Laughter is a normal nonverbal interactional resource, which occurs not only in native language interaction (Jefferson, 1979; Glenn, 2013; Haakana, 2001) , but in L2 settings (Cekaite, 2007; Petitjean & Gonzalez-Martinez, 2015) . In this study of a Chinese university’s oral proficiency interview which takes English as the foreign language and determines the admission into the target college in Britain, laughter happens in a great frequency, particularly in the responding positions, yet it is not any of usual reactions to humor. Rather, it accomplishes certain actions and plays various roles in terms of its prosodic features, places and sequential environment. Therefore this study collects cases of laughter that occur as major responses to various actions and intends to identify the actions they perform, and further explores the reasons why laughter, rather than the verbal responses, happens. By doing so, the study contributes to the proposition that laughter, by means of doing alignment or affiliation, turns out to be one of the skillful interactional resources of L2 interactions.
1.1. Laughter in Institutional Settings
Since Jefferson’s pioneer work on laughter’s analysis (Jefferson, 1979) , trouble’s talk (Jefferson, 1984) and its transcription (Jefferson, 1985) , there have been a great number of researches keeping on, and among them many have proved that there’s not a simple stimulus-response relationship between laughter and humor. Potter and Hepburn (2010) discover that laugh particles inserted into words can be used to mark one or more lexical items as having some problem or insufficiency, and as a resource for action formation to modulate the nature or strength of the action. In Holt’s (2010) finding, both the occurrence and absence of laughter at an appropriate juncture may result from orientation to interactional considerations such as the ongoing trajectory of the talk. Clift (2012) finds out that laughter, through precisely placed, is intrinsically connected to reported complaint. It plays a key role in the way what is reported is formed up for its circumstances at that time. Those innovative studies show that laughter was not only something worthy of careful description but something that is organized in its fine detail to coordinate with and sometimes sustain ongoing actions. In Jefferson’s words, laughter has done “a recognizable sort of job” (Jefferson, 1984: p. 351) .
Such an idea of laughter connected to an action formation could be more evident in institutional researches. Haakana (2001, 2002) and Wilkinson (2007) take ordinary patients and the particular aphasic patients in medical interaction as the participants and discover that they both employ laughter as a device to deal with delicate issues. Glenn (2013) explores the potential employee’s use of laughter in handling nervousness in interviews, which involves self-directed assessing and knowing, self-boasting, and managing insufficient answer. Walker (2017) examines laughter produced by young children to their mother’s questions and argues that by laughing the children are showing an orientation to the conditional relevance of a second pair part on the production of a first pair part and the possible bases for young children using laughter to respond to questions are the children’s inability to answer a question in full, and their unwillingness to do so.
Along with this idea, this article continues to explore what sort of job laughter does, given the particular circumstance of oral proficiency interview in which participants are characterized by L2 learners in the college. To be more exact, this study aims to explore what action laughter performs or what role laughter plays in action formation in oral proficiency interview.
1.2. Laughter in Responding Positions and Its Stance
Given the institutional interaction, a salient feature of “asymmetry” has been found in terms of the use of laughter, such as such as in interviews (Glenn, 2010) , doctor-patient interactions (Haakana, 2002) , interactions with aphasic speakers (Wilkinson, 2007) , adult-children interactions (Walker, 2017) , and instructional settings of interaction (Petitjean & Gonzalez-Martinez, 2015) . Laughter in those situations is majorly produced by the participants who do not have any institutional identity, and normally described as volunteered laughter (Glenn, 2013) and used to handle variant delicate situations or play certain roles in the progressivity of sequence.
In terms of the responding positions, laughter received much attention in medical settings. The most significant are Ticca’s (2011) and Haakana’s (2001) work. The former detects the patients’ use of laughter as responses in medical settings to resist the engagement with or replying to doctor’s question or criticism and the latter discovers that laughter marks the talk “delicate” or problematic and displays patients’ refusal to reply to doctors’ questions or actions, particularly in isolation, as response to prior actions. It is also recently identified in the previously-mentioned Walker’s (2017) similar findings on children’s sole laughter response to questions due to inability or unwillingness.
Most of those studies display the functional disalignment or disaffiliation of laughter as responses to prior actions. However, laughter is generally regarded as a marker of affiliation (Glenn, 2013) . It is “a coincidence of thought, attitudes, sense of humor and the like” as quoted in Glenn’s (2003: p. 29) work. Bell (2013) has ever provided evidence in a quantitative study. Clift (2016) has demonstrated as well that laughter from the audience can be affiliative with its recipient and disaffiliative to its prior turn in broadcasting interview setting. Some studies (e.g. Stivers, 2008 ) make a distinction between alignment and affiliation by conceptualizing alignment as the structural level of cooperation and affiliation as the affective level of cooperation, while this study treats them basically in the same line and does a little distinction according to the prosodic features. I tend to employ Dings’ (2014: p. 744) basic explanation on alignment, which refers to “the ways in which interlocutors demonstrate their intersubjectivity by showing each other that they are understanding each other and are being understood” and affiliative responses cooperate at the level of action and affective stance which are maximally pro-social when they match the prior speaker’s evaluative stance, display empathy and/or cooperate with the preference of the prior action (Stivers et al., 2011: p. 21) .
Given this discrepancy that laughter has demonstrated and the insufficient studies of laughter in the alignment orientation, this study proposes that laughter is one of the alignment or affiliation expressions in this oral proficiency interview and is supposed to deal with laughter as the major constituents of the response to various actions, except the question-formed inquiry action (which is the major action in the interview and will be discussed in other study).
