia. (‘{you} used to go with your family.’ ) (line 7), which receives a language-concordant confirmation sí (line 8). Excerpt 4. (F12.15:00) 01
JOE:
02
03
siempre íbamos los fines.= always go.IPFV.1PL the ends we always used to go on weekends ANA:
04
05
hace mucho tiempo que no vamos. make.PRS.3SG much time that no go.PRS.1PL it’s been a long time since we’ve gone there
=oh my go::d.=us too. como cada fin de sema[na. like each end of week like every weekend
JOE:
06
[ajá:, ((chewing)) uh huh (1.5)
((both participants chewing))
07
ANA:-> andaba(h) con tu familia.= walk.IPFV.2SG with your family {you} used to go with your family
08
JOE:=> =sí.
09
ANA:
nosotros también. we also us too
10
.hh a Magda le gustaba to M DAT.3SG please.IPFV.3SG .hh Magda used to like
11
un restaurante que tenían allí,= a restaurant that have.IPFV.3PL there a restaurant that they had there
12
= este::=el (.) el… uhm the the uhm the (.) the…
In this case, Ana’s question (line 7) draws an inference from Joe’s use of firstperson plural references in lines 1–2 (and her own such use in line 3), issuing a request for confirmation that he was referring to visits to the center with his family. Through his language-concordant sí, Joe acquiesces to the questioner’s “sequential and thematic agency, allowing [her] to set the agenda of the local conversational sequence” (Enfield, et al. 2019: 296). He does not pursue any expan-
Chapter 8. Answer possibility space of Spanish-English bilinguals
sion of his turn or this sequence (Heritage & G. Raymond 2012: 186), but rather allows Ana to do so; and indeed, in the context of this acquiescent answer, Ana immediately retakes the floor (line 9) and proceeds to launch a story about her daughter, Magda, which took place at one of the center’s restaurants (lines 10–12). In short, then, here a language-concordant particle is produced and understood as affirming the hypothesis put forth in the question, including the stance the questioner enacted through its design. Joe is not oriented to as having, with his response, taken issue with Ana’s asking of the question, or her having asked it in a certain way; rather, both participants orient to the language-concordant response particle as a pragmatically unmarked, acquiescent answer type. Further support that language-concordant particles constitute the unmarked polar answer option can be found in the predominance of such answers in response to other-initiations of repair (Schegloff, Jefferson, and Sacks 1977). This is in line with earlier claims that the answer types used in such subordinate action sequences cross-linguistically are designedly acquiescent and non-agentive, produced in the service of “allowing the dialogue to get quickly back on track” (Enfield, et al. 2019: 291; see Hakulinen 2001; Keevallik 2010; Sorjonen 2001; Stivers 2010). While this occurs in both Spanish and English insert sequences, I present just an English example here, in Extract (5). In this case, Tamara and Sancho are discussing restaurant options for dinner with a large group of people. The primary issue is finding a suitable restaurant that can accommodate both the size of the group as well as the participants’ various dietary restrictions. Tamara proposes the “Vagabond’ in line 1, which Sandro resists, claiming he also had the Vagabond in mind but withheld proposing it in light of its extremely small size (lines 4–7) (Küttner & C. W. Raymond 2022; C. W. Raymond 2022). Our focus is on Tamara’s initiation of repair on the number of tables the Vagabond has, which continues with Sancho’s switch to English (line 7) and indicates a problem of hearing: “Two you said,” (line 9).7 Excerpt 5. (2013.Sum.08:22:20) 01
TAM:
02
03
qué tal el Vagabundo. what about the vagabond what about the Vagabond. ((restaurant name)) nu[nca]=he ido. never have.PRS.1SG go-PPT I’ve never gone (there)
SAN:
[yo-]
7. While the focus in this study is on responsive actions, initiating actions likewise merit systematic examination in this way (e.g., code-switching in other-initiations of repair vis-à-vis the turn they initiate repair on) (see Wei & Milroy 1995).
247
248
Chase Wesley Raymond
I 04
SAN:
yo iba a decir el Vagabundo pero I go.IPFV.1SG to say.INF the vababond but I was gonna say the Vagabond but
05
este:: (.) la parte interior es this the part interior be.PRS.3SG uhm the interior is
06
chiquita pero chiquiTIta. small.DIM but small.DIM.DIM small like really small.
07
.hh they only have like e-two tables.
08
(.)
09
TAM:-> two you said,
10
SAN:=> yeah.=
11
TAM:
12
SAN:
13
SAN:
14
TAM:
=yeahbuen(o)=entonce’ no.< okay then no okay then no
In response to Tamara’s English initiation of repair (line 9), Sancho issues a language-concordant particle yeah (line 10). This provides for the immediate (latched) resumption of the main sequence in that Tamara moves directly on to provide the agreement that had been due at line 8, prior to the initiation of repair: yeah piensas comprarlas? think.PRS.2SG buy.INF=CLI.ACC.3PL.F are {you} thinking about buying them (.)
8. While the negative particle ‘no’ is orthographically the same in Spanish and English, the vowel quality is distinct: Spanish’s monophthongal [no] vs. English’s diphthongized/labialized [nəʊ]. See also Extract (7).
249
250
Chase Wesley Raymond
13
PED:=> no:.
14
(.)
15
RON:
[por-] wh(y)
16
PED:
[sí- ] (.) sí quisiera:, yes yes want.IMP.SBJV.1SG I would indeed like to
17
pero ya tengo tres mil ig(h)uales.= but already have.PRS.1SG three thousand same.PL but I already have three thousand of the same
18
RON:
19
PED:
=£o(h)h:[:£=hah hah hah [hah hah hah
In a context where Ron has topicalized these sunglasses and is showing them to Pedro, and where both have agreed that they are ‘beautiful’ and ‘inexpensive’, Ron’s question draws an inference in light of the preceding talk and puts forth a hypothesis that Pedro may have at least some intention to buy them (Heritage & C. W. Raymond 2021; Robinson 2020). What is particularly noteworthy about this case is that, notwithstanding the disconfirmation, Pedro goes on to legitimize the hypothesis that Ron’s question design intimated: After issuing his languageconcordant particle no:., but prior to launching the account itself, he produces sísí quisiera:, (‘I would indeed like to:,’ ) (line 14), with polar emphasis via sí (cf. C. W. Raymond 2017, 2019). With this, Pedro effectively endorses the stance Ron took in asking his question – i.e., this was an appropriate inference to make and inquire about in this way – as opposed to resisting or otherwise challenging Ron’s inferencing or stance. I propose that what binds language-concordant response particles together as an answer type is their acquiescent pragmatic force. Such answers are produced and understood as accepting the terms of the question as it was asked, acquiescing to the questioner’s agency within the local sequence and thereby legitimating the question’s action and its design. This reveals language choice to be one of the relevant terms oriented to by participants in the formulation of polar questions, in that acquiescence to questioners’ choice of language is demonstrably implicated in the design of answers that acquiesce to the terms of the question more broadly. This argument becomes clearer as we compare the use and contexts of occurrence of language-concordant particle answers with those of language-non-concordant ones, to which we now turn.
Chapter 8. Answer possibility space of Spanish-English bilinguals
3.3 Language-non-concordant particles Interactional research on code-switching in bilingual conversation has convincingly shown that “the sequential organization of alternative choices of language” is what itself “provides a frame of reference for the interpretation of functions or meanings of ” a code-switch in situ (Wei 1998: 164; see also Auer 1984; Cashman 2008; Wei 2002). In this way, moments of language alternation, regardless of their directionality, serve as “contextualization cues” (Gumperz 1982) that work to “indicate otherness” (Auer 1995: 123–124), which participants interpret for meaning within the local environment of their production. Under what interactional conditions do we find this “otherness” indexed within our focal sequential context (answers to polar questions), and how do participants interpret its meaning there? In contrast with the use of language-concordant response particles, language-non-concordant particles (14% of all answers; Table 8.1) – e.g., an English particle in response to a Spanish question (recall Extracts [2] and [3], above) – constitute a marked, agentive answer type within the bilingual answer possibility space of this community of practice. In breaking with the language of the question, these answers are produced and understood as resisting the terms established more broadly through the questioner’s turn. That is, through their very asking or through some aspect(s) of their design, these questions are oriented to as somehow misaligned to the state of affairs relevantly at play in the interaction. In such cases, the “otherness” (Auer 1995: 124) indexed through language non-concordance in the answering turn is agentive and emphatic in enacting this stance from a sequence-responsive position. Language-non-concordant response particles therefore go beyond simply answering the question as it was put, instead actively introducing some “pragmatic turbulence” (Enfield, et al. 2019: 291) into the local action sequence via language choice. As a first exemplar of this sort, consider the language-non-concordant disconfirmation in (7). The extract begins as roommates Carmen and Fabiola are finishing looking at something on Fabiola’s cell phone while seated together at the dining table. At line 2, as Carmen disengages from looking at the phone and leans back in her chair, she takes a large in-breath and produces a shift-implicative anyway (Park 2010). Then, in reference to some friends from out of town who will be arriving later in the day, she announces that the two of them (Carmen and Fabiola) will need to leave before three o’clock to head to the airport to pick them up (lines 3–5), with which Fabiola agrees (line 6). Following a silence, Fabiola produces las dos llegan más o menos a la vez,=verdad? ‘both of them arrive more or less at the same time, right?’ (lines 8–9), the [declarative + tag] format indexing a shallow epistemic gradient (Heritage 2012; C. W. Raymond 2015) that strongly invites confirmation of the state of affairs presented in the question.
251
252
Chase Wesley Raymond
Excerpt 7. (2014.Kitchen1.15:00) 01 02
(4.0) ((participants looking at phone together))
CAR:
.hhhhh anyway,=hhh ((leaning back in chair))
03
.hh hay que irnos have.PRS.3SG that go.INF=REFL.1PL .hh we have to leave
04
ya antes de las tres, already before of the three already before three
05
pa’ recoger a las chicas. for pick-up.INF ACC the girls tuh pick up the girls.
06
FAB:
07 08
(1.0) FAB:-> las dos llegan the two arrive.PRS.3PL they both arrive
09
10
‘tá bien. be.PRS.3SG well okay
-> más o menos a la vez,=verdad?= more or less at the time truth more or less at the same time right CAR:=> =[no::(w),=I already to:ld you,
10.1 car:=> =[((eyebrow furrow)) 11
we’re gonna >have to< wait at
12
the roundy round.
13
FAB:
oh,=(th)at’s right.
In response to Fabiola’s request for confirmation, Carmen issues a disconfirming no::(w), (line 10) which, to be sure, is pronounced as an English diphthongized/ labialized [nəʊ] and is thus designedly non-concordant with the language of the question. Moreover, it is produced on the heels of the question to which it responds (i.e., latched), and thus occurs quickly, particularly for a dispreferred answer (Pomerantz & Heritage 2013; Sacks 1987 [1973]; Schegloff 2007). Carmen continues using English as she expands her turn, but importantly, before providing the relevant information regarding the pick-up, she includes a preface that flags the epistemic inappositeness of the question. With the admonishment I already to:ld you, Carmen follows up her language-non-concordant response particle by explicitly holding Fabiola accountable for asking a question that she
Chapter 8. Answer possibility space of Spanish-English bilinguals
should already have known the answer to, and moreover, for in fact requesting confirmation of that incorrect hypothesis. In contrast with what we saw in (6) above, then, where a language-concordant disconfirmation prefaced more talk from the answerer that legitimized the asking of the question, here Carmen’s language-non-concordant disconfirmation is followed by an expansion that specifically highlights the question’s inappropriateness. Note that Fabiola quickly concedes in light of Carmen’s stance toward her question, responding with oh,=(th)at’s right., indicating a ‘just now’ recollection (Heritage 1984a; Küttner 2018; see also Koivisto 2013; Seuren, Huiskes, and Koole 2016). Consider another disconfirmation case in (8). Here, Selma is telling Brian about the laptop loan program that her university’s bookstore offers: Students can check out laptops for free and use them to complete schoolwork. After positively assessing the program together (lines 1–8), Brian asks if the store is open every day (line 9), interdicting the onset of Selma’s response (line 11) to expand his question with so you can like- (.) take it to cla:ss ‘n stu:ff ? (lines 12–13). Selma’s answer is presented as an affirmation of Brian’s hypothesis (yeah), immediately following up the answer particle by commenting that the store is open five days a week (line 14). The difference between everyday and five days a week then becomes topicalized as potentially relevant for assessing the program (cf. C. W. Raymond & White 2022). Brian produces a change-of-state token oh (Heritage 1984a), followed by a cut-off but-, before restarting his turn twice to produce: so then, (.) so then they can’t use ‘em on the weekend then. (lines 16–17). This question draws an inference from Selma’s mention of the five-days-per-week schedule – namely that this must preclude students from using laptops on the weekend (i.e., the two days per week the bookstore must be closed). Excerpt 8. (F15.RA.10:10:00) 01
SEL:
it’s great for students who ca- can’t
02
like afford one.
03
(0.2)
04
SEL:
05
BRI:
06
y’kno[w? [to:tally. (0.2)
07
BRI:
and it’s cool the school offers it for free:.
08
SEL:
yeah for sure.
09
BRI:
are they open every da:y too:?
10
(.)
253
254
Chase Wesley Raymond
11
SEL:
12
BRI:
fi[ve[so you can like- (.) take it to cla:ss=
13 14
=‘n stu:[ff? SEL:
[yeah,=they’re open five days a week.
15
(0.2)
16
BRI:-> oh but- (.) so then, (.) so then they
17
-> can’t use ‘em on the weekend then.
18
(.)
19
BRI:
20
SEL:=>
21
SEL:
22
er- (.) I gue[ss unless they [ha[Sí↓:↑: .h
[↑sí:,
El viernes se the friday REFL.3SG on Fridays you pue[de sacar y ] gua-= can.PRS.3SG take-out.INF and (keep-) can check (them) out and (keep-)
23
BRI:
[
el viernethe Friday on Frida-
24
SEL:
=ajá, =hasta el lunes. uh-huh until the Monday uh huh,=until Monday.
25
BRI:
that’s cool then,
]
Brian designs his question as a declarative request for confirmation – incorporating so then twice turn-initially (see Bolden 2009), as well as then again turnfinally, negative polarity with can’t, and falling intonation. Thus, he not only draws an inference here and inquires about it, but moreover shows himself grammatically to be strongly committed to the likelihood of that inference (Heritage 2012; Heritage & C. W. Raymond 2021). In addition, here the question also implicates an incipiently negative evaluation of the program (note but-) if in fact the laptops turn out not to be available to students on the weekend (C. W. Raymond & Heritage 2021). Following a brief silence, Brian begins revising the design of his question (line 19) which, although left incomplete, is on its way to a more epistemically cautious question format with I guess (Drew 2013, 2018). The questioner himself thereby orients to his prior questioning turn as too oversupposing in its construc-
Chapter 8. Answer possibility space of Spanish-English bilinguals
tion, and he begins to reconfigure this dimension of the turn, in addition to others, in pursuit of agreement (Sacks 1987 [1973]; Heritage & C. W. Raymond 2021). Before he can bring this alternative design to completion, however, Selma produces a language-non-concordant response to the initial version of the question: Sí↓:↑: .h ↑sí:, (line 20). This response particle sí, despite its affirmative semantics, is used twice here in a language-non-concordant environment to disconfirm the negative proposition formulated in the question – something along the lines of ‘yes, they can’ in English. With this answer format, Selma resists not just the inference that Brian has drawn, but the overassuming stance with which he offered it up as a hypothesis for confirmation. The agency of her language-nonconcordant answer is further demonstrated through the overlapped entry (Drew 2009; Jefferson 1984a; Sacks, Schegloff, and Jefferson 1974) and marked prosody (cf. Thompson, Fox & Couper-Kuhlen 2015: 172–3), as well as a repeat of the response particle itself. Selma continues on in Spanish to begin issuing a correction of Brian’s incorrect inferencing, thereby also blocking any potential negative assessment the question may have intimated, and Brian acquiesces to this shift in language in his own overlapped self-correction (line 21–23). Once this disconfirmation and correction are issued, Brian switches back to English to offer a positive assessment of the laptop program (line 25). Researchers investigating other bilingual speech communities have previously noted that dispreferred actions (of various sorts) are a frequent site for code-switches (e.g., Auer 1995; Milroy & Wei 1995; Wei 1994). Disconfirmation turns – in which the hypothesis presented by the question’s polarity turns out to have been incorrect – certainly fit within that umbrella, and thus the plurality of such instances in the present dataset is in parallel with these findings; however, there are a range of language-concordant disconfirmations in the present dataset as well (as seen in [6] above). Language non-concordance is recurrently implicated in those disconfirmations which specifically work to flag the additional elements of inappositeness in the question’s design, beyond simply issuing disconfirmation. So in (7), the answerer not only disconfirms and corrects the proposition in the questioner’s question, but also takes the stance that the questioner should have known better; she should have drawn upon what she had been told earlier to design a more appropriate request for confirmation here-and-now in the interaction. Likewise, in (8), Selma not only issues disconfirmation and correction, but with language non-concordance works to resist a question whose design asks her to confirm a hypothesis that undermines the quality of the program she’s been positively assessing. Selma’s answer is moreover done in overlap with the questioner’s own orientation to his question as inappropriately designed, demonstrated through his subsequent incipient transformation thereof. Thus, while undoubtedly sharing family resemblances with language-concordant dis-
255
256
Chase Wesley Raymond
confirmations due to their disconfirming of the question’s proposition, languagenon-concordant disconfirmations go beyond simply issuing disconfirmation, to enact an emphatic form of disagreement that agentively resists additional terms, constraints, and implications established by the question’s design. It is not only as part of turns issuing disconfirmation that answerers make use of language non-concordance; in confirmations, too, language-non-concordant response particles routinely serve to resist the terms and constraints depicted in the question’s design. The most recurrent context in which this can be observed in the present dataset is in questions that challenge the epistemic authority of the recipient. In these cases, speakers assert a proposition wholly from their knowledge domain – e.g., by making an announcement, or telling an event from a story – and in response, recipients produce a polar question that enacts a disbelieving stance toward that prior claim. Whether this incredulous stance is a ‘genuine’ one, or whether it is more ritualistic and affiliative in its function (e.g., to index surprise, newsworthiness), the question nonetheless disputes the recipient’s rights to make the claim they have just made. By breaking with language concordance in responding to such ‘challenges’, answerers do not acquiesce to the stance embodied by the question; instead, they go beyond simply affirming the proposition of the question to produce an emphatic form of agreement, reasserting authority over the territory of knowledge at play in the sequence. In so doing, answerers show themselves to be “‘hold[ing] a position’ against a query which [they] reflexively constitute, at least in a minimal way, as embodying inapposite doubt” (Heritage 1998: 297–299). Through such sequences, then, the participants collaboratively construct the surprise, intrigue, newsworthiness, etc. of the announcement or telling in progress, and provide for further expansion thereupon (see Heritage 1984a: 339–344; Jefferson 1981: 62–66; C. W. Raymond & Stivers 2016; Thompson, Fox & Couper-Kuhlen 2015: 80–82; Wilkinson & Kitzinger 2006). Consider Example (9), in which Alejandra enters the kitchen and finds her aunt there cooking. The aunt is using three of the stovetop’s four gas burners in preparing a meal, currently focusing on heating up tortillas on a comal (a smooth, flat griddle over a flame): She places and repeatedly flips the tortillas on the hot comal by hand, quickly so as to not burn herself. At the beginning of the extract, as Alejandra enters the room, she announces that she recently felt a chill while she was lying down (line 1, 4).9 Given how hot it is in the kitchen, and the 9. The semantics of ahora ‘now’, and its diminutivized version ahorita ‘now.DIM’, are intensely complex; for recent studies that include detailed bibliographies of prior work, see Granados (2021: 2–7) and Malaver (2017: 28–45). This colloquial English translation (‘just a bit ago’) captures its sense here.
Chapter 8. Answer possibility space of Spanish-English bilinguals
aunt’s current activity, the aunt immediately takes a disbelieving stance toward this claim with the repetition (‘’) (line 5), the prosody of which, combined with a dramatically furrowed brow, giving the turn a particularly incredulous hearing (cf. Selting 1996). Excerpt 9. (Cell.Cam.04:03:30) 01
ALE:
02
ahorita:: I was laying dow:n, now.DIM just a bit ago I was lying down (1.0) ((ALE puts bowl in sink))
03
AUN:
04
ALE:
05
AUN:-> [ cold
05.1 aun: 06
mm[::, [y me dio un frío, and DAT.1SG give.PFV.3SG a cold and I suddenly felt a chill
[((eyebrow furrow))
ALE:=>
[YE:S.=
07
=por eso ahorita que: (.) for that now.DIM that =that’s why right now when
08
la
oí, que vino? hear.PFV.1SG that come.PFV.2SG I heard you, that you came {in} ACC.2SG
09
AUN:
10
ALE:
11
12
no[:. elthe [y me puse un swea(hhh)s::w and REFL.1SG put.PFV.1SG a (sweater) and I put on a (sweater) pa’ s(h)ali[hhhh heh heh for leave.INF to (come out)
AUN:
[heh heh heh
While Alejandra has primary rights to assert her own prior experience with coldness, her aunt’s question in line 5 takes a stance that calls these rights into question. By issuing a language-non-concordant response particle YE:S. (line 6), with intensified prosody matching that of the questioning turn, Alejandra reasserts her primary authority to make this assertion; she thus ‘holds her position’ as to her prior expression of coldness, collaboratively constituting it as newsworthy. Moreover, she immediately goes on to cite further evidence in support of her claim –
257
258
Chase Wesley Raymond
namely that this was the reason she put on a sweatshirt before coming into the kitchen, which both participants laugh about in the context of the hot kitchen (lines 10–12). In cases like this, then, the questioner is oriented to not as initiating repair (recall Example [5]), but rather as expressing disbelief (and perhaps thereby surprise) with regard to what was just said, and with a language-nonconcordant particle response, the answerer is heard as emphatically ‘doubling down’ on her stance in the face of that incredulity, thereby aligning with the recipient’s surprised reaction. Extract (10) offers another confirmation example of this sort. Here, Román is talking to his friend Eve about a class he took and particularly enjoyed. As he explains, although students watched films as part of the class, the course was actually offered by the Political Science Department (lines 1–8). In lines 9–16, Román comments that the reason he enjoyed the class was due to the guest lectures by scholars from, for example, Harvard and MIT (Massachusetts Institute of Technology). Eve’s partial repeat in line 21 is treated as a simple initiation of repair through Román’s mm hm response (line 22) (Robinson 2013; cf. C. W. Raymond & Stivers 2016). Immediately following this, though, Eve pursues further expansion with: en una clase de polít- (‘in a class on polit-’) (line 23). It is in response to this pursuit that Román then issues a language-non-concordant £y(h)eah.£. (line 25). Excerpt 10. (F12.21:8:30) 01
EVE:
02
era una: clase de película, be.IPFV.3SG a class of film was it a film class hablan[do de qué. speak.GER of what speaking about what ROM:=>
25.1 rom:=> 26
ROM:
27
28
[£y(h)eah.£ [((eyebrow raise)) u::hm de:: como matemáticas of like mathematics uhm about like math y política y[: and politics and and politics and
EVE:
[O::h.
