Dealing with Moments of Crisis Interculturally in Educational Virtual Exchanges: A SinoFinnish Case Study

Being able to deal with “moments of crisis” is crucial in intercultural communication. Using identities as an analytical lens, this paper examines different types of “moments of crisis” identified in educational virtual exchanges between Chinese and Finnish university students. The study shows that the participants use soothing and code-switching as strategies in these moments of crisis to “do” interculturality, i.e., to achieve interactional arrangements (e.g., moving the discussion forward) and self-other alignments (e.g., saving face). Focusing on soothing as a positive strategy and code-switching as an avoidance strategy in dealing with “moments of crisis”, we also scrutinize the emergence of the identities of mediator, fence-sitter, and facilitator in the students’ behaviors. Finally, students’ working experience, language use, and physical surroundings are identified as having a potential influence on their use of soothing and code-switching during the online intercultural interactions. Recommendations for preparing students for international virtual exchanges conclude the paper.


Introduction
The United Nations [1] reported that the COVID-19 pandemic has created the largest disruption of global education systems in history, affecting nearly 1.6 billion learners in more than 190 countries. The COVID-19 situation has also hindered the internationalization of higher education and exacerbated existing problems [2] (p. 129). It is thus urgent to unveil the realities of online education in the circumstances of COVID-19 when internetbased learning with the application of various tools appears to have become mainstream.
"Virtual exchange", as a component of intercultural educational programs, refers to the engagement of groups of learners in online interactions with partners from other cultural contexts or geographical locations [3] (p. 1). Virtual exchange can help students "enter into a new realm of collaborative enquiry, viewing their expanding repertoire of identities and communication strategies as resources in the process" [4] (p. 21). During the process, i.e., interacting with people who have potentially different life experiences, identities are (re-)negotiated interculturally.
Identity construction in computer-mediated communication (CMC) has been increasingly explored by scholars [5,6]. However, there have been limited studies on identity construction in synchronous CMC [7,8]. In order to fill this gap and offer pedagogical implications for intercultural communication education, we examine how students communicate with each other in online situations and how identities are co-constructed and re-negotiated in the process. We pay specific attention to how the students deal with what we refer to here as "moments of crisis", triggered by interactional incidents.
A qualitative case study from a Chinese-Finnish project serves as an illustration. The choice for this project is based on the fact that online interactions between (Mainland) Chinese and European students are rare in education and that, to our knowledge, no previous study has focused on this specific context of CMC. Furthermore, considering the specific position of China during the 2020-2022 pandemic (e.g., questions around the origins of COVID-19, the "new Cold War"), the ways the students deal with "moments of crisis" can serve as good and original examples for promoting interculturality.

Identities in Interculturality
Research has highlighted the complexity of identities in intercultural interactions [9]. Many researchers agree on the fact that identities are context-dependent [9]. For example, Clift [10] indicates how the identity of "complainer" attempts to forestall negative attributions of character materializes by laughing while complaining. In this sense, identities are regarded as constituted as individuals go about doing whatever they do with others [11]. These identities are referred to as "interactionally occasioned identities" [11], which resonates with our understanding of identities in this research: identities are created in a given context, which is shaped, maintained, and changed jointly by participants and contexts continually in the course of interaction [12].
From this post-modern perspective on identity, individuals constantly (re)negotiate and (re)construct their identities together while "doing" interculturality. "Doing" interculturality here focuses on the prefix "inter-", hinting at in-betweenness, reciprocity, and co-construction, as well as on the suffix "-ality", emphasizing the flexible and never-ending characteristics of intercultural interaction [13]. In other words, interculturality is a process in which individuals (re)negotiate their identities endlessly when encountering the other, placing forward, accepting, refusing, and modifying certain identity markers such as cultural/national affiliations, race, religion, but also social class and gender. This goes well beyond, e.g., the ideology of acculturation, whereby one would need to become like the other "culturally" to be able to communicate with them. For example, a study by Dervin [14] examines how students from different backgrounds co-constructed and negotiated their identities via online chat sessions, viewing self as dialogically and socially constructed well beyond cultural and national identities. In this paper, identities are "the products of joint actions and the intersubjective organization of verbal interaction" [15] (p. 203).
Drawing on a critical perspective on identities and interculturality, our study highlights that individuals have the agency to create identities in interaction with others by utilizing, e.g., social resources at hand (e.g., self-experience and social presuppositions) [16]. In what follows, we examine the details of talk-in-online-interaction [17] to demonstrate how identities are (co)constructed by strategically dealing with problematic situations in online intercultural interactions. Our specific focus is on what triggers "moments of crisis" and how to deal with them.

