1.Introduction

Schegloff, Jefferson, and Sacks (1977)Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53: 361–382. initiated a long and continuing line of research in Conversation Analysis (CA) on repair in talk-in-interaction. The findings of this research with regard to the organization of repair do not need to be reviewed here, except to say that repair completion can be distinguished from its initiation and that, because problems with hearing, speaking, and/or understanding can always occur, some means of maintaining intersubjectivity is necessary in any system of communication and repair is a primary means through which this is accomplished in talk-in-interaction. On the other hand, there may be good reason in some situations for participants to try to avoid initiating repair. One such situation may be second language (L2) conversations-for-learning, in which one or more participants is an L2 user and in which the ostensible purpose of the conversation is for the L2 user(s) to practice using the L2. If the initiation of repair is sometimes avoided, then there is a need for other means to handle threats to intersubjectivity. In this paper, I examine several instances in a series of L2 English conversations-for-learning in which the first language (L1) user of English avoids initiating repair. The practice of avoiding repair is one means through which the L1 user contributes to the construction of the L2 user as interactionally competent.

2.Avoiding initiation of repair

Since any turn-at-talk is, in principle, vulnerable to being treated as a trouble source and targeted for repair (Schegloff, Jefferson, & Sacks 1977Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53: 361–382. ; Schegloff 2007Schegloff, Emanuel A. 2007 Sequence Organization in Interaction: A Primer in Conversation Analysis. Cambridge: Cambridge University Press. ), when repair is not initiated, this can be understood as passing up the opportunity to do so. Schegloff (1992)Schegloff, Emanuel A. 1992 “Repair after Next Turn: The Last Structurally Provided Defense of Intersubjectivity in Conversation.” American Journal of Sociology 97: 1295–1345. http://​www​.jstor​.org​/stable​/2781417 demonstrates how each turn-at-talk creates in its wake a “repair space” of multiple opportunities to initiate repair – by self (i.e., the speaker of a turn on which repair may be initiated) during or immediately following the completion of the turn, by other in the next turn, and by self (and once again by other) following the next turn. Therefore, with any turn-at-talk on which repair is not initiated by either self or other, multiple opportunities to initiate repair have been passed up.

However, passing up opportunities to initiate repair is not the same as avoiding initiation of repair, with the latter being drastically more restricted. It involves a first speaker producing a turn-at-talk followed by a second speaker saying something which clearly demonstrates that he or she has not understood either the content or the action (or both) of the prior turn. The first speaker then has an opportunity either to self-initiate repair on his or her own misunderstood turn (Schegloff 1992Schegloff, Emanuel A. 1992 “Repair after Next Turn: The Last Structurally Provided Defense of Intersubjectivity in Conversation.” American Journal of Sociology 97: 1295–1345. http://​www​.jstor​.org​/stable​/2781417) or to other-initiate repair on the turn following the misunderstood turn, such as with an open-class repair initiator like “huh?” (Drew 1997Drew, Paul 1997 “ ‘Open’ Class Repair Initiators in Response to Sequential Sources of Troubles in Conversation.” Journal of Pragmatics 28: 69–101. ). Instead, though, the first speaker does what Liberman (1980Liberman, Ken 1980 “Ambiguity and Gratuitous Concurrence in Inter-Cultural Communication.” Human Studies 3: 65–85. , 2012Liberman, Ken 2012 “Semantic Drift in Conversations.” Human Studies 35: 263–277. ) has termed “gratuitous concurrence” and treats the turn following the misunderstood turn as a sequentially appropriate next turn which does not display any misunderstanding. Both Long (1983)Long, Michael H. 1983 “Native Speaker/Non-Native Speaker Conversation and the Negotiation of Comprehensible Input.” Applied Linguistics 4: 126–141. and Liberman (1980Liberman, Ken 1980 “Ambiguity and Gratuitous Concurrence in Inter-Cultural Communication.” Human Studies 3: 65–85. , 2012Liberman, Ken 2012 “Semantic Drift in Conversations.” Human Studies 35: 263–277. ) have demonstrated this for interaction between L1 and L2 users of English.11.Liberman’s (1980Liberman, Ken 1980 “Ambiguity and Gratuitous Concurrence in Inter-Cultural Communication.” Human Studies 3: 65–85. , 2012Liberman, Ken 2012 “Semantic Drift in Conversations.” Human Studies 35: 263–277. ) work is Ethnomethodological and is not framed as Conversation Analytical. Long’s (1983)Long, Michael H. 1983 “Native Speaker/Non-Native Speaker Conversation and the Negotiation of Comprehensible Input.” Applied Linguistics 4: 126–141. is quantitative Second Language Acquisition research on L2 interaction. In addition, Long states that “the native speaker repairs the discourse by treating the inappropriate response” (p. 136) as appropriate. That is, Long labels as repair what I am referring to as avoidance of repair initiation. It should be noted, though, that Long’s notion of repair is somewhat different from how it is understood within CA and that he makes no attempt to differentiate repair initiation from repair completion or to identify specific turns-at-talk that either initiate or complete repair. Long found that this way of dealing with “a communication breakdown” (p. 136) was too rare to be worth quantifying. However, while avoidance of repair initiation may be more likely in interaction involving L2 users, it is certainly not limited to such interaction and Schegloff (1992)Schegloff, Emanuel A. 1992 “Repair after Next Turn: The Last Structurally Provided Defense of Intersubjectivity in Conversation.” American Journal of Sociology 97: 1295–1345. http://​www​.jstor​.org​/stable​/2781417 provides one example from a conversation between two L1 users of English.