With her additional pursuit in line 23, Eve is heard to adopt a more incredulous stance: Román has already stated that this is a political science class (line 3), that the course included guest speakers from MIT (line 16), and then reaffirmed that there were indeed guest speakers from MIT (line 22). So when Eve yet again calls this information into question in the immediately subsequent line 23, and with a prepositional phrase format that is grammatically parasitic on Román’s talk, her question indexes not just incredulity, but persistent incredulity in the face of the prior sequence, and makes relevant further explanation to rectify this. Indeed, Eve’s abandonment of this question design suggests an orientation on her part to its continued expression of doubt as inappropriate, as she shifts instead to a more concessionary wh-question as the means to invite further elaboration (line 24) (recall also [8] above). Notwithstanding her abandonment of the polar question, though, Román produces a language-non-concordant £y(h)eah.£ in overlap, the infusion of a laughter particle and smile voice (in addition to the eyebrow raise) seeming to likewise orient to the disbelieving stance enacted by Eve’s question (see Jefferson 1984b). In this case, then, as Eve upgrades her incredulous stance over the course of the sequence, so too does Román upgrade the agency indexed through his confirmation answers by incorporating language non-concordance. Through these stance-takings, the interactants collaboratively construct these guest speakers as a surprising or otherwise remarkable aspect of the class. Consider one final case (11), which offers two additional language-nonconcordant confirmations. Here, Raquel is telling Berta about what happened the previous day at the hospital where the two work: One of their friends (Mercy) was called in by a supervisor (whom they all share in common) and accused of violating policy by discussing work-related matters outside of the workplace. While the telling has been going on for several turns, Raquel pauses in lines 1–2 to qual-
Chapter 8. Answer possibility space of Spanish-English bilinguals
ify her narration as second-hand, admitting that she only has access to Mercy’s side of the story, which Berta acknowledges (line 3). The direct reported speech climax of the telling (Holt 1996) then comes when, in the context of being interrogated by the supervisor (lines 4–7), Mercy issues a snappy retort that the supervisor herself has been involved in the very activities she’s attempting to sanction Mercy for (lines 8–9). Observe Berta’s use of a newsmark (line 11) and a request for confirmation (lines 15–16) as means of orienting to the climax and affiliating with Raquel’s stance toward it. Excerpt 11. (2013.Sum.03:04:30) 01
RAQ:
02
porque yo no más estoy oyendo because I no more be.PRS.1SG hear.GER because I’m only hearing la parte de Mercy. the part of M Mercy’s side
03
BER:
mm hm,
04
RAQ:
>(y luego)< pienso que le and later think.PRS.1SG that DAT.3SG and then I think that she ((supervisor))
05
preguntó like (.) Quién era. (.) ask.PFV.3SG who be.IPFV.3SG asked her ((Mercy)) like (.) who was it (.)
06
Quién es la que está who be.PRS.3SG ACC.3SG.F that be.PRS.3SG who is the one ((woman)) who is
07
hablando, y ella dice, talk-GER and she say.PRS.3SG talking, and she ((Mercy)) says
08
↑oh. .tch es que tú también oh be.PRS.3SG that you also oh .tch it’s that you too
09
estabas involucrada en esto. be.IPFV.2SG involved.F.SG in that were involved in that
10
(0.3) ((Berta takes a bite of food))
11
BER:-> [de veras?] ((with mouth full, eyes wide)) of truth really
12
RAQ:=> [p e r o -]=[YEa::h. but
12.1 raq:=>
[((single nod + eyebrow raise))
261
262
Chase Wesley Raymond
13
(.)
14
RAQ:
15
BER:-> =así le dijo like-that DAT.3SG say.PFV.3SG =that’s how she ((Mercy)) said it
16
17
a su c[ara? to her face to her ((supervisor’s)) face? RAQ:=>
17.1 raq:=> 16
y dijo, .hh= and say.PFV.3SG and she said .hh
RAQ:
[Yea:h. [((single nod)) y dice que ((four employee names)), and say.PRS.3SG that and she says that ((four employee names)),
17
que todas las cuatro son, that all the.F.PL four be.PRS.3PL that all four of them,
18
.hh s-son las que siempre be.PRS.3PL those.F that always .hh are the ones who always
19
están shop- shop talking. be.PRS.3PL are shop- shop talking
After taking a bite of food, but before starting to chew it, Berta first orients to the climax with a newsmark in line 11 de veras? (‘really?’), spoken through her just-filled mouth and accompanied by a widening of her eyes. This newsmark ‘marks’ the prior claim as newsworthy by calling into question its veracity (note the literal translation ‘of truth?’), and in this way, as in (9) and (10) above, the questioner adopts a ‘disbelieving’ stance toward something she has just been told. Raquel’s immediate, prosodically emphatic, language-non-concordant response particle – YEa::h. (line 12) – delivers a confirmation that effectively reasserts the claim she just made, with single nod and raised eyebrows reinforcing this stance non-vocally. That she cuts off her own incipient turn in Spanish (pero- ‘but’) to deliver this confirmation in English – i.e., with language non-concordance – is particularly noteworthy. Raquel then briefly returns to the telling in line 14, in Spanish, before Berta interdicts with así le dijo a su cara? (‘That’s how she said it to her face?’) (lines 15–16). This request for confirmation, like the newsmark in line 11, is strongly affiliative with the overall stance that Raquel is taking toward the events
Chapter 8. Answer possibility space of Spanish-English bilinguals
of her telling; but again it accomplishes this affiliation by ritualistically calling into question this part of Raquel’s report, and here doing so at a point where Raquel has clearly projected more talk. The language-discordant yea:h., offered in terminal overlap and again with a single nod, agentively issues emphatic reconfirmation of this proposition, thereby contributing to the co-construction of this reported speech event as a particularly newsworthy story climax. Whether questioners are issuing an ‘actual’ challenge, or indexing doubt in the service of more affiliative aims, answerers use the “otherness” (Auer 1995: 124) indexed by language-non-concordant response particles to resist the terms established by such questions, and are thereby heard to be emphatically ‘holding their position’ in the face of disbelief (ritualistic or otherwise). Especially in the case of confirmations, the agency of language non-concordance is routinely implicated in the co-construction of newsworthiness, where unmarked answer types (i.e., a language-concordant particle) would seem to be pragmatically insufficient – a point to which we will return in the Discussion.
4. Discussion and conclusions The aim of this chapter has been to explore the ‘answer possibility space’ (Stivers 2022) of a community of Spanish-English bilinguals. Close inspection of interactional data from this speech community reveals choice of language to be one of the terms established by a polar question’s design. Accordingly, languageconcordant response particles are found in contexts where answerers show themselves to accept the terms and constraints of the question as it was put forth, acquiescing to the questioner’s agency in designing and producing the question. By contrast, language-non-concordant response particles are produced in contexts where answerers take issue in some way with the terms established by the question’s design. Such answers make use of language non-concordance as a “contextualization cue” (Gumperz 1982) to locally resist the terms of the question and thereby issue an agentive, emphatic answer. In this way, it is the participants themselves who, through their very conduct in designing question-answer sequences, display their orientation to language choice and language (non-)concordance as resources for social action. This is thus one concrete means through which the members of this speech community effectively ‘“do being” bilingual’ (cf. Sacks 1984) with one another – through the shared-in-common, systematic deployment and interpretation of options from their questioning and answering repertoire.10 10. This sort of participant-orientation evidence therefore illustrates that ‘English’ and ‘Spanish’ are not wholly theoretical categories imposed by analysts – i.e., nothing more than “ideal-
263
264
Chase Wesley Raymond
How might the present study’s findings about a bilingual community of practice be situated within the broader literature on (monolingual) polar questionanswer sequences in conversation? Because polar question-answer sequences – and the stances enacted in and through their design – are “accomplice to” (Heritage 1984a: 299; Jefferson 1984c: 216) a diverse array of local interactional aims and actions, one way forward is to compare some of these action environments cross-linguistically. For example, in many of the same environments as Spanish-English bilinguals produce language-non-concordant particle answers, monolingual speakers of other languages likewise deploy designedly agentive and emphatic formats from within their answering repertoire. In monolingual English, for instance, ritualized displays of disbelief and newsmarks are often responded to with answer formats that are epistemically agentive and go beyond just answering the question (e.g., Heritage 1998; Heritage & G. Raymond 2012; C. W. Raymond & Stivers 2016; G. Raymond 2003; Stivers 2011, 2022). In his analysis of oh-prefacing, for instance, Heritage (1998) summarizes one subset of cases as those in which “the item which is queried – if only…in a minimal, ‘newsmarking’ fashion – has already been clearly asserted or presupposed”; in these cases, “the oh-prefaced confirmation achieves a reassertive quality and thereby ‘holds a position’ against a query which it reflexively constitutes, at least in a minimal way, as embodying inapposite doubt” (299). A brief illustration is found in (12): Excerpt 12. (NB:II:2.R) 01
Emm:
Are you th:e ol:dest one'n the cla:ss?
02
Nan:
°Oh: w- by fa:r.°
03
Emm:-> ↑Are yih reall[y¿↑
04
Nan:=>
[°Oh: ya:h.°
While monolingual English speakers often produce oh-prefaced answers in this action environment, in the bilingual Spanish-English interactions analyzed here, speakers frequently mobilize language-non-concordance in such contexts. In both the monolingual and bilingual answer possibility spaces, then, answerers are seen to issue agentive answer formats in response to such ‘incredulous’ questions. Consider also that, in monolingual English, a negatively formatted question like so then they can’t use them on the weekends (example (3/8)) typically cannot
ized and ‘ideologized’ abstraction[s] with a name” (Gafaranga 2005: 290; see also Auer 1984: 28; C. W. Raymond 2018: 166–167). Rather it is the participants themselves whose systematic production of particular turn designs in particular sequential contexts reveals their own interpretations of the notions of language concordance and non-concordance, and therefore, by extension, of languages as constructs.
Chapter 8. Answer possibility space of Spanish-English bilinguals
be disconfirmed only with yes, but rather requires the use of a repetition (e.g., yes, they can) or some other assertive practice (e.g., yeah huh!) (Couper-Kuhlen & Selting 2018: 247–248; and on negative polarity in question design, see Heritage & C. W. Raymond 2021; C. W. Raymond & Heritage 2021). German is similar in that it uses a designated particle doch to disconfirm negatively polarized interrogatives and B-event statements (Couper-Kuhlen & Selting 2018: 248; Heinemann 2009; Gubina 2021, p.c.), which we saw accomplished with a language-non-concordant particle in (8). But there is further functional overlap between doch and languagenon-concordant response particles as described here: Similar to the languagenon-concordant confirmations we saw in (9)–(11), doch is routinely found in response to ritualized displays of disbelief and newsmarks, such as Ist nicht wahr! (lit. ‘Is not true!’) and Nicht dein Ernst! (lit. ‘You’re not serious’). Moreover, Gubina (2021) finds that doch can in fact also be used in response to turns that are not overtly grammatically negative, but nonetheless call into question or otherwise challenge speakers’ rights to claim what they’ve just claimed. It seems then that there is considerable overlap between what German speakers accomplish with a particular agentive particle, and what Spanish-English bilinguals accomplish with language non-concordance. While true cross-linguistic comparisons along these lines lie outside the scope of the present study, it is nonetheless noteworthy that many of the specific action environments in which bilingual answerers produce language non-concordance look – to borrow a phrase from Levinson (2006: 46) – so “very similar, in some cases eerily similar” to the environments in which monolingual systems likewise produce agentive answers from their repertoire. Interestingly, then, although the specific repertoire of answer options certainly differs across speech communities, there is at least some evidence that, within each speech community (bilinguals included), answerers regularly deploy more acquiescent vs. more agentive answer types in response to some of the same sorts of questions (see Enfield, et al. 2019; Heritage & G. Raymond 2012; C. W. Raymond et al. 2021; Stivers 2022). This suggests that participants cross-linguistically are using the indexation of (certain forms of ) agency to deal with similar, perhaps in some way basic, interactional needs. Future comparative work along these lines – in addition to comparative studies of the different agentive options within the answer possibility space of a given bilingual community of practice – will allow us to further refine our conceptualization of agency in such sequences, including especially how it is used and oriented to as a participants’ resource for action.
265
266
Chase Wesley Raymond
Finally, while there is evidence that the claims put forth here do extend to participants in other bilingual communities,11 the methodological perspective proposed here offers a means to discover the orderliness of such practices in each language community: Standardizing a collection of cases sequentially allows the options within the answer possibility space of each community of practice to be examined internally, on its own terms, thereby ensuring that our analyses of biand multilingual practices are always grounded in the orientations of those who use them.12
Acknowledgements Preliminary versions of the analysis offered here were presented at the NCA Annual Meeting, the Conference on Spanish in the United States, ICCA-18, the University of Colorado, Boulder, UCSB, and UCLA; I am grateful to the audiences on these various occasions for their comments and suggestions. My thanks to Galina Bolden, John Heritage, Marja-Leena Sorjonen, and Sandy Thompson for detailed feedback on previous written versions of this chapter. I am also indebted to Betty Couper-Kuhlen, Barbara Fox, Uwe Küttner, and especially Alexandra Gubina, for generously sharing their cross-linguistic expertise with me, which is reflected in Section 4. Any errors in this report are of course my own.
References Atkinson, J. Maxwell, and John Heritage (eds). 1984. Structures of Social Action. Cambridge: Cambridge University Press. Auer, Peter. 1984. Bilingual Conversation. Amsterdam: John Benjamins. Auer, Peter. 1995. “The Pragmatics of Code-Switching: A Sequential Approach.” In One Speaker, Two Languages: Cross-Disciplinary Perspectives on Code-Switching, ed. by Lesley Milroy, and Pieter Muysken, 115–135. Cambridge: Cambridge University Press. Auer, Peter. 2007. “The Monolingual Bias in Bilingualism Research, or: Why Bilingual Talk is (Still) a Challenge for Linguists.” In Bilingualism: A Social Approach, ed. by Monica S. Heller, 319–339. New York: Palgrave-Macmillan.
11. See Bolden (2018: 154–155) for a particularly interesting parallel case produced amongst Russian-English bilinguals in the U.S. See also C. W. Raymond (2014: 244–251), and references cited therein, for a sampling of additional candidate cases from the prior published literature on bilingual conversation involving other languages, including English and Greek, English and Hebrew, French and Kinyarwanda, and Italian and Sicilian. 12. This approach can also, importantly, reveal any participant-specific competence, preference, or usage issues that may be relevant to the analysis at hand (e.g., a speaker who only produces yeah, to the total exclusion of sí).
Chapter 8. Answer possibility space of Spanish-English bilinguals
Bolden, Galina B. 2009. “Implementing Incipient Actions: The Discourse Marker ‘so’ in English Conversation.” Journal of Pragmatics 41 (5): 974–998. Bolden, Galina B. 2018. “Speaking ‘Out of Turn’: Epistemics in Action in Other-Initiated Repair.” Discourse Studies 20 (1): 142–162. Bolinger, Dwight. 1978. “Yes-No Questions are Not Alternative Questions.” In Questions, ed. by Henry Hiż, 87–105. Dordrecht: Reidel. Cashman, Holly R. 2005. “Identities at Play: Language Preference and Group Membership in Bilingual Talk in Interaction.” Journal of Pragmatics 37 (3): 301–315. Cashman, Holly R. 2008. “Conversation and Interactional Analysis.” The Blackwell Guide to Research Methods in Bilingualism and Multilingualism, ed. by Li Wei and Melissa G. Moyer, 275–295. Malden, MA: Blackwell. Cashman, Holly R., and Chase Wesley Raymond. 2014. “Doing Gender in Soccer (Football) and Beyond: Making Gender Relevant in Spanish-Language Broadcast Discourse.” Gender & Language 8 (3): 311–340. Clayman, Steven E., and Virginia Teas Gill. 2004. “Conversation Analysis.” In Handbook of Data Analysis, ed. by Alan Byman and Melissa Hardy, 589–606. Beverly Hills, CA: Sage. Clift, Rebecca. 2016. Conversation Analysis. Cambridge: Cambridge University Press. Clift, Rebecca, and Chase Wesley Raymond. 2018. “Actions in Practice: On Details in Collections.” Discourse Studies 20 (1): 90–119. Couper-Kuhlen, Elizabeth, and Margret Selting. 2018. Interactional Linguistics. Cambridge: Cambridge University Press. Drew, Paul. 2009. “‘Quit Talking While I’m Interrupting:’ (A Comparison between) Positions of Overlap Onset in Conversation.” In Talk in Interaction: Comparative Dimensions, ed. by Markku Haakana, Minna Laakso, and Jan Lindström, 70–93. Helsinki: Finnish Literature Society. Drew, Paul. 2013. “Turn Design.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell and Tanya Stivers, 131–49. Malden, MA: Wiley-Blackwell. Drew, Paul. 2018. “Epistemics in Social Interaction.” Discourse Studies 20 (1): 163–87. Enfield, N. J., Stivers, T., Brown, P., Englert, C., Harjunpää, K., Hayashi, M., Heinemann, T., Hoymann, G., Keisanen, T., Rauniomaa, M., Raymond, C. W., Rossano, F., Yoon, K.-E., Zwitserlood, I., & Levinson, S. C. 2019. “Polar answers.” Journal of Linguistics 55 (2): 277–304. Gafaranga, Joseph. 2005. “Demythologising Language Alternation Studies: Conversational Structure vs. Social Structure in Bilingual Interaction.” Journal of Pragmatics 37: 281–300. Gafaranga, Joseph, and Maria-Carme Torras. 2002. “Interactional Otherness: Towards a Redefinition of Codeswitching.” International Journal of Bilingualism 6 (1): 1–22. Granados, Daniel. 2021. “Atenuación y expansión de ahorita en la Ciudad de México, un cambio analizado desde el tiempo aparente [Reduction and expansion of ahorita ‘now.DIM’ in Mexico City, an apparent-time study].” Cuadernos de Lingüística de El Colegio de México 8: e202. Grosjean, François. 1998. “Studying bilinguals: Methodological and conceptual issues.” Bilingualism: Language and Cognition 1: 131–49. Gubina, Alexandra. 2021. “Functions and operational levels of the response particle ‘doch’ in spoken German.” Talk given at Arbeitstagung zur Gesprächsforschung, March 24–26.
267
268
Chase Wesley Raymond
Gumperz, John J. 1982. Discourse Strategies. Cambridge: Cambridge University Press. Hakulinen, Auli. 2001. “Minimal and Non-minimal Answers to Yes-no Questions.” Pragmatics 11: 1–16. Hänggi, Philipp. 2022. “Language Choice and the Multilingual Soundscape: Overhearing as a Resource for Recipient-Design in Impromptu First-Time Encounters.” Research on Language and Social Interaction 55 (4): 299–325. Hayano, Kaoru. 2013. “Question Design in Conversation.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell and Tanya Stivers, 395–414. Malden, MA: Wiley-Blackwell. Heinemann, Trine. 2009. “Two Answers to Inapposite Inquiries.” In Conversation Analysis: Comparative Perspectives, edited by Jack Sidnell, 159–86. Cambridge: Cambridge University Press. Heller, Monica S. 1982. “Negotiations of Language Choice in Montreal.” In Language and Social Identity, edited by John J. Gumperz, 108–18. New York: Cambridge University Press. Heritage, John. 1984a. “A Change-of-State Token and Aspects of Its Sequential Placement.” In Structures of Social Action, ed. by J. Maxwell Atkinson and John Heritage, 299–345. Cambridge: Cambridge University Press. Heritage, John. 1984b. Garfinkel and Ethnomethodology. Cambridge: Polity Press. Heritage, John. 1998. “Oh-Prefaced Responses to Inquiry.” Language in Society 27 (3): 291–334. Heritage, John. 2010. “Questioning in Medicine.” In “Why Do You Ask?”: The Function of Questions in Institutional Discourse, ed. by Alice F. Freed and Susan Ehrlich, 42–68. New York: Oxford University Press. Heritage, John. 2012. “Epistemics in Action: Action Formation and Territories of Knowledge.” Research on Language and Social Interaction 45 (1): 1–29. Heritage, John, and Chase Wesley Raymond. 2021. “Preference and Polarity: Epistemic Stance in Question Design.” Research on Language and Social Interaction 54 (1): 39–59. Heritage, John, and Geoffrey Raymond. 2012. “Navigating Epistemic Landscapes: Acquiescence, Agency and Resistance in Responses to Polar Questions.” In Questions: Formal, Functional and Interactional Perspectives, ed. by J. P. De Ruiter, 179–92. Cambridge: Cambridge University Press. Holt, Elizabeth. 1996. “Reporting on talk: The use of direct reported speech in conversation.” Research on Language and Social Interaction 29 (3): 219–245. Jefferson, Gail. 1981. “The abominable ‘ne?’: A working paper exploring the phenomenon of post-response pursuit of response.” Occasional Paper No. 6, Sociology, University of Manchester. Jefferson, Gail. 1984a. “Notes on Some Orderlinesses of Overlap Onset.” In Discourse Analysis and Natural Rhetoric, ed. by V. D’Urso and P. Leonardi, 11–38. Padua: Cleup Editore. Jefferson, Gail. 1984b. “On Stepwise Transition From Talk About a Trouble to Inappropriately Next-Positioned Matters.” In Structures of Social Action, edited by J. Maxwell Atkinson and John Heritage, 191–221. Cambridge: Cambridge University Press. Jefferson, Gail. 1984c. “On the Organization of Laughter in Talk About Troubles.” In Structures of Social Action, ed. by J. Maxwell Atkinson and John Heritage, 346–69. Cambridge: Cambridge University Press.