1.3. Interactional Competence
Interactional competence, since its birth (Kramsch, 1986) , has been argued by He and Young (1998) to be incorporated into the framework of communicative competence (Bachman & Palmer, 2010; Canale & Swain, 1980) , for the conviction that different interactional styles can lead to misunderstandings and pragmatic failure in conversations between speakers from differing linguistic backgrounds. L2 interactional competence has recently been investigated in l2 classroom contexts (e.g. Cekaite, 2007 ), in language proficiency interview (e.g. Galaczi, 2014 ; Lee, Park, & Sohn, 2011 ) and in study abroad contexts (e.g. Ishida, 2009 ; Masuda, 2011 ).
Young (2008) defines interactional competence as a “relationship between the participants’ employment of linguistic and interactional resources and the contexts in which they are employed”. IC includes resources related to turn and topic management, knowledge of rhetorical scripts, knowledge of the pragmatics of specific syntactic structures and lexis, and the means for signaling boundaries (He & Young, 1998) , and identity resources related to participation frameworks, linguistic resources such as register and modes of meaning, and interactional resource including speech acts, turn taking, repair, and boundaries (Young, 2008) . Markee (2008) also proposed three components of interactional competence: 1) language as a formal system including pronunciation, vocabulary and grammar; 2) semiotic systems, including turn-taking, repair and sequence organisation; 3) gaze and paralinguistic features. This framework is highly generalized, but do have a place for paralinguistic features.
1.4. Laughter in This Study
In China, nearly all the college students except those major in non-universal language in foreign language colleges, take English as the foreign language. The study-abroad orientation leads them to participate in many oral proficiency interviews, in which their performance is irresistibly influenced by the pressure of intense competition and test tension, let alone their sufficient oral interactional competence (Walker, 2017) .
Unpresumably in this study, there happen quite a number of laughter from the candidates’ side which is incompatible with the tension of the test, but accords with Glenn’s (2010) findings on the asymmetries of the institutional setting. Specifically there appear many instances in which laughter occur as the major constituent of the response to first pair part (FPP), which performs various actions, involving appreciation, wish, telling, compliment, repair solutions and formulating (except questions, which is the most frequent action according to data and more complicated, and therefore will be investigated alone). By using major, I mean the response is composed of a single laughter or laughter together with other verbal response, but laughter is heard as dominant. Here are three typical instances which occur at a stretch near the end of one interview.
Extract 1 (Yu cong)
38 → IR: Yes, quite, uhm, yes > (what set me things) < you sound a very interesting one
39 → IE: e hehh e [.hhhhhh
40 → IR: [and I’d like to wish you good luck with it.
41 → IR: and I’d like to wish you good luck with it.
42 → IE: huh£th(h)ank(h) you h£
43 → IR: So thank you very much
44 → IE: e he hu
45 → IR: that’s-that’s the end of your interview.
46 → IE: oh
The three extracts occur near the end of the interview. Following them is the announcement of the end by the interviewer. The three FPPs in these extracts perform compliment (line 38) in Extract 1, wish (line 41) in Extract 2, and appreciation action (line 43) in Extract 3 respectively, yet the same laughter occur as major responses to these actions. Though in the wish sequence, the response involves not merely the laughter, it is discerned acoustically louder and prevalent due to its strong explosiveness of “h” and therefore more obvious and dominant than the verbal expression “thank you”. Hence laughter in this case is still considered as the major constituent of the responding turn. Since laughter is prominent in the responding turn, it becomes the objective of this study. Basically three questions will be investigated:
1) what actions laughter is employed to accomplish or concurrently construct?
2) why laughter is employed in this particular situation?
3) what is the relations of laughter and L2 Interactional Competence(IC)?
2. Data Collection and Methodology
The data are taken from the audio recordings of around 260 minutes’ interviews (which were not conducted in one day and each interview, involving one interviewer and one candidate, lasted for different length) on 23 candidates’ oral language proficiency, as a gate-keeping test for the interviewer’s college. Each interview involves three parts: free conversation, self-presentation and the following questions about the self-presentation. The interviewer is from the target college in Britain, and the interviewees are from a university of China, who are mostly in their third year of college life with much diversified majors.
This study specifically selected those sequences in which laughter is employed as a major constituent in the responding turn to various actions except question-formed inquiring and transcribed them systematically and strictly according to Jefferson’s (1974, 1985) and Glenn’s (2013) work, under the aid of “audacity” software. The data transcription has been recognized and agreed on details by the Chinese DIG workshop colleagues.
CA, as one of the empirical research methods, starts from a stance that language, especially in talk-in-interaction, delivers social action instead of meaning, and what actions are conducted depends mostly on their sequential positions in the conversation. The minimal sequence in interaction consists of two paired utterances: the adjacency pair. Schegloff (2007) has characterized the adjacency pair as composed of two adjacently placed turns by different speakers, and ordered as first pair part (FPP) and second pair part (SPP). And most importantly is that those two adjacency pair should be pair-type related. That is, the FPP and SPP should come from the same pair type, such as greeting-greeting, question-answer, offer-accept/decline, and the like. In other words, in doing an action by means of a turn-at-talk, a speaker standardly projects the relevance of a particular “next” action to be done by a subsequent speaker and in doing a next action, a speaker normally addresses the immediately preceding talk (Clift, 2016) . For instance, a request as a FPP will make the next turn a granting or a decline. Anything that is not conformed in type will still be considered as relevant and understood as a special form of granting or a decline. That is basically how laughter in this study is treated, considering its responding position, turn design and prosodic features.
Apart from the adjacency pair, the attention has to be paid to the preference of SPP to the FPP, which refers to the different alignment that alternative responses embody. The preferred SPP is the response to the FPP which embodies or favors furthering or the accomplishment of the activity (Schegloff, 2007: p. 59) in the sequence. For example, in terms of a request, a granting as the SPP is a preferred response, while a decline a dispreferred one. The issue of “preference” and “dispreference” concerns the alignment in which a second action stands to a first.
The methodology of CA imposes analysts to regards all the action, the pairs or the preference organization as in the whole sequence of the activity. Any conclusion cannot be reached in a rough and haste way without considering their sequential positions and roles in the whole structure.
For the presented data, I use arrows to mark the relevant prior actions, and their responding turns, and capitalized IR to refer to the interviewer, IE to the interviewee.
3. Data Analysis
Question-formed inquiring as FPP are the main action type in each interview and constitute the major activity of this data as well and will be discussed in another article. The rest of the prior action types, ranging from appreciation, wish, telling, compliment, repair solution, to formulation, all share the same laughter as SPP. Although those non-question type FPP occupy only a small percentage in number, they are nearly all responded with laughter, which indicate a salient problem from candidates’ side and justify our attention to that phenomenon. The data of this study will be analyzed and illustrated qualitatively based on the CA methodology, in terms of various FPP.
3.1. When FPP Is an Appreciation or a Wish
It is discovered that near the end of the interview, as a pre-closing sequence, the interviewer is very likely to make a wish and a ritual gratitude to the candidates, as in Extract 2 and 3 which happen at a stretch, and after that the interviewer finishes the sequence and the interview.
41 → IR: and I’d like to wish you good luck with it.
42 → IE: huh £th(h)ank(h) you h£
43 → IR: So thank you very much
44 → IE: e he hu
45 → IR: that’s-that’s the end of your interview.
46 → IE: oh
Both the responses in these two extract consist mainly of laughter. For the former, the candidate delays her answer by means of laughing in a short and closed sound of the vowel, “u”, together with the out breath, which displays an understanding process, or searching for the words. The subsequent verbal response “thank you” is produced with explosive “h” within the words to reveal the simultaneous laughter. It is what Potter and Hepburn (2010) has defined as IPA, the “interpolated particles of aspiration”, which can be used to “mark one or more individual lexical items as having some problem or insufficiency and are a resource for action formation”. That is, they may signal “possible trouble in the use of the word or that there is more going on than the mere use of the word would indicate”. They can be inserted into lexical items within turns to finesse or modulate the action that is produced in some way. So the initial delay of laughter and IPA within “thank you” mark the recipient’s hesitance in understanding and searching for the words and insufficiency in delivering the answer, and when delivering, they fitnesses the delicacy of being verbally incompetent and concurrently construct an appreciation to the wish and relational friendliness.
For the latter, the normal practice for speaker to respond to thank you is to acknowledge that gratitude or downgrade the efforts or the hardship of what is done, by saying “that’s all right”, “it doesn’t matter” or the like, or upgrade the position of the other side, by saying “you’re welcome”, or “with my pleasure”. However, the gratitude of this case is not the hard labor or help that candidates have offered to the interviewer. The interviewer is just being courtesy and doing something ritual (Goffman, 1967) , which is very common for the interviewer or the lecturer to pre-finish the conversation or the reporting, and in return, the recipient has an appreciative or grateful nature too, and is very likely to provide verbal “thank you”, or bodily feedback, like nod, smiling, or applause by clapping their hands, etc. Similarly in this case, the candidate resorts to the three laugh particles of all closed sounds for a returned appreciation. This laughter features the response as delicate due to the speaker’s potential lack of linguistic and pragmatic competence, but skillfully fulfills a combination of acknowledging the gratitude, downgrading the favor, and displaying a weak gratitude to the interviewer, thus an alignment stance.
Another instance indicates the idea of the laughter as an affiliative contributor to a returned appreciation, as in Extract 4. It is the last reciprocal turn in the interaction of the interview.
17 → IR: That sounds very optimistic way > of looking at things≤
18 → IE: =hehhh
19 → IR: o::k, thank you very much.
20 → IE: £th(h)ank you very m(h)uch£
The candidate in line 20 of this case appears friendlier, more enthusiastic, and more grateful to the interviewer by laughing along with the verbal expression and with he’s inserted into “thank” and “much”. As Potter and Hepburn (2010) has suggested, IPA indicates more than a mere returned appreciation, it might reveal in this position a willingness to turn their relationship into an intimate one (Rubino, 2016) , if the solidarity has been built through the interview. Laughter, consequently, together with the verbal utterance, constructs a strengthened appreciation in return, and demonstrates the speaker’s affiliation to the prior turn.
The use of laughter response instead of literal utterance may very well result from the failure to take the prior utterances as social actions and the candidates’ incompetence to respond linguistically due to insufficient knowledge of language use. Laughter serves as a disguise to their incompetence and displays a good intention to build the solidarity. It is commonly acknowledged that the vocal laughter, when in an incompetent linguistic state in the testing setting, compared with silence and unvoiced smiling, shows more about candidates’ participation and good intentions. When it is produced saliently but singly in terms of prosodic features or served as an emphasis on the verbal expression, it constructs the response as an affiliative one.
3.2. When FPP Is Telling
The action of telling in this data of interview, normally derives from the interviewee’s side, touched off (Jefferson, 1978) by the interviewer’s topic-proffer question. In this sequence, the interviewee’s telling is also touched off by the interviewer’s question about the pet, and elaborates on her experience and feeling about pets, though not type-conformed. Whereas in this sequence, it is the interviewer who performs the telling action, which is reciprocal and touched off by the interviewee’s telling. The interviewee now turns to be the story recipient, as the example shows:
12 → IR: Oh, They are sweet, aren’t they? Do you have a:(.) a pet, yourself?