3.Interactional competence as locally constructed

Over the last few decades, the concept of interactional competence has become widely used in research on L2 use, learning, and testing. (See, for example, the volume edited by Hall, Hellermann, and Pekarek Doehler (2011)Hall, Joan Kelly, John Hellermann, and Simona Pekarek Doehler (eds.) 2011 L2 Interactional Competence and Development. Bristol, UK: Multilingual Matters. .) In theory, interactional competence is distinct from communicative competence (Hymes 1974Hymes, Dell 1974 Foundations in Sociolinguistics: An Ethnographic Approach. Philadelphia: University of Pennsylvania Press.) in that the former “is not an attribute of an individual participant” but rather is “something that is jointly constructed by all participants” (He & Young 1998He, Agnes Weiyun, and Richard Young 1998 “Language Proficiency Interviews: A Discourse Approach.” In Talking and Testing: Discourse Approaches to the Assessment of Oral Proficiency, ed. by Richard Young, and Agnes Weiyun He, 1–24. Amsterdam: John Benjamins. , 7). In other words, while communicative competence can be seen as the possession of the individual, interactional competence is not, at least in theory, seen in this way. However, research on interactional competence rarely seems to treat interactional competence as wholly constructed within the interaction. For example, Erickson (2004)Erickson, Frederick 2004 Talk and Social Theory. Cambridge: Polity Press. looks at interactional competence of students as it is made manifest in classroom interaction, focusing on turn-taking and turn stealing, while Walsh (2012)Walsh, Steve 2012 “Conceptualising Classroom Interactional Competence.” Novitas-ROYAL 6: 1–14. looks at the interactional competence of teachers and students in language classrooms to manage the interaction in ways that create opportunities for meaningful language use and language learning. In both cases, while the authors analyze specific instances of interaction and interactional competence, there also seems to be some residue individual ability – of what the individual brings to the interaction – in how interactional competence is actually conceptualized in practice. Wootton’s (1975)Wootton, Anthony 1975 Dilemmas of Discourse: Controversies about the Sociological Interpretation of Language. London: George Allen & Unwin. observation from the 1970s about how the “erosion … of the distinction between competence and performance” (p. 23) had been incomplete continues, apparently, to hold true for the concept of interactional competence today.

This is not necessarily a problem, but it does raise the issue of whether interactional competence can actually be distinguished from communicative competence as conceptualized by Hymes (1974)Hymes, Dell 1974 Foundations in Sociolinguistics: An Ethnographic Approach. Philadelphia: University of Pennsylvania Press.. Rather, it may simply be another name for, or perhaps as Johnson (1995)Johnson, Karen E. 1995 Understanding Communication in Second Language Classrooms. Cambridge: Cambridge University Press. sees it, a component of, communicative competence. In this paper, though, I attempt to adopt in practice the view of interactional competence made in theory by He and Young (1998)He, Agnes Weiyun, and Richard Young 1998 “Language Proficiency Interviews: A Discourse Approach.” In Talking and Testing: Discourse Approaches to the Assessment of Oral Proficiency, ed. by Richard Young, and Agnes Weiyun He, 1–24. Amsterdam: John Benjamins. and others. That is, I will treat interactional competence as locally and wholly constructed by participants within interaction and try to completely erase the distinction between competence and performance. On this view of L2 interactional competence, there is no direct relationship between it and L2 proficiency, as limited proficiency L2 users may nevertheless be constructed as interactionally competent while, conversely, highly proficient L2 users may be constructed as interactionally incompetent (cf., Hauser 2016Hauser, Eric 2016 “The Construction of Interactional Incompetence in L2 Interaction.” Paper presented at The Teaching and Testing of L2 Interactional Competence Symposium, Rice University, Houston, TX.). In this paper, though, as I analyze a practice of the L1 user, it is specifically a part of this participant’s contribution to the construction of the L2 user’s interactional competence which is in focus.

4.Data and method

I will now move to presentation and analysis of the data. In this section, I will briefly introduce the data and how it has been transcribed and analyzed. In the next section, I will present two instances in which the L1 user does not avoid repair initiation. In the following section, I then analyze several instances of the focal practice of avoiding repair initiation. Finally, I will discuss the maintenance of intersubjectivity when initiation of repair is avoided and the local construction of L2 interactional competence.

The data are drawn from a series of nineteen meetings between the researcher, an L1 user of English and proficient L2 user of Japanese named Eric, and an adult L1 user of Japanese and L2 user of English (of somewhat limited proficiency) who I will call by the pseudonym of Nori. The ostensible purpose of the meetings, which took place semi-regularly over a seven-month period, was to provide Nori, who had immigrated with his wife and children from Tokyo to Honolulu a few months prior to the first recording, with opportunities to practice English.22.Meeting length was not uniform, but the average length was 43 minutes. The total time of recorded interaction was approximately thirteen and a half hours. For more complete information, see Table 1 in Hauser (2013a)Hauser, Eric 2013a “Stability and Change in One Adult’s Second Language English Negation.” Language Learning 63: 463–498. . The interaction at these meetings can thus be considered conversations-for-learning. A secondary purpose was to provide Eric with data for research and each meeting was audio-recorded with the permission and knowledge of Nori, who wore a clip-on microphone. Each meeting has been transcribed in whole following CA conventions (Jefferson 2004Jefferson, 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. ). (See Appendix A for transcription conventions.) Nori made fairly frequent use of Japanese, which has been transcribed in italics. In the excerpts below, when Nori uses Japanese, a morpheme-by-morpheme gloss (based largely on symbols introduced in Nguyen and Kasper (2009)Nguyen, Hanh thi, and Gabriele Kasper (eds.) 2009 Talk-in-Interaction: Multilingual Perspectives. Honolulu, HI: University of Hawaiʻi, National Foreign Language Resource Center.; see Appendix B for symbols) appears in the following line and an idiomatic translation in the line after that. The entire dataset is longitudinal and I have elsewhere published research that investigates development over time.33.In particular, see Hauser (2013aHauser, Eric 2013a “Stability and Change in One Adult’s Second Language English Negation.” Language Learning 63: 463–498. , bHauser, Eric 2013b “Expanding Resources for Marking Direct Reported Speech.” In Pragmatics & Language Learning, Vol. 13, ed. by Tim Greer, Donna Tatsuki, and Carsten Roever, 29–53. Honolulu, HI: University of Hawaiʻi, National Foreign Language Resource Center.). The longitudinal nature of the data, though, is not part of the analysis in this paper.