Chapter 8. Answer possibility space of Spanish-English bilinguals
Jefferson, Gail. 2004. “Glossary of Transcript Symbols with an Introduction.” In Conversation Analysis: Studies from the First Generation, ed. by G. H. Lerner, 13–31. Amsterdam: John Benjamins. Keevallik, Leelo. 2010. “Minimal Answers to Yes/No Questions in the Service of Sequence Organization.” Discourse Studies 12 (3):283–309. Koivisto, Aino. 2013. “On the Preference for Remembering: Acknowledging an Answer with Finnish ai nii(n) (“Oh that’s right”).” Research on Language and Social Interaction, 46: 277–297. Küttner, Uwe-A. 2018. “Investigating inferences in sequences of action: The case of claiming ‘just-now’ recollection with Oh that’s right.” Open Linguistics 4: 101–126. Küttner, Uwe-A., and Chase Wesley Raymond. 2022. “I was gonna say…: On the Doubly Reflexive Character of a Meta-Communicative Practice.” In Sprachreflexive Praktiken: Empirische Perspektiven auf Metakommunikation, ed. by Florian Busch, Pepe Droste, and Elisa Wessels, 51–73. Springer. Levinson, Stephen C. 2006. “On the Human ‘Interactional Engine.’ ” In Roots of Human Sociality: Cognition, Culture, and Interaction, ed. by N. J. Enfield and Stephen C. Levinson, 39–69. London: Berg. Malaver, Irania. 2017. “Ahorita: Lexicalización y cambio lingüístico en la comunidad de habla caraqueña [Ahorita (now.DIM): Lexicalization and Lingusitic Change in the Speech Community of Caracas].” Nueva Revista de Filología Hispánica LXV (1): 27–57. Milroy, Lesley, and Li Wei. 1995. “A Social Network Approach to Code-Switching: The Example of a Bilingual Community in Britain.” In One Speaker, Two Languages: CrossDisciplinary Perspectives on Code-Switching, ed. by P. Muysken and L. Milroy, 136–57. Cambridge: Cambridge University Press. Mondada, Lorenza. 2018. “Greetings as a Device to Find Out and Establish the Language of Service Encounters in Multilingual Settings.” Journal of Pragmatics 126: 10–28. Park, Innhwa. 2010. “Marking an Impasse: The Use of anyway as a Sequence-Closing Device.” Journal of Pragmatics 42: 3283–3299. Pomerantz, Anita M., and John Heritage. 2013. “Preference.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell and Tanya Stivers, 210–28. Malden, MA: Wiley-Blackwell. Raymond, Chase Wesley. 2014. On the Sequential Negotiation of Identity in Spanish-Language Discourse: Mobilizing Linguistic Resources in the Service of Social Action. Ph.D. dissertation, University of California, Los Angeles. Raymond, Chase Wesley. 2015. “Questions and Responses in Spanish Monolingual and Spanish-English Bilingual Conversation.” Language & Communication 42: 50–68. Raymond, Chase Wesley. 2017. “Indexing a Contrast: The Do-construction in English Conversation.” Journal of Pragmatics 118: 22–37. Raymond, Chase Wesley. 2018. “On the Relevance and Accountability of Dialect: Conversation Analysis and Contact Linguistics.” Journal of Sociolinguistics 22 (2): 161–89. Raymond, Chase Wesley. 2019. “Intersubjectivity, Normativity, and Grammar.” Social Psychology Quarterly 82 (2): 182–204. Raymond, Chase Wesley. 2020. “Negotiating Language on the Radio in Los Angeles.” In Spanish in the Global City, edited by Andrew Lynch, 406–29. New York: Routledge. Raymond, Chase Wesley. 2022. “Suffixation and Sequentiality: Notes on the Study of Morphology in Interaction.” Interactional Linguistics 2 (1): 1–41.
269
270
Chase Wesley Raymond
Raymond, Chase Wesley, Holly R. Cashman. 2021. “Institutional Roles as Interactional Achievements: The Epistemics of Sports Commentary.” In Contexts of Co-Constructed Discourse: Interaction, Pragmatics, and Second Language Applications (Festschrift in Honor of Dale Koike, ed. by Lori Czerwionka, Rachel Showstack, and Judith Liskin-Gasparro, 23–45. New York: Routledge. Raymond, Chase Wesley, Rebecca Clift, and John Heritage. 2021. “Reference without Anaphora: On Agency through Grammar.” Linguistics: An Interdisciplinary Journal of the Language Sciences 59 (3): 715–55. Raymond, Chase Wesley, and John Heritage. 2021. “Probability and Valence: Two Preferences in the Design of Polar Questions and Their Management.” Research on Language and Social Interaction 54 (1): 60–79. Raymond, Chase Wesley, and Luis Manuel Olguín. 2022. Análisis de La Conversación: Fundamentos, Metodología y Alcances. New York: Routledge. Raymond, Chase Wesley, Jeffrey D. Robinson, Barbara A. Fox, Sandra A. Thompson, and Kristella Montiegl. 2021. “Modulating Action through Minimization: Syntax in the Service of Offering and Requesting.” Language in Society 50: 53–91. Raymond, Chase Wesley, and Tanya Stivers. 2016. “The Omnirelevance of Accountability: OffRecord Account Solicitations.” In Accountability in Social Interaction, ed. by Jeffrey D. Robinson, 321–53. Oxford: Oxford University Press. Raymond, Chase Wesley, and Anne Elizabeth Clark White. 2022. “On the Recognitionality of References to Time in Social Interaction.” Language & Communication 83: 1–15. Raymond, Geoffrey. 2003. “Grammar and social organization: Yes/no interrogatives and the structure of responding.” American Sociological Review 68: 939–967. Robinson, Jeffrey D. 2013. “Epistemics, Action Formation, and Other-Initiation of Repair: The Case of Partial Questioning Repeats.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 261–92. Cambridge: Cambridge University Press. Robinson, Jeffrey D. 2020. One type of polar, information-seeking question and its stance of probability: Implications for the preference for agreement. Research on Language and Social Interaction, 53(4):425–442. Sacks, Harvey. 1984. “Notes on Methodology.” In Structures of Social Action, edited by J. M. Atkinson and J. Heritage, 21–27. Cambridge: Cambridge University Press. Sacks, Harvey. 1987 [1973]. “On the Preferences for Agreement and Contiguity in Sequences in Conversation.” In Talk and Social Organisation, ed. by Graham Button and John R. E. Lee, 54–69. Clevedon, England: Multilingual Matters. Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson. 1974. “A Simplest Systematics for the Organization of Turn-Taking for Conversation.” Language 50 (4): 696–735. Schegloff, Emanuel A. 1993. “Reflections on Quantification in the Study of Conversation.” Research on Language and Social Interaction 26: 99–128. Schegloff, Emanuel A. 1996. “Confirming Allusions: Toward an Empirical Account of Action.” American Journal of Sociology 102 (1): 161–216. Schegloff, Emanuel A. 2007. Sequence Organization in Interaction: A Primer in Conversation Analysis, Volume 1. Cambridge: Cambridge University Press. Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks. 1977. “The Preference for SelfCorrection in the Organization of Repair in Conversation.” Language 53: 361–82.
Chapter 8. Answer possibility space of Spanish-English bilinguals
Selting, Margret. 1996. “Prosody as an Activity-Type Distinctive Cue in Conversation: The Case of So-Called ‘A stonished’ Questions in Repair Initiation.” In Prosody in Conversation, ed. by E. Couper-Kuhlen and M. Selting, 231–70. Cambridge: Cambridge University Press. Seuren, Lucas M., Mike Huiskes, and Tom Koole. 2016. “Remembering and Understanding with Oh-Prefaced Yes/No Declaratives in Dutch.” Journal of Pragmatics 104: 180–92. Sidnell, Jack, and Tanya Stivers. 2013. The Handbook of Conversation Analysis. Malden, MA: Wiley-Blackwell. Sorjonen, Marja-Leena. 2001. “Simple Answers to Polar Questions: The Case of Finnish.” In Studies in Interactional Linguistics, edited by Margret Selting and Elizabeth Couper-Kuhlen, 405–31. Amsterdam: John Benjamins. Stivers, Tanya. 2010. “An Overview of the Question-Response System in American English.” Journal of Pragmatics 42 (10): 2772–2781. Stivers, Tanya. 2011. “Morality and Question Design: ‘Of Course’ as Contesting a Presupposition of Askability.” In The Morality of Knowledge in Conversation, edited by Tanya Stivers, Lorenza Mondada, and Jakob Steensig, 82–106. Cambridge: Cambridge University Press. Stivers, Tanya. 2022. The Book of Answers: Alignment, Autonomy, and Affiliation in Social Interaction. Oxford: Oxford University Press. Stivers, Tanya, and N. J. Enfield. 2010. “A Coding Scheme for Question-Response Sequences in Conversation.” Journal of Pragmatics 42 (10): 2620–26. Stivers, Tanya, Nicholas J. Enfield, and Stephen C. Levinson. 2010. “Question-Response Sequences in Conversation across Ten Languages.” Journal of Pragmatics (Special Issue: October 2010) 42: 2615–2860. Stivers, Tanya, and Makoto Hayashi. 2010. “Transformative Answers: One Way to Resist a Question’s Constraints.” Language in Society 39: 1–25. Thompson, Sandra A., Barbara A. Fox, and Elizabeth Couper-Kuhlen. 2015. Grammar in Everyday Talk: Building Responsive Actions. Cambridge: Cambridge University Press. Wei, Li. 1994. Three Generations, Two Languages, One Family: Language Choice and Language Shift in a Chinese Community in Britain. Clevedon, England: Multilingual Matters. Wei, Li. 1998. “The ‘Why’ and ‘How’ Questions in the Analysis of Conversational CodeSwitching.” In Code-Switching in Conversation: Language, Interaction and Identity, ed. by Peter Auer, 156–76. New York: Taylor & Francis Group. Wei, Li. 2002. “‘What Do You Want Me to Say?’ On the Conversation Analysis Approach to Bilingual Interaction.” Language in Society 31 (2): 159–80. Wei, Li, and Lesley Milroy. 1995. “Conversational Code-Switching in a Chinese Community in Britain: A Sequential Analysis.” Journal of Pragmatics 23: 281–99. Wilkinson, Sue, and Celia Kitzinger. 2006. “Surprise as an Interactional Achievement: Reaction Tokens in Conversation.” Social Psychology Quarterly 69: 150–82.
271
chapter 9
Post-confirmation modifications Trajectories of un-initiated responses to polar questions in Japanese Kaoru Hayano & Makoto Hayashi
Japan Women’s University | Nagoya University
It has long been acknowledged that, in response to polar questions, disagreements are normatively delayed with agreements such that they are presented as modifications or exceptions rather than outright disagreements. While there is a large body of literature on how the initial element in the answer projects upcoming modifications or disagreements to come, little attention has been paid to the grammatical format in which delayed modification is presented. This study thus attempts to document how grammatical formats contribute to the interactional outcomes of modifications in this position. The focus is on three grammatical formats that are used to make such modifications following the minimal confirmation particle un in Japanese: the coordinate clause marked with demo (“but”), the subordinate clause, and clauses that are not grammatically tied to the preceding confirmation. The findings suggest that these grammatical formats are used systematically to modify the preceding confirmation to varying degrees and in different ways. Modifications made in coordinate clauses modify the degree of assessment presented in the question and/or preemptively deny a possible implication that follows the state of affairs confirmed by un without retracting or conditioning the confirmation. Modifications made in subordinate clauses, on the other hand, retract or significantly qualify the confirmation that has been conveyed with un. Finally, modifications in clauses that are not grammatically marked for their semantic relation to the confirmation address an issue with an assumption underlying the question and retroactively portray the confirmation as ostensible. It is concluded that post-confirmation modification is one environment in which a fine division of labor is assigned to different grammatical formats. Keywords: conversation analysis, “yes but” answers, Japanese, un, responses to polar question, confirmation, modification, grammar
https://doi.org/10.1075/slsi.35.09hay © 2023 John Benjamins Publishing Company
Chapter 9. Post-confirmation modifications in response to polar questions
1.
Introduction
Sacks (1987, 1992) revealed an organization that operates in sequences of polar questions and answers, adopting as his starting point the observation that in everyday conversation, yes, or an answer that agrees with the expectation expressed by the questioner, is produced far more often than no. Sacks then provided evidence verifying that this outcome is achieved through the interactional work performed by questioners and answerers, both of whom are oriented toward the preference for agreement (yes) over disagreement (no). One manifestation of interactants’ orientation toward preference for agreement is the time taken to produce an answer: While agreements are produced contiguously with the question, disagreements are ordinarily delayed (Pomerantz 1984; Sacks 1987; Stivers et al. 2009). The most obvious means of delaying a disagreement may be to wait before beginning to respond, leaving a gap between the question and answer, but there are various other ways to do so. As Sacks (1987: 59) writes, “[I]nsofar as disagreements are pushed into the back, then there is a variety of things that go in front of them, that then can get treated as ‘going in front of disagreements’, and that may have an import in signaling the future forthcomingness of a disagreement.” Accordingly, in analyzing the “variety of things” that precede disagreement, conversation analysts find each of them signaling and foreshadowing the different ways in which what follows may be dispreferred or may address some issue with the question (Bolden 2018; Hayashi and Hayano 2018; Heritage 1998, 2015; Kim 2013, 2018; Morita and Takagi 2018; Raymond 2018; Weidner 2018, among others).1 The present chapter aims to contribute to this line of research by focusing on answers to polar questions in which what Schegloff (2007) called a pro forma agreement is followed by different forms of modification. Among the “variety of things” that can precede disagreement, pro forma agreements form a special class because they set up a particular environment by reference to which the subsequent disagreement is heard. That is, when disconfirmation is delayed by a pro forma agreement, “there will be a position in which to put in elements that could otherwise appear as disagreements, which can, then, be cast in the agreement mode as being modifications or exceptions” (Sacks 1987: 62). Excerpt (1) is an example that Sacks discussed to make this point. Although B virtually disconfirms A’s question, B starts with a pro forma agreement (“Yuh it’s
1. See also Kendrick and Torreira (2015). They study the timing of responses to requests, offers, invitations, proposals and suggestions to show that different objects that come between an FPP and its response (i.e., gap, inbreath, turn-initial particles, etc.) are associated with a dispreferred SPP to different degrees.
273
274
Kaoru Hayano & Makoto Hayashi
permanent”), and disconfirmation is incorporated into or embedded within the “agreement mode.” Excerpt 1. (Sacks 1987: 62) 1 2 3
A: B:
So is this permanent? Yuh it’s permanent; permanent until I get moved again.
Schegloff (2007: 70) uses the expression “yes, but…” format as an umbrella term for answers of this nature. However, “but”-clauses are by no means the only format used to present disconfirmations or modifications in this position. In the above case, B does not use the contrastive conjunction “but,” instead qualifying the word “permanent” with the subordinate, adverbial clause initiated by “until” to specify the time period during which the proposition is true. In other cases, modifications and exceptions are produced without any markers that indicate their syntactic relations to what precedes them. To the best of our knowledge, no systematic study has yet investigated interactional outcomes of these syntactic variations in the ways in which modifications and exceptions are presented. In this study, then, we examine how modifications and exceptions following pro forma agreements are grammatically packaged. Based on the analysis of cases drawn from Japanese ordinary conversation, we demonstrate how the grammatical packaging is consequential for the interactional work achieved by the modifications and exceptions.
2.
Responses to polar questions in Japanese
A notable feature of responses to polar questions in Japanese, and one that is particularly relevant to this discussion, is that Japanese polar answers deploy the socalled “agree-disagree system” (Sadock and Zwicky 1985); that is, in Japanese, an affirmative response particle (e.g., un (‘yes’)) is used to convey the respondent’s agreement with the proposition presented by the prior question, regardless of the polarity of that proposition. A negative particle (e.g., iya (‘no’)) on the other hand is used to convey the respondent’s disagreement with the prior question’s proposition, again regardless of the proposition’s polarity. Consider the following constructed exchange: [Constructed] A:
kinoo ame huranakatta? yesterday rain did.not.fall Did it not rain yesterday?
B-1:
un huranakatta. yes did.not.fall Yes [=Yes, I agree with and confirm the proposition you presented], it did not rain.
Chapter 9. Post-confirmation modifications in response to polar questions
B-2:
iya hutta. no fell No [=No, I disagree with and disconfirm the proposition you presented], it rained.
Thus, an affirmative particle is used to construct a confirming answer, irrespective of the polarity of the proposition presented by the prior question. Of the various affirmative particles available in Japanese, un is the most commonly used (Hayashi 2010). Given that “type-conforming answers” (i.e., particletype answers) are preferred over “non-conforming answers” (e.g., repetition-type answers) across languages including Japanese (Enfield et al. 2019; Raymond 2003), and that confirming answers are preferred over disconfirming answers (Sacks 1987; Stivers 2010; Hayashi 2010, among others), we may then conclude that un is the most preferred (i.e., the least pragmatically marked) answer to polar questions in Japanese. Now, while un can occur as a stand-alone token that occupies an entire answer turn, it is often followed by further components within the same answer turn. While several studies have examined additional turn components within answer turns and explored how they work vis-à-vis the initial component (e.g., Ford, Fox, and Hellermann 2004; Hakulinen 2001; Raymond 2003; Seuren and Huiskes 2017; Sorjonen 2001; Steensig and Heinemann 2013), little attention has been paid to the interactional consequences of grammatical variations observed in the ways in which modifications and exceptions are presented. It is from this perspective that the present study investigates answers to polar questions in Japanese that consist of un followed by further components that in some ways modify the confirmation provided by the initial un. Through this examination, we show how different grammatical formats employed in the further components work to achieve different interactional ends in modifying the confirmation provided by un.
3.
Data
The data for our study were derived from 36 hours of video-recorded, face-toface conversations and nine hours of audio-recorded telephone conversations among native speakers of Japanese. These recordings include eight hours of telephone data from the CallFriend and CallHome corpora available at TalkBank ⟨www.talkbank.org⟩ (MacWhinney and Wagner 2010), six hours of face-to-face data from the Corpus of Everyday Japanese Conversation (Koiso et al. 2018), and 31 hours of data from the authors’ personal collections.
275
276
Kaoru Hayano & Makoto Hayashi
The present study is based on 30 cases of answers to polar questions in Japanese that consist of un (with varying degrees of elongation and prosodic coloring) as the first component, followed by further turn components that in some ways modify the initial confirmation with un. From our data, we have identified three types of turn-constructional formats that were recurrently observed in these answers: (i) un + demo (‘but’) modification (e.g.) un demo koko hodo samukunai un but here extent not.cold Un, but it’s not as cold {over there} as it is here.
(ii) un + subordinate clause (e.g., kedo ‘though’-marked clause) modification (e.g.) un soko made wa hanashi kikanakatta kedo un that extent PRT story didn’t.hear though Un, though {I} didn’t hear about that much detail.
(iii) un + modification (e.g.) un docchi demo ii un either even good Un, either is fine.
We will address each type of answer turn in the next section and explore the relationships between the grammatical formats employed in the components following the initial un and the interactional work that is accomplished in modifying the confirmation provided by un. Before proceeding to the data analysis, it is important to note two points about un. The first concerns the spelling used herein to represent the particle. While the particle is often phonetically produced as nn or mm, it is a common practice to orthographically render these phonetic variants as un, and we adopt the latter to represent the token in our texts and English translations of the utterances in the transcripts. The second concerns the prosody associated with un. Un, as a confirmation particle, is typically produced with a pitch fall in its intonation contour, as seen in Figure 1. When it is elongated and produced with relatively flat intonation, it becomes increasingly indistinguishable from the hesitation display u::n, which is similar to uh::m in English. Since u::n as an expression of hesitation is typically produced with flat intonation (see Figure 2, for example), we made an operational decision to treat the presence of a pitch fall in the intonation contour as a defining feature of un’s use as a confirmation particle. That is, no matter how elongated un may be, as long as it is produced with a recognizable pitch fall (see Figures 3 and 4, for example), it is hearable and thus is treated as a confirmation particle, though with an indication of hesitation conveyed via its delivery. Let us now proceed to data analysis.
Chapter 9. Post-confirmation modifications in response to polar questions
Figure 1. un as a confirmation token
Figure 2. u::n as a hesitation token
Figure 3. u:n. in line 9 of Excerpt (4)
277
278
Kaoru Hayano & Makoto Hayashi
Figure 4. u::↓n in line 19 of Excerpt (7)
4.
Analysis
4.1 un + demo (‘but’)-clause modification2 First, let us consider the type of post-confirmation modification that takes the format of ‘but’-clauses – the format that is treated as representative of the class (Schegloff 2007: 70). In Japanese, the contrastive conjunction that is most commonly used in casual conversation is demo. When this format is used, the confirmation that un conveys is neither retracted nor conditioned by the ensuing components. Rather, the ensuing components modify the degree of assessment presented for confirmation or disconfirmation in the question and/or preemptively deny a possible implication that follows the state of affairs confirmed by un. We may observe both of these features in Excerpt (2). Hana (han), Eri (eri) and Ami (ami) are chatting over tea. They all live in a town in the Midwestern United States, and prior to the excerpt, Hana and Eri, who have lived in the area for over a year, have warned Ami, who has not yet experienced the winter there, that it gets very cold in the winter. They then briefly talk about how cold it gets in Kanazawa, Ami’s hometown in Japan. It is after the discussion of Kanazawa has wound down that Hana mentions Kyoto, where Eri is from, and asks her a polar question about the winter climate there: >anata< ↑kyoo:to samui n janai no? (‘You, isn’t it cold in Kyoto?’). After a delay of 1.2 seconds before Eri’s response, Ami and Hana pursue it by building onto the question with the impressions they have about the climate in Kyoto (lines 3–7). Eri’s answer (line 8) starts with un but then is followed by a demo (‘but’)-clause in which Eri adds that it is not as cold as it is here, i.e., in the town in the Midwestern U.S. 2. For a comparable analysis of confirmation answers followed by “but”-initiated clauses in Brazilian Portuguese, see Harjunpää and Ostermann (this volume).