13 → IE: mm::. Because(.) my: classes are always, er, is, are always very(.)full, and my parents don’t
14 → like me (.) have these animals. But(.) in my chil-, childhood, I have many (.)er, pe-< I have
15 → many pets> and I love them very much=
16 → IR: =mm::↑
17 → IE: er, Looking them grow bigger and bigger, that’s-er, I feel very happy,
18 → [˚hihhh˚
19 → IR: [Yes, yes. uh, I have a lovely cat=
20 → IE: =˚huhh˚
21 → IR: a black’n white cat.
22 → IE: ˚huh [hhh˚
23 → IR: [uh-huh
24 → IE: £Beautif(h)ul?£
25 → IR: Yeah, he is. He is very sweet.
The interviewer presents the topic in the beginning (line 12), and the interviewee keeps telling about the topic from line 13 to 15 with interviewer’s go-ahead feedback in intervals (line 16). While in line 19, the interviewer does the reciprocal telling, after providing the information-receipt and agreement token “yes” twice and a trivial delay “uh”. The telling “I have a lovely cat” is straight-out and is produced in a single and simple, turn-constructional unit of a declarative sentence (line 19). Following that, the interviewee employs a single-beat and low-voiced laughter to encourage the speaker to continue (line 20), which functions nearly the same as brief expressions of overt language support like “uh huh”, “mhmm”, “yeah”, and other non-verbal signals like head nods, facial expression and direction of gaze (Stubbe, 1998) . It could be evidenced through line 21 with more information telling about the cat. The same thing happens in line 22, which even demonstrates an emotional amazement and excitement with a slightly longer aspiration after the same vowel “u”, and therefore responded with a confirmation in line 23. Both the laughter are the information receipt and implicit supportive responses to the action of telling (Yu, 2003) . By laughing in these two positions, the interviewee first projects herself as an aligning story recipient and then turns herself to be an affiliative story collaborator (Yu, 2008) . The latter role is more active evidenced by the next-turn omitted question “beautiful?” and the simultaneous mild laughter for repair initiation in line 24.
The candidate’ voice of laughter is very low and weak, with closed and reserved vowel “u” and aspiration, which is a symbol of tension for the test. Glenn (2013) discovers that in employment interview, laughter is a frequent device for interviewees to release the extra energy and solve the delicacy of nervousness. Although not an explicit supportive feedback, laughter serves as an information-receipt and a go-ahead token, and is undoubtedly treated by the interviewer as such, hence an alignment to the preceding telling under such a test setting.
3.3. When FPP Is a Compliment or a Repair Solution
Compliment is an elevated supportive action (Downes, 1998) , compared with a continuer to the telling FPP and it is usually accomplished based on the assessment action. Subsequent to the action of compliment are two systems of acceptance/rejection and agreement/disagreement. Pomerantz (1978: pp. 87-88) has elaborated on its interrelatedness as follows:
“While appreciations and agreements are affiliated components (as evidenced by their co-occurrence in local environments), they are not sequentially interchangeable. Agreements tend to occur less frequently than appreciations and seem to have more restrictive conditions for their productions. When agreements do co-occur with appreciations, they are proffered routinely after initial appreciations. In short, while appreciations and agreements are interrelated, appreciations over agreements seem to be preferentially selected for accepting compliments. Rejections are routinely performed with disagreements.”
However, the subsequent action will not simply be an initial appreciation plus an agreement for acceptance and a disagreement for rejection. Another system of self-praise avoidance will make it incompatible to provide preferred answer to the compliment and at the same time avoid self-praise. This conflicting preferences lead to a challenge for recipients to respond to prior turns. Sensitive to the delicate situation, Chinese candidates of this study resort to laughter as the solution, shown in Extract 1.
38 → IR: Yes, quite, uhm, yes > (what set me things) < you sound a very interesting one
39 → IE: e hehh e [.hhhhhh
40 → IR: [and I’d like to wish you good luck with it.
This sequence occurs near the end of the interview for this candidate. Line 38 is a brief assessment of the interview and a delayed start to close this sequence. Although it is quite ritual, it performs as a compliment to the candidate with the word “very” strengthening its directness. In response to the compliment, the interviewee delivers a three-beat laughter, which are not as open as “/a/” in Jefferson et al.’s (1977) division, but relatively a louder and clearer in the interview, lengthened by the inbreath and outbreath.
It does not only occur at the end, but in the beginning of the interview about the introduction of the candidate, as in Extract 6.
1 → IR: Now(.), can you(.) tell me your name, pronounce it for me, please.
2 → IE: Chinese name?
3 → IR: Yes.
4 → IE: er Yu Cong
5 → IR: Yu:
6 → IE: Cong
7 → IR: cung
8 → IE: cong [huhh .h
9 → IR: [cong=
10 → IE: =yeah
11 → IR: o:k
12 → IE: Which means smart he he [.hh
13 → IR: [>Which means what↑<
14 → IE: A smart and intelligent
15 → IR: Really↑
16 → IE: £I like a: er-a small’n little f(h)ish(h)£
17 → IR: ↑O::[:::: h↓
18 → IE: [he hee
19 → IR: well, that’s a nice name [to got those. So (.) you have a lot to live up to that
20 → IE: [he hehh
21 → Yeah.
The candidate self-initiates the meaning of her own name as positive and the interviewer has achieved an understanding after repair. Subsequently she made a compliment to the candidate who got the name (line 19), which is achieved by an assessment of the name. The interviewee has taken it as a compliment and demonstrates an acknowledgement and somewhat an acceptance (line 20) with a softened appreciation through two laughter particles, which is also a relatively open contour, voiced out immediately after “nice name”, as onset overlapping with “to got those”. “Yeah” in line 21 is the minimal agreement response to the declarative-formed formulation of the latter half of line 19.
In these two instances, laughter is similarly employed to mark the situation delicate and mitigate the conflict of self-praise avoidance and giving a preferred answer. However, laughter is not the only solution to mitigate the nature of acceptance. Pomerantz (1978) presents two types of verbal responses to compliment: evaluation shifts and referent shifts, while the former downgrades the praise and the latter shifts the referent. Holmes (1986) summarizes four types of response for acceptance: appreciation or agreement token, agreement utterance, downgrading or qualifying utterance, and return compliment. Among the first type, smile is involved. Yuan (2002) also presents even more types of verbal responses, and detects the presence of smiling as responses to compliment as well in Chinese. Similar to the roles of smiling and other preferred linguistic responses, laughter is also aligned with the prior turn, but more affiliated than smiling, considering its vocal feature. It may simultaneously involve appreciation, agreement, acceptance and emotional display due to its ambiguous nature (Glenn, 2003, 2010) . Therefore, the delicate-solving (Glenn, 2013) and ambiguous nature that laughter reveals are the potential explanation for the uses of laughter in response to compliment action.
In interviews between native speakers and second language learners, repair sequences enjoy a big frequency. In response to repair solution of this data, laughter consists of a major turn, which shares the similar alignment feature with the SPP of the compliment, combining an agreement and acceptance, and simultaneously marking an emotional embarrassment. See Extract 7.
63 → IR: So: have you thought about studying biologist yourself?
64 → IE: mm:: I don’t (0.2) very interested in this=
65 → IR: =You’re->you’re not that interested.<
66 → IE: ˚huh-huh.˚
The candidate has trouble in using the negative form of the adjective and made a short pause within the turn in line 65. The interviewer initiates and made the repair solution directly in the next turn, for all a little repetition in the beginning. Sequentially next to the repair solution, the candidate voiced a low and closed laughter (line 66) in two particles, an indicator of embarrassment for making grammatical mistakes. This embarrassment is made stronger by the other-initiated other-solution, given the preference to the self-correction (Pomerantz & Heritage, 2013) . Therefore, laughter, in this awkward situation, is employed to be delicacy-solving and aligned with the prior repair solution, by means of doing agreement and acceptance.
3.4. When FPP Is Formulating
Formulating is “an action that involves summarizing, glossing, and developing the gist of the speaker’s earlier statements” (Heritage & Watson, 1979: p. 100) . As Bolden (2010: p. 8) points out, ‘‘formulations are sequence-initiating actions: they function as requests for confirmation, making a response (a confirmation or disconfirmation) conditionally relevant from the addressee’’. In oral proficiency interview, formulating is repeatedly made by the interviewer in a topic-proffer sequence, when the interviewee has provided several turns of telling on the topic, encouraged by the interviewer’s go-ahead signals, as shown in Extract 8:
1 → IR: so, uh::m >tell me a bit about yourself, what-what do you like to do when you are not
2 → studying<
3 → IE: er: I have er, many hobbies.
4 → IR: uh-huh
5 → IE: I like animals, actually I raise a dog, in my spare time, I-I’m crazy about fashion design,
6 → and er, I’m fond of doing sports,
7 → IR: uh-huh
8 → IE: say, swimming, playing badminton, table tennis ball,
9 → IR: uh-huh
10 → IE: and I like watching movies,
11 → IR: uh-huh
12 → IE: there are so many=
13 → IR: =so:: you are very busy
14 → IE: hu he hehe [he,
15 → IR: [do you have time to study?
16 → IE: yes,
The interviewer presents a topic about hobby and the interviewee mentions a lot more hobbies rather than one. After four turns of telling, the interviewer in line 13, starting with extended “so”, summarizes her opinion of the candidate’ life and evaluates her in an extreme case, based on the previous talk. The subsequent and relevant response should be a confirmation or a disconfirmation from the recipient in terms of adjacency pair. When it is performed by five loud and slightly fading-in laugh particles (line 14), we may get clues to what is missed or replaced, a confirmation “yes”. And the next line (15) “do you have time to study” is also based on such a recognition of prior laughter as a confirmation “yes”.
The reason why the speaker delivers laughter rather than the verbal expression might be the linguistic incompetence of the candidates who are more used to question-typed confirmation check in English. When the question turns to be a declarative sentence, the candidate may have a slow understanding and delay his response with linguistic token like “uhm”, “well”, or paralinguistic devices in searching for the correct answer and replace it with interactional instinct and nonverbal devices. In other words, due to the insufficient linguistic competence, the interviewee picks up laughter to mark the delicacy and take the place of the type-confirmed “yes”, to show the alignment or the affiliation. Since it is prosodicaly loud and clear and long, the laughter is more evidently served as an affiliation to the formulation.
The affiliation will be better illustrated in the next example (Extract 9).
22 → IR: yes, yeah, hm now tell me first of all, is English important for your：for your future.
23 → IE: yes, quite important, you know, previously I hear there’s a saying in China, that is,
24 → there are three keys in twenty-first century. The first one is English, the other one-
25 → er, the other one is computer and er, er, driving license.
26 → IR: Oh, [driving license.
88 → IR: Have you got computing skills?
89 → IE: er, yeah↑the basical one wor:d, er-o-off-officet,
90 → IR: [mm hm↓
91 → IE: [you know [basic stuff.
92 → IR: [yes, yes, yes.((itch))
93 → So it’s-it’s English that you’re working on now
94 → IE: e he heheh↑yeah.