5.Initiation of repair in next turn and third position

As is shown in the first two excerpts, when Nori produces a turn that shows that he has not understood Eric’s prior turn, Eric sometimes initiates repair. In Excerpt 1, Eric starts telling Nori about two acquaintances who visited the restaurant where Nori works as the assistant manager. Nori, though, thinks that Eric is talking about something different. In line 14, Eric initiates repair on Nori’s turn.

01  E: las’ week uh: (.) las’ Friday_ (0.2)

02  N: n.

03  E: two of my: ↓uh ↑my teachers at >thee University

04     of Hawaii,<  [went to Ryooritei.

                             restaurant name ((pseudonym))

05  N:                 [ah=↓ah ↑ah=ah=ah ah=↓ah

06  E: right?=

07  N: =okay okay.

08  E: so one was Japanese,

09  N: m.

10  E: an’ thee other one ↓German.

11  (1.6)

12  N: soo  soo. kae- ↑kaechatta  yo ne?

       yeah yeah       go-home-as ip  ip

       ‘That’s right. You wen- you went home, right?’

13  (0.9)

14  E: hm?

15  N: jikan ga nakatta.

       time  sb ng-as

       ‘You didn’t have time.’

16  (0.4)

17  N:  °uh-°  (0.2)  uh:: wai- waiting time lo- long

18      time waiting?

In lines 01 and 03–04, Eric mentions a visit to the restaurant where Nori works by two of his teachers. In overlap, Nori claims recognition of what Eric is talking about in line 05 and then again in line 07. (Though it is not transcribed in italics, “ah” can be considered similar to English “oh” (Tanaka 1999Tanaka, Hiroko 1999 Turn-Taking in Japanese Conversation: A Study in Grammar and Interaction. Amsterdam: John Benjamins.).)44.It seems to me that users of Japanese often do not consider “ah” to be specifically Japanese. I have therefore not used italics for the transcription. Eric then elaborates on who visited the restaurant in lines 08 and 10. Following line 08, Nori receipts the information about one of the teachers with “m” said with falling intonation. Following line 10, though, there is a relatively long gap, which may be indicative of some sort of trouble, in that there is no information receipt as in line 09 and in that Eric waits for Nori to respond. However, when Nori does respond in line 12, nothing that he says indicates that he sees himself as not understanding what Eric has said. However, what he says does indicate that he thinks that Eric is talking about something else. In response to this, following a gap in line 13, Eric targets Nori’s talk for repair with an open class next turn repair initiator in line 14, which can indicate trouble understanding how Nori’s turn is a sequentially appropriate next to what Eric has mentioned (Drew 1997Drew, Paul 1997 “ ‘Open’ Class Repair Initiators in Response to Sequential Sources of Troubles in Conversation.” Journal of Pragmatics 28: 69–101. ).

Excerpt 2 shows how Eric can also initiate repair on his own talk in third position (Schegloff 1992Schegloff, Emanuel A. 1992 “Repair after Next Turn: The Last Structurally Provided Defense of Intersubjectivity in Conversation.” American Journal of Sociology 97: 1295–1345. http://​www​.jstor​.org​/stable​/2781417) when what Nori says indicates that he has not understood Eric’s prior turn. In this excerpt, Nori and Eric have been discussing places in Nagoya, where both of them lived a few years.

01  E:  =oh. (.) so where are the good shops.

02  N:  nyeah. [↓two.

03  E:         [where.

04  (1.5)

05  N:  uh: ↑Tonki °(to yuu).°

             restaurant name (qt say)

06  (0.3)

07  E:  Ton [ki_

08  N:      [heh heh Tonki.  [.hh h

09  E:                       [where’s that.

10  (0.2)

11  N:  uh: ↑name. uh ↑shop name.   [Tonki:.

12  E:                              [yeah,


13  (0.2)

14  N:  .t oishii.

           delicious

        It’s good.

15  E:  what part of Nagoya is  [it.

16  N:                          [n n Na- no place Nagoya.

17  (0.5)

18  E:  but where (.) in Nagoya.

19  (1.1)

20  N:  Shinsakae.

        place name

In line 01, Eric starts a sequence by asking a question with “where.” While this is a new sequence, it is designed as connected to the prior talk, which has been about places to eat a particular kind of food for which the Japanese city of Nagoya is famous. While Nori moved to Honolulu from Tokyo, before going to Tokyo, he had lived for several years in Nagoya. In addition, Eric had also lived in Nagoya before moving to Tokyo and then to Honolulu, something which Nori is aware of. Therefore, an answer to the question in line 01 can be expected to be recipient-designed for someone who is familiar with different locations in Nagoya. (See Schegloff (1972)Schegloff, Emanuel A. 1972 “Notes on a Conversational Practice: Formulating Place.” In Studies in Social Interaction, ed. by David Sudnow, 75–119. New York: The Free Press. on place formulation.) However, rather than provide some sort of place formulation, Nori answers with a confirmation token and then a number (line 02). Eric takes the confirmation token as indicating that Nori has not understood his question and, in overlap with the number, initiates repair on his question in third position by saying “where.”

I will return to what happens in lines 03 to 15 below, but in response to Eric’s question in line 15, which again asks for information which is recipient-designed for someone familiar with Nagoya, Nori simply reiterates that the place he is referring to is in Nagoya (line 16). Following a gap (line 17), Eric then initiates repair on his question in third position in line 18, placing stress on the word “where,” indicating that the place formulation “Nagoya” is insufficient as an answer. Finally, and after another gap (line 19), in response to this, Nori provides a place formulation (line 20) which is recipient-designed for someone familiar with Nagoya, in that it is the name of a well-known neighborhood in central Nagoya.