Chapter 9. Post-confirmation modifications in response to polar questions
Excerpt 2. (DEM10: 7) 01 HAN:
[>anata< kyoo:to samui n janai no? you Kyoto cold NOM TAG PRT You, isn’t it cold in Kyoto?
02
(1.2)
03 AMI:
>kyooto< samu soo. Kyoto cold seem Kyoto sounds cold.
04 HAN:
atsukute samui, hot.and cold {Kyoto} is hot {in the summer} and then cold {in the winter}?
05 AMI:
atsukute samui. hot.and cold {Kyoto} is hot {in the summer} and then cold {in the winter}.
06 HAN:
ne::= Yeah=
07 AMI:
=un =Yeah
08 ERI:=>
un::↓: >demo< wa samuku nai. un but here as.much PRT cold not Un but not as cold as it is here.
09 AMI:
h h h h
10 ERI:
ko[ko wa [kaze ga tsumetai kara:, here PRT wind SBJ cold so Here, the wind is cold.
11 HAN:
12 AMI:
[n:::n. hmmm [koko wa here PRT Here…
Given that the question nominates Kyoto as the next place to talk about in their discussion of places with notable winter climate, the question putatively suggests that Hana suspects Kyoto to be not just a little cold but remarkably cold. It is this intensity of assessment implied in the question that Eri’s un-initiated question modifies: while it confirms with un that it is cold in Kyoto, it subsequently specifies that it is not as cold as a simple, unmodified confirmation would imply. The demo-clause does not deny that it is cold in Kyoto, but it resists juxtaposing the coldness of Kyoto and that of the Midwestern town as if they were comparable.
279
280
Kaoru Hayano & Makoto Hayashi
It should be noted that in line 8, un itself is not delivered as an unproblematic confirmation. Its production is substantially delayed (lines 2–7), and the elongation of un at the beginning of the turn conveys some hesitation. Thus, it is not that the demo-clause here contradicts or cancels an unqualified confirmation that a decisively produced token of un has conveyed. Rather, it accounts for the hesitation that has already been evident in the way that un was produced. Similarly, in Excerpt (3), the speculative assessment that the questioner presents for confirmation or disconfirmation receives an un-initiated answer followed by a demo-clause. This is an excerpt from a telephone conversation between husband (Hide) and wife (Mika). Hide is temporarily studying in the U.S., living apart from the family. Having discussed what is new with their children, Mika begins her report of the previous day, when she went to Nagoya with the children to see her mother. Before Mika starts to describe the day, however, Hide asks a polar question in line 7: nagoya sugoku atsukatta desho: ‘It must have been very hot in Nagoya, wasn’t it?’. Mika’s answer to this question starts with an indication of hesitation followed by un and then a modification marked with demo.3 Excerpt 3. (CallHome JPN 1041 0m50s)4 01 MIK:
hhha [de kinoo and yesterday hhha and yesterday,
02 HID:
03 HID:
[de: and And u:n doo shita [no: so- so how did PRT Yeah, how did it go? ITJ
04 MIK:
05 HID:
[ma- mama to atta. mom with saw {I} saw m- mom. a:n juuhachi n'chi kita uhm 18 date came Hm {she} came on the 18th?5
3. In line 09 of Excerpt (3), the conjunction demo is placed within the clause, instead of at the beginning of a clause. This non-canonical positioning does not appear to change the interactional import of the clause in so far as its relation to the preceding confirmation is concerned. 4. The abbreviation “cop” is used in our transcripts to gloss various forms of copula. 5. Juuhachi n'chi ‘18th’ in line 05 refers to the day that Mika refers to as ‘yesterday’ in line 01. The use of a date in this context may strike a reader as a strange choice, but it has to do with the fact that there is a large time difference between the U.S. (where Hide is) and Japan (where Mika and the rest of the family are). It is likely that Hide has to keep track of what day it is in
Chapter 9. Post-confirmation modifications in response to polar questions
06 MIK:
u:n= U:n
07 HID:
=nagoya [(honto-) sugoku atsuka tta desho: Nagoya really very hot PFV TAG =It must have been (really-) very hot in Nagoya.
08 MIK: 09 MIK:=>
10 HID:
[a #u:#:nto u:n maa ki↑noo wa demo ne:, uhm un well yesterday PRT but PRT Uhm un well but yesterday was… nn mm-hm
11 MIK:=>
ano: chotto: (0.5) ano:: suzu- ↓suzushii well a.little well coo- cool …well a little (0.5) well coo- cool
12
tte iuka [ima made ni kurabe tara= QUOT or now till to compare if or rather compared to the days before…
=>
13 HID:
14 MIK:
15 HID:
16 MIK:=>
[nn mm-hm =suzushii [tte kanji da tta kedo[:, cool QUOT feel COP PFV but …it was cooler or something like that but [nn mm-hm
[nn mm-hm
atsuka tta yo? [demo hora. hoteru ittari hot PFV FP but see hotel went it was hot. But see, {we} were going back and forth between …
17 HID:
[nn mm-hm
18 MIK:
kitari shiteta kara:, came doing.PFV so the hotel ((and other places)) so,
The polar question-answer sequence initiated by Hide’s question (line 7) follows the pattern that we observed in Excerpt (2). Hide’s question presents a strong assessment (‘It was very hot in Nagoya, wasn’t it?’). Mika’s answer starts with un,
Japan when he talks with Mika and therefore used the date instead of the relative time reference (‘yesterday’) in this exchange.
281
282
Kaoru Hayano & Makoto Hayashi
delivered with an indication of hesitation (#u:#:nto ‘#uh:#:m’) and, in the prolongation of the particle un, conveying some trouble in confirming the question. The subsequent demo-clause accounts for the hesitation by modifying the intensity of the assessment presented in Hide’s question: Mika explains that it was cooler compared with earlier days. Here again, the demo-marked clause modifies the assessment that un confirms, yet it does not go as far as to disconfirm the question. On the contrary, Mika’s orientation toward confirming the answer is conveyed throughout the turn. After explicating that ‘yesterday’ was cooler than earlier days, she confirms that it was hot (line 16), although she continues to stress that it was not too bad because they went back and forth between their hotel and other places (rather than staying outside in the heat) (lines 16/18). It is intriguing that Mika makes this much interactional effort to push back against the assessment presented in Hide’s question. The final part of her turn, in which Mika claims that the heat was bearable (lines 16/18), gives us a clue as to what the issue is. Hide’s question pertains to the heat on the day that Mika spent with her mother. If it were very hot, the implication would then be that the day was unpleasant, rather than enjoyable. Thus, although it looks genuinely empathic and other-attentive on the face of it, Hide’s question overshadows the experience of spending a day with her mother, opening up a possibility for Mika to complain about it. It is this implication that Mika attempts to block through the design of her response. While demo-clause modifications in the previous two cases both modify the assessment presented in the question and deny its possible implication, the answer in Excerpt (4) achieves only the latter of these two tasks. Nana (nan), Shin (shn), Eita (eit), and Kana have been jokingly criticizing people who do not wake up early enough to be on time for appointments on account of low blood pressure unless the occasion is too consequential, and Shin has been teased for being one such person. Shin tries to defend himself in line 5 (“That is because …”), but his utterance is overlapped by Nana’s announcement that she has once been told that her blood pressure is low (lines 3/6). Here, she describes her blood pressure with a predicative adjective (ketsuatsu hikui, ‘(my) blood pressure was low’). After a gap of 1.0 seconds, Shin asks for confirmation (line 8), reformulating it with the noun (teiketsuatsu ‘low blood pressure’). Nana’s response (lines 9–10) is constructed with un followed by a demo-clause. Excerpt 4. (DEM6: 1) 01 ETT:
sooiu hito wa kekkyoku(h)= such person PRT eventually Such people are, after all,
02
=[( )=
Chapter 9. Post-confirmation modifications in response to polar questions
03 NAN:
[demo atashi mo suggoi= but I also very [But I was also…=
04 EIT:
=[(£ £)]
05 SHN:
=[( ) sore wa dakara ] that PRT because =[( ) That is because ]
06 NAN:
=[ketsuatsu hikui (toki ari) mashi ta] yo. blood.pressure low time be HON PFV FP =[…once told that my blood pressure was very] low.
07
(1.0)
08 SHN:
teiketsuatsu n’ no low.blood.pressure COP P {You} have low blood pressure?
09 NAN: => u:n. =demo sore o shiru mae made wa:, mecha un but that OBJ know before until PRT very U:n.=but before {I} found that out, {I}… 10
=> (.) a- hayaoki ya tta.= early.rise COP PFV … used to be quite an (.) early riser.=
11 SHN:
=ah:::::[::::] =ah:::::::::
12 NAN: =>
13
[demo] sore o shitte kara= but that OBJ know since But once {I} found that out=
=> =ii n ya moo tte [(omo(h)tte)] okay NOM COP now QUOT thought ={I} thought, “It’s okay now ((to not wake up early)).”
14 EIT:
[(
)]
15 SHN:
[a::::::: ] naruhodo.= ITJ I.see A::::::: I see.
Unlike the previous two cases, Shin’s confirmation request simply paraphrases what Nana has just said and does not present a speculative assessment that poses an issue for the answerer. Accordingly, when Nana modifies her confirmation, the demo-clause does not address any issue with the proposition of the question per se. However, it negates the possible implication of the proposition—that if one has low blood pressure, they cannot wake up early in the morning: although she turned out to have low blood pressure, she had been an early riser until she learned about it (lines 9–10). The implied message (or mockery) is that those
283
284
Kaoru Hayano & Makoto Hayashi
people, including Shin, who claim that low blood pressure makes it difficult to wake up early in the morning should not be taken seriously. Nana’s announcement about her blood pressure in lines 3 and 6 was produced as a contribution to the preceding discussion on how people (and Shin) use their low blood pressure as an excuse for tardiness. Thus, it may well be that the demo-clause that emerges in lines 9–10 (as well as the subsequent punch line in lines 12–13, where she reveals that she stopped being an early riser once she found out about her low blood pressure) would have been produced even if Shin had produced, for example, a news receipt instead of a confirmation request in line 8. Nonetheless, Nana exploits Shin’s question in line 8 as a launch pad to produce the next part of her announcement, in which she joins the others to tease Shin by denying the claimed outcome of having low blood pressure. In fact, the construction of Shin’s question proves convenient for Nana’s project: the categorical noun (teiketsuatsu, ‘low blood pressure’) that he uses to replace the predicative formulation in Nana’s turn (ketsuatsu hikuii, ‘blood pressure is low’) makes it sound more like a serious syndrome and thus implicative of accompanying symptoms. The demo-clause in Nana’s response resists the serious consequences of the condition implied in the question. Another way in which this excerpt differs from the preceding two is evident in the delivery of un. In Excerpts (3) and (4), un was delayed and produced with indications of hesitation, whereas the token of un in line 8 of this excerpt is produced without such features. Although our collection is not sufficiently large to determine the factors contributing to this difference, it is worth noting the possibility that the degree of hesitation conveyed in the delivery of the confirmation particle may reflect the degree of complexity that will be addressed in the remainder of the answer turn and/or the seriousness of the nature of the modification to come.
4.2 un + subordinate clause modification We saw in the last section that the demo-clause refutes an implication that is likely to follow the state of affairs that has just been confirmed by un while leaving the confirmation itself unqualified. In this section, we now turn to modifications that are constructed as subordinate clauses, which, by contrast, do qualify or even cancel the confirmation.6 6. As Maeda (2006) points out, the distinction between the subordinate and independent clauses may not be necessarily solid in Japanese grammar. She suggests that a clause should be characterized within a continuum with the undoubtedly subordinate clause at one end and the undoubtedly independent clause at the other. In this paper, we use the term subordinate clause loosely to characterize a clause that necessitates or alludes to another clause to which it is subordinate.
Chapter 9. Post-confirmation modifications in response to polar questions
The most common of the conjunctive particles observed in our collection is the contrastive particle keredomo and its variants (e.g., kedo, the most frequently used form in casual speech, keredo, kedomo, etc.), which can roughly be glossed as though. Keredo(mo)/kedo/kedomo can be used as a conjunctive adverb when placed at the beginning of a sentential unit (e.g., kedo wakan nai: ‘But I don’t know.’ ). However, when placed after a predicate, it functions as a conjunctive particle (e.g., wakan nai kedo: ‘Though I don’t know/ I don’t know, but’) and presents the clause to which it is attached as subordinate to the main clause.7 The data suggest that subordinate clause modification is used differently from demo-clause modification: unlike demo-clause modification, subordinate clause modification is used to retract or significantly qualify the confirmation that has been conveyed with un. Excerpt (5) is a case in which the confirmation conveyed by un is qualified with a kedomo (‘though’)-clause that follows it. This exchange is part of a conversation between old friends Goro (gor) and Taku (tak). Taku is a barber and is giving Goro a haircut. Prior to line 1, Goro reported to Taku that he had recently learned that their mutual friend had undergone a bone marrow transplant to treat leukemia. In response to this report, Taku made some comments, mentioning two celebrities who had leukemia: while one of them recovered after a bone marrow transplant, the other did not because she could not find a matching donor. Taku then asks whether or not it is a family member who was the donor (lines 3–4). Goro begins his response with the confirmation particle un (hun) but this is followed by a kedomo-clause modification that qualifies the epistemic ground of his confirmation. Excerpt 5. (NINJAL T15_008 3m05s) 01 TAK:
a
demo ko↑tsuzui: ishoku ↓ka:: but bone.marrow transplant Q Oh but it is bone marrow transplant. ITJ
02
(0.6)
03 TAK:
>are na no-< (0.5) miuchi no hito:::nthat COP N family L person Is it that (0.5) It’s not like a family member…
04
toka nantoka ↑tteyuu wake ja nai n da etc. something COMP NOM PRT not N COP … ((was the donor)), right?
7. Keredomo (or a variant) often occurs in utterances without clauses to which they are subordinate, leading some linguists to argue that it has attained usage as a final particle (Kinjo 2001). However, this is not to deny that the use of keredo produces the impression that something has been left unsaid, which may constitute the main clause, inviting the recipient to infer what that unsaid thing may have been.
285
286
Kaoru Hayano & Makoto Hayashi
05 GOR:=>
hun soko made wa hanashi kikanakatta kedomo un that up.to PRT story hear.not.PFV but Un though {I} didn’t hear {his story/ ask {him} that far.
06
(0.2)
07 TAK:
aa:kore de< kintooni natta ka↓na: this with even became Q {I} wonder if it is now even.
02
(2.6)
03 AKI:
kono hito ga ooi this person SBJ much This person has got too much.
04
(6.0)
05 AKI:
kintoo da to omou[:? even COP QUOT think Do {you} think it’s even?
06 EMI:=>
[un. Un.
07
(0.2)
08 EMI:=>
>watashi< sukuname de mo ii shi I little COP also fine and Since it is fine if my portion is smaller.
09
(0.2)
10 AKI:
n::
11
(1.2)
12 AKI:
konna mo:n? this NOM Something like this ((is okay))?
13 EMI:
ookini Thank you.
(b)
moo kurai shi kae roo. already dark shi go.home let.us Since {it} is already dark, let’s go home.
In (a), shi is attached to an item that it presents as one of the items in the list. In (b), it is attached to a reason for a suggested act, which is stated in the following, main clause. For the analysis of shi in Excerpt (6), see Footnote 10.
287
288
Kaoru Hayano & Makoto Hayashi
Aki’s question in line 5 invites Emi to look at the plates and judge whether or not the rice has been evenly dished out. Emi confirms the question with un in line 6, but this confirmation is substantially qualified or contradicted when Emi says that it is fine if her portion is smaller (line 8). The conjunction particle shi formats this modification into a subordinate clause stating the reason or basis for the judgment that Emi has expressed with un in line 6. We can thus see the shi-marked modification cannot be directly tied to that proposition that has been confirmed with un: It does not make sense to say, “Since it is fine if my portion is smaller, the rice is evenly distributed.” Instead, the implied message is more in line with the following: “Since it is fine if my portion is smaller, the way you distributed the rice is fine (whether it is even or not).” To put it another way, the modification in line 8 reinstates the possibility that the rice may not be dished out evenly and, thus, significantly qualifies the confirmation that Emi has conveyed with un or her commitment to it.9 Indeed, this appears to capture Aki’s analysis of Emi’s answer: Aki does not stop adjusting the amount of rice on the plates after line 8 and asks Emi again to confirm that the portions are even (line 12), which suggests that Aki does not understand lines 6/8 to be an endorsing response to act upon. Emi neither confirms nor disconfirms the second question by Aki in any way but instead thanks Aki for her work (line 13). Although this response by Emi is disaligning in so far as it does not provide the response that the question has made relevant (Stivers and Robinson 2006), it is a concerted response to appreciate and support the activity with which Aki is engaged. The confirmation conveyed by un in the above two examples is not entirely negated by the clauses that follow it, although it is substantially qualified. By contrast, in Excerpt (7), the confirmation expressed by un is contradicted by what follows it. Excerpt (7) is taken from a telephone conversation between Rie (rie) and Junko (jun). Rie knows that Junko was seeing a psychiatrist, and Rie appears to attempt to topicalize it when she says that Junko seems “to be doing well recently” (line 1). However, Junko’s answer is confined to how she was feeling that day (lines 2 and 4, plus subsequent lines omitted from the transcript), which can be an indication that she does not want to talk about how she has been doing in general. Then, Rie more explicitly addresses the topic by asking if Junko is not seeing the doctor anymore (lines 17–18). As we mentioned earlier, un confirms the ques9. We cannot altogether dismiss the possibility that shi in line 8 of Excerpt (6) is used to introduce an item in the list of manners (see Footnote 9) the participants use in distributing steamed rice: “It is fine if my potion is smaller or we can make it evenly distributed.” Whether shi in this case is used as a subordinate conjunction particle or a coordinate conjunction particle, however, we can still say that line 8 is designed to be grammatically parasitic with this conjunction particle attached, inviting the recipient to infer what it is parasitic to, and it is in this broad sense that we call shi-marked clauses subordinate clauses.
Chapter 9. Post-confirmation modifications in response to polar questions
tioner’s expectation regardless of the polarity of the question. The agreement particle un in line 19 thus confirms that she is not seeing the psychiatrist any longer. However, some trouble is indexed by the way in which un is delivered (line 19): it is delayed by an inbreath (.hh) followed by 0.2 seconds of pause, and the sound stretch on un is hearable as an indication of hesitation or reluctance (see Figure 4). The two clauses that follow un, which cancel the confirmation, clarify the nature of the trouble. Excerpt 7. (CallFriend JPN 6666 14m26s)10 01 RIE:
>teka< saikin genki soo da ne.(I) mean< (you) seem to be doing well recently.
02 JUN:
u↑:↓n. maa (0.5) ne: kyoo wa yoku un well PRT today PRT well un. Well (0.5) {I} slept well today(h),…
03
neta kara(h)hh slept so …so hh.
04 RIE:
soo na no? that COP PRT {You} did?
05 JUN:
na:nka- ichido- ↑itsumo sa like once usually PRT Like once- usually, {I} wake up at… ((11 lines of talk about how JUN had trouble waking up that day omitted))
17 RIE:
.hh sonde: nani:? saikin wa moo and what recently PRT EMP .hh And what. {You} are not seeing…
18
oishasan itte nai n↑o doctor go not PRT … the doctor any ↑more?
19 JUN:=>
.hh (0.2) u::↓n maa itte n da kedo ne, un well go N COP but PRT .hh (0.2) u::↓n well though {I} am.
20 RIE:
n:n, Mm-hm, (0.3)
21 JUN:=>
demo maa ch'tto socchi no hoo wa but well a.bit that L area PRT But well, since things have not changed…
10. The abbreviation “emp” is used in our transcripts to gloss emphasis markers.
289
290
Kaoru Hayano & Makoto Hayashi
22
taishite kawari nai shi ↓sa much change not since PRT … much in that area,
23
(0.2)
24 RIE:
ke- kaware nai? kawara nai? change.can not change not ch- Can’t change? Doesn’t change?
25
(.)
26 JUN:
nn Un
27
(0.2)
28 RIE:
↑so’ na n da: that COP N COP I see.
29
(0.5)
30 RIE:
.hh da:mn! .hh Da:mn!
31
(0.4)
32 JUN:
Hhh
33 RIE:
mo(h)o yoku naka- hh moo £↑moo sorosoro£ already well already already by.now Already well- hh already £{I} thought£…
34
£yoku natteru kana£ to omotta n da kedo. well has.got Q QUOT thought N COP but … £{you} would have been well already£ but
Un in Junko’s answer in line 19 is followed by two clauses: a clause marked with kedo (“though”; line 19) and then one marked with shi (‘since’; line 21). The adverb maa is used in both, which in this context conveys that the speaker is withholding information (Togashi 2002: 27). Thus, while Junko reveals her situation in these clauses, she does so while showing some signs of reluctance. Let us now consider the modification that is added by each of the two clauses. In the first clause, marked with kedo (line 19), Junko admits that she is still seeing a doctor. The confirmation denoted by un and the information that this clause carries contradict one another, with the latter effectively canceling the former. In Excerpts (5) and (6), the semantic relations between the confirmation particle and modification clauses are relatively transparent: In Excerpt (5), the modification clause downgrades the epistemic certainty of the speaker; in Excerpt (6), the modification clause provokes a reanalysis of the confirmation,
Chapter 9. Post-confirmation modifications in response to polar questions
insinuating that what the questioner had done was fine, whether or not the rice was evenly distributed. However, in this case, the ways in which the confirmation and the modification can be consolidated are not immediately clear, and that may be why Rie does not treat Junko’s answer to that point as complete: in line 20, Rie produces a continuer (Schegloff 1982), rather than a change-of-state token (Heritage 1984). The shi-marked clause, which also builds on the answer initiated with un, provides a clue to how the contradicting pieces of response should be understood. In lines 21–22, Junko says that things “have not changed much in that area,” and this turn starts with demo and ends with the conjunction particle shi. The conjunction demo here connects line 21 not to the confirmation un but to the kedo-clause that follows un, so this may be put aside for our purposes. As we discussed earlier, shi presents the clause to which it is attached as the basis for an outcome or judgment, which may or may not be stated. Here, again, the outcome of, or judgment based on, what Junko says in lines 21–22 is not clearly stated, and Rie thus has to infer the upshot of “things” not having changed much in that area. Given the delicateness of the topic, the plausible inference to be made is that since things have not changed much in that area, there is not much to say about it. Taken together, the response conveys Rie’s unwillingness to discuss the matter further. This analysis of the shi-marked clause in lines 21–22 fits with other aspects of Junko’s conduct in this exchange: avoidance of discussing the agenda behind Rie’s question in line 1; turn-initial delay (line 19); the use of the adverb maa (lines 19 and 21); the choice of a covert form socchino hoo (‘that area’) in referring to her psychological condition (line 21); and the lack of substantial utterances thereafter (lines 25–27, 29, 31–32). Furthermore, the analysis of the shi-clause as an indication of Junko’s reluctance to discuss the matter accounts for the production of un at the beginning of Junko’s answer. That is, it is only after having once confirmed that she is not seeing a doctor anymore that she admits that she is still seeing a doctor. This delayed revelation can be analyzed as another indication that Junko is not willing to discuss her condition. In this section, we have attempted to demonstrate that modifications are formatted into subordinate clauses when they significantly qualify or cancel the pro forma confirmation, which stands in contrast to what we observed with demo-clause modifications. We believe that this functional difference lies in the grammatical status of these two constructions. While coordinate clauses marked with demo leave un and the confirmation it conveys as a separate unit and present the modification as additional, subordinate clauses marked with keredomo (‘though’) or shi (‘since’) revise the preceding confirmation, qualifying or negating it. Thus, although we cannot uncritically presume that grammatical categories provide for legitimate units of interactional orders, the division of labor between
291
292
Kaoru Hayano & Makoto Hayashi
the grammatical formats we have examined can be understood as a case in which grammar finds its place in its “home environment” (Schegloff 1996: 112).