95 → IR: yes, yeah, (.hh) ok. Now(.) I’m going to ask you to talk about a topic (.) for one minute
In the question-answer sequence, the candidate provides more than sufficient answer on the importance of English, and talk about driving license and computing skills. When those two things have been illustrated and understood as finished, the interviewer makes inferences and proceeds to formulate her understanding that the candidate is working on English at that moment (line 93), and hence requests for a confirmation. The non-linguistic feature “itch” and the verbal “so” in the initial positions are both the signal to start the formulating action. Similar to the previous laughter, the laugh particles of this case (line 94) share the similar beats and gradual fade-in characteristics, which could be explained as a realization process or a delay to decide on the exact word to reply. The differences are that in the second case the laughter is relatively more fast-paced, high-pitched, hence more mirthful and projects a confirmation at first, and that the interviewee did produce the word “yeah” for confirmation after the laughter. The gradual change of laughter to the definite verbal “yeah” displays increasing certainty about his answer. The mirthful laughter, together with the verbal “yeah” accomplishes a strengthened confirmation action and shows an strong affiliative stance to the prior turn, given the candidate’s current linguistic competence.
This study focuses on three major questions: first, what actions laughter is employed to accomplish or concurrently construct? And second, why laughter is employed in this particular situation? And third, what is the relation of laughter and L2 Interactional Competence?
4.1. What Action Laughter Is Employed to Accomplish or Concurrently Construct?
Laughter in this study has performed various actions, involving appreciation to appreciation and wish, acknowledgement to telling, acceptance to compliment and repair solution, and last, confirmation to formulating action.
This study considers the data in L2 oral proficiency interview mostly made of single laughter as the major constituent of the second pair part in response to different actions of the first pair part, thus performing different actions or concurrently construct different responding actions. This adjacency relation between the FPP and the SPP makes the SPP conditionally relevant and constraint, although laughter takes different forms in different pairs. Herbert (1990: p. 201) captures this structure as ‘‘a two-unit in which Utterance I and Utterance II are linked by both temporal and relevancy conditions’ such that Utterance II is conditionally relevant and sequentially dependent on I’’. This is how to determine what action that laughter has performed in this study. When the FPP is a ritual appreciation or a wish, the single laughter in the next turn conducts a relevant acceptance and an appreciation. The prosodic contour is relatively clearer and stronger than the laughter together with the verbal expression as in Extract 2. When the FPP is a telling, the low-voiced laughter as a response constitutes or at least regarded by the interviewer as a continuer, a sort of supportive feedback. When the FPP is a compliment or a repair solution, the laughter in response performs an agreement or acceptance, while it mitigates the strength of the acceptance for compliment in accordance with self-praise avoidance. Last for formulating action, the laughter as a SPP means a confirmation “yes”, though the laughter contour is a little longer and in Extract 8, it does a small change from the vowel “u” to “e”.
What needs to be mentioned more is that the interviewer does not join in any of these laughter situations. That is exactly why these volunteered and standalone laughter could successfully perform those actions in this particular situation, only as responses to the FPP. Shared laughter (Holt, 2010) may have very distinct roles. The asymmetry in the institutional interaction is displayed and explained well with reference to the non-participance of the interviewer in the laughter, which facilitate the progress of the sequence and hold the conversation in the interviewer’s hand.
4.2. Why Laughter Is Employed
The reason why laughter is used instead of verbal utterances or other paralinguistic devices has two main sources. One is the delicacy of all those hard moments, i.e. the sequential environment in which laughter is used. The other is the intrinsic features that laughter possesses.
When dealing with wish, appreciation, formulating action and repair solution, the candidates display insufficient language competence. For the first three, the candidates find it hard in searching for the proper words to reply and for the last action, the candidate finds it embarrassed to be repaired by the other side. Glenn (2013) regards incompetence as one of the delicate situations and potential employees use laughter to mark the delicacy and handle the insufficient utterances, although the insufficiency does not only refer to the linguistic incompetence. Walker (2017) detects children’s linguistic incompetence and unwillingness in resorting to laughter as the response to adults’ questions. Similar to the children in that study, L2 language learners as participants of this study have just arrived at certain points along the continuum of interlanguage (Selinker, 1972) and are far from being sufficient to interact smoothly with native speakers. This study proves in particular that Chinese students in their English learning process, are not as accustomed to responding to social actions, like formulating, appreciation and wish, as responding to question-formed inquiring which are normal and typical in L2 language study, and they may not even have a clear realization of social actions as the basic unit of interaction and therefore lack the knowledge of language use.
When dealing with telling, the delicacy derives from candidates’ tension which makes a low and soft “u” sound and long aspiration. Freud (1938) advanced the argument that laughter results from an excess of energy needing release. Einhorn (1981) found that unsuccessful job applicants were more likely to exhibit nervous behaviors, including “inappropriate” laughter. Glenn (2013) uses nervousness to represent the interviewee’s mental state in employment interview and displays that they make use of laughter in various positions to modulate their actions. Consequently, tension-release is an unignorable function from the perspectives of the interviewees and an important explanation for why those laughter occur instead of verbal responses.
When responding to compliment, interviewees suffer from the conflict of being preferred and avoiding self-praise and similarly resort to laughter for the delicate moment. However, laughter as a nonverbal response to compliment is seldom discussed in previous studies (Chen, 1993; Golato, 2003; Holmes, 1986) except for in Holmes’ (1986) and Yuan’s (2002) study, which both detect the presence of smiling as acceptance. Laughter is similar to smiling in roles, but specific to the L2 interactional setting of this study as a nonverbal resource for the Chinese candidates to be affiliative to prior compliments.