6.Instances of avoiding initiation of repair

Excerpts 1 and 2 show that when Nori says something which indicates that he has not understood Eric’s prior turn, two ways that Eric may respond are to use an open class repair initiator to other-initiate repair on Nori’s turn and to self-initiate repair on his own turn in third position. Neither of these ways of responding is unusual. In addition, though, Eric often responds by avoiding the initiation of repair and not treating what Nori has said as sequentially inappropriate.55.In all instances found in the data of repair initiation being avoided, it is Nori who appears not to understand a prior turn and Eric who avoids initiation of repair. This can be seen, in minimal form, in Excerpt 2 above. In response to Eric’s repaired question in line 03, and following the gap in line 04, Nori responds with the name of a restaurant, rather than a location formulation recipient-designed for someone familiar with Nagoya. Instead of again initiating repair, though, Eric receipts the name by repeating it (line 07). He then asks the question again in line 09. The design of this question does not indicate that there is any problem with the sequential appropriateness of Nori’s talk in line 05. Rather, it is simply designed as a next question, with the name of the restaurant being replaced by “that.” Again, rather than provide a recipient-designed place formulation, Nori reiterates that “Tonki” is the name of the restaurant (line 11), perhaps misunderstanding “where’s that” as “what’s that.” Again, rather than initiating repair, Eric accepts the answer in line 12 and then asks another next question in line 15, this time using “it” to refer to the restaurant and providing a clue as to what sort of answer would be appropriately recipient-designed by saying “what part of Nagoya is it.” There is nothing about how the question is articulated, though, that would indicate that “what part of Nagoya” is designed as a replacement of “where” in any of the prior questions. That is, the question is not designed to initiate repair on Eric’s previous questions. Nor does it treat what Nori has said as sequentially inappropriate.

In Excerpt 2, Nori appears to understand that he has been asked an information question, but does not seem to understand what sort of information he has been asked to provide. When Eric avoids initiation of repair, he does this by first accepting Nori’s answer and then redoing the question as a next question, rather than as a repaired version of his prior question. The acceptance of Nori’s answer, though, is minimal, in each case consisting of only one word. In Excerpt 3, Nori again appears to understand that he has been asked an information question, but not the sort of information that he has been asked to provide. Prior to this excerpt, Nori and Eric have been discussing golf, a popular recreational sport in both Japan and Hawaiʻi which Nori likes to play.

01  E:  =do y- do you play golf very offen?

02  (1.5)

03  N:  m.

04  (0.4)

05  E:  how offen do you play golf.

06  N:  m:::↑:: ten- ten years?

07  (0.5)

08  E:  oh you’ve played for ten [years?

09  N:            [↑ah=ah ↓m.

10  E:  how offen do  [you play.

11  N:                [.hh (.) uh:  [tuh-

12  E:                              [do you

13      play every week?  [or

14  N:                    [wah::  no?

15  E:  ever  [y month?

16  N:        [o:ne       one month: one month ↑month:

17      ° ↓oo° (1.6) ° ih°  ↑one time ↓two time

18  E:  oh  [once or twi  [ce a month,  ]

19  N:      [(xx)         [↑once or t   ]en ↑yes:.

In line 05, Eric asks a question about the frequency with which Nori plays golf. This can be understood as a follow-up question for his question in line 01, which is a yes-no question also about Nori’s frequency of play. As a yes-no question, if the answer is “yes,” it makes possibly relevant more information, namely, information about how often Nori plays. It is the possible relevance of this additional information which allows the question in line 05 to be heard as a follow-up question. Following a long gap (line 02)66.The reviewer points out that according to Sacks, Schegloff, and Jefferson (1974)Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson 1974 “A Simplest Systematics for the Organization of Turn-Taking for Conversation.” Language 50: 696–735. (below, SSJ), this silence is not a gap, but a pause after the next speaker has been selected. The reviewer is correct with regard to what SSJ say, but it seems to me that there is a contradiction in SSJ and that it is more consistent to label this sort of silence a gap. In note 26 of SSJ, pauses are roughly characterized as intra-turn and gaps as inter-turn. This is a rough characterization because silences can be transformed. Further, in the so-called short turn-taking paper, written as an earlier version of SSJ but published as Sacks (2004)Sacks, Harvey 2004 “An Initial Characterization of the Organization of Speaker Turn-Taking in Conversation.” In Conversation Analysis: Studies from the First Generation, ed. by Gene H. Lerner, 35–42. Amsterdam: John Benjamins. , one technique is spelled out explicitly through which a selected next speaker can transform a possible gap into a pause, by saying “uh” and then pausing. This implies that a silence before a selected next speaker starts is a gap, rather than a pause. after the first question, Nori answers with only a minimal confirmation token and Eric’s follow-up question in line 05 pursues the possibly relevant frequency information. Nori then responds in line 06 by doing thinking with an elongated nasal and then answers the question with a formulation of time duration, rather than frequency.

As in Excerpt 2, Nori appears to have understood that he has been asked an information question, but not the content of the question. While the gap in line 07 may indicate trouble on Eric’s part with Nori’s answer, rather than initiate repair, in line 08 Eric uses “oh,” often called a change-of-state token (Heritage 1984Heritage, 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.), and a reformulation to receipt what Nori has said as informative and uses rising intonation to elicit confirmation of the reformulation, which Nori provides in line 09. He then re-asks the question (line 10) as the next question in a series. When Nori displays trouble answering (lines 11 and 14), Eric then initiates repair on the question in line 10 by providing candidate answers (lines 12–13, 15). This helps Nori to understand the question and formulate an answer in terms of frequency in lines 16 to 17. This is then accepted by Eric as informative in a manner similar to how line 08 accepted Nori’s previous answer as informative – it receipts the answer as informative with “oh” followed by a reformulation of the information, with the reformulation also providing an embedded correction (Jefferson 1987Jefferson, Gail 1987 “On Exposed and Embedded Correction in Conversation.” In Talk and Social Organisation, ed. by Graham Button, and John R. E. Lee, 86–100. Clevedon, UK: Multilingual Matters.) of Nori’s English. Though it does not end with rising intonation, this reformulation also elicits confirmation.