Identifying Moments of Crisis
According to Goffman [18] (p. 103), a moment of crisis is defined as "the little social system participants created in interaction collapses, and they hurriedly try to assume a new set of roles". This means that individuals can sense a crisis in interactions and they can deal with it strategically. Based on this understanding, we define "moments of crisis" as moments when individuals deal with the unexpected and uncontrollable strategically and purposefully in online intercultural interaction. In what follows, we present two ways of linguistically identifying such moments of crisis. This is followed by a discussion of two strategies used for dealing with such crises.
Goffman [19] has problematized how people react to moments of crisis. He finds that individuals tend to use blurted vocalizations such as semiwords (Uh-oh!), imprecations (My God), and filled pauses (ah, er, um, en, etc.) to respond to such moments. For example, drivers might often shout "Oh My God" to respond to a crisis moment when some motorists and pedestrians suddenly pass out of range. Individuals' responses to crisis moments are manifested through these blurred vocalizations, which Goffman defines as "response cries" [19]. By doing so, individuals align themselves with what is happening. The cries display evidence of "interactional arrangement between self and other" [19] (p. 83). In this sense, individuals' identities are (re)negotiated and (re)constructed by uttering these "response cries" in the moments of crisis to align the self and the other in interaction.
Occasionally, code-switching, defined here as "alternating use of two or more "codes" within one conversational episode", might also be triggered when individuals are confronted with crises in interaction [20] (p. 1). Myers-Scotton [21] (p. 1263) explains that in code-switching processes "the most expected and feasible language used in the interaction is unmarked choice" and "what is not predicted is marked choice" [22] (p. 5). The marked choice indicates that some "crisis" needs to be "dealt with" in this context. By making one code choice over another, individuals negotiate a change of roles in interactions [21]. Since our context of research is multilingual (Chinese, English, Finnish, and other "underground" languages and dialects not necessarily apparent in the virtual exchanges), code-switching appears to be a potentially relevant approach to identify moments of crisis. Code-switching, alongside with the aforementioned "response cries", also provides indications for moments of crisis in this paper.

Reacting to Moments of Crisis in Interculturality
In order to deal with moments of crisis, individuals employ strategies for different purposes [17] (p. 33). For example, in the specific context of virtual exchanges, Liaw and English [23] found that addressivity and politeness were useful strategies for online selfpresentation in a Taiwan-France telecollaborative project. Vinagre and Corral Esteban [24] show that students construct positive identities (e.g., good images in front of others) by expressing their own feelings rather than judging their partners' behavior in telecollaborative emails in a Spanish-American context.
From a psychotherapy perspective, Goldman and Greenberg [25] (p. 66) claim that self-other soothing is essential in helping people deal with identity threats. According to Sutherland et al. [26] (p. 744), (self-)soothing [is] done by crediting [interactants] with "positive attributes, or accomplishments, as well as expressing the speakers' caring feelings to the interactants" (e.g., offering sympathy or supportive advice). Bearing these in mind, soothing serves to comfort interactants' nerves or mitigate the tensions in interactions. Goffman's [18] (p. 120) concept of interaction-consciousness explains why individuals may use soothing as a strategy in interaction. According to the scholar, interaction-consciousness is related to the special responsibility that an individual may have for the interaction "going well".
Code-switching, which was presented earlier as a way of identifying moments of crisis but which will also serve as a strategy in this study, is the most "direct strategy when individuals face unresponsive interlocutors" [27] (p. 313). Code-switching has often been described as "a conversational strategy for stylistic effects" such as mitigating complains [28] (p.66). It also serves as a membership categorization device to (re)align relationships with other interlocutors as ingroup or outgroup (e.g., including Chinese students but excluding Finnish students by switching to Chinese) [29]. Cashman [27] (p. 307) maintains that code-switching is "a resource used by speakers to ascribe and accept or reject membership in groups". This resonates with our understanding of codeswitching as a membership categorization device used for dealing with moments of crisis in interculturality.
In this paper, we draw on critical interculturality [30], putting communication front and center in interculturality. This enables us to examine how identities are constructed in interactions influenced by who individuals interact with, for what purposes, and within what contexts. Since interculturality is an unstable process consisting of identity in-/consistencies [9], moments of crisis are unavoidable in individuals' interaction. "Response cries" [18] and code-switching [20] help us identify moments of crisis in the study. Among many strategies of interactional management employed by individuals, soothing [26], and code-switching [27] demonstrate how individuals react to moments of crisis to (re)negotiate their identities in intercultural interaction.
Based on the discussion above, the following research questions guide us to explore how identities are constructed by employing soothing and code-switching as strategies in moments of crisis online: (1) What triggers moments of crisis?
(2) How do the students make use of soothing and code-switching in order to deal with them?
(3) What does the use of these two strategies reveal about the identities constructed by the students in these specific online intercultural encounters and, at the same time, about the type of interculturality that they "do"?

Data Description
We follow a qualitative case study approach [31] to describe what happens during the students' online discussions and investigate how their identities are (co)constructed. According to Willