Information and affiliation: Disconfirming responses to polar questions and what follows in third position

+ Models

PRAGMA-4132; No. of Pages 14

Available online at www.sciencedirect.com

ScienceDirect Journal of Pragmatics xxx (2015) xxx--xxx www.elsevier.com/locate/pragma

Information and affiliation: Disconfirming responses to polar questions and what follows in third position Seung-Hee Lee * Department of English Language and Literature, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, South Korea

Abstract Polar questions offer a candidate proposition as a likely possibility, which can be disconfirmed in two different ways. They can be disconfirmed with what this paper terms ‘negation’ responses that merely negate the questioner’s candidate proposition, or alternatively, with what this paper terms ‘replacement’ responses that assert a revised state of affairs in replacement of the questioner’s proposition. This paper examines questioners’ conduct in third position following these two forms of disconfirmation in Korean conversation. In about 53% of 70 sequences from audio-recordings of ordinary calls and calls to an airline service, questioners produce a question in alignment with the prior disconfirming response in third position. In airline service calls, questioners produce repetitional questions in third position following both negation and replacement responses. They enact registration, establishing correct information on record. In ordinary calls, by contrast, questioners may offer another, revised question following negation responses, whereas producing repetitional questions following replacement responses. They orient to (re-)establishing a common ground with respondents. This paper shows that the different ways in which third-position questions are used can be conditioned by the nature of interaction as well as the form of disconfirmation. © 2015 Elsevier B.V. All rights reserved.

Keywords: Conversation analysis; Polar question; Preference; Response; Third position; Korean

1. Introduction A polar question, according to Bolinger (1978:104), sets up a ‘‘hypothesis for confirmation’’. It offers a candidate proposition as a likely possibility, incorporating a candidate answer that encodes the questioner’s expectations concerning the matter at hand (Pomerantz, 1988). It formulates a most probable, ideal or legitimate scenario about the matter that is primarily within the respondent’s epistemic domain, based on the questioner’s knowledge and expectations (Pomerantz, 1988). A polar question thus proposes the questioner’s candidate understanding and invites the respondent to affirm or reject that as a party with epistemic authority, while constraining the terms in which the matter at hand can be (dis)confirmed (Heritage and Raymond, 2012). Overwhelmingly a polar question gets a confirming response (Stivers et al., 2009). According to Sacks (1987), confirmation is a preferred action. It is produced not only more frequently but also faster than disconfirmation (Sacks, 1987), as evidenced in a recent 10-language study (Stivers et al., 2009). This preference for confirmation is observed by questioners as well as by respondents who design their response so as to maximize elements of confirmation and avoid or minimize disconfirmation (for review, see Pomerantz and Heritage, 2013). Questioners in particular, in constructing a polar question, design and propose their candidate proposition so as to permit the respondent to produce confirmation (Pomerantz and Heritage, 2013; Raymond, 2003; Sacks, 1987). In order to

* Tel.: +82 2 2123 5302; fax: +82 2 392 0275. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.pragma.2015.10.003 0378-2166/© 2015 Elsevier B.V. All rights reserved.

Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

2

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

permit confirmation, questioners have to take into account the respondent’s situations, behaviors, etc., and offer a particular state of affairs that is appropriate for the respondent being addressed. They have to design their polar question for the respondent they target and enable the respondent’s confirmation (see Raymond, 2003). Questioners can be held accountable for achieving such a question, particularly given the principle of recipient design (Sacks and Schegloff, 1979) along with the preference for agreement (Sacks, 1987). Thus, when a disconfirming response to a polar question is due, it [2_TD$IF]can not only require the respondent to provide a relevant elaboration and explain how the state of affairs is different, minimizing elements of disconfirmation. It can also indicate a ‘failure’ on the part of the questioner in achieving a recipient-designed question that permits confirmation. A disconfirming response can show that the questioner has failed to (correctly) take into account the respondent’s situations and offer a most likely hypothesis that is appropriate for the respondent. This paper examines two ways in which a polar question is disconfirmed and what questioners do in that context indicating their failure to propose a correct hypothesis. Consider extract 1 from a primary care visit, in which the doctor’s candidate proposition is disconfirmed with a mere negation. The patient is a middle-aged woman with an adult daughter. At line 1, the doctor produces a polar question concerning the patient’s marital status, incorporating ‘married’ as a candidate answer (Pomerantz, 1988). This shows the doctor’s understanding that, for this middle-aged patient with an adult daughter, being married can be a most likely or desirable possibility (Heritage, 2010).

At line 3, the patient disconfirms the doctor’s hypothesis with ‘‘No’’ alone. This response, although type-conforming (Raymond, 2003), is in fact uninformative and uncooperative in that it does not deliver information for which the question was asked. The doctor’s question is proposing one candidate marital status out of several options, such as divorced, widowed, etc., and the patient merely negates the single option proposed by the question without providing her marital status. This can make it necessary for the doctor to try again, e.g. by proposing another, revised hypothesis, especially given that the doctor has to get the information as part of a record keeping process. In third position at line 5, the doctor revises his initial hypothesis, proposing another marital status. The doctor renews the question to get the information, permitting confirmation at the second try. Alternatively, a questioner’s hypothesis can be disconfirmed by its replacement. In extract 2 from an opening of a school officer’s call to the parent of a student, the respondent replaces the questioner’s hypothesis with a new assertion. At line 3, the officer proposes the identity of the other party, inviting confirmation through the use of a rising intonation (Schegloff, 1979). In response, the mom produces a replacement of the identification proposed by the officer, displaying rejection of the officer’s proposal (line 5).

Given the mom’s replacement, the officer is not obligated to make another guess about the identity of the other party. At line 6, the officer instead enacts her registration of the new, replaced information by repeating the replacement (‘‘Missus Williams’’). Then the officer produces an apology, ‘‘I’m sorry’’, which shows an understanding that her ‘wrong’ guess about Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

3

the identity can be sanctioned or treated in some way problematic. The officer moves to identifying herself after dealing with the failure of her initial hypothesis. Thus, questioners’ candidate propositions can be disconfirmed in two different ways, with a mere negation (extract 1) or replacement (extract 2); and depending on the method of disconfirmation, questioners perform different actions in third position. They may re-try another hypothesis (extract 1) or register the new, replaced information (extract 2). This paper examines these two ways in which polar questions are disconfirmed and how questioners deal with them in Korean conversation. It will focus on questioners’ conduct in third position following each of the two disconfirmation forms: what I term ‘negation’ responses that merely negate the questioner’s candidate proposition (as in extract 1) and ‘replacement’ responses that assert a revised state of affairs in replacement of the questioner’s proposition (as in extract 2). In audio-recordings of ordinary calls and calls to an airline service, questioners tend to produce another polar question in third position that is built in alignment with the prior disconfirming response. Particular ways in which questioners deploy those third-position questions are different in the two contexts, however. This paper will show that questioners’ thirdposition questions can be used and understood differently by reference to the form of disconfirmation as well as the underlying purposes and activities of the sequence at hand -- which is closely tied to the nature of ordinary vs. airline service interaction. 2. Background In conversation analysis, disconfirming responses to a polar question are considered a dispreferred alternative over confirming responses (Sacks, 1987; Schegloff, 2007; see Pomerantz and Heritage, 2013). They disagree with a candidate proposition formulated in the question, and are constructed with practices associated with dispreferred responses. Disconfirming responses are normally produced with practices breaking their contiguity with the question (e.g. inter-turn gap; Sacks, 1987), and accompanied by elaborations that minimize elements of disconfirmation, such as mitigations, accounts, corrections, etc. (see Heritage, 1984a; Schegloff, 2007; Pomerantz and Heritage, 2013). More recent studies suggest the normativity of elaborations accompanying a disconfirmation. For instance, Ford (2001) showed that in English, a negation expressing disagreement with prior talk (e.g. a polar question) is regularly followed by elaboration such as accounts, alternatives, or corrections. If not, interactants pursue relevant elaboration and treat unelaborated negations as problematic. Similarly, Robinson and Bolden (2010) showed that dispreferred and disaffiliative second pair part actions -- such as disconfirming responses to a polar question -- normatively require the production of accounts both in English and Russian conversation. In the absence of accounts, first pair part speakers may tacitly or explicitly solicit accounts. Finally, Walker et al. (2011) showed that participants regularly pursue further information when a type-conforming no answer alone is produced in response to a polar question, treating such a disconfirmation as uninformative or inadequate. These normative elaborations accompanying a disconfirmation have been discussed as a work toward achieving a common ground and promoting affiliation between the parties, in line with the literature on preference structure (see Pomerantz and Heritage, 2013, for a review). While disconfirmation itself conducts a dispreferred action of disagreement, elaboration can provide a resolution of the disagreement (Ford, 2001; Robinson, 2009; Robinson and Bolden, 2010) and display a level of affiliation with their interlocutors by addressing the agenda or activity being pursued (Walker et al., 2011). Thus, prior studies suggest that respondents, in constructing disconfirming responses, may be required to provide affiliative or reconciliatory grounds which will provide some sort of resolution of the disagreement. This paper will show that questioners may likewise work to provide or display resolution when a disconfirming response is produced. It will do so by examining questioners’ conduct in third position following negation and replacement responses in Korean conversation. Based on data from audio-recordings of ordinary telephone calls and calls to an airline service, the analysis will show that questioners tend to produce another polar question in revision of the initial hypothesis in third position. Particular ways in which those third-position questions are deployed and understood, however, may be different in the two contexts, as they are closely tied to the nature and purpose of activities involved in the question-answer sequence. Before presenting the analysis, Section 2.1 will provide a brief background on responses to polar questions in Korean. 2.1. Responses to polar questions in Korean Korean is an ‘agree--disagree’ language: a positive ‘yes’ particle affirms the proposition and a negative ‘no’ particle denies it, regardless of the polarity of the question (Sadock and Zwicky, 1985). Affirmative ‘yes’ particles can confirm both positively and negatively formulated questions; and they involve several forms, such as ey, yey, ney for polite forms, and ung, um[1_TD$IF], e, eng for non-polite forms (Yoon, 2010). These various particles comprise a type-conforming answer in Korean (Yoon, 2010). Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

4

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

Disconfirming responses to a polar question in Korean occur significantly less than ‘yes’ answers, in only about 12% of responses (Yoon, 2010; see also Park, 2009). Negative ‘no’ particles involve several forms, for instance aniyo, anipnita for polite forms, and ani, aniya for non-polite forms. These ‘no’ particles always reject the proposition at hand regardless of the polarity of the question, and thus are dispreferred responses. They comprise a type-conforming answer in Korean (Park, 2009; Yoon, 2010). Although more systematic research is in need, type-conforming ‘no’ answers may not comprise the predominant form of disconfirming response in Korean. Prior studies report an uncommon use of ‘no’ particles, whether singly or in combination with an elaboration providing a replacement, account, etc. (Park, 2009; Yoon, 2010). Instead, nonconforming responses are more commonly used, which can take the form of a repetition of the question with negation, or of an assertion that displays disconfirmation. In this paper, the distinction between ‘replacement’ and ‘negation’ responses lies on whether or not a revised state of affairs replacing the questioner’s hypothesis is provided in the response, rather than on the form of type-conforming and nonconforming responses. 3. Data and method Data of this paper were drawn from two sets of audio-recorded Korean telephone calls: ordinary calls, and calls by customers to an airline service. Ordinary telephone calls mostly involved conversation between friends and family. The collection of these calls is referred to as ‘ordinary calls’ in this paper. Calls to an airline service were collected by the company and involved reservation services such as booking and purchasing airline tickets. This collection is referred to as ‘airline service calls’ in this paper. Seventy sequences from these two types of collection formed the core corpus of sequences in which a disconfirming answer is produced in response to a polar question. Out of the 70 sequences, 34 sequences were from ordinary calls and 36 sequences from airline service calls. The data were transcribed and analyzed by using the method of conversation analysis (Heritage, 1984a). Pseudonyms were used to identify interactants in the transcript. Transcripts have three lines. The first, italicized line provides romanized Korean according to the Yale system, representing actual sounds produced by the speaker. The second line provides a literal English translation of each word, with a morpheme-by-morpheme gloss. The third, boldfaced line presents an idiomatic English translation. Square brackets [ ] at the beginning of the idiomatic English translation indicate an occurrence of overlapping talk. See brackets in the italicized line for the positioning of overlap onset. 4. Disconfirming responses and questioners’ conduct in third position This section examines questioners’ conduct in third position following two forms of disconfirming response to a polar question: ‘replacement’ and ‘negation’ responses. Replacement responses assert a revised state of affairs in replacement of the questioner’s candidate proposition, which may be preceded by a turn-initial ‘no’ particle. By contrast, negation responses merely negate the questioner’s candidate proposition. They can be constructed with a negative ‘no’ particle, a clause or sentence negating the proposition (e.g. ‘I’m not busy’ in answering ‘Are you busy?’), or both in combination (e.g. ‘No I’m not busy’). Negation responses only claim that the questioner’s hypothesis is mistaken and do not show how it is to be revised, unlike replacement responses. As shown in Table 1, replacement responses were more frequently produced than negation responses in the dataset. In about 53% of the 70 sequences under analysis, questioners produced another polar question in third position following disconfirming responses. They revised their initial hypothesis and constructed a question in alignment with the prior disconfirming response, seeking confirmation. This paper focuses on such third-position questions following disconfirming responses. The analysis will show that in airline service calls, questioners tend to produce a repetitional question in third position following both negation and replacement responses. In ordinary calls, by contrast, questioners may offer another, revised question following negation responses, whereas producing a repetitional question following

Table 1 Distribution of replacement and negation responses. Genuine polar question

Proxy polar question

Total

Replacement response Negation response

22 (51.2%) 21 (48.8%)

24 (88.9%) 3 (11.1%)

46 (65.7%) 24 (34.3%)

Total

43

27

70

p = 0.0016 (Fisher’s exact test).

Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

5

replacement responses. The analysis will suggest that these ways in which questioners deal with disconfirming responses are shaped by the form of disconfirmation as well as the underlying purpose and activity of the sequence at hand. Below I will first provide a preliminary observation on the kinds of polar questions that relate to the selection between negation and replacement responses, which ultimately shapes the terrain for questioners’ third-position questions. 4.1. Preliminary: two kinds of polar questions While polar questions constrain their answers to ‘yes’ and ‘no’, there seem to be two kinds of polar questions that make relevant different choices when the answer is ‘no’. In polar questions such as ‘Are you busy?’ and ‘Are you a member?’, ‘yes’ and ‘no’ answers exhaust the possible choices. A ‘no’ answer itself is informative, although it may be followed with relevant elaborations such as accounts, mitigation, etc. By contrast, in response to polar questions such as ‘Are you married?’ and ‘Are you Korean?’, the choice is between ‘yes’ and, if not ‘yes’, multiple alternatives (e.g. ‘divorced’, ‘widowed’, etc., and ‘American’, ‘Japanese’, etc.). A ‘no’ alone is thus uninformative, leaving the questioner no more informed than s/he was before. Thus, when the answer is ‘no’, respondents may be required to provide a replacement, a correct alternative out of the several possibilities. In these cases, polar questions seem to be used as a ‘proxy’ for a wh-question (e.g., for ‘What is your marital status?’ and ‘Where are you from?’). These two kinds of polar questions can thus make relevant different forms of disconfirmation. In response to the former, ‘genuine’ polar questions in which ‘yes’ and ‘no’ exhaust the possibilities, negation responses as well as replacement responses may be relevant and informative. By contrast, the latter, ‘proxy’ polar questions may be relevantly responded with replacement responses, but not with negation responses. Negation responses may be uninformative or inadequate when used in response to a ‘proxy’ polar question, because they do not provide information for which the question was produced. In the dataset under analysis, negation and replacement responses showed a different distribution by reference to these two kinds of polar questions (Table 1). Across ordinary and airline service calls, the majority of ‘proxy’ polar questions were responded with replacement responses (88.9%). Only one question from airline service calls and two questions from ordinary calls were responded with negation responses. By contrast, in response to ‘genuine’ polar questions in which ‘yes’ and ‘no’ exhaust the possible choices, both negation and replacement responses were commonly used (Table 1). The common production of replacement responses (51.2%) seems to suggest that respondents orient to the relevance of providing a replacement or elaboration in constructing dispreferred responses. Thus, when constructing disconfirming responses, question recipients may make a selection between negation and replacement responses in the context of this distinction between two kinds of polar questions. This can ultimately shape ways in which questioners deal with disconfirming responses in third position, as will be examined below. 4.2. Third-position questions in airline service calls In airline service calls under analysis, customers make a request for a flight reservation and agents process two kinds of information in making the reservation. First, agents process several elements of flight information: date, itinerary, time, number of passengers, and the date and time of a return trip, if applicable -- mostly in this order due to the constraints of a computer system. Second, when the availability of a requested flight is established, agents move to processing several elements of the customer’s identification information, such as the customer’s membership, name, and phone number. These several elements of flight and identification information are normally specified over a set of question--answer sequences initiated by agents (see Lee, 2009). Question--answer sequences in these calls are thus primarily task-oriented, processing particular pieces of flight and identification information for the purpose of making a flight reservation. In these sequences, parties engage in eliciting and providing information that is necessary for making a reservation. The purpose of questions and the nature of activities performed through question--answer sequences are oriented to processing and transferring reservation-related information in a correct way, so as to bring about an outcome -- a flight reservation -- the customer requested. When a disconfirming response is produced in this context, questioners orient to establishing the correct information on record. They tend to produce another question repeating the response and enact its registration in third position. Consider extract 3 in which a question recipient produces a replacement response. Prior to the segment below, the customer specified the date and itinerary of his trip. At line 1, the agent seeks confirmation of the itinerary the customer specified. The customer produces a confirmation (yey ‘yes’) and asks about departure times (line 2). Before responding to the question, the agent initiates an insert sequence by producing a question about the number of passengers at line 4. The departure time is thus made contingent on the number of passengers (see line 10). Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

6

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

In asking about the number of passengers at line 4, the agent constructs a polar question, proposing ‘one person’ as a possibility. This may be a default, most likely, or even the only possible number of passengers the agent can propose in choosing the form of a polar question. Note that the question leaves open multiple alternatives when the answer is ‘no’. At line 6, the customer produces a new assertion that replaces the number of passengers proposed in the agent’s question. This replacement response, although displaying disconfirmation, is informative and cooperative in that it provides the information pursued by the question. In third position (line 8), the agent produces a polar question that almost repeats the replacement. Note that the difference is in the agent’s use of an honorific form referring to ‘persons’ and deferential sentence-ending, which is not translated in English. In producing the (partially) repetitional question (line 8), the agent explicitly articulates her receipt of the new, replaced information that can be consequential for the availability of particular flights (cf. line 10). The agent enacts her registration (Goldberg, 1975; Schegloff, 1997), displaying her revised understanding that the situation is different than what she had proposed in the initial question. The agent thus commits her turn to registering the revised information, establishing the ‘correct’ information on record. In this process, the agent also revises her initial question in Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

7

alignment with the disconfirming response so as to secure confirmation. At line 9, the customer produces a confirmation in response. The agent then provides the information inquired by the customer, marking it as adjusted according to the revised number of passengers with kulem ‘then’ (line 10). Agents likewise construct a repetitional question in third position when their initial question is disconfirmed with a negation response. As described in Section 4.1, except in only one case, negation responses in this context are produced in response to ‘genuine’ polar questions in which a ‘no’ answer is informative. Consider extract 4. The agent and the customer processed all elements of flight information. As the availability of a particular flight is established for reservation, the agent moves to the next step of processing the customer’s identification information. At line 1, the agent produces a polar question asking about membership in the airline service. This is normally the first, ‘default’ question agents ask in processing the customer’s identification information in airline service calls. The customer responds by producing a statement negating the proposition (line 3). This negation response is adequate and informative in this context.

In third position at line 4, the agent produces a question (partially) repeating the negation response. Note that the difference is in the agent’s use of an honorific suffix -si, which is not translated in English, along with a connective ending -kwu ‘and’. By repeating the negation, the agent articulates her receipt of, and registers, the disconfirmation. She thus puts the correct, revised information on record. In so doing, the agent also seeks confirmation, in alignment with the disconfirming response. The customer produces a confirmation in response at line 5. Thus, in airline service calls questioners produce a repetitional question in third position following both negation and replacement responses. They articulate their receipt of the disconfirming response, enacting its registration. As part of the process of enacting registration, agents also (re-)construct their question so as to get a confirming response. By asking a question that repeats what has just been told, agents engage in registering and establishing the correct information on record in running the sequence. Agents’ use of repetitional third-position questions may relate to the nature of activities involved in question--answer sequences in airline service calls. Questions are produced for the purpose of getting reservation-related information. In question--answer sequences participants are geared to processing information for a reservation and doing so correctly,

Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

8

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

because incorrect information can result in adverse outcomes such as a reservation for a ‘wrong’ flight, a ‘wrong’ passenger, etc. Agents may thus orient to registering and establishing the correct information in an explicit way, especially when their initial question is disconfirmed (and replaced with a new assertion). They focus on doing getting the information right in third position, orienting to the larger reservation activity in progress. There is one case that departs from the general pattern, in which a questioner invites the respondent to continue when a negation response is produced. This is when a negation response is produced in response to a ‘proxy’ polar question that leaves open multiple possibilities for a ‘no’ answer (Section 4.1). Extract 5 illustrates the case. Prior to the segment below, the agent asked about the date and itinerary using wh-questions, and the customer only provided the itinerary. At line 1, the agent seeks confirmation of the itinerary. The customer produces a confirmation ( yey: ‘yes:’) and moves to introduce a particular departure time for the itinerary, without providing the date (line 2). At line 3, the agent constructs a polar question to get the information about the date of departure. Note that the date and itinerary need to be first processed in the computer system to locate other elements of flight information such as departure times.

Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

9

In constructing the polar question at line 3, the agent proposes ‘today’ as a possibility. This is a default, most likely day the agent can propose in using the form of a polar question, especially given that the customer had not previously specified the date. This polar question leaves open multiple possibilities concerning the date of departure when the answer is ‘no’, and can thus be relevantly responded with a replacement response. At line 5, however, the customer merely produces a negation. The negation response is uninformative and uncooperative, in that it does not provide the date of departure. Following the negation response, the agent does not construct a (repetitional) question or otherwise register the response, unlike in extracts 3--4 above. The agent rather produces a continuer, yey ‘yes’ (line 7), and prompts the customer to see that the prior negation was inadequate. The agent invites and mandates the customer to continue and provide a revised, correct date. At line 9, the customer provides a revision. Given the replacement, the agent constructs a question repeating the revised day, enacting its registration (line 10). With the revised, correct information, the agent orients to establishing that on record, as in extracts 3--4. What is to be noted is that the agent does not register the response or offer a revised hypothesis given the mere negation (cf. extract 1), but invites the customer to provide a revision. This is distinctive from questioners in ordinary calls, as will be examined in Section 4.3. In this airline service context, there is a sense in which the agent cannot propose a revised, alternative date of departure. When the agent chooses to use the form of a polar question in asking about a departure date, ‘today’ is the default, or even the only possible day the agent can propose. The agent relevantly lacks knowledge about the respondent’s personal situations, experiences, etc., in this task-oriented context, and thus cannot provide or revise a hypothesis based on shared, intimate knowledge about the respondent. The agent may only offer a question based on default expectations from a probabilistic point of view, as in the initial question (line 3). Thus when such a default question is disconfirmed, the agent produces a continuer. This shows that the agent’s conduct is closely tied to the nature of interaction, including the nature of relationship between the parties. Particular ways in which questioners deal with disconfirming responses when they are informative and adequate (extracts 3--4) vs. when they are not (extract 5) can be shaped by the task-oriented nature of interaction as well as the activities involved in the sequence at hand. 4.3. Third-position questions in ordinary calls Ordinary calls under analysis involve conversation between friends and family. In such interaction between intimates, questions and question--answer sequences are not or not only constructed for the purpose of processing information as in airline service calls. They can be deployed for various objectives and activities. In addition, questioners have a ‘‘responsibility to make use of what they know about their interlocutors’’ and display their shared knowledge in the design of questions (Stivers et al., 2011:18). Thus, when disconfirming responses are produced in this context, questioners may not only orient to the underlying purposes and activities involved in the sequence but may also seek to (re-)establish a common ground and affiliation with their interlocutors. In contrast with airline service calls, questioners in ordinary calls tend to deal with negation and replacement responses differently in third position. They may produce a different kind of third-position question following replacement vs. negation responses. First, when their initial question is disconfirmed with a replacement response, questioners tend to construct a question repeating the replacement in third position. While this form of question on the surface is similar to that in airline service calls, particular ways in which such a third-position question is deployed and understood are different in ordinary calls. Consider extract 6. Min called her sister cousin who is visiting home during the vacation, and Min’s aunt answered the phone. At line 3, Min produces a polar question in launching a switchboard request. While the question is produced for the purpose of reaching the sister, its form with no interrogative markings can encode Min’s expectation that the sister is likely to be home (cf. Lee, 2015), especially given that the call occurred in a weekend’s morning and that the sister is visiting home just for the vacation. Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

10

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

After the repair initiation (line 4) -- possibly due to the overlap (lines 2--3) -- and Min’s repair (line 5), the aunt constructs a replacement response at line 7. She responds to the question by making an assertion concerning whereabouts of the sister. By constructing the replacement instead of a mere negation that may be adequate in this context, the aunt not only provides an account for the sister not being home but also implies the length of time the sister would be unavailable. The aunt thus orients to the purpose of the questioner in constructing the replacement response. In third position at line 8, Min registers the new, revised assertion. She first claims her change of state from not knowing to knowing about the matter with a, which is equivalent to a change-of-state token ‘oh’ in English (Heritage, 1984b). She then produces a repetitional question, showing her receipt of the new piece of information (line 8). In so doing, Min seems to re-enact her earlier misapprehension and display revised understanding, which, as a consequence, implicates a need to try an alternative in order to accomplish her purpose of reaching the sister. Thus at line 9, the aunt does not merely confirm the repetitional question (e ‘yes’) but offers that the sister has her cell phone. This shows the aunt’s understanding of the repetitional question as indicating Min’s urgent need to contact the sister. Thus, while the form of third-position question is nearly identical to that in airline service calls (cf. extract 3), the third-position question is understood differently by reference to the underlying purpose and activity involved in the sequence at hand. What is particularly distinctive in ordinary calls is when polar questions are disconfirmed with negation responses. As examined in Section 4.2, questioners in airline service calls repeat and register the negation when it is produced in response to a ‘genuine’ polar question and is informative (extract 4), whereas inviting the respondent to provide a revision when a negation response is uninformative (extract 5). In ordinary calls, questioners may alternatively offer a revised Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

11

hypothesis when their initial question is disconfirmed with a mere negation, regardless of whether the negation is informative or not for the purpose of the question. Consider extract 7. A negation response here is produced in response to a ‘genuine’ polar question, in which a negation alone can be an informative response. Jay called her friend, Mia, who is a school teacher. At the beginning of the call, Jay produces a polar question about whether Mia is busy at the moment (line 1). This question serves as a preliminary that is addressed to checking whether Mia would be available to talk further. It also displays Jay’s understanding that Mia, who is a teacher, is likely to be busy given that the call occurred during the day. In response, Mia produces a negative ‘no’ particle followed by a negation of the questioner’s candidate proposition (line 3). This negation response is informative and adequate in this context, serving as a ‘go-ahead’ response that allows the questioner to proceed to the action projected, i.e. to talk further (Schegloff, 2007). By constructing the mere negation, Mia treats the question as a preliminary and invites Jay to move on.

(7) Marriage 1

JAY: ->

2

ya pappa cikum? VOC busy:IE now Hey you’re busy now? (0.8)

3

MIA: ->

ani an pappa. no not busy:IE No I’m not busy.

4

JAY: =>

cikum swuep kkuthna-ss:-ci. now class finish-PAST-COMM Now classes are don:e, right.

5

(1.0)

6

JAY:

an [kkuthna-ss-e¿ not finish-PAST-IE [] Not done¿

7

MIA:

[e ani swuep kkuthna-n ke-n ani-ntey:¿ yes no class finish-ATTR thing-TOP not-CIRCUM [] Yes no it’s not that classes are done but:¿

8

JAY:

ung: yes Yes:

9 10

(0.5) MIA:

ai na kongkang-iya DM I between.classes-IE Well I’m between classes

At line 4, however, Jay does not move on to the projected action, although given the go-ahead response. Jay rather retries another polar question in third position, proposing a revised hypothesis concerning a possible account for the negation. In so doing, she demonstrates her shared knowledge, making an attempt to be on the same page with Mia (cf. Smith, 2013). Jay uses and displays her knowledge about the respondent’s situations and life calendar, and re-tries her question so as to receive confirmation. Thus, given the disconfirming response, although it is a go-ahead, Jay orients to re-establishing a common ground with the respondent (cf. Clark, 1996; Enfield, 2006). Jay’s attempt, however, is not successful. There is no immediate response from Mia (line 5), and Jay in the end reverses the polarity of the question (line 6). Jay thus makes another, third attempt to receive confirmation. Mia finally states that she is between classes, replacing Jay’s candidate understandings (lines 7, 10). Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

12

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

Questioners may similarly offer a revised hypothesis in third position following a negation response that is uninformative, i. e. when a mere negation is produced in response to a question that leaves open multiple possibilities for a ‘no’ answer. As noted in Section 4.1, a mere negation was produced in response to such ‘proxy’ polar questions in only two cases of ordinary calls. Both questioners in the two cases offered another, revised polar question in third position following negation responses. Extract 8 illustrates one of the two cases. Kim called his older friend, Lee, after some time since they talked to one another. In the call, Kim has been updating Lee on his life experiences for a while. Near the end of the call, Kim turns to Lee’s life, first producing a how-are-you inquiry (line 1). Lee produces a ‘neutral’ response at line 2 and does not further elaborate on his life (Sacks, 1975). At line 3, Kim seeks to have an update on Lee, introducing and seeking confirmation of Lee’s recent life. In so doing, Kim uses an evidential marker -tamye ‘I hear’, indicating the matter at hand as information he acquired through hearsay and thus inviting Lee, who has epistemic authority, to elaborate on the information (Kim, 2011). However, Lee is not cooperative in this business of offering updates (cf. Smith, 2013) and merely produces two tokens of confirmation, e e ‘yes yes’ (line 4).

Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

13

At line 5, Kim further pursues the inquiry and the activity of updating. He produces a polar question asking about whether Lee is in a graduate program at KIST. KIST is a research institute that offers graduate programs as well as various research-related positions. The question thus proposes one possibility out of several options in terms of Lee’s status at KIST. Note that Kim uses animyen ‘or’ in ending the question, which leaves open alternative options (line 5). In response, however, Lee produces a mere negation (line 6), and does not provide the information pursued by the question. In third position (line 7), Kim offers a revised question, proposing another likely status of Lee. He thus makes another attempt to get things right, trying to have a shared knowledge about Lee. Kim’s use of the revised question may be occasioned by Lee’s uncooperativeness in the business of updating. Lee has been uncooperative in answering the question (line 6) as well as in the ongoing activity of providing updates on his life throughout the sequences. Thus, by using the revised question Kim not only tries to deal with the trouble and pursue an update, but also seeks to achieve a common ground with Lee. At line 9, Kim’s revised hypothesis is in the end disconfirmed with a replacement. Thus, questioners in ordinary calls may offer a revised, another question in third position following negation responses. Alternatively to registering the response (e.g., with repetitional questions) or inviting the respondent to provide a revision (e.g., with continuers), questioners themselves can make another guess and revise their initial question. In choosing to use a revised question in particular, questioners seek to display their shared knowledge and to re-establish a common ground and affiliation with respondents. Questioners’ production of a revised question in third position seems to relate to the nature of interaction in ordinary calls. In interaction between intimates, participants are expected to not only share their own experience and but also know such shared experience of others (Smith, 2013; Stivers et al., 2011). They have the responsibility to make use of what they know about one another, and the use and pursuit of such shared knowledge in the talk can be crucial in managing interpersonal intimacy and affiliation (Enfield, 2006; Smith, 2013; Stivers et al., 2011). Questioners orient to such expectations and responsibilities when offering a revised hypothesis in third position. While they may well produce repetitional questions or even continuers following negation responses, questioners may alternatively choose to propose a revised understanding, working to demonstrate and re-assert shared knowledge. Through the use of revised questions in third position, questioners strive to re-establish a common ground with respondents, orienting to interpersonal intimacy and affiliations (cf. Enfield, 2006; Smith, 2013). This may be essentially distinctive from airline service calls in which the interaction is task-oriented in nature and does not pursue interpersonal relationships. The different nature of interaction can influence ways in which questioners deal with disconfirming responses in third position. 5. Conclusion Section 4 examined two ways in which polar questions are disconfirmed and how questioners deal with them in third position. It focused on third-position questions following replacement and negation responses in ordinary and airline service calls. In airline service calls, questioners tend to produce repetitional questions following both negation and replacement responses. They register the new information, seeking to establish the correct, revised information on record. In ordinary calls, questioners tend to produce repetitional questions following replacement responses. Those repetitional third-position questions are deployed and understood by reference to the underlying purpose and activity of the sequence at hand. Following negation responses, questioners may alternatively come up with another, revised polar question in third position. They may make another attempt to get things right, seeking to achieve a common ground with respondents. Thus, particular ways in which third-position questions are used and understood can be conditioned by the nature of interaction as well as the form of disconfirmation. As described in Section 4, interaction in airline service calls is task-oriented in nature. Interactants are engaged in processing several kinds of information that is necessary for flight reservations, and question--answer sequences are largely used for such purposes of processing information. This nature and purpose of question--answer sequences shape questioners’ conduct in third position that focuses on doing getting the correct information. By contrast, interaction in ordinary calls is not oriented to processing information primarily. Interactants can be engaged in various actions and activities, and question--answer sequences are not only used to process information. In addition, in the design of questions and answers, interactants get to display and make use of mutual knowledge about one another, and their management of such shared knowledge and common ground can be crucial in interpersonal relationships (Clark, 1996; Enfield, 2006; Smith, 2013; Stivers et al., 2011). Thus, questioners may produce different kinds of question in third position depending on how respondents design their disconfirmation; and when their respondent does not provide a common ground in producing a negation response, questioners may seek to re-establish shared knowledge and affiliation. Questioners’ conduct in third position is thus closely tied to the nature of interaction. This shows one way in which the two different contexts are realized in question--answer sequences and their aftermath. This paper suggests that the terrain for questioners’ third-position questions can be shaped and conditioned by the interplay of multiple aspects such as the form of disconfirming response, the kind of polar question, and the nature of interaction and relationship between the parties. Questioners design and deploy their third-position questions in particular Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003