Laughter could be used to handle all those previous delicacy in that laughter, in whatever forms, share the common ambiguity and implicitness. Whereas we could be relatively certain about the action that laughter has performed, depending on adjacency relations between FPP and SPP and the sequence to follow, which gives us a clue about how the other speaker understands the previous turn, we have not yet enough evidence about its meaning and referent, for instance, whether the laughter in Extract 8 in formulating action is a searching-for-word process, and whether the laughter reacts to the assessment “very busy”. Glenn (2013) proposes that laughter can be relevant to the presence of any linguistic, semantic, paralinguistic, embodied, or pragmatic features. Possibly due to this very ambiguous nature of laughter, the participants in this study employ laughter to modify or mitigate the acceptance of wish, appreciation and praise, represent an implicit support to telling, modulate the embarrassment of being repaired, and strengthen or weaken the confirmation to formulating, and dealing with all those language deficiency.
4.3. Laughter and L2 Interactional Competence (IC)
Laughter in all those prior cases not only marks the delicacy, but displays a specific stance, or marks that the speaker has a stance (Potter & Hepburn, 2010) . The laughter demonstrates, on one hand, “a stance of being lightening or showing the speaker is not ‘serious’ or they are being ‘playful’ or similar” (Schegloff, 2007) . For example, in extract 8, the candidate presents his hobbies in a playful and cheerful way by laughing in line 13 and projects that he has a rich life rather than a formally and seriously busy life, so he laughs to modulate the seriousness of “busy” (Schegloff, 2001; Schenkein, 1972) .
On the other hand, they demonstrate a stance of being aligned or affiliated with the prior turns. How laughter is shaped in terms of its prosodic features and understood in the sequence determines its stance. The laughter in the telling and the repair sequence display an alignment stance due to its obviously low and stressed contour, which Goffman (1956) once described as “hollow laugh” to conceal embarrassment, while the laughter in the formulation and compliment sequence demonstrate an stance of affiliation for its relatively open and loud contour with more laugh particles, though the degree of affiliation may vary. For instance, in responding to the concluding assessment “you sound a very interesting one” at the end of the interview (Extract 1), the laughter sounds clearer and louder with longer aspiration, but in responding to the assessment of the candidate’s name in the beginning of the interview in Extract 6, the recipient produced just two laugh particles in weaker voice. These two cases of laughter obviously represent different degrees of affiliative stance. In expressing appreciation to the wish and ritual appreciation, the laughter presents varied stance. In Extract 2 and 4, the laugh particles are generally accompanied with verbal utterance and therefore concurrently construct an affiliative stance, though the laugh particles are not as strong as in the formulation and compliment action. But Extract 3 presents an alignment laughter in expressing appreciation to appreciation due to the three low and restrained laugh particles.
Alignment represents a stance of the recipient to the prior speaker, and the resources that speaker uses to show the alignment are a representative of their interactional competence levels (Dings, 2014) . It refers to the means interlocutors use to demonstrate their intersubjectivity, which involves moves such as assessments, collaborative contributions, and collaborative completions index shared understanding, the ability to adopt the other’s point of view, and the ability to speak in the other’s voice. It is included as another important grouping of interactional resources relevant to models of interactional competence (IC). Laughter in this study performs different types of responding action and fully shows alignment and different degrees of affiliation, therefore could be proposed in this study as a constituent of IC models. Why and how the speaker uses laughter could demonstrate part of their interactional competence, rather than the linguistic competence, in their talk-in-interaction. This proposal further extends Dings’ (2014) expression of alignment activity in L2 to the paralinguistic resources, and set a place for laughter in the IC model.
Laughter of this study could be recognized as an indicator of Chinese candidate’ IC levels for another remarkable argument made by Hall and Doehler (2011) . “Central to competent engagement in our interactions is our ability to accomplish meaningful social actions, to respond to co-participants’ previous actions and to make recognizable for others what our actions are and how these relate to their own actions” and that IC is “the context-specific constellations of expectations and dispositions about our social worlds that we draw on to navigate our way through our interactions with others. It implies the ability to mutually coordinate our actions” (p. 1). It is exactly for this essential factor of social actions relevant to prior actions that laughter is considered as a resource of IC. In respective cases of appreciation, wish, telling, compliment, repair solution, and formulation, the candidates respond to prior actions and make themselves recognizable as doing appreciation to appreciation, acknowledging to telling, acceptance to compliment and repair solution, and confirmation to formulation. Meanwhile the laughter displays candidates’ specific perception of testing environment and their own identities as test takers who aspire to potential access to the target college through the gatekeeping interview. Consequently what laughter has accomplished accords with the central understanding of IC. Hence laughter is firmly regarded as a constituent of IC and an indicator of IC level, as long as it accomplishes the actions and facilitates the interaction.
Through doing alignment or affiliation by means of laughter, the interviewees 1) respond to prior action; 2) construct themselves as goal-oriented and well-intentioned Chinese candidates who aspire to potential admission; 3) demonstrate their insufficiency of linguistic expressions and certain IC levels; 4) display their engagement in testing activity (Seedhouse & Supakorn, 2015) ; and 5) builds the solidarity between the interlocutors and maintains social concordance (Glenn, 2013; Lavin & Maynard, 2001) . Due to the delicate situation, the laughter achieves the intersubjectivity and progressivity of interaction, and proves a skillful resource parallel to linguistic expressions in testing environment. Although many researchers (e.g. Glenn, 2003 ; Greatbatch & Clark, 2003 ; Holt, 2012 ) argue that laughter could be disaffiliative or resisting and hence do not necessarily contribute to the solidarity in certain situations, at least the first four roles that laughter has achieved will not be denied.
This study also orients us to have the conclusion that, in performing social actions, Chinese candidates do not have sufficient linguistic and pragmatic competence (Bachman & Palmer, 2010) , but do have corresponding interactional consciousness and succeed in demonstrating their interactional competence by employing the paralinguistic resource of laughter to mark those delicate moments and perform alignment or affiliative actions to prior turns in talk-in-interaction. Therefore this study contributes to our understanding of Chinese candidates’ linguistic and pragmatic competence level, and of laughter as a vital paralinguistic resource of IC in L2 interaction. It not only sheds lights into the qualitative study of speaking test, but attributes to the application of Conversation Analysis into various domains of applied linguistics.