4.3 un + modification So far, we have examined two basic sets of formats that are used to modify confirmation: demo-clause modifications and subordinate clause modifications. The analysis has revealed that these two grammatical formats are used to effect different interactional functions: demo-clauses modify the intensity of an assessment presented by the questioner and/or to take issue with the implication that a mere confirmation would accept; subordinate clauses serve as a means of substantially qualifying or negating the confirmation such that they retroactively explicate how and why the initial confirmation needs to be modified. In this section, we discuss the third format used to modify confirmation: clauses that are not marked for their grammatical relationship to what precedes them. Modifications in this format are found to address an issue with an assumption underlying the question, which makes neither confirmation nor disconfirmation a good option for the answerers. Modifications of this type are disaligning on some level; in some cases, they constitute non-answer responses (Stivers and Robinson 2006), while in others, they constitute transformative answers, conveying confirmation or disconfirmation only evasively (Stivers and Hayashi 2010). With such disaligning responses, the turn-initial un is analyzable as a display of orientation to alignment, which may then serve to prepare a context in which the subsequent modification is heard as cooperative and/or non-confrontational. Excerpt (8) is an exchange taken from a family dinner conversation among an elderly couple, Sawa (saw) and Eiji (eij), and their daughter, Miki. Several minutes prior to the exchange in the excerpt, Sawa offered to prepare tea by asking a question: ocha atsui hoo ga ii hito tsumetai hoo ga ii hito (‘Who wants hot tea and who wants iced tea?’). Although this question was designed to invite both Miki and Eiji to tell Sawa their preferences, only Miki answered, saying that she would prefer hot tea. Sawa then stood up to boil water for hot tea. When the hot tea is almost ready, the following exchange is initiated by Sawa’s question: Otoosan wa tsumetai n h- (0.2) ocha no’ ga ii? (‘Would cold tea be better for you?’). Considering the preceding context, the question is produced to pursue Eiji’s response to Sawa’s earlier question. This time, however, Sawa’s question is designed as a polar question, inviting Eiji to confirm that he would prefer iced tea, conveying the presupposition that Eiji should be able to answer by either confirming or disconfirming the question. Eiji’s non-answer response rejects this presupposition and states that whichever would be fine for him. Our analytical interest lies in what function is served by the response-initial un.
Chapter 9. Post-confirmation modifications in response to polar questions
Excerpt 8. (NINJAL C001–007: 30:36) 01 SAW:
otoosan wa tsumetai n h- (0.2) father(=you) PRT cold Would iced tea be…
02
ocha no’ ga ii? tea better SBJ like better for you?
03 EIJ:=>
(y)un ↓docchidemo ii (yo). un whichever fine FP Un whichever is fine.
04
*(1.5) *pours iced tea into Eiji’s cup
saw
When it is produced as a response to a polar question that invites the recipient to accept the offered item (iced tea in this case), the answer docchidemo ii (yo) (‘whichever is fine’) can be understood as a harbinger of rejection (as we will see in Excerpt (10) below), unless the context suggests otherwise. However, Sawa does not orient to this possibility; she simply goes ahead and pours Eiji iced tea (line 4). One factor that appears to contribute to this is the complication with the preference: although Sawa’s question is syntactically designed to prefer acceptance of the offer of iced tea, given that Miki earlier requested hot tea and that requesting iced tea may cause Sawa additional trouble, iced tea may not be a socially preferred option to select. Thus, Eiji’s indecisive answer does not unequivocally convey reluctance to accept iced tea. In addition, the turn-initial un plays a role to present the response as a whole as aligning. That is, while docchidemo ii (‘whichever is fine’) by itself is a disaligning response that can be heard as an implicit rejection of the offered item, or even a dismissive stance, the confirmation particle un prepares a context in which a possibly dismissive response can be heard as innocently entrusting the choice to the interlocutor. A similar case, in which a non-aligning response of the same nature is not preceded by un, lends some support for the above analysis. In Excerpt (9), Yui asks Aki if green tea would be fine for her. In response, Aki says, nandemo ii, arigatoo (‘Anything would be fine, thank you.’ ). This question-answer sequence is thus comparable to the one examined in Excerpt (8): a polar question implements an offer of a particular drink, which receives a response entrusting the questioner with the choice. However, Aki’s answer here does not start with the confirmation particle un, and the subsequent trajectory is in clear contrast with what we saw in Excerpt (8).11
11. We do not have visual access to what Yui and Emi are doing during the excerpt since they are out of frame. However, given that Aki, who is in the frame, is not looking toward them,
293
294
Kaoru Hayano & Makoto Hayashi
Excerpt 9. (ThaiDinner 1: green tea) 01 YUI:
ryokucha de ii kai? green.tea with fine Q Would green tea be fine {by you}?
02 AKI:=>
nandemo ii, arigatoo. anything fine thank.you Anything would be fine, thank you.
03
(3.5)
04 AKI:=>
ryokucha tte sa::, kafein ooi tte honto:? green.tea QUOT PRT cafein much COMP true Is it true that green tea contains a lot of caffeine?
05 EMI:
nani ga ooi, what SBJ much A lot of what?
06 AKI:
kafe[in. Caffeine.
07 EMI:
08 YUI:
[un ooi [yo, un much FP Un, it does. [ooi.ooi:. much much It does, it does.
09 Aki:=>
honto:? ja- [tkafein sukuname no ocha ga really then caffeine little L tea SBJ Really? Then, tea without much caffeine
10
ii kedo:, fine but would be nice, but
After 3.5 seconds of lapse in line 3, Aki asks a question regarding the amount of caffeine contained in green tea and uses Emi’s and Yui’s responses to this question as an opportunity to request something other than green tea (line 9), like barley tea, which they in the end decide to have. Retrospectively, then, we can see that Aki was not content with what Yui offered in line 1. The non-conforming, disaligning answer in line 2 may facilitate this understanding. When contrasted with this response by Aki, the token of un in Eiji’s response in Excerpt (8) appears to lend itself to being analyzed as an indication of cooperative stance to suggest that the producer does not mean to reject the option presented by the questioner.
focusing on the food on the table that she is preparing, it is reasonable to consider that there is not much happening with their bodily behavior that is critical to our analysis of the segment.
Chapter 9. Post-confirmation modifications in response to polar questions
Excerpt (10) is another example of an un-initiated answer followed by a modification that is not grammatically connected to un. Kumi has been telling Yuna about her boyfriend, who would not commit to their relationship. Having complained about that, Kumi says, doo shi yo kkana: mitaina ne (‘It’s like, “What am I gonna do?”’; line 1), most likely insinuating that she is considering whether or not she should stay in that relationship. In response, Yuna produces what sounds like the beginning of advice (koko ga (‘Here is’)), but then changes the trajectory and asks a polar question: demo datte Kumi mada suki na n desho::? Mochiron (‘But you still love him, right? Of course.’ ).12 The conjunctive datte roughly means “because” and is used to preface a reason or an account behind an action or decision. In this case, datte contributes to the hearing of the question as a pre-advice: the question addresses a possible reason why Kumi should (or should not) stay in the relationship. Although Kumi starts her response with the confirmation particle un, what follows it constitutes a transformative answer (lines 5–6). Excerpt 10. (JPN1684: 12:53) 01 KUM:
mo::: [hh doo shi yo kkana: mitaina ne= how do VOC Q like P It’s like, “What am {I} gonna do,”= EMP
02 YUN:
[tch.hhh
03 YUN:
=koko ga- demo datte Kumi mada here SBJ but because Kumi(=you) still =Here is- but you still…
04
suki na n desho::? mochiron. like COP N TAG of.course …love {him}, right? Of course.
05 KUM:=>
.hh u:n. o↑tagai suki tte itteru n desu un mutually like QUOT say NOM COP .hh Un {we} say to each other that {we} love each other.
06
↓yo.
[demo ↑kare wa ↓kare-= but he PRT he But he, he-= FP
07 YUN:
08 KUM:=>
[dakaso So=kare wa ne, [suggoi warikireru kara:, he PRT P very separate.can so =as for him, (he) can rationalize (things), so
12. Indeed, shortly after Excerpt (10), Yuna does start advice with this format, saying that here is the time when Kumi should hold out. The pre-advice question in lines 3–4 in Excerpt (10) can thus be analyzable as already heading for that direction.
295
296
Kaoru Hayano & Makoto Hayashi
09 YUN:
[n: mm-hm
Given that Yuna’s question in lines 03–04 implements a pre-advice, a simple confirmation would be treated as a go-ahead response to encourage the production of the advice and would accept that whether Kumi still loves the boyfriend constitutes a good basis for the decision of whether to stay with him. Although Kumi starts her response with the confirmation particle un, the turn-initial inbreath conveys some reluctance, which will be explicated in the rest of the turn: it is not just that Kumi still loves her boyfriend but that the boyfriend also loves her, and she knows it because he says so (lines 5–6). The answer as a whole, therefore, takes issue with the assumption conveyed by the question – that Kumi can/should decide whether to stay with the boyfriend based on her feelings for the boyfriend, which may be all that is left in the relationship. On the basis of our analysis of un in relation to Eiji’s answer in Excerpt (8), we can discuss what the turn-initial un in Kumi’s answer accomplishes. We observed in Excerpt (8) that un may work to preempt the possibility that the ensuing modification is heard as disagreement-implicative and uncooperative. In Excerpt (10), Kumi challenges Yuna’s assumption with the subsequent modifying clause, and the turn-initial un appears to be a means of mitigating the challenging tone that the ensuing clause potentially carries. We may then consider that this is the environment in which un really functions as a pro forma agreement and nothing else. Once the rest of the response has been produced, the answerer can see that it does little other than framing a disaligning response with a displayed orientation to alignment and cooperation.
5.
Discussion and conclusion
Our findings point toward the conclusion that various interactional tasks and outcomes are accomplished by modifications following turn-initial confirmation and that there is a fine division of labor assigned to different grammatical formats. The demo-clause modifications decline the questioner’s assessment and/or its possible implications without canceling the confirmation itself. The subordinate-clause modifications qualify the confirmation by downgrading the epistemic reliability of the confirmation or the answerer’s commitment to it or even cancels it altogether. Modifications that are not grammatically marked for their semantic relations to confirmation are found when they address the inadequacy of both confirmation and disconfirmation as responses to the question. It follows that, while all of these are issues addressed by answerers as “modifications” presented
Chapter 9. Post-confirmation modifications in response to polar questions
in “agreement mode” (Sacks 1987), the differences in format are meaningful and consequential. Throughout this chapter, we have remarked on the various ways in which un is delivered and their possible relations to what is said in the rest of the turn. Sacks (1987) suggests that the initial agreement is not necessarily delivered in a way that projects incipient disagreement. In Excerpt (1), for example, the turn-initial “yuh” does not accompany prosodic or other paralinguistic cues of disagreement to come. Indeed, in some of the cases presented in this chapter Excerpts (4, 5, 6, and 8), the turn-initial pro forma agreements were produced immediately, without any indication of incipient modifications. In the other cases, however, un was delayed or produced with signs of hesitation, providing an environment in which the following modifications may be analyzed as accounting for the hesitation indicated by un. We believe that the availability of these two methods of delivery itself – whether or not the pro forma agreement particle is delivered to foreshadow upcoming modifications – is an interactional resource that is available to the answerers. On one hand, by not displaying hesitation through the ways of delivering un, answerers may present the subsequent modification as insignificant (in the case of epistemic downgrading in the discussion of a friend’s situation (Excerpt (5))), non-serious (in the discussion of low blood pressure and its relation to tardiness (Excerpt (4))), or as an embodiment of prosocial consideration for an interlocutor (in support of an interlocutor who is trying to distribute rice evenly or who is serving tea to the family (Excerpts (7) and (9))). On the other hand, by foreshadowing some problem through the delivery of un, answerers may be able to prepare a ground by reference to which the turn will be heard, whether to adumbrate some issue with the question (Excerpts 2, 3) or to indicate the delicacy involved in answering the question (Excerpt (7), (10). To conclude, we hope to have contributed to the developing body of literature on how recipients of polar questions maneuver the sequence under all of the constraints that the questions impose. Modifications, which have been uniformly treated as “real answers” that answerers sneak into their responses, may merit further exploration in their own right.
Acknowledgements An earlier version of this paper was presented at the 5th International Conference on Conversation Analysis (ICCA) at Loughborough University. We thank the editors of this volume and an anonymous reader for their insightful comments.
297
298
Kaoru Hayano & Makoto Hayashi
References Bolden, Galina. 2018. “Nu-prefaced Responses in Russian Conversation.” In Between Turn and Sequence, ed. by John Heritage, and Marja-Leena Sorjonen, 25–58. Amsterdam: John Benjamins. Enfield, Nick J., Tanya Stivers, Penelope Brown, Christina Englert, Katariina Harjunpää, Makoto Hayashi, Trine Heinemann, Gertie Hoymann, Tiina Keisanen, Mirka Rauniomaa, Chase Raymond, Federico Rossano, Kyung-Eun Yoon, Inge Zwitserlood, and Stephen C. Levinson. 2019. “Polar Answers.” Journal of Linguistics 55(2): 277–304. Ford, Cecilia E., Barbara A. Fox, and John Hellermann. 2004. “‘Getting Past No’: Sequence, Action and Sound Production in the Projection of No-initiated Turns.” In Sound Patterns in Interaction: Cross-linguistic Studies from Conversation, ed. by Elizabeth Couper-Kuhlen, and Cecilia E. Ford, 233–269. Amsterdam, John Benjamins. Hakulinen, Auli. 2001. “Minimal and Non-minimal Answers to Yes-no Questions.” Pragmatics 11(1): 1–15. Hayashi, Makoto. 2010. “An Overview of the Question-Response System in Japanese.” Journal of Pragmatics 42(10): 2685–702. Hayashi, Makoto, and Kaoru Hayano. 2018. “A-Prefaced Responses to Inquiry in Japanese.” In Between Turn and Sequence: Turn-initial Particles across Languages, ed. by John Heritage, and Marja-Leena Sorjonen, 193–224. Amsterdam: John Benjamins. Heritage, John. 1984. “A Change-of-State Token and Aspects of Its Sequential Placement.” In Structures of Social Action: Studies in Conversation Analysis, ed. by J. Maxwell Atkinson, and John Heritage, 299–345. New York: Cambridge University Press. Heritage, John. 1998. “Oh-prefaced responses to inquiry. Language in Society, 27: 291–334. Heritage, John. 2015. “Well-Prefaced Turns in English Conversation: A Conversation Analytic Perspective.” Journal of Pragmatics 88: 88–104. Heritage, John, and Geoffrey Raymond. 2012. “Navigating Epistemic Landscapes: Acquiescence, Agency and Resistance in Responses to Polar Questions.” In Questions: Formal, Functional and Interactional Perspectives, ed. by Jan Peter de Ruiter, 179–192. Cambridge: Cambridge University Press. Kendrick, Kobin H. and Francisco Torreira. 2015. “The Timing and Construction of Preference : A Quantitative Study.” Discourse Processes 52(4): 255–289. Kim, Stephanie Hyeri. 2013. “Reshaping the Response Space with Kulenikka in Beginning to Respond to Questions in Korean Conversation.” Journal of Pragmatics 57: 303–317. Kim, Stephanie Hyeri. 2018. “Two Types of Trouble with Questions: A Comparative Perspective on Turn-initial Particles in Korean.” In Between Turn and Sequence: TurnInitial Particles Across Languages, ed. By John Heritage, and Marja-Leena Sorjonen, 97–118. Amsterdam: John Benjamins Publishing. Kinjo, Katsuya. 2001. “Bunmatsu hyoogen toshiteno keredomo no kinoo ni tsuite [On the sentence final expressions keredomo/ga].” Scripsimus 10: 87–101. Koiso, Hanae, Yasuharu Den, Yuriko Iseki, Wakako Kashino, Yoshiko Kawabata, Ken’ya Nishikawa, Yayoi Tanaka, and Yasuyuki Usuda. 2018. “Construction of the Corpus of Everyday Japanese Conversation: An Interim Report.” Proceedings of the 11th Edition of the Language Resources and Evaluation Conference, 4259–4264.
Chapter 9. Post-confirmation modifications in response to polar questions
MacWhinney, Brian, and Johannes Wagner. 2010. “Transcribing, Searching and Data Sharing: The CLAN Software and the TalkBank Data Repository.” Gesprachsforschung 11: 154–173. Maeda, Naoko. 2006. “Gendai nihongo ni okeru setsuzokujoshi shi no imi, youhou: Heiretsu to riyuuno kankei wo chuusin ni [Usage of conjunctive particle shi in modern Japanese].” Jinbun 4: 131–144. (Gakushuin University, Research Institute for Humanities) Morita, Emi, and Tomoyo Takagi. 2018. “Marking ‘Commitment to Undertaking of the Task at Hand’: Initiating Responses with Eeto in Japanese Conversation.” Journal of Pragmatics 124: 31–49. Pomerantz, Anita. 1984. “Agreeing and Disagreeing with Assessments: Some Features of Preferred/Dispreferred Turn Shapes.” In Structures of Social Action: Studies in Conversation Analysis, ed. by J. Maxwell Atkinson, and John Heritage, 57–101. Cambridge: Cambridge University Press. Raymond, Chase. 2018. “Bueno-, Pues-, and Bueno-Pues-Prefacing in Spanish Conversation.” In Between Turn and Sequence: Turn-initial Particles across Languages, ed. by John Heritage, and Marja-Leena Sorjonen, 59–96. Amsterdam: John Benjamins Publishing. Raymond, Geoffrey. 2003. “Grammar and Social Organization: Yes/no Interrogatives and the Structure of Responding.” American Sociological Review 68: 939–967. https://www.jstor .org/stable/1519752. Sacks, Harvey. 1987. “On the Preferences for Agreement and Contiguity in Sequences in Conversation.” In Talk and Social Organisation, ed. by Graham Button, and John R. E. Lee, 54–69. Clevendon: Multilingual Matters. Sacks, Harvey. 1992. Lectures on Conversation, Vol. 2. Oxford: Blackwell. Sadock, Jerrold M., and Arnold M. Zwicky. 1985. “Speech Act Distinctions in Syntax.” In Language Typology and Syntactic Description, Volume I, Clause Structure, ed. by Timothy Shopen, 155–196. Cambridge: Cambridge University Press. Schegloff, Emanuel A. 1982. “Discourse as an Interactional Achievement: Some Uses of ‘Uh huh’ and Other Things that Come between Sentences.” In Georgetown University Roundtable on Languages and Linguistics, ed. by Deborah Tannen, 71–93. Washington, D.C.: Georgetown University Press. Schegloff, Emanuel A. 1996. “Turn Organization: One Intersection of Grammar and Interaction.” In Interaction and Grammar, ed. by Elinor S. Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 52–133. Cambridge: Cambridge University Press. Schegloff, Emanuel A. 2007. Sequence Organization in Interaction: A Primer in Conversation Analysis. Cambridge: Cambridge University Press. Seuren, Lucas M., and Mike Huiskes. 2017. “Confirmation or Elaboration: What Do Yes/No Declaratives Want?” Research on Language and Social Interaction, 50(2): 188–205. Sorjonen, Marja-Leena. 2001. Responding in Conversation: A Study of Response Particles in Finnish. Amsterdam: John Benjamins. Steensig, Jakob, and Heinemann, Trine. 2013. “When ‘Yes’ Is Not Enough as an Answer to a Yes/no Question.” In Units of Talk−Units of Action, ed. by Beatrice Szczepek Reed, and Geoffrey Raymond, 207–241. Amsterdam: John Benjamins. Stivers, Tanya. 2010. “An Overview of the Question-response System in American English Conversation.” Journal of Pragmatics 42(10): 2772–2781.
299
300
Kaoru Hayano & Makoto Hayashi
Stivers, Tanya, Enfield, N. J., Brown, Penelope, Englert, Christiana, Hayashi, Makoto, Heinemann, Trine, Hoymann, Gertie, Rossano, Federico, de Ruiter, Jan P. Yoon, Kyung-Eun, and Levinson, Stephen C. 2009. “Universals and Cultural Variation in Turntaking in Conversation.” Proceedings of the National Academy of Sciences 106(26): 10587–92. Stivers, Tanya, and Hayashi, Makoto. 2010. “Transformative Answers: One Way to Resist a Question’s Constraints.” Language in Society, 39: 1–25. Stivers, Tanya, and Robinson, Jeffrey D. 2006. “A Preference for Progressivity in Interaction.” Language in Society, 35: 367–392. Togashi, Jun’ichi. 2002. “Danwa hyooshiki maa ni tsuite [On the danwa hyooshiki maa].” Tsukuba Japanese Linguistics, 7: 15–31. Weidner, Matylda. 2018. “Treating Something as Self-evident: No-prefaced Turns in Polish.” In Between Turn and Sequence: Turn-initial Particles across Languages, ed. by John Heritage, and Marja-Leena Sorjonen, 225–250. Amsterdam: John Benjamins Publishing.
chapter 10
Responding to polar questions without a polarity item ‘yes’ or ‘no’ in Finnish Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen University of Helsinki
Polar questions and their responses have been a subject of increasing interest for conversation analysts. This paper contributes to this research by studying answers to interrogatively formatted polar questions in Finnish that do not contain a separate element for expressing positive or negative polarity. We will show that the answers are used for a range of practices that form a continuum from cases where the polarity is implied to ones where the recipient, in their response, leaves it to the questioner to evaluate the quality of the response. Moving beyond the polarity of the question, these answers focus on what the recipient inferred to be the core relevance of the question, furthering the progressivity of the interaction by sustaining the ongoing project. Keywords: polar question, polar interrogative, answer, response, implicit polarity, evaluation, co-operation, context, Finnish
1.