In some instances which involve Eric asking an information question, Nori does not seem to understand that he has been asked to provide information. Excerpt 4, in which Nori and Eric are discussing a recent fireworks show in celebration of Independence Day, a U.S. holiday, provides an example.

01  N:  look (.) .h mi  ni itta?

                    see pp go-as

                    ‘Did you go to see them?’

02  E:  ah we- we saw them from here.

03  N:  ah

04  (0.5)

05  E:  did you go an’ watch the firewor   [ks?

06  N:                                    [↑ah=↓ah

07      ↑(oo wah) ((you watch?))?

08  E:  yeah we watched them  [here. h heh

09  N:                        [oh=oh:.

10  (1.1)

11  N:  [↑hen (.) .h ↑hen     datta deshoo.=

         strange      strange cp-as  cp-as

                      ‘It was strange, right?’

12  E:  [did you

13  N:  =.hhh boh boh boh bohm. .h ssssss=

14  E:  =yeah there was an  [there was an=

15  N:                      [↑are heh

16  E:  =accident.  [right?

17  N:              [accident yeah.

18  E:  so (0.8) they:  they went off one time,

19  N:  n=n.

20  E:  but then they stopped,

21  N:  m=m.

22  E:  right, (0.4)  [an’ then a second time they=

23  N:                [(x)

24  E:  =star[ted again,

25  N:       [soo   soo.

              right right

              ‘Exactly.’

26  E:  then they stopped again

27  N:  yeah Ala Wai    de=

             place name pp

             ‘At Ala Wai.’

28  E:  =an’ then a third time once  [more.

29  N:                               [n:  Ala Wai Harbor?

30  E:  yeah

31  N:  (oide) (.) wuh- (.) one pe(h)rso(h)n ha

32      hi(h)to(h)ri de.   [.hhh

        one-person   pp

        ‘By myself.’

33  E:                     [you went there by

34      your  [self?

35  N:        [yeah .h ((continues))

In the sequence in lines 01 to 03, Nori asks if Eric went to see the fireworks, Eric answers that he and (presumably) his wife saw them from their apartment, and Nori claims understanding with an “ah.” At this point, this sequence (though not necessarily the topic) can be considered closed. After a half-second gap (line 04), Eric then asks Nori in line 05 basically the same question in English that Nori had asked in Japanese. However, Nori either does not notice or simply ignores the fact that Eric asks this question and instead produces two more “ah”s (line 06), the first one in overlap with the very end of Eric’s question, and, apparently, asks for confirmation that Eric watched the fireworks (line 07). This is how Eric understands line 07, as he confirms that they watched the fireworks from their apartment (line 08), to which Nori responds in line 09.

Nori’s question in line 07 indicates that he has not completely understood Eric’s answer (in line 02) to his first question, in spite of his claim of understanding in line 03. It also indicates that he does not recognize that he has also been asked a question (in line 05), that Eric has initiated a new sequence, or if he does recognize this, he ignores it. Perhaps the reason that he does not understand that he has been asked a question is the overlap in lines 05 and 06. However, most of Eric’s question in line 05 is produced in the clear, following the silence in line 04, and the overlap only occurs at the very end. Rather than respond to this lack of understanding by either initiating repair in third position on his earlier question or initiating repair on Nori’s question in the next turn through, for example, an open class repair initiator (Drew 1997Drew, Paul 1997 “ ‘Open’ Class Repair Initiators in Response to Sequential Sources of Troubles in Conversation.” Journal of Pragmatics 28: 69–101. ), Eric goes along with the direction set by Nori’s question by answering it, thus avoiding initiation of repair. After another gap, Eric starts to ask another question (line 12), possibly, and in this sequential context likely to be, the same as or similar to his question in line 05. However, at the same time, Nori moves the discussion in a different direction and begins talking about something that happened at the fireworks show (line 11). Eric again goes along with this and the two of them enter a discussion of what happened at the fireworks show that continues beyond the end of this excerpt. In lines 31 to 32, Nori also incidentally provides information that entails the answer to Eric’s question in line 05 and this question never gets re-asked. While in some cases (e.g., Excerpts 2 and 3) Nori understands that he has been asked a question, and in others (e.g., Excerpt 4) he does not, this does not seem to have an effect on the practice of avoiding repair initiation.

In the instances of avoiding initiation of repair looked at above, Eric’s misunderstood turn has been a first pair part and, more specifically, a question. There are also a few instances in which Eric seems to be developing a topic, but then Nori produces an abrupt topic shift. While these instances cannot really be said to involve avoidance of repair initiation, similar to the instances above, Eric goes along with the topic shift and then later returns to the topic that he was developing earlier. An example is shown in Excerpt 5, in which they are discussing the first place Eric lived in Tokyo, which was the neighborhood of Ikebukuro.

01  E:  it was very noisy around where I lived.=

02  N:  =ah::

03  E:  so I moved to uh Higashi Nagasaki.

                         place name

04  N:  Japanese ramen ↑Taishoken?

                        restaurant name

05  (1.0)

06  E:  [oh I’ve never [been there.

07  N:  [I know?       [you know?

08  N:  ah=Taishoken waka   [nnai?

                     understand-ng

           ‘You don’t know Taishoken?’