+ Models

PRAGMA-4132; No. of Pages 14

14

S.-H. Lee / Journal of Pragmatics xxx (2015) xxx--xxx

ways, considering those multiple aspects, while at the same time revising their initial question so as to get a confirming response and thus obeying Sacks’ (1987) message on the preference for agreement. Disconfirming responses and what questioners do in third position can provide a lens through which one can see the nature of activity, interaction, and interactants’ interpersonal relationships. References Bolinger, Dwight, 1978. Yes--no questions are not alternative questions. In: Hiz, H. (Ed.), Questions. Reidel, Dordrecht, pp. 87--105. Clark, Herbert H., 1996. Using Language. Cambridge University Press, Cambridge. Enfield, N.J., 2006. Social consequences of common ground. In: Enfield, N.J., Levinson, Stephen C. (Eds.), Roots of Human Sociality: Culture, Cognition, and Interaction. Berg, Oxford, pp. 399--430. Ford, Cecilia E., 2001. At the intersection of turn and sequence: negation and what comes next. In: Selting, M., Couper-Kuhlen, E. (Eds.), Studies in Interactional Linguistics. John Benjamins, Amsterdam, pp. 51--79. Goldberg, Jo Ann, 1975. A system for the transfer of instructions in natural settings. Semiotica 14, 269--296. Heritage, John, 1984a. Garfinkel and Ethnomethodology. Polity Press, Cambridge. Heritage, John, 1984b. A change-of-state token and aspects of its sequential placement. In: Atkinson, J.M., Heritage, J. (Eds.), Structures of Social Action: Studies in Conversation Analysis. Cambridge University Press, Cambridge, pp. 299--345. Heritage, John, 2010. Questioning in medicine. In: Freed, A.F., Ehrlich, S. (Eds.), ‘‘Why Do You Ask?’’: The Function of Questions in Institutional Discourse. Oxford University Press, New York, pp. 42--68. Heritage, John, Raymond, Geoffrey, 2012. Navigating epistemic landscapes: acquiescence, agency and resistance in responses to polar questions. In: de Ruiter, J.P. (Ed.), Questions: Formal, Functional and Interactional Perspectives. Cambridge University Press, Cambridge, pp. 179--192. 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. Lee, Seung-Hee, 2009. Extended requesting: interaction and collaboration in the production and specification of requests. Journal of Pragmatics 41, 1248--1271. Lee, Seung-Hee, 2015. Two forms of affirmative responses to polar questions. Discourse Processes 52, 21--46. Park, Jiseon, 2009. Pre-verbal negation yes/no question--answer sequence in conversation: action formation and sequence organization. Linguistics [enehak] 30, 75--105. Pomerantz, Anita, 1988. Offering a candidate answer: an information seeking strategy. Communication Monographs 55, 360--373. Pomerantz, Anita, Heritage, John, 2013. Preference. In: Sidnell, J., Stivers, T. (Eds.), The Handbook of Conversation Analysis. Wiley-Blackwell, Boston, pp. 210--228. Raymond, Geoffrey, 2003. Grammar and social organization: yes/no interrogatives and the structure of responding. Am. Sociol. Rev. 68, 939--967. Robinson, Jeffrey D., 2009. Managing counterinformings: an interactional practice for soliciting information that facilitates reconciliation of speakers’ incompatible positions. Human Communication Research 35, 561--587. Robinson, Jeffrey D., Bolden, Galina B., 2010. Preference organization of sequence-initiating actions: the case of explicit account solicitations. Discourse Studies 12, 501--533. Sacks, Harvey, 1975. Everyone has to lie. In: Blount, B., Sanches, M. (Eds.), Sociocultural Dimensions of Language Use. Academic Press, New York, pp. 57--79. Sacks, Harvey, 1987. On the preferences for agreement and contiguity in sequences in conversation. In: Button, G., Lee, J.R.E. (Eds.), Talk and Social Organization. Multilingual Matters, Clevedon, pp. 54--69. Sacks, Harvey, Schegloff, Emanuel A., 1979. Two preferences in the organization of reference to persons and their interaction. In: Psathas, G. (Ed.), Everyday Language: Studies in Ethnomethodology. Irvington, New York, pp. 15--21. Sadock, Jerrold, Zwicky, Arnold, 1985. Speech act distinctions in syntax. In: Shopen, T. (Ed.), Language Typology and Syntactic Description. Clause Structure, vol. 1. Cambridge University Press, Cambridge, pp. 155--196. Schegloff, Emanuel A., 1979. Identification and recognition in telephone conversation openings. In: Psathas, G. (Ed.), Everyday Language: Studies in Ethnomethodology. Irvington, New York, pp. 23--78. Schegloff, Emanuel A., 1997. Practices and actions: boundary cases of other-initiated repair. Discourse Processes 23, 499--545. Schegloff, Emanuel A., 2007. Sequence Organization in Interaction: A Primer in Conversation Analysis. Cambridge University Press, Cambridge. Smith, Michael Sean, 2013. ‘‘I thought’’ initiated turns: addressing discrepancies in first-hand and second-hand knowledge. Journal of Pragmatics 57, 318--330. Stivers, Tanya, Enfield, N.J., Brown, Penelope, Englert, Christina, Hayashi, Makoto, Heinemann, Trine, Hoymann, Gertie, Rossano, Federico, de Ruiter, Jan Peter, Yoon, Kyung-Eun, Levinson, Stephen C., 2009. Universals and cultural variation in turn-taking in conversation. Proceedings of the National Academy of Sciences 106 (26), 10587--10592. Stivers, Tanya, Mondana, Lorenza, Steensig, Jakob, 2011. Knowledge, morality and affiliation in social interaction. In: Stivers, T., Mondana, L., Steensig, J. (Eds.), The Morality of Knowledge in Conversation. Cambridge University Press, Cambridge, pp. 3--24. Walker, Traci, Drew, Paul, Local, John, 2011. Responding indirectly. Journal of Pragmatics 43, 2434--2451. Yoon, Kyung-Eun, 2010. Questions and responses in Korean conversation. Journal of Pragmatics 42, 2782--2798. Seung-Hee Lee (Ph.D., UCLA) is an associate professor at the Department of English Language and Literature, Yonsei University, Seoul, Korea. Her current research interests include practices of talk and action in ordinary conversation as well as in commercial and health service contexts. She studies various aspects of sequence structure that relate to agendas and activities pursued in interaction, including those that represent particular institutional goals, procedures, and outcomes.

Please cite this article in press as: Lee, S.-H., Information and affiliation: Disconfirming responses to polar questions and what follows in third position. Journal of Pragmatics (2015), http://dx.doi.org/10.1016/j.pragma.2015.10.003