Introduction
Polar questions and their responses in interaction have received increasing interest in research for quite some time already. One impetus to this was an influential study by Raymond (2003) who, focusing on answers to polar interrogatives in English interactions, broadened the conversation analytic concept of preference. He presented (ibid.: 944–946) that a polar interrogative as a turn and action design sets a preference, and a normative expectation, for the form that the responding action should take. This preferred form would include a polarity item, in English the particle yes or no, or an equivalent form (e.g., yep, nope, or uh huh). A response initiated with such an element conforms to the grammatical design of the question (“type-conforming” response) while producing a preferred response relative to the action type of the question. On the other hand, a response withhttps://doi.org/10.1075/slsi.35.10vep © 2023 John Benjamins Publishing Company
302
Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen
out such a form is “nonconforming”, displaying trouble with or resistance to the terms of the question (ibid.: 949). The evidence Raymond (ibid.: 947) provides for the preference for the type-conforming responses is three-fold: (1) they are more frequent, (2) they are treated as default responses by the questioner, and (3) they advance the progress of the on-going activity. In this paper, we will investigate a type of responses that would be treated as non-conforming in Raymond’s system in Finnish everyday interactions. Nonconforming responses have been investigated further in subsequent research. Stivers and Hayashi (2010) examined such responses in American English and Japanese interactions, analyzing responses to both interrogative and declarative polar questions (requests for confirmation). They describe the responses in focus as a subclass of nonconforming answers, terming them “transformative” (as opposed to “direct nonconforming answers” that are full-clause repeats of the question). Stivers and Hayashi propose two main ways in which answers adjust (transform) the terms of the questions: (1) by specifying the design of the question or replacing its component (question term transformations), and (2) by shifting the focus, bias or presupposition of the question (agenda transformations). Both of these answer types resist the question’s constraints, where resistance can be understood as arising from the formal relationship between the question and the answer, that is, from the fact that the answer does not contain a positive or a negative particle (yes or no). Of the two types, the agenda transforming responses are more resisting in “evading” the question, whereas the term-transforming responses adjust merely the question design. Stivers and Hayashi (ibid.: 7) also provide quantitative information on the relative frequencies of direct conforming, direct non-conforming, and transformative nonconforming answers. Some subsequent studies have pointed out that whether an answer is aligning or disaligning with the question is related to issues other than the relationship between the grammatical design of a question and its answer. In their study on English interactions, Walker, Drew, and Local (2011) show that recipients design their response with respect to their interpretation of the agenda of the question and the reason for asking it in the ongoing activity. They distinguish three types of response to (mainly polar) questions: direct, ‘direct’ and indirect: Direct responses correspond to what Raymond (2003) terms type-conforming, whereas ‘direct’ responses are “clear and obvious responses”, and all that prevents them from being a type-conforming (or in Searlean terminology direct) response is the lack of a positive or negative polarity particle. Walker, Drew, and Local state (2011: 2347) that the questioners sometimes expect answers which are more elaborate than a mere response particle would be, that is, “it seems that they were counting on their co-participants’ ability to respond to ‘why that now’ rather
Chapter 10. Responding to polar questions without a polarity item
than the response explicitly provided for by the prior turn’s syntactic structure” (ibid.: 2347). The indirect responses, by contrast, are non-type-conforming, not initiated with a polarity item, and do not contain typical structural elements that tie back to the prior turn as in more direct responses (e.g., repetition, ellipsis and pronominalization). The indirect responses also refer to non-contiguous talk having occurred before the question, or to shared knowledge, generating inferences providing an answer to the question (Walker, Drew, and Local 2011: 2441). Responses to polar questions not starting with a polarity item have also been studied in institutional settings. In her study of Korean airline customer service phone calls, Lee (2011), building on Raymond’s (2003) schema of typeconformity and nonconformity, shows that when orienting to the questioner’s higher-level goals, customers may depart from type-conformity in order to further the progress of the activity. This is treated as an exception to Raymond’s schema. Taking a step further, Robinson and Heritage (this volume) show how patients orient to the ongoing activity when responding to the doctor’s polar question with a response that does not contain a positive polarity item. Relatedly, Pomerantz (2017) examined how recipients infer the purpose of the question and how they deal with the relationship between the literal meaning of the question and the inferred purpose of the question (see also Pomerantz 1988, and 2021: 241–244, her commentary on the article). She notes that the significance of the questioner’s agenda and the role of the context have long been acknowledged (e.g., Levinson 1981, questioner’s “higher level goals”), and she calls for further detailed analyses of the specific aspects of the form (design) of the question, the immediately preceding interaction, and the information that the participants share about the situation they are in. Pomerantz focuses on three types of sequences: (1) sequences in which the inferred purpose of a polar question or a fixed choice question is to seek explanation, (2) sequences in which the inferred purpose of a request for confirmation is seeking response to an associated query, and (3) sequences in which the inferred purpose of an open query (question-word question) is to seek information about a specific matter. Pomerantz (ibid.: 75) discusses how recipients address the question why that now in their answers. She emphasizes the relevance of understanding “specifically which prior bits of talk, action, gesture, shared understanding, etc. are relevant for making sense and inferring the purpose of a current utterance” (ibid.: 75). In this paper, we will continue the line of thinking represented by Walker, Drew, and Local (2011), and by Pomerantz (2017). Our aim is to qualitatively analyse answers that do not contain a separate affirmative or negative element (a repeat of the verb of the question, or a positive or negative particle) in everyday conversations in Finnish. We will show that in formulating an answer to a question, speakers rely on the contextual circumstances and make inferences about the
303
304
Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen
purpose of the question. Accordingly, the grammatical design of the answer is only one of the factors that should be taken into account in considering the ‘success’ in the outcome of the Q-A pair. We will focus on answers to polar interrogatives. Thus, no claims are made about answers to other types of polar questions (such as, declaratives) nor to question word questions. Interrogative polar questions are morpho-syntactically marked in Finnish: unlike declaratives, they are verb initial, and the question clitic -kO is attached to the verb.1 Polar interrogatives are answered minimally with a repetition of the verb used in the question (Example (1)), or in certain sequential contexts with a particle (Example (2)) (on question types and their responses, see Raevaara 1993; Sorjonen 2001a, b; Hakulinen 2001; Laury 2018).2 A negative response is implemented by the negation verb (an auxiliary verb). For all the minimal response forms (verb repeats, negative auxiliaries, and particles) we will use the term polar item. Answers that include any of these items align grammatically with the polarity of the question. These answers would be interpreted as type-conforming in terms of Raymond’s schema. Excerpt 1. (invented) 01 QUE: ->
Mene-t-kö konsertti-in? go-2SG-Q concert-ILL Are you going to the concert?
02 ANS: =>
Mene-n. | E-n (mene). go-1SG NEG-1SG go Yes. | No.
Excerpt 2. (Tiina: With whom I am talking, 3–4; Sorjonen 2001b: 412–413) 01 ANS:
Voi-da-an me tul-la. can-PASS-4 we come-INF We can come
02 QUE: ->
Voi-tte-ko, can-2PL-Q Can you
03 ANS: =>
Joo::. PRT
Yes
1. In the marking -kO, the capital letter denotes either of the two possible vowels, o or ö. This variation is caused by vowel harmony: the quality of the stem vowels determines the quality of the clitic vowel (front vs. back vowels, e.g. mene-t-kö ‘are you going’, tule-t-ko ‘are you coming’). 2. Notice that in these cases the questions do not contain a separate 2nd person personal pronoun, as is common in polar interrogatives (in Example (1) the singular pronoun sinä, or colloquial sä; in Example (2) the plural te). However, the person is indicated with the person ending at the end of the verb: in Example (1) (line 1), sg2 indicates that the t is the 2nd person singular person ending, in Example (2) (line 2), pl2 indicates that tte is the 2nd plural person ending.
Chapter 10. Responding to polar questions without a polarity item
Our data consist of interrogative polar questions and their answers in nearly a hundred hours of telephone conversations in Finnish. We have examined questions designed as grammatically positive interrogatives, since negative questions are based on different kinds of presuppositions; we have also ruled out interrogatives that are used as requests for action or as rhetorical remarks. The focus in this paper lies on approximately 40 instances where a polar question receives an answer that contains no separate explicit polarity item. We will investigate what kinds of inferences the recipient has made about the purpose of the question in its given context of occurrence. The questions we examine rarely occur at the beginning of a larger activity. Consequently, for understanding the purpose and the action done by the co-participant with their question, the recipient makes use of the prior talk and the shared history with the co-participants concerning the matter at hand. Neither does the question-answer sequence typically end a larger activity. Rather, it may be followed by exchanges that pursue the ongoing activity for various reasons. We will begin by examining answers that are unequivocally interpretable as either affirmative or negative, even though they contain no polar response form (cf. what Walker, Drew, and Local (2011) call ‘direct’: you cried to him on the phone – not on purpose.). These answers form two groups: one in which the polarity is implied (Section 2), and another in which the answer is produced as explicitly negative or affirmative but in which the polarity is presented as relativized (Section 3). We then turn our attention to cases in which the response to a polar question does not express polarity. The first set of these, discussed in Section 4, consists of cases where the response leaves the evaluation sought by the question to the questioner to make. In the second set (Section 5), the response treats the presuppositions of the question as invalid.
2.
Implied polarity
One context for an answer that implies ‘yes’ or ‘no’ without explicitly stating it is found in sequences where the question asks for information that can be understood as a precondition for a main action to come. In the following example, the question (line 3) is located towards the end of making arrangements about a present for a friend. Excerpt 3. (HY0094_Secondphase (SG112) 00:02:49, telephone) 01 ANN:
kiitos et te hommaa-tte ni .hh mää tua-r thanks PRT you.PL get-2PL PRT I bring-1SG thanks for getting (‘it’) so .hh I’ll bring the
305
306
Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen
02
raha-a sit vaikka< (0.3) money-PAR then PRT money then say, (0.3)
03
>oo-k-sä huomenna koto-nawill you be at home tomorrow oo-k-s sää soit[ta-nu Jari-lle,] be.SG2-Q-CLI you.SG call-PPCP 1nameM-ALL have you cal [led Jari], [o.vhhhhhhhhho
04 ERI: =>
]
05
= ööhh hh Jari o-n hankki-nu mei-lle 1nameM be-3SG get-PPCP we-ALL = um hh Jari has got us
06
ne
lipu-t? thyhh ticket-PL the tickets? thyhh DEM3.PL
07 PIK:
[o-n hankki-nu?] be-SG3 get-PPCP [he has? ]
08 ERI:
[thhhhhh [thhhhhh
09 PIK:
okei? okay?
10 ERI:
.mthhh mut tota mu-l o-n nyt vaan sit se but PRT I-ADE be-3SG now just then DEM3 .tch but like now I just have that
11
kulku-yhteys semmone ongelma. travel-connection DEM3.ADJ problem transportation kinda problem.
12 PIK:
no.
] joo. ] yeah.
PRT
Pike’s turn in line 3 is her second question to Eri, posed after she receives no immediate answer to her first one. The first question, which asks about the program of the following day, is in the place of the reason for the call. In posing her second question, and by using the verb in the present perfect tense, Pike displays her understanding of the current relevance of Eri calling Jari, and treats it as recognizable by Eri. In formulating her answer (lines 5–6), Eri, in turn, shows her understanding of the motivation of Pike’s question and their shared understanding of the reason for Eri to contact Jari: it was Jari who was supposed to get tickets for all of them (the determiner ne ‘the’, line 6, Laury 1997). The fact that Eri begins the full clause response with the name Jari, and not with an anaphoric pronoun, makes the response more independent from the prior turn. The form, a locally initial reference form in a locally subsequent reference position, implies that Eri not only gives the information Pike requested but also begins a new activity, in which she eventually brings out her own obstacles in coming along (see Schegloff 1996; also Sacks & Schegloff 1979; C. W. Raymond, Clift and Heritage 2021).
Chapter 10. Responding to polar questions without a polarity item
By not initiating her answer with a polarity item, Eri responds to the main reason underlying the question of whether Jari had bought the tickets or not, and not to the less relevant issue of whether it was she who had called him (see also Example (14) in Walker, Drew, and Local 2011). The rising intonation towards the end of her response (line 6) can be heard as implying continuation on the matter that her response began to deal with (Ogden & Routarinne 2005). Orientation towards this can also be seen in Pike’s responses. She first requests confirmation (line 7), possibly seeking further elaboration without getting it. She then receives the response with a third position okei ‘okay’ (line 9), which also has a rising intonation that solicits further elaboration (cf. Koivisto & Sorjonen 2021). In the two examples in this section, the question is understood, in its context, to seek information that is crucial for whether a projected main action or course of action will be pursued or not. In the first case, the recipient made use of the immediately preceding talk in making sense of the question. In the latter case, the knowledge shared by the participants rested on their previous encounters. These extracts bear a resemblance to the sequences discussed by Pomerantz (2017) in which the inferred purpose of a request for confirmation is to seek response to an associated query. Instead of initiating their answer with a polar response, the recipients proceed straight on to providing the relevant information and thereby sustain the progressivity of the ongoing activity to which they are both committed. Whether the answer to the question is positive or negative is implied by the answerer, and it is inferable by the questioner.
3.
From polarity to scalarity
We now turn to answers that are designed in a way that evokes polarity through an element that is scalar. These answers can be heard as unequivocally positive or negative, but instead of deploying the binary frame, as proposed in the question, they treat the absolute positive and negative set by the question’s design as two points on a scalar cline. One means of creating scalarity in our data is using an adverbial phrase (AdvP) as a response, as in the following extract (line 7). In this example, the question ‘did you play’ by Niilo (line 5) is triggered by the immediately preceding talk about ‘casino games’ which were part of the economic turbulence in the last years of the Soviet Union. Excerpt 5. (KTA SG095_03_Uncle 00:20:43, telephone) 01 NII:
=ehh ehe s(h)ie-l(h) £nuole-vat kasino-peli-n ehh ehe DEM3.LOC-ADE lick-3PL casino-game-GEN =ehh ehe the(h)re they are licking the casino
309
310
Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen
02
noin [noi-ta p-£ [haavo-j-a. PRT DEM2.PL-PAR wound-PL-PAR game’s [kinda [wounds
03 VIK:
[joo,
si[inä tul-i miljoona-tappio-t come-PST.3SG million-loss-PL [yes, millions’ worth of losses were made there PRT
DEM3.INE
04
[(---)
05 NII:
[pela-si-t-ko sinä. play-PST-2SG-Q you.SG [did you play(‘gamble’)
06
(0.4)
07 VIK:
ihan pikkusen, just a.little just a little bit
08 NII:
hävi-si-t-köh, loose-PST-2SG-Q did you lose
With his question (line 5), Niilo seeks to discover whether his nephew Viki was involved in this large-scale casino business, that is, whether he gambled. Instead of using a positive or negative polar response, Viki answers with the adverbial phrase ihan pikkusen ‘just a little bit’. This phrase implies a positive answer but simultaneously depicts Viki’s own playing as less serious. In the context where being involved in ‘casino gaming’ and its results have been talked about disparagingly, with a positive answer, the answerer might have presented himself as one of the players who now ‘lick the wounds of the casino games’ (lines 1–2). On the other hand, opting for ‘no’, he would have denied having played at all. Through interpreting the question as involving a scale and answering with an adverbial instead of a polar response, he can be seen to deal with the implications of both of these answers. The sequence leads to a brief discussion about Viki’s success in ‘playing’. Example (6) illustrates another instance where a polar question is answered with a scalar AdvP. Mari and Kati, two students in teacher training, are talking about demonstration lessons, one of which Kati has ahead of her. The segment below is part of talking about their respective problems in preparing for their lessons and complaining about comments from their supervisors. Having received encouragement for the sketch of her performance (line 7), Mari turns to Kati with a question (line 8).
Chapter 10. Responding to polar questions without a polarity item
Excerpt 6. (KTA SG s18b_02_Blabbers2 00:06:20, telephone) 01 MAR:
.he
heh
£mm:.£ mm
.nh joo >sit mu-n täyty-y sit PRT then I-GEN must-3SG then .nh yes then I must then
02
hirvee-sti skarpa-ta ett-#e-n mä# (.) et terrible-ADV watch.out-INF that-NEG-1SG I that terribly watch out so that I don’t (.) that
03
mä näytä-n sillai .hh myönteise-ltä sie-llä (.) I look-1SG like positive-ABL DEM3.LOC-ADE I look like .hh positive there (.)
04
Arvo-n oppila-i-den ede-ssä.= 1nameM-GEN pupil-PL-GEN front-INE in front of Arvo’s pupils.=
05 KAT:
=£joo:.£ PRT
06 MAR:
.joo:,
mä panosta-n I invest-1SG I’ll invest in that, PRT
sii-hen, DEM3-ILL
07 KAT:
#panosta.# invest.IMP.2SG #do that.#
08 MAR:
.mt no selvä pyy. jännittä-ä-k-s sua. PRT clear case be.nervous-3SG-Q-CLI you.SG-PAR .tch well okay. are you nervous.
09 KAT:
no vä↑hän ↓taas. a.little again well a litt↑le bit ↓again. PRT
10 MAR:
£no:: ei PRT
11 KAT:
mitään syy-tä.£ any reason-PAR no reason at all.£
NEG.3SG
no: ei-pä nii. NEG.3SG-CLI PRT not indeed.
PRT
12 MAR:
niin.= PRT
right.= 13 KAT:
=mm:.
14 MAR:
nonii:. PRT
15 KAT:
[no mutta näh-dä-än ehkä huome-nna. PRT but see-PASS-4 maybe tomorrow-ESS [well but we’ll see maybe tomorrow. [( )( )
311
312
Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen
16
näe-mme, see-1PL we’ll see,
In line 8, within the same turn where she finishes the sequence that dealt with her own future plans, Mari moves to ask Kati if she is nervous about her shortly upcoming demonstration lesson. This shift, turning to the co-participant, can be seen as a step toward closing the phone call; the problems, in particular Mari’s own that have been the reason for her call, have been amply dealt with. Kati orients to the question as acknowledging her situation as causing nervousness. She formulates her implicitly positive answer with three items: the initial particle no ‘well’, the adverb vähän ‘a little bit’, and the clausal adverb taas ‘again’. This last item implicates that the topic of being nervous is something that they have discussed previously in similar situations. Answering the question merely ‘yes’ or ‘no’ would ignore this shared experience. The particle no, in this particular environment, implies that bringing up the topic of being nervous was not expected as the next relevant course of action by Kati (Vepsäläinen 2019: 135–140). Furthermore, the adverb vähän presents the nervousness as not only commonplace in this kind of situation but something with which one can cope, indicating that Kati is not seeking extensive consolation from Mari. The response has a ‘no news’ character also due to its stylized prosody, where the second syllable of the word vähän ‘a little’ is produced with a step-up (indicated with an arrow up), and the pitch comes down when the last word taas ‘again’ is uttered (indicated with an arrow down; Ogden, Hakulinen, and Tainio 2004). Together with the turn-initial no (cf. Vepsäläinen 2019: 135–140), the turn implies ‘no big deal’ – a stance that the issue is not worth pursuing. Through responding with the three items instead of just an ‘unmarked’ polarity item, Kati both maintains the collegial bond between them by admitting her recurrent nervousness and shows trouble resistance and readiness to close the topic of her imminent plans by downplaying the feeling. Subsequently, the topic and the exchange are jointly brought to an end. Contrary to Examples (5) and (6), in the following extract the scale in the answer is extended beyond the reference point in the question. Niilo works in a charity organisation that collects donations to a children’s hospital in Soviet Estonia, and Viki has been donating to the cause in the past. The topic has turned to the charity’s continuing quest to provide Soviet hospitals with supplies. At the beginning of the segment, Niilo urges Viki to save any superfluous office supplies at his workplace for charity (lines 1–6). Excerpt 7. (KTA SG095_03_Uncle 00:11:11, telephone) 01 NII:
nii että jos, (.) jos jos noin o-n: (.) kuorma that if if if PRT be-3SG load so that if, (.) if if there is: (.) a load PRT
Chapter 10. Responding to polar questions without a polarity item
02
kaatopaika-lle meno-ssa tällas-ta nin noin landfill-ALL going-INE DEM1.ADJ-PAR PRT PRT of this kind of stuff going to the dump pit
03
noi-ta ph £paneta syrjään noi(n)£ DEM2.PL-PAR ph put-CUR.IMP.SG2 aside PRT then have (it) £put aside£ .hh#h (.)