09  E:                      [.hh  actually my- (.) my

10      apartment was closer to thee uh: kuyakusho,

                                         ward-office

11  N:  ah kuyakusho. ↑oh:::  ↓oh=↓oh:. ↑oh ↓hoh

           ward-office

12      ↓hoh. ((not laughter))

13  (0.2)

14  N:  .hh moo    ↑s:ugoku near.

                        really  very

15  E:  oh really?

16  N:  n.  [.hh ↑my:  ↓uh ↑every day (0.5)↓a:no .h=

                                            sf

17  E:      [° oh.°

18  N:  =(2.0)↑baikku     de totteta.

               motorcycle pp go-through-as-as

               ‘I went by there.’

19  E:  ah ↑okay.

20  N:  n.

21  (1.1)

22  E:  then ↑after that I moved to uh ↑Higashi

23      Nagasaki.

In line 01, Eric negatively assesses the place in Ikebukuro where he lived as “very noisy” and Nori responds with an elongated “ah” (line 02). Eric then states that he moved to a different location, using “so” to formulate this as done because his prior neighborhood was “very noisy” (line 03). While Eric has thus shifted the topic slightly from Ikebukuro to a different neighborhood, Higashi Nagasaki, Nori abruptly shifts back to Ikebukuro and asks in line 04 about a famous restaurant in the Ikebukuro area. Without anything to mark it as tied back to earlier talk about Ikebukuro, it comes across as responsive to Eric’s talk in line 03. However, as what Eric says in this line is about a different neighborhood, it appears that Nori does not realize that Eric has shifted the topic away from Ikebukuro. There are two indications of trouble on Eric’s part with what Nori has said, namely, the gap (line 05)77.See discussion of pauses and gaps in note 6. following Nori’s turn and the oh-prefacing (Heritage 1998Heritage, John 1998 “Oh-Prefaced Responses to Inquiry.” Language in Society 27: 291–334. ) of Eric’s eventual response (line 06). However, Eric goes along with the topic shift back to Ikebukuro and responds with a comment about his lack of experience with this restaurant (line 06). He then formulates more specifically where within Ikebukuro he lived (lines 09–10), after which Nori talks about his own experience in that area (lines 14, 16, and 18). Finally, after they close the topic in lines 19 to 20 and the gap in line 21, Eric states once more that he moved to a different neighborhood (lines 22–23). This is formulated not as a consequence of Ikebukuro being “very noisy,” but rather as simply what he did next.

In most instances of avoiding initiation of repair, Eric eventually attempts a return to earlier talk, generally by re-asking his misunderstood (or non-understood) question or a similar question. Along a conversation-internal, or enchronic (Enfield 2009Enfield, N.J. 2009 The Anatomy of Meaning: Speech, Gesture, and Composite Utterances. Cambridge: Cambridge University Press. ), time scale, when it does occur, this return to earlier talk can be almost immediate, as in Excerpts 6 and 7, or much later, as in Excerpts 8 and 9. Following discussion of where Eric lived in Tokyo, in Excerpt 6 Eric asks Nori where he lived.

01  E:  where did you live in Tokyo.

02  N:  .hhh (1.0) ° where do you live in Tokyo:° (.)

03      ↑eight year.

04  E:  ah.=↑di- did you live in Ikebukuro?

                                 place name

05  (0.5)

06  N:  m ↑ah=m. ↑hotondo. ↑ano::: no? .hh first

                  mostly    sf

07      ti:mu (.) Itabashiku.

                  place name

08  E:  Itabashi   [ku.

09  N:             [yes,

10  E:  oh ↑okay.

Even though in line 02, Nori eventually repeats, with a slight change, the question Eric has asked, he does not seem to understand it, as in line 03 he answers with a time formulation, rather than a place. Eric receipts this as informative in line 04 with “ah” and then immediately asks a similar question, with some self-initiated self-repair, in line 04. While it is possible to understand the second question as a repaired version of the first question, especially as it is changed to a presumably easier to answer yes-no question, it should be noted that Eric starts not by initiating repair but by receipting the time formulation as informative. It should also be noted that with the shift up in pitch as Eric starts the question, the question is hearable as separate from the information receipt in spite of the latching. He thus, at least initially, avoids repair initiation, though it is probably better to say that repair initiation is not so much avoided, as just very slightly delayed. (That Nori goes on to answer the question with a place formulation, rather than a type-conforming (Raymond 2003Raymond, Geoffrey 2003 “Grammar and Social Organization: Yes/No Interrogatives and the Structure of Responding.” American Sociological Review 68: 939–967. ) yes or no, may be related to the fact that Itabashiku is very close to Ikebukuro.)

In Excerpt 7, Nori and Eric are discussing a sushi chef at the restaurant where Nori works, who apparently spends a lot of time surfing.

01  E:  .hh  [↑does uh ↑does Minami: go surfing:=

                             personal name ((pseudonym))

02  N:       [.hhh

03  E:  =very offen?

04  (0.2)

05  N:  n.

06  (1.2) ((N sips coffee))

07  N:  m (0.4) ((swallows)) offen?

08  E:  does he go surfing: a lot?

09  (0.3)

10  N:  does he (0.7) .h does he ↓n. (0.8) does

11      ↑go [ing sus  ]

12  E:      [does: ↑Mi]nami,

13  N:  ↑Minami. yes.

14  E:  go surfing

15  N:  n.

16  E:  a lot?

17  (1.3)

18  E:  does ‘e go surfing (0.3) offen?

19  (0.5)

20  N:  .h off offen ↓nani?

                      what

21  (0.2)

22  E:  all the time?

23  N:  oh oh ↓oh oh. n↑: .hh m::: ↑doo daro.

                                    how cp-as

                           How is it?

24     (1.3) fuh- five year,

25  (0.8)

26  E:  oh he’s been surfing for a [long time.

27  N:                             [n. (.)

28      ↑long time.

29  E:  oh ↑okay.

30  (0.2)

31  E:  .h but ↑do you know .h uh: (1.3) how offen

32      he goes surfing?

After responding to Eric’s question in lines 01 and 03 with “n” and then sipping his coffee, Nori initiates repair in line 07 by repeating the last word of the question with rising intonation. This is the start of fairly extensive repair work, as Eric reformulates the question (line 08), Nori attempts to repeat parts of the reformulated question (lines 10–11), Eric breaks down the question into smaller pieces (lines 12, 14, and 16), Nori responds to these smaller pieces (lines 13 and 15), Eric reformulates the question again (line 18), Nori explicitly asks what meaning of “often” is (line 20), Eric provides a definition of “often” (line 22), and Nori finally claims understanding (line 23). Nori then does thinking with an elongated “m” and a Japanese expression, followed by a silence, before providing an answer. The answer he provides, as a formulation of duration, indicates that he has not actually understood the question, which is about frequency and is formulated as a yes-no question. Rather than initiate (further) repair, Eric receipts the answer as informative and reformulates “five year” as “a long time” (line 26). Nori confirms the reformulation through repetition in line 28, after which Eric once again receipts what Nori has said as informative (line 29). After a brief gap (line 30), Eric produces another version of his question about frequency, though this is done with a fair amount of disfluency. In this excerpt, unlike in Excerpt 6, Eric first closes the sequence of which Nori’s answer in line 24 is a part before re-asking a question related to frequency.

At the other extreme, there is approximately 10 minutes of talk about several different topics between the avoidance of repair in except 8 and the return to earlier talk in Excerpt 9. Prior to Excerpt 8, Nori and Eric have been talking about a visit by a carpenter to Nori’s apartment to fix the front door.

01  E:  =how- how did it break.

02  (0.2)

03  N:  yeah. .hh ↑the las- las- last day ↑n=↓n

04      ↑last week? .hh  [accident t- telephone::=

05  E:                   [uhhuh

06  N:  =cable (0.6) cut.

07  E:  yeah,

In line 01, Eric asks an information question about how the door broke, but Nori appears to understand the question as a yes-no question (e.g., “Did it break?”) and replies with a confirmation token in line 03. He then introduces an unrelated event which resulted in the interruption of telephone service for several days in his neighborhood. That he is introducing an unrelated event, though, does not start to become evident until the end of line 04. Eric does not initiate repair either on his question or on Nori’s answer and they move stepwise through a series of topics (not shown), discussing such things as whether the interruption of phone service was in the local news, Nori’s inability to check email as a result of the interruption, and the costs of phone service in the U.S. and Japan, among other things.

Excerpt 9 starts just after this last topic has been closed, approximately ten minutes after Excerpt 8.

01  (1.1) ((N sips coffee))

02  E:  so what happened to your door.

03  (1.4)

04  N:  happen to door?

05  E:  yeah you said ↓uh (.) ↑a carpenter

06      came to your place last week.

07  N:  oh=oh=oh

08  E:  to  [fix the door.

09  N:      [.hhh         yeah (0.2) ah: (1.3)

10      (goor) (kagagu) (.) (nan). hhh=

11  E:  =the:  hinge?

With the prior topic having been closed, Nori starts to sip his coffee in line 01. Eric then returns to earlier talk in line 02 by again asking about the problem with the door. This question is designed quite differently from the question in Excerpt 9, being marked as topically-disjunctive in this location but also something that is already “on the conversational agenda” (Bolden 2006Bolden, Galina B. 2006 “Little Words that Matter: Discourse Markers “so” and “oh” and the Doing of Other-Attentiveness in Social Interaction.” Journal of Communication 56: 661–688. ) through the use of “so,” containing a reference to the door as “your door” rather than “it,” and presupposing only that something happened to the door, not that it broke. Nevertheless, it can still be understood as a version of the earlier question. This is followed by some repair work related to what is being asked and the talk turns to the problem with the door.

7.Discussion – intersubjectivity and construction of interactional competence

Repair is necessary in order to maintain an adequate degree of intersubjectivity within interaction. As shown in Shegloff (1992), when problems with intersubjectivity are not recognized and, as a result, not repaired, the consequences for the interaction and the relationship of the participants can be disastrous.88.See Tyler (1995)Tyler, Andrea 1995 “The Coconstruction of Cross-Cultural Miscommunication: Conflicts in Perception, Negotiation, and Enactment of Participant Role and Status.” Studies in Second Language Acquisition 17: 129–152. for another example of what happens when problems with intersubjectivity are not recognized and Hauser (2003)Hauser, Eric 2003 ‘Corrective Recasts’ and Other-Correction of Language Form in Interaction among Native and Non-Native Speakers of English: The Application of Conversation Analysis to Second Language Acquisition. Unpublished doctoral dissertation, University of Hawaiʻi at Mānoa. for a reinterpretation of Tyler’s data in terms of failure to recognize problems with intersubjectivity. Nevertheless, there are also cases in which some problem with intersubjectivity seems to be recognized by at least one participant, but that participant does not initiate repair. In addition to those found in the dataset for this paper, a subset of which are presented above, Schegloff (1992)Schegloff, Emanuel A. 1992 “Repair after Next Turn: The Last Structurally Provided Defense of Intersubjectivity in Conversation.” American Journal of Sociology 97: 1295–1345. http://​www​.jstor​.org​/stable​/2781417 presents an instance of this and they can also be found in Liberman (1980Liberman, Ken 1980 “Ambiguity and Gratuitous Concurrence in Inter-Cultural Communication.” Human Studies 3: 65–85. , 2012Liberman, Ken 2012 “Semantic Drift in Conversations.” Human Studies 35: 263–277. ) and Long (1983)Long, Michael H. 1983 “Native Speaker/Non-Native Speaker Conversation and the Negotiation of Comprehensible Input.” Applied Linguistics 4: 126–141. . In these instances, whatever the motivation may be for not initiating repair, and even for avoiding initiation of repair, a separate means of maintaining intersubjectivity becomes visible. This involves what Liberman (1980)Liberman, Ken 1980 “Ambiguity and Gratuitous Concurrence in Inter-Cultural Communication.” Human Studies 3: 65–85. terms gratuitous concurrence. One participant recognizes that there is a problem with intersubjectivity in the form of another participant not understanding what has been said, but this problem is dealt with and intersubjectivity is maintained by the first participant treating the second participant’s understanding as appropriate, that is, going along with the other participant as a way of circumventing the need for repair. One thing that neither Liberman and Long consider (but that Schegloff does) is that the misunderstood talk may remain available to be returned to. While repair is certainly necessary and central for the maintenance of intersubjectivity, in some circumstances, there may be other ways to accomplish this.