This data collection and transcript are greatly supported by Professor Yu (Guodong Yu) and Professor Wu (Yaxin Wu) and collectively discussed and analyzed in Chinese’s Dig group.
*This article is the periodical achievement of the 13th Five-Year Education Planning Program (GH-18004) of Shanxi Province in China.
 Chen, R. (1993). Responding to Compliments: A Contrastive Study of politeness strategies between American English and Chinese Speakers. Journal of Pragmatics, 20, 49-75.
 Galaczi, E. D. (2014). Interactional Competence across Proficiency Levels: How do Learners Manage Interaction in Paired Speaking Tests. Applied linguistics, 35, 553-574.
 Glenn, P. (2013). Interviewees Volunteered Laughter in Employment Interviews: A Case of “Nervous” Laughter? In P. Glenn, & E. Holt (Eds.), Studies of Laughter in Interaction (pp. 255-275). London: Bloomsbury.
 Goffman, E. (1956). Embarrassment and Social Organization. In E. Goffman (Ed.), Interaction Ritual: Essays on Face to Face Behaviour (pp. 97-112). Penguin: Harmondsworth.
 Greatbatch, D., & Clark, T. (2003). Displaying Group Cohesiveness: Humour and Laughter in the Public Lectures of Management Gurus. Human Relations, 56, 1515-1544.
 Hall, J. K., & Doehler, S. P. (2011). L2 Interactional Competence and Development. In J. K. Hall, J. Hellermann, & S. P. Doeher (Eds.), L2 Interactional Competence and Development (pp. 1-15). Bristol/Buffalo/Toronto: Multilingual Matters.
 He, A. W., & Young, R. (1998). Language Proficiency Interviews: A Discourse Approach. In R. Young, & A. W. He (Eds.), Talking and Testing: Discourse Approaches to the Assessment of Oral Proficiency (pp. 1-24). Philadelphia/Amsterdam: John Benjamins.
 Ishida, M. (2009). Development of Interactional Competence: Changes in the Use of Ne in L2 Japanese during Study Abroad. In H. T. Nguyen, & G. Kasper (Eds.), Talk-in-Interaction: Multilingual Perspectives (pp. 351-385). Honolulu, HI: University of Hawai’i National Foreign Language Resource Center.
 Jefferson, G. (1978). Sequential Aspects of Storytelling in Conversation. In J. Schenkein (Ed.), Studies in the Organization of Conversational Interaction (pp. 219-248). New York: Academic Press.
 Jefferson, G. (1979). A Technique for Inviting Laughter and Its Subsequent Acceptance/Declination. In G. Psathas (Ed.), Everyday Language: Studies in Ethnomethodology (pp. 79-96). New York: Irvington.
 Jefferson, G. (1984). On the Organisation of Laughter in Talk about Troubles. In J. Atkinson, & H. J. Maxwell, (Eds.), Structures of Social Action: Studies in Conversation Analysis (pp. 346-369). Cambridge: Cambridge University Press.
 Jefferson, G., Sacks, H., & Schegloff, E. (1977). Preliminary Notes on the Sequential Organization of Laughter. (Pragmatics Microfiche). Cambridge: Cambridge University, Department of Linguistics.
 Lavin, D., & Maynard, D. W. (2001). Standardization vs. Rapport: Respondent Laughter and Interviewer Reaction during Telephone Survey. American Sociological Review, 66, 453-479.
 Lee, S., Park, J., & Sohn, S. (2011). Expanded Responses of English-Speaking Korean Heritage Speakers during Oral Interview. In G. Pallotti, & J. Wagner (Eds.), L2 Leaning as Social Practice: Conversation-Analytic Perspectives (pp. 87-104). Honolulu, HI: NFLRC.
 Masuda, K. (2011). Acquiring Interactional Competence in a Study Abroad Context: Japanese Language Learners’ Use of the Interactional Particle Ne. The Modern Language Journal, 95, 519-540.
 Petitjean, C., & Gonzalez-Martinez, E. (2015). Laughing and Smiling to Manage Trouble in French-Language Classroom Interaction. Class Discourse, 6, 89-106.
 Pomerantz, A. (1978). Compliment Responses: Notes on the Co-Operation of Multiple Constrains. In J. Schenkein (Ed.), Studies in the Organization of Conversational Interaction (pp. 79-112). New York: Academic Press.
 Potter, J., & Hepburn, A. (2010). Putting Aspiration into Words: “Laugh Particles”, Managing Descriptive Trouble and Modulating Action. Journal of Pragmatics, 42, 1543-1555.
 Stivers, T. (2008). Stance, Alignment, and Affiliations during Storytelling: When Nodding Is a Token of Affiliation. Research on Language and Social Interaction, 41, 31-57.
 Stivers, T., Mondada, L., & Steensig, J. (2011). Knowledge, Morality and Affiliation in Social Interaction. In: T. Stivers, L. Mondada, & J. Steensig (Eds.), The Morality of Knowledge in Conversation (pp. 3-24). Cambridge: Cambridge University Press.
 Stubbe, M. (1998). Are You Listening? Cultural Influences on the Use of Supportive Verbal Feedback in Conversation. Journal of Pragmatics, 29, 257-289.
 Wilkinson, R. (2007). Managing Linguistic Incompetence as a Delicate Issue in Aphasic Talk-in-Interaction: On the Use of Laughter in Prolonged Repair Sequences. Journal of Pragmatics, 39, 542-569.