04
niin< niin< noin,(0.2) se: se se o-n noi-ta pPRT PRT PRT DEM3 DEM3 DEM3 be-3SG DEM2.PL-PAR ? so so so (0.2) it it it is those p- for it
05
si-lle si-lle kyllä löyty-s, (0.2) sillon DEM3-ALL DEM3-ALL PRT find-COND.3SG then for it would surely be found (0.2)
06
käyttö°-mahdollisuuks-i-a°, use-possibility-PL-PAR then possibilities to use
07 VIK:
.hh#h (.)
no
mite-s se o-n< o-o-n-k-s se how-CLI DEM3 be-3SG b-be-3SG-Q-CLI DEM3 well how is it< ha-has it been PRT
08
ote-ttu hyvin vastaan sie-llä sitte, take-PPPC well DEM3.LOC-ADE then well-received there then,
09 NII:
.mt .hhh kuule e:ttä noin< toi o-n, .hh se listen.IMP.2SG that PRT DEM2 be-3SG DEM3 .tch .hhh listen like< that has, .hh it
10
o-n (0.2) ol-lu hei-lle ihan henki ja elämä be-3SG be-PPC they-ALL quite spirit and life has (0.2) meant the world to them
11
noin, (0.2) mone-ssa-kin kohta-a PRT many-INE-CLI place-PAR like (0.2) in many ways ((further talk ensues about the conditions in the children’s hospital))
In lines 7–8, Viki inquires whether the donations have been well received. The recipient (Niilo) does not respond with a polarity response but construes the question as scalar, producing a response that can only be interpreted as a positive answer. However, while in Examples (5) and (6) the answer was located somewhere on the scale from ‘no’ to ‘yes’, in this one, the scale is extended beyond the reference point ‘well’ offered in the question (line 8). Niilo responds by stating that the previous donation has ‘meant the world’ (line 10) to the donees. In fact, he uses the idiom henki ja elämä, ‘spirit and life’, which can be traced to the Gospel of John (6:63). This well-established biblical idiom produces an extreme and thus expands the frame of evaluation proposed in the question, implying that the recipient finds it insufficient for an accurate depiction of the situation. In this
313
314
Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen
sense, the idiom has a similar function as the adverbial phrases in the previous examples. Moreover, it should be noted that in this case, due to the context in which the conversation is being conducted, a simple verb repeat on ‘has (been)’, comparable to the English particle ‘yes’, might be perceived as inadequate not only by the recipient himself, but also by the questioner. With his question about the previous donation (lines 7–8), Viki draws attention to the context of charity. Charity as a social concept can be seen as reciprocity between need and giving as well as gratitude and gratification. Arguably, in order for the donation to take place, the potential donor must first regard the need as worthy of action. Gratitude, expressed in advance or afterwards, results in a positive experience vital to the continuation of the exchange. Viki’s question (lines 7–8) could even be heard as fishing for gratitude. Niilo recognizes Viki’s action and proceeds accordingly (line 10): he expresses gratitude by emphasising the significance of Viki’s donation. Within this frame, a simple acknowledgement of the fact that the donation was well received might have come across as insufficiently grateful, even to the point of rudeness. The initial, commending statement is followed and endorsed by Niilo’s further account of the dismal shortage of medical supplies in Tallinn’s hospitals. As in answers that imply polarity, these answers reveal a pursuit of cooperation that stems from the recipient’s interpretation of the underlying agenda of the question. Scalarisation allows the answerer to refine or expand the frame in which the subject matter is being discussed, and thus reconstruct the participants’ mutual understanding of it. Even if the reconstruction is often to the answerer’s own benefit (as in Examples (6) and (7)), its primary function in our data appears to be co-operative: scalarised answers display the answerer’s effort to reinforce the participants’ mutual understanding of the on-going action. They are treated as valid by the questioner, in many contexts arguably even more so than a response with a polar item would be. The orientation to context and co-operation continue to play an important role as we turn our attention to answers offered to evaluation-seeking questions.
4.
Answering requests for evaluation
The answers we will examine in this section are given to questions that contain an evaluation of a referent or a state of affairs, that is, properties or dimensions of an object, such as distance or price (e.g. far, expensive). Grammatically the question is designed as polar interrogative, whereas its focus is on a relative and scalar evaluation, typically expressed with adjectives (e.g. Is it cold there?). Scalarity can also be expressed by adverbs, such as much or often. (Hakulinen et al. 2004: 603–605).
Chapter 10. Responding to polar questions without a polarity item
Accordingly, requests for evaluations pursue no ‘objective’, absolute answers (e.g. −15 °C) but typically rely on subjective assessments or opinions (e.g. terribly cold). However, evaluation requests designed as polar interrogatives may also receive factual, objective responses whose implied polarity, if existent, is obscured (cf. “bias transformations” in Stivers and Hayashi 2010). Responses like this are, in fact, the most typical ones to requests for evaluation in our data. Asking about the relative value of the price of something appears to be a common, forward-looking practice in the service of some future action (cf. Schegloff 2007: 106–109 on pre-seconds). Example (8) is a case in point. In line 4, Make returns to the topic of a masquerade which Elli has mentioned earlier. The extract is from the beginning of the call (the turns in lines 1–3 concern the technical quality of the phone call). Excerpt 8. (HY0079_Call me (SG111_2b3) 00:00:21, telephone) 01 MAK:
no niin. nyt ei säre. .heh[hh PRT now NEG.3SG crackle okay. now (it’s)not crackling. Hehhh PRT
02 ELL:
[°°no ni. minä PRT PRT I [okay. I’m
03
[kuiskaa-n nytte.°° ] whisper-1SG now going to whisper now]
04 MAK:
[ni mitä, £n#e:# naa]miaise-t o huomenna£. PRT what DEM3.PL masquerade-PL be.3SG tomorrow so how (was it), the masquerade is tomorrow
05 ELL:
joo. hh £tuu-t-sä katto-o£. come-2SG-you.SG watch-ILL yes. hh are you coming to watch (it) PRT
06
(0.4)
07 MAK:
mm::, on-k-s sinne hirvee-t lipu-t. mm::, be-Q-CLI DEM3.LOC.ALL awful-PL ticket-PL mm::, are the tickets awful(ly expensive)
08 ELL:
.ehh ö viistoista. uhm fifteen
09
(0.4)
10 MAK:
viistoista.[.hh ou n]ou::, h fifteen. .hh oh no. .hh
11 ELL:
[mmh.
12 MAK:
.hh
]
.mh Ari soitt-el-i just etotäo 1NAMEM call-FRE-PST.3SG just that .mh Ari just called so he
315
316
Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen
13
pyys tänää pyörähtä-än to-ssa ask.PST.3SG today twirl-INF.ILL DEM2-INE asked to come around today
14
kaupungi-lla o iltapäivä-llä?o city-ADE afternoon-ADE to downtown in the afternoon?
With his question in line 7 Make asks whether the tickets to the masquerade are ‘awfully expensive’. His question is a response to Elli’s inquiry (line 5) about Make attending the ball, which, in this context, might be heard as a possible invitation. Instead of responding with a polarity item, Elli states the exact numerical price ‘fifteen’ (line 8). Make then formulates his turn in line 10 as a repetition of the number followed by a display of his affective stance (‘oh no’) to the price of the tickets. This kind of post-answer evaluation by the questioner demonstrates that both parties orient to the matter as negotiable. Almost immediately after this response, Make changes the topic (line 12), although he later returns to it several times by bringing up other potential obstacles to him attending the masquerade. The design of Elli’s response (line 8) is determined by the context in which it is uttered. She treats the numerical information, fifteen, as a relevant answer, drawing on the preceding interaction: Instead of immediately responding to Elli’s possible invitation (line 5), Make initiates an insertion sequence (pre-second) to ask about the price of the tickets. With the lexical design of his question, Make displays his stance to the preferred price value: the choice of the modifier, an extreme formulation hirveet ‘awful’, displays a stance that ‘awfully expensive tickets’ would be undesirable. By responding with a polarity item, Elli would treat herself as having the knowledge and authority to evaluate the expensiveness of the tickets on behalf of Make. Furthermore, by responding with on ‘yes’, she would run the risk of discouraging Make from attending the party, and by responding with ei ‘no’, she would align with what the design of the question seems to suggest (‘no’ to ‘awfully expensive tickets’), and encourage Make to attend. In this context, her response type can be seen as showing other-attentiveness, rather than being a way of ‘pushing back’ on the terms of the question: she leaves it to the recipient to make his own evaluation with respect to the price and its effect on his attendance. A numeral is offered as a response to a scalar question also in the following example. The extract is from a telephone call in which the participants, Viki and Niilo, talk and complain about the current economic situation in the late 1980s. Niilo’s question in lines 7–9 ensues from talk about the prevalent process of discharging employees and of the unemployment caused by the recession.
Chapter 10. Responding to polar questions without a polarity item
Excerpt 9. (KTA SG095_03:2–3_Uncle 00:01:21, telephone) 01 NII:
ai jaa >joo-joo< .hh nii että kato, (.) si-tä PRT PRT PRT that watch.IMP.2SG DEM3-PAR oh >yeah yeah< .hh so you see, (.) PRT PRT
02
o-n ä: käytä-vä työ-ssä niin kauan ku be-3SG ? go-PTCP work-INE PRT long PRT one has to uhm go to work as long as
03
työ-tä on, work-PAR be there is work
04
(0.4)
05 VIK:
se: on nykyään(h) (.) pak(h)koh >y[hyhy< be nowadays necessity it is these days an obligation DEM3
06 NII:
[juu, PRT
yea 07
on-k-s muuten noin tei-llä #tää::: tää# be.3SG-Q-CLI by.the.way PRT youPL-ADE DEM1 DEM1 has this by the way (‘at your workplace’)
08
(.) vaihe (te-) teh-ny paljo halla-a phase do-PPCP much night.frost-PAR #this this# (.) phase (wre-) wreaked much havoc
09
tää,(0.6) L:AMA. DEM1 recession this,(0.6) recession.
10 VIK:
.mt kyl sie-ltä::: (0.4) sa-i kakskyt-kaks PRT DEM3.LOC-ABL get-PST.3SG twenty-two .tch indeed from there (0.4) twenty-two:::
11
henke-ä ken°kä-ä. °person-PAR boot-PAR people got the boot
12 NII:
°kaks-kaks,° twenty-two
13 VIK:
joo, se o aika paljo,= be quite much yeah, it’s quite a lot PRT DEM3
14 NII:
nii niin niin [niin ei-hän TEI-Llä nii]ö-vöö-ö PRT PRT PRT NEG.3SG-CLI you.PL-ADE yes yes yes yes youPL don’t uhm PRT
15 VIK:
[tois-ta viikko-o sitte,] second-PAR week-PAR ago over a week ago,
317
318
Heidi Vepsäläinen, Anna Sundqvist, Marja-Leena Sorjonen & Auli Hakulinen
16 NII:
henkilökunta-määrä niin suuri oo,= staff-amount PRT large be have that large amount of staff
With his question in lines 7–9 Niilo inquires whether the recession has wreaked much havoc in Viki’s workplace. The answer (lines 10–11) is given by stating the number of employees who have been fired. The turn, including the numeral twenty-two, is initiated with the particle kyl which, in a full sentence answer, projects that the subsequent talk will provide a response that leans towards a positive answer (see Hakulinen 2001). Therefore, the particle may give Niilo a hint about Viki’s stance despite Viki not explicating it. Given the preceding talk and their shared knowledge about Viki’s workplace (see also Niilo’s subsequent statement in lines 14 and 16), Viki may expect Niilo to share his stance. Nevertheless, he gives Niilo an opportunity to evaluate the situation for himself and possibly to express his affective stance regarding his evaluation. In the same way as in Example (8), the questioner repeats the number in the answer (line 12) and in doing so, treats it as yielding remarkable information: here, Niilo expresses his affective stance by repeating the number sotto voce (cf. oh no in Example (8); see also Stivers & Hayashi 2010, Example (5), Stevanovic, Hakulinen, and Vatanen 2020, Example (6)). The answerer Viki responds to the repeat (line 13) with a turn that is designed as a confirmation ‘yes, it is quite a lot’, which explicates the stance he has heard in Niilo’s repeat. In contrast to the previous case, this time the questioner proceeds to discuss the extent of the havoc by asking how many employees Viki’s workplace had before the layoff, and the topic continues for a while after that. In the following Extract 10, a type of behaviour is described and its normality placed under evaluation. Before the excerpt, Eva has spent quite a while describing the behaviour of someone she knows to be a member of Jehovah’s Witnesses. Aki has mentioned that his brother is a member of the same denomination, and he has shared his knowledge of the differences between Jehovah’s Witnesses’ beliefs and those of Evangelical Lutherans, the largest religious group in Finland. They have now launched into a disparaging evaluation of the differences in the Bible script used by the Jehovah’s Witnesses. Eva’s question in line 1 initiates a return to the topic of Aki’s brother, asking whether the brother is ‘a normal person’. Excerpt 10. (HY0065_Diver (SG098) 00:15:44, telephone) 01 EVA:
no
miten se su-n velje-s how DEM3 you.SG-GEN brother-POSS.2SG well how does the brother of yours PRT
02
käyttäyty-y o-n-k-s se: behave-3SG be-3SG-Q-CLI DEM3 behave is he
Chapter 10. Responding to polar questions without a polarity item
03
normaal(h)i ih hi[h [ihminen, normal ih hi[ [human being?
04 AKI:
[.thh[h
ne pidä mitään yhteyt-tä Georgi:na. 03 (1.0) 04 HV: --> An:d you’re spelling that, 05 (0.5) 06 M: Er g-e-o-r-g-i-n-a. 07 (1.1) 08 M: °Yeah,° 09 (0.3) 10 HV: --> An:d (.) what’s your (.) um: big girl’s name, 11 M: .hh Alison,
Chapter 11. Renewing a social action in US primary care
Each subsequent question (lines 4 and 10) – which are part of a form-filling process directed to the acquisition of institutional data – is prefaced by the word And. In this way, the HV links these questions to the previous question-answer sequence, constituting the series of question-answer sequences as a larger-scale course of action (Heritage & Sorjonen 1994). Second, linkages across turns and sequences are consolidated through what Sacks (1992: 716–721) termed ‘tying techniques’ that build cohesion between turns. While there are many resources for the management of cohesion (Halliday & Hasan 1976), anaphoric and cataphoric reference are prominent among them. For example, in (1) above, the HV’s use of that in line 4 references the mother’s name for her new baby in line 2 and, together with the turn-initial And, consolidates the second question as an expansion of the first. Relatedly, the HV’s contrastive reference in line 10 to the mother’s big girl’s name trades off the previous reference to the new-born (lines 1–2) and, again together with the And-preface, sustains the overall focus of the activity as the collection of ‘face sheet information.’ As noted above, the significance of these supra-sequence activity structures is that they influence recipients’ understandings of, and thus responses to, the particular social actions implemented by the polar questions that constitute such activities (Robinson 2013; Montiegel & Robinson 2022). For example, in (2) below, where a child has an upper-respiratory infection, and the child’s grandmother is asked if the child has a history of wheezing (line 1), the following transpires: Excerpt 2. (3.6:506:Stivers 2007: 58) 01 02 03 04 05 06
DOC: GMA: GMA: DOC:
Does he have uh history of wheezing? (0.5) No. (.) He doesn’t have asthma. (Okay.)
In addition to providing a disaffirmative response No (line 3), the grandmother expands her response to rule out a diagnosis, which is one that would not normally be treated with antibiotics (line 5). In doing so, she demonstrates that she understands that the question is being asked in pursuit of making both a diagnosis and a related treatment recommendation (see also Lee 2011). In what follows, we show that participants’ orientations to the activity context of a specific polar question, and the action it implements, is central to the design of their responses. In our data, some of our cases accord with Raymond’s (2003) prior research in that patients respond to physicians’ unspecified offers to help – implemented through polar interrogatives, or sometimes, as in (3) below as a phrasal question (line 1) – with polar yes-/no-type responses (line 2) and a subsequent specific request (line 2), as in (3):
331
332
Jeffrey D. Robinson & John Heritage
Excerpt 3. (PCT: 23–01) 01 02 03
DOC:
Any other questions Yeah what’s causing that severe rash It’s probably a yeast infection…
PAT: DOC:
However, our data also contain a substantial number of cases in which questions formed as polar interrogatives (lines 1–2) are responded to, as in (4), with just a specific request without a polar response (lines 5–6): Excerpt 4. (Swollen Lips; MC.18.11) 01> 02> 03 04 05 06 07
DOC: PAT: PAT: DOC:
‘S=there anything else that you wanna disc[uss with me tod[ay [((compress lips)) [((opens mouth)) (0.6) ((780 ms.)) I don’t know if I- (0.4) I been having swelling of thuh lips, my li[ps really swe:ll, [((Nods))
Our research question concerns when, or under what circumstances, patients respond with or without a type-conforming turn-initial response. We compare responses to the question Is there something/anything else you’d like to address in the visit today? under circumstances we describe below. In a nutshell, we use these unspecified offers to help as a case study to determine if polar questions always make yes-/no-type answers conditionally relevant and, if not, when, why, and with what consequences?
3.
Data
Our data are drawn from a previous study (Heritage et al. 2007) and involve the openings of US primary care visits. As depicted in the schema below, these visits are begun with the medical activity of opening (Robinson 1998, 2003, 2006) and then progress to that of problem presentation (Heritage & Robinson 2006a; Robinson 2001, 2003). Ordering of trained events 1. Activity of Opening 2. Activity of Problem Presentation a. Physician solicits initial reason-for-visit with Wh-question b. Patient presents initial reason-for-visit c. Physician solicits additional reasons-for-visit with polar question In our study, we trained physicians to do two things. First, physicians were trained to initiate the problem presentation activity by soliciting patients’ reasons-for-
Chapter 11. Renewing a social action in US primary care
visit with ‘open’ Wh-questions (see 2a in the above schema; Thompson, Fox, & Couper-Kuhlen 2015). These include questions such as What brought you here today?, What can we do for you today?, What’re you here for?, and How can I help you?. While most physicians did this, in a minority of cases they diverged from our training and instead initiated problem presentation by asserting patients’ reasons-for-visit for confirmation (Alright, sore throat and runny nose for four days?, I know you had like a pair of migraine headaches?; Heritage & Robinson 2006b).2 We will argue that these two solicitation practices (i.e., Wh-questions vs. assertions for confirmation) have different interactional repercussions for how physicians’ subsequent polar questions (described next) are understood. Second, once patients finished presenting their initial reason-for-visit, physicians were trained to solicit additional reasons-for-visit with one of two polar questions (2c in the schema above): Is there something (or anything) else you’d like to address in the visit today? This polar question implemented an implied and unspecified offer to help (Robinson 2001, 2003).
4. Results and analysis A substantial number of patients did not present additional reasons-for-visit and responded to physicians’ solicitations of additional reasons with no-type answers, which were treated as sufficient (2c in schema). The observations motivating this chapter are made on the 39 visits in which patients responded affirmatively with additional reasons. Table 1 presents the distribution of these 39 cases according to whether or not: (1) physicians solicited initial reasons-for-visit with Wh-questions (2a in schema) or assertions for confirmation; and (2) patients’ answered physicians’ polar questions with yes-type particle components (note that nodding was considered to be a yes-type answer).3 Our initial observation was that, in 22/39 (56%) of the visits, patients’ affirmative responses did not include yes-type answers, vocal or embodied (e.g. with a head nod). This distribution, in and of itself, is at variance with Raymond’s (2003) principle of type-conformity. Our initial observation was elaborated by a second, which is that, in cases where physicians initially solicited patients’ reasonsfor-visit with Wh-questions, patients’ responses to physicians’ subsequent polar questions did not include yes-type answers 83% (20/24) of the time, which clearly departs from the principle of type-conformity. However, in cases where physi2. These questions were based on information from pre-consultation interviews conducted by practice nurses. 3. Just over 50% (9/17) of the yes-type responses were accompanied by nods.
333
334
Jeffrey D. Robinson & John Heritage
Table 1. The presence of physicians soliciting reasons-for-visit with Wh-questions and patients responding to subsequent polar questions with yes-type answers Did physicians initially solicit reasons-for-visit with Wh-questions or assertions?
Did patients’ affirmative responses to physicians’ subsequent polar questions include Yes-type particles (including nodding)? Included
Not included
Wh-question
4 (17%)
20 (83%)
Assertion
13 (87%)
2 (13%)
17
22
Total
cians initially asserted reasons-for-visit for confirmation, patients’ responses to physicians’ polar questions did include yes-type answers 87% (13/15) of the time, which very clearly supports the principle of type-conformity. Given that these distributions are extremely unlikely to have arisen by chance (Fisher’s exact test, p < .0001), we investigated further. Our chapter begins with two sections where we examine the bulk of polar questions that did or did not receive yes-type answers (i.e., the upper-right and lower-left quadrants of Table 1, respectively), and ends with a section wherein we examine discrepant cases (i.e., the upper-left and lower-right quadrants of Table 1).