What, though, is accomplished by using other means? One thing that is accomplished is the continued progressivity (Stivers & Robinson 2006Stivers, Tanya, and Jeffrey D. Robinson 2006 “A Preference for Progressivity in Interaction.” Language in Society 35: 367–392. ) of the interaction. When a prior turn-at-talk is targeted for repair, progressivity is interrupted (Kitzinger 2013Kinzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Malden, MA: Blackwell.). By avoiding initiation of repair, progressivity is prioritized, with the possibility that, as the interaction progresses, it may be brought back to the talk that was misunderstood but not repaired. While the need for repair is by no means eliminated, the practice of avoiding initiation of repair allows for a decrease in the frequency with which progressivity is interrupted. Being able to participate in interaction without causing an excessive amount of trouble for progressivity can be considered one of the main elements of interactional competence. In the conversations-for-learning that form the dataset for this paper, Nori is a fairly limited proficiency user of L2 English, though his proficiency in at least some areas does slowly improve over the seven-month recording period (Hauser 2013aHauser, Eric 2013a “Stability and Change in One Adult’s Second Language English Negation.” Language Learning 63: 463–498. , bHauser, Eric 2013b “Expanding Resources for Marking Direct Reported Speech.” In Pragmatics & Language Learning, Vol. 13, ed. by Tim Greer, Donna Tatsuki, and Carsten Roever, 29–53. Honolulu, HI: University of Hawaiʻi, National Foreign Language Resource Center.). Indeed, the ostensible purpose of the meetings between Nori and Eric is to provide Nori with opportunities to practice English, opportunities which are only necessary because of his limited proficiency with the language. Problems with understanding can therefore be expected to be frequent, making progressivity especially vulnerable. Through Eric’s avoidance of initiation of repair, the two participants are able to maintain progressivity and Nori (as well as Eric, for that matter) is constructed within the interaction as interactionally competent to participate in L2 conversations-for-learning. This is not to say that Nori has some sort of independent interactional competence to participate in these conversations-for-learning, but rather that one way in which Eric contributes to the construction of Nori as interactionally competent is by avoiding initiation of repair.

8.Concluding remarks

An old idea related to talking with L2 users was the concept of foreigner talk (Ferguson 1975Ferguson, Charles A. 1975 “Toward a Characterization of English Foreigner Talk.” Anthropological Linguistics 17: 1–14.), that is, how L1 users adjust their language when they talk to L2 users whose L2 proficiency is perceived to be limited. Quantitative work on interaction within Second Language Acquisition (e.g., Long 1983Long, Michael H. 1983 “Native Speaker/Non-Native Speaker Conversation and the Negotiation of Comprehensible Input.” Applied Linguistics 4: 126–141. ) moved beyond the language adjustments associated with foreigner talk to look at conversational adjustments, that is how L1 users talking with L2 users may modify their interactional behavior. Though this may not have been the intention of researchers looking at either foreigner talk or conversational adjustments, it is not difficult to conclude on the basis of this research that there is something unusual or even abnormal about interaction involving L2 users, in that it is presented in terms of how it is different from normal interaction. Even Liberman’s (1980)Liberman, Ken 1980 “Ambiguity and Gratuitous Concurrence in Inter-Cultural Communication.” Human Studies 3: 65–85. Ethnomethodological approach comes across as treating what he calls “inter-cultural communication” as abnormal in comparison to what can be expected when communication is not “inter-cultural.” However, in concluding this paper, I would like to emphasize three ways that interaction involving L2 users, even of extremely limited proficiency, is normal. First, at the broadest level, interaction involving one or more participants using an L2 is by no means an infrequent occurrence. Around the world, there are people who interact on a regular basis in an L2. This is not only true of interaction in our modern, ‘globalized’ society. Linguistic homogeneity is not and probably has never been a common feature of society and is certainly not a feature of Hawaiʻi, where these data were collected.99.Nori, of course, if from Japan, where linguistic homogeneity tends to be assumed. Even Japan, though, is more linguistically diverse than is commonly recognized (Heinrich 2012Heinrich, Patrick 2012 The Making of Monolingual Japan: Language Ideology and Japanese Modernity. Bristol, UK: Multilingual Matters. ). Second, participants in interaction work to “normalize” it (Firth 1996Firth, Alan 1996 “The Discursive Accomplishment of Normality. On ‘Lingua Franca’ English and Conversation Analysis.” Journal of Pragmatics 26: 237–259. ). This can be seen in instances of avoiding initiation of repair, as turns-at-talk which demonstrate misunderstanding are treated as unproblematic. And third, there is nothing that happens in interaction involving L2 users that cannot be found in interaction with no L2 users, though there may be differences in relative frequency. As Schegloff (1992)Schegloff, Emanuel A. 1992 “Repair after Next Turn: The Last Structurally Provided Defense of Intersubjectivity in Conversation.” American Journal of Sociology 97: 1295–1345. http://​www​.jstor​.org​/stable​/2781417 demonstrates, for example, initiation of repair may be avoided when there are no L2 users. This claim is impossible either to verify with complete certainty or to falsify, as on the one hand there may always be something which occurs only in L2 interaction which simply has not been found yet, while on the other something which is claimed to occur only in L2 interaction may simply be rare and difficult to locate in interaction which does not involve L2 users. Nevertheless, it is a claim which is supported by the growing body of research on L2 interaction.