4.1 Polar questions that typically do not receive Yes-type answers In the vast majority of cases in which patients’ responses to physicians’ polar questions do not include yes-type answers (i.e., the upper-right quadrant of Table 1), physicians had initially solicited patients’ reasons-for-visit with Wh-questions. Three points can be made about the social action implemented by such questions and its relationship to physicians’ subsequent polar questions. First, the action of these Wh-questions makes conditionally relevant patients’ presentations of their reasons-for-visit (Heritage & Robinson 2006b) but, in contrast to polar questions, they do not make yes-type answers conditionally relevant. Second, physicians and patients orient to this action as soliciting a ‘main’ or ‘chief ’ reason-for-visit. A consequence is that, while US primary-care patients frequently have multiple discrete reasons-for-visit (e.g., sinus infection and foot rash), they tend to restrict themselves to answering this question by presenting a single, discrete reason (e.g., only sinus infection; Heritage et al. 2007). Additional reasons (e.g., foot rash) may or may not emerge during subsequent parts of the visit (e.g., history taking, treatment, closing; Robinson, Tate, & Heritage 2016). Thus, third, patients’ presentations of a single reason-for-visit do not exhaust the
Chapter 11. Renewing a social action in US primary care
possible relevance of additional reasons. These observations highlight the fact that physicians’ subsequent polar questions (e.g., Is there something/anything else you’d like to address in the visit today?) expand the offer of help by inquiring into a set of relevant possibilities (i.e., additional medical issues) without naming one. While a yes-type answer affirms the relevance of such a possibility it does not sufficiently answer the question (i.e., as an action) because it does not name the issue itself, which is a requirement for the physician to progress the visit (Pomerantz 2017). We propose that, in visits where physicians begin by soliciting patients’ reasons-for-visit with ‘open’ (Wh-) questions, their subsequent solicitations of additional reasons-for-visit with polar-question-formatted offers to help (i.e., Is there something/anything else you’d like to address in the visit today?), were normally hearable as renewing the action implemented by the prior, ‘open’ (Wh-) questions. In this circumstance, the polar questions make conditionally relevant one of two different answer types: (1) disaffirmation of additional concerns; or (2) the nomination of additional reasons-for-visit, thus continuing or renewing the action implemented by the initial Wh-question). The former answer type, disaffirmation, is normally enacted by no-type answers. However, the latter answer type, the nomination of an additional reason-for-visit, is normally enacted by the nomination of additional concerns without the use of yes-type particle responses. In these cases – that is, where the polar question is understood as renewing the social action implemented by the prior Wh-solicitation – this context overwhelms or overshadows otherwise normal constraints of the question’s grammatical format (Schegloff 1984; Raymond 2003), and thus yes-type answers are not conditionally relevant. There are two ways in which physicians’ polar questions are understandable as renewing the social action implemented by prior Wh-solicitations. First, polar questions frequently contain turn-design features (e.g., word repeats) that ‘tie’ them back to physicians’ Wh-solicitations (Sacks 1992: 716 et seq.). Second, polar questions are commonly positioned immediately after patients’ presentations of their initial reasons-for-visit are possibly complete. The adjacent positioning of physicians’ polar questions to this activity provides a type of context in which to understand such questions as renewing or continuing the activity, and thus as soliciting further presentation of reasons-for-visit. Excerpts (5–8) provide examples of our 20 cases in which, following an open initial question, the physician’s polar follow-up did not receive a type-conforming response. In (5), after the visit’s opening, the physician initiates the activity of problem presentation by soliciting the patient’s reason-for-visit with an Wh-question (line 1):
335
336
Jeffrey D. Robinson & John Heritage
Excerpt 5. (Swollen Lips; MC.18.11) 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19
DOC: PAT:
PAT: DOC: PAT: DOC: DOC: PAT: PAT: DOC:
So what brought you here today. I think I have a si:nus infection? .hhhh I’ve had=a sore thro:at, a::n’ (.) s:tuffiness an’ tha:t for like thuh la:st week an’ a ha:lf, .hhh an’ I’m startin’=a cough up thuh yellow real dark an’ bro:wnish, (.) .h[hhhhhhh] [Okay, ] An’ I’m just ti:red all thuh ti:me. Okay. (0.6) ‘S=there anything else that you wanna disc[uss with me tod[ay [((compress lips)) [((opens mouth)) (0.6) ((780 ms.)) I don’t know if I- (0.4) I been having swelling of thuh lips, my li[ps really swe:ll, [((Nods))
Immediately after the patient possibly completes her problem presentation at line 6 (see also line 10) with a description of current symptoms (using present tense; Robinson & Heritage 2005), the physician acknowledges it with a shiftimplicative okay, (line 9) tempered by slightly rising intonation (Heritage & Clayman 2010: 113–115), followed by a downward intoned, completion-relevant okay. (line 11; see also Beach 1993; Mondada & Sorjonen 2021), and then solicits additional concerns with a polar question (lines 13–14). In this question: (1) anything else is understood relative to, and thus ties back to, the immediately prior reason-for-visit presented at lines 2–10; (2) discuss with me explicitly reinvokes the immediately prior activity of problem presentation insofar as it is in the service of ‘discussion’ in the form of history taking (Robinson 2003); and (3) the person and temporal reference forms you and today, as repeats from the physician’s prior Wh-question (line 1), contribute to the polar question being understood as an extension or continuation of the social action implemented by the physician’s prior Wh-question. In these ways, the physician’s polar question is understandable as renewing the conditional relevance of the action implemented by their prior Wh-question. Indeed, the patient answers by presenting an additional reason-forvisit (lines 17–18) and does so without a yes-type answer (vocal or embodied). Similarly, in (6), the physician initiates the activity of problem presentation by soliciting the patient’s reason-for-visit with an ‘open,’ Wh-question (line 1). The patient’s response (line 2) is solely embodied: she extends her right arm toward the physician and uses her left hand to raise her right sleeve, revealing and presenting her rash.
Chapter 11. Renewing a social action in US primary care
Excerpt 6. (Dermatologist; MC.13.14) 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17
DOC: PAT: DOC: PAT: PAT: DOC: PAT: DOC: DOC: PAT:
DOC:
.hhhh Well what can we do for you today. (1.4) ((bodily ‘shows’ or ‘presents’ her rash)) ] [ (S]o) (.) Hh=[Heh heh ( ] ) [ ] (0.6) .hhh
(.) .hhhh (0.4) Uh: other than thuh ra:sh are there any other issues (0.4) to disc[uss today?] [Uh : : ]m (2.0) I wan’ed duh get a referral, to a dermatologi[st,] ((re. freckles)) [ Mm] hm,
There are remarkable similarities between (5) and (6), even though they are conducted by different physicians operating in different clinics. In (6), immediately after the patient possibly completes her problem presentation (line 2), the physician acknowledges it with a range of okay responses while gazing at the patient’s rash (lines 3, 7, and 10), and then solicits additional reasons-for-visit with a polar question (lines 12–13). In this question: (1) other than thuh ra:sh contributes to the hearing of this question as extending/continuing the social action of presenting reasons-for-visit by raising the relevance of an additional, unmentioned category of the collection ‘reasons-for-visit’ (membership categorization, see Sacks 1972); (2) other issues is understood relative to, and thus indexes, the immediately prior reason-for-visit, the rash; (3) discuss explicitly reinvokes the activity of problem presentation; and (4) today, as a repeat of the physician’s today (see line 1) contributes to the polar question being understood as an extension/continuance of the social action implemented by the physicians’ prior Wh-question. In these ways, the physician’s polar question is understandable as renewing the conditional relevance of the previously initiated action of soliciting patients’ reasons-for-visit. Once again, the patient answers by presenting an additional reason (lines 14–16) and does so without a yes-type answer (vocal or embodied). Again in (7), which involves yet a different physician operating in a different clinic, the physician initiates problem presentation by soliciting the patient’s reasons-for-visit with an ‘open,’ Wh-question (line 1). This action makes conditionally relevant the presentation of reasons-for-visit, which the patient does at line 2: pap smea:r::. Excerpt 7. (Blood Tests; MC.08.10) 01 02
DOC:
((Opening)) Goo:d >what are we< doing for you toda:y. >what’re y’ here< fo:r.
337
338
Jeffrey D. Robinson & John Heritage
03 04 05 06 07 08 09 10 11 12 13
PAT: DOC:
PAT: PAT: DOC: PAT: DOC:
Pap smea:[r:: ] [Pa:p sm]e:ar okay before we go on with thuh pap smear are there some other issue you ( )’nna tal[k tuh me] about [(.) toda:y.] [Hh hh ] [.hhhh ] Uh::m .tch I wan’ed duh get some blood tests done for uh:m (.) cholestero::l= =Okay,= =Ess tee dee:(h) ((STD)) huh huh .h[h ] [Wh]y new- (.) new partner?
After acknowledging the patient’s reason-for-visit, pa:p sme:ar okay, the physician solicits additional concerns with a polar question (lines 5–6). In this question: (1) other issue is understood relative to, and thus indexes, the immediately prior reason-for-visit (i.e., pap smear), and this is made explicit by the physician’s preface: before we go on with thuh pap smear…; (2) before we go on with thuh pap smear contributes to understanding this question as extending/continuing the immediately prior activity of problem presentation; (3) talk tuh me about explicitly reinvokes the immediately prior activity of problem presentation (similar to discuss in Extracts 5–6, above); and (4) toda:y., as a repeat of the physician’s toda:y. (>what are we< doing for you toda:y.; line 1), contributes to this polar question being understood, once again, as an extension/continuation of the action of soliciting patient reasons-for-visit implemented by the physician’s prior Wh-question. In these ways, the physician’s polar question is understandable as renewing the social action implemented by the physician’s prior Wh-question. The patient answers by presenting an additional reason (lines 8–9) and does so without a yes-type answer (vocal or embodied). In Excerpts 5–7 above, physicians’ polar questions are sequentially adjacent to the sequence of action initiated by physicians’ Wh-solicitations of patients’ reasons-for-visit. However, (8) demonstrates that such positioning is not necessary for polar questions to be understood as renewing the relevance of this social action. The physician initially solicits the patient’s reason-for-visit with the open question how (could/can) I help you. (line 1) and the patient begins to answer by disclosing his reason in a narrative format (line 2; Heritage & Clayman 2010: 127–132; Heritage & Robinson 2006a; Robinson & Heritage 2005; Ruusuvuori 2000: 115–142). Over the next almost two minutes (data not shown), the patient completes his problem presentation (regarding his injured hand/ wrist) and the physician takes a medical history of that problem, including a physical exam (Robinson 2003). Excerpt 8. (Wart on Thumb; MC.12.14) 01 02 03
DOC: PAT:
Bri:an how (could/can) I help you. .hhhh Uh: about two weeks ago:=hh I: slipped on some ice an’ fell.
Chapter 11. Renewing a social action in US primary care
04 05 06 07 08 09 10 11 12 13 14 15 16 17
DOC: PAT: DOC: DOC: PAT: DOC:
PAT: DOC: PAT:
. .((1 minute, 52 seconds)4 . .mtch (0.9) Wh:at t=uh: I would recommend is=‘at we (.) send you upstairs and get an ex ray.= =Mkay. (.) Okay, which is probably what- (0.3) [( ) you thought ] [What I=w’s expecting,] Yeah. .hhhh (.) are there some other issues you’d like me duh discuss as well. (.) ((200 ms.)) .hh ((132 + [222 ms.)) Uh::v (0.2) nothing you can [((thumb up, head down)) do about it now. but=I just (0.3) pro’ly (.) two months ago go:t a wa:rt on my thu:mb,…
At lines 4–5, the physician makes a test recommendation regarding the patient’s wrist, which the patient accepts (line 6). Close to this activity boundary (Robinson 2003), the physician solicits additional reasons-for-visit with a polar question (lines 11–12). In this question: (1) other issues is understood relative to, and thus indexes, the patient’s initially presented reason-for-visit (i.e., his injured hand/wrist); (2) as well contributes to the hearing of this question as extending/continuing the social action implemented by the physicians’ original Wh-question; (3) discuss explicitly reinvokes the earlier activity of problem presentation; and (4) the second-person pronoun you (of the contraction you’d; line 12), as a repeat of the physician’s original you (how (could/can) I help you.; line 1), contributes to this question being understood as extending/continuing the social action implemented by the physicians’ original Wh-question. In these ways, the physician’s polar question is understandable as renewing the social action of open-endedly soliciting the patient’s reasons-for-visit. Again, the patient answers by presenting an additional reason (lines 14–17) and does so without a yes-type answer (vocal or embodied). In sum, participants’ processes of action ascription (Deppermann & Haugh 2022; Levinson 2013) involve, first and foremost, situating actions’ turns in their local sequential contexts, which not only includes the (immediately) prior (sequence of ) talk, but also activities (i.e., sequences of sequences) and the overall-structural organization of encounters (Heritage & Atkinson 1984; Robinson 2013). As a result, in Excerpts 5–8, physicians’ polar-question-formatted offers to help are understood by patients as renewing the conditional relevance of a prior sequence-initiating action (e.g., What can I do for you today?) that specif-
4. During this time the patient continues with the problem presentation, and the physician takes a history of the problem and engages in a physical examination.
339
340
Jeffrey D. Robinson & John Heritage
ically made conditionally relevant the presentation of reasons-for-visit. Ascribing this action to these polar questions appears to overwhelm or overshadow the otherwise normal constraints of their grammatical format (Schegloff 1984). Indeed, patients only answered these polar questions with yes-type answers 17% of the time (4/24 cases), and there are reasons to believe that the four apparently discrepant cases align with our analysis (see below).
4.2 Polar questions that typically do receive Yes-type answers In the vast majority of cases in which patients’ responses to physicians’ polar questions do include yes-type answers physicians initiate the activity of problem presentation not by soliciting patients’ reasons-for-visit with Wh-questions, but rather by asserting a reason-for-visit for confirmation (Heritage & Robinson 2006b). As a social action, rather than making conditionally relevant the presentation of reasons-for-visit as a first order of interactional business (as do Wh-solicitations), these assertions are B-event statements (Heritage 2010; Labov 1972), that is, declaratives that make conditionally relevant (dis)confirmation. Due to their sequential position immediately after visit openings, these assertions sequentially implicate problem presentation and, thus, at least some elaboration of the physician-asserted reason-for-visit and there are occasional cases, such as (9), where patients move immediately from confirmations (i.e., yes-type answers; e.g., mm hm, line 2) to such elaborations (lines 2–4; e.g., right above the (.) left cheek…).5 Excerpt 9. (Back Pain; MC.04.05) 01 02 03 04 05
DOC: PAT:
DOC:
So you’re here for your back pain? Mm hm (‘s) right above the (.) left cheek you know like to the left- to the left of the spine. It’s not the spine. Okay.
However, in the wake of physicians’ assertions, if patients have additional reasonsfor-visit they generally do not assert them then and there. It is more common
5. In contrast to cases like Excerpt 9, there are also cases where patients produce confirmations as transition-relevant answers, and where physicians understand such answers accordingly by progressing to the next activity of history taking. These types of cases, such as Excerpt A, make the point that physicians’ assertions (line 1) per se do not make problem elaboration conditionally relevant. Excerpt A. (P3:60:10) 01 02 03
DOC: PAT: DOC:
You’re having knee problems since Ju::ne. Yes. Okay what have you done for that. Since then.
Chapter 11. Renewing a social action in US primary care
for patients to confirm and wait for physicians to speak next (Excerpts 10 and 12 below), or for physicians to continue speaking and solicit additional reasons-forvisit with polar questions (which is a unique feature of the training discussed in the Data section, above; see Excerpt 11). In contrast to the polar questions in Excerpts 5–8, above, those in (10–12) are not hearable as tying back to, and renewing the relevance of, a prior social action that made solely conditionally relevant the presentation of reasons-for-visit, but rather to an action that made conditionally relevant at least (dis)affirmation. The polar questions in (10) (lines 5–6), (11) (line 4) and (12) (lines 33–34) have – as they did in Excerpts 5–8 – an information-seeking purpose of soliciting additional reasons-for-visit, and thus yes-type answers are insufficient. However, these polar questions make conditionally relevant at least (dis)affirmation and, contingent upon yes-type answers, the nomination of additional reasons-for-visit. We provide three examples. In (10), after opening the visit, the physician asserts a reason-for-visit for (dis)confirmation (lines 1–2). The patient affirms by nodding (line 3) and saying that’s right, (.) yeah, (line 4), thereby treating the assertion as having made conditionally relevant (dis)confirmation. Excerpt 10. (Nail Puncture; MC.07.07) 01 02 03 04 05 06 07 08 09 10 11
DOC: PAT: PAT: DOC: PAT:
DOC:
.hh See I- I know you’re he:re for=a follow up up from thuh hospitalization everyth[ing,= [((nods)) =That’s [right, (.) yea[h,] [.hh [ any: other issue:s th[at we need duh cover t[uh [day.] [((unlocks hands)) [((nods)) [ Yea]h, (.) uh (.) .h I have a (0.2) yesterday I helped my daughter a bit=h .h (.) and I (0.4) got=h (0.2) a na:il, (.) puncture [(in here)] [Oh my:. ]
The physician continues by asking a polar question (lines 5–6). The patient initially responds in an embodied way by beginning to unlock his hands (line 7) in preparation for rolling up his right sleeve to present another reason-for-visit (a na:il, (.) puncture, lines 8–10), thereby treating the polar question as making conditionally relevant the nomination of a reason-for-visit. Despite the fact that the patient is already on his way to presenting another reason-for-visit in an embodied fashion, he nonetheless additionally (and initially) answers by nodding and verbally affirming, yeah, (line 7), highlighting his orientation to the polar question as also making conditionally relevant (dis)affirmation. In (11), prior to the physician’s entry, the patient was seen briefly by a medical student who recorded the patient’s reason-for-visit as a follow up (line 2) on a prior medical problem. After opening the visit, the physician asserts this reason-
341
342
Jeffrey D. Robinson & John Heritage
for-visit (lines 1–2). The physician ends this assertion gazing at the patient (who is gazing back; Robinson 1998). Rather than waiting for the patient to respond – that is, after a brief pause of only 62 milliseconds (line 3), which is far less than a normal transition space of 200 ms (Levinson & Torreira 2015) – the physician quickly continues talking and asks a polar question (lines 4–5), relegating his assertion (lines 1–2) to a question-preface. Excerpt 11. (Pain and Irritation; MC.02.11) 01 02 03 04 05 06 07 08 09 10 11 12 13
DOC:
DOC:
PAT: PAT: PAT: DOC: PAT:
.hhhh Uh:m thuh studen’ doctor tells me that you’re in fer=a follow up, (062 ms.) Are there some other issues you’d like to discuss. (.) today. (250 ms) [(64 ms) [((begins to nod)) Yeah. (173 ms) [(95 ms) [((prepares to touch shoulder)) °°Okay.°° Gotta little bit duh: pain that I noticed (0.3) in my shoulder:…
The patient answers the polar question initially by affirming with a nod (line 7) and yeah (line 8), and subsequently by beginning to present another reasonfor-visit in an embodied way by raising his hand to touch his painful shoulder (line 10), thereby treating the polar question as making conditionally relevant both types of answers (the patient verbally presents his additional reason at lines 12–13). Excerpt 12 is slightly more complicated. During intake, the patient indicated that her reason-for-visit is a recurrence of back pain. During the visit opening (lines 1–21), the patient mentions that she did not work (i.e., she rested) during her holiday (line 6), which ultimately engenders positive evaluations from the physician (lines 7, 14–15, and 20). These evaluations represent possible closings to the opening phase of the visit (Schegloff 2007: 123–7). The physician then initiates problem presentation by asserting what the patient might have otherwise thought about the prognosis of her back given her restful holiday (lines 22–24; see the contrast marker but, which is replaced by the upshot marker so; Bolden 2006). This assertion, clearly in the patient’s epistemic domain (Labov 1972; Heritage 2012), invites confirmation from the patient (which the patient provides at lines 25–26 and 29) and projects the relevance of a ‘problematic realization’ (which the patient begins at lines 31–32; re. ‘At first I thought, but then I realized,’ see Jefferson 2004). Excerpt 12. (Blood Tests; MC.14.07) 01 02 03
DOC: PAT: DOC:
((knocks)) .hh .tch [Hi:: ] [Hi: Donna] how are you
Chapter 11. Renewing a social action in US primary care
04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
PAT:
33 34 35 36
DOC:
DOC: PAT: DOC: PAT: DOC: PAT: PAT: DOC: PAT: PAT: DOC: PAT: DOC: PAT: DOC: PAT:
DOC: PAT: PAT:
PAT:
I’m fi:ne how are you Goo:d how was your [holiday] [Goo:d ] it was uh: restful Good Didn’t go inda thee office once, -ehhhhh [heh] ((laughing)) [ Oh][: m[y::. ] [.hh [We had bets on i]t. (.) Yeah. -e[hhh heh [(heh) ] [eY:e::ah [but th]at’s goo:[:d. ] [It w]as good. (.) Yeah. (0.4) Ve:r[y [goo:d.] [.tch [Ye:ah.] .h [But then you though]t so you thought [( )] your back w[ould feel BE]tter.] [Yeah ] -hh ]hh (h)y(h)eah. (0.4) [Oh: my:[:] [.h .hhh [Y]e:ah. (.) .h But just thuh last couple=a weeks: it’s been more of a constant=h du::ll (.) pa:in …. . .((Patient finishes presenting her concern; 51 seconds)) . .mm=.h=.tch (0.2) Are there some other issues you[’d like to di]scuss [today? ] [ ] [.tch=.hh] Uh:m: hh (1.5) .mtch I just (.) feeling ti:red. …
Adjacent to the completion of the patient’s problem presentation, the physician produces a polar question (lines 33–34). Similar to (Excerpts 10–11, above), the patient treats this question as having made conditionally relevant both (1) (dis)affirmation, by saying, (line 35); and (2) the presentation of an additional reason-for-visit (lines 35–36). In sum, compared to Excerpts 5–8 (above), the sequential context for the polar questions in Excerpts 10–12 (above) is distinctive. In (5–8), the polar questions, embodying a variety of cohesion practices that tie them to the prior, are clearly hearable as continuations of a course of questioning that was begun with the physician’s initial reason-for-the visit question. In (10–12), by contrast, the polar questions follow a sequence in which the reason-for-the-visit was asserted by the physician for (minimal) confirmation. The subsequent polar question is thus heard as the first question the physician has asked and, therefore, an appropriate candidate for a polar response particle.
343
344
Jeffrey D. Robinson & John Heritage
5.
Discrepant cases
We now turn to deal briefly with one set of cases that are discrepant relative to our argument.6 In these four cases (upper-left quadrant of Table 1), subsequent to physician’s Wh-questions, patients do respond to physicians’ polar questions with yes-type answers. In two of these cases, patients’ yes-type answers are solely embodied in the form of head nodding, as seen in (13) at lines 16–17. The patient’s additional reason-for-visit (lines 16–20) involves how to lessen the chance of contracting her young son’s perpetual colds and flus. Excerpt 13. (Germ Vector; MC.18.09) 01 02 03
DOC:
14 15 16
DOC:
PAT:
PAT:
What brought you in here today. .mtch=.h uhm (.) I’ve ha:d (.) (w)- what I think is just a nasty cold,… . .((10 lines omitted; patient presents problem)) . Okay. .hh is there anything else that you wan’ed=a talk tuh m[e about toda[y ] [((1 Nod)) [.m]tch [uhm (.) jus’ more ge]nerally
6. The lower right quadrant in Table 1 represents 2 cases in which there was not an initial reason-for-the-visit Wh-question and the response did not contain an affirmative, yes-type answer. However, in one of these cases (Excerpt B below), the patient initially responds with a non-committal ‘epistemic hedge’ (Weatherall 2011) I don’t know (line 20). Excerpt B. (Stomach Flu; MC.09.17) 01 02 03 04
DOC: PAT:
So you’re here because for about a week you’ve been sick, huh? Right. Well, (.) it just started off as like a cough,… ((13 lines omitted; patient presents concern))
18 19 20 21 22 23 24 25 26
DOC: PAT:
DOC: PAT:
Is there anything else that uh we can discuss today? (.) >I don’ know< actually, today, (.) I started getting like- I feel like I’m getting a stomach flu. kind of. I don’t know. I’m kinda nauseous? Same time frame? About a week? or just today that stuff. No just today.
Perhaps due to the fact that the physician’s polar question – “Is there anything else that uh we can discuss today?” (line 18) – contains the negative-polarity item Any, which can conduce or bias the answer toward disaffirmation (Heritage et al. 2007), the patient – who does have an additional concern to present – initially answers with “>I don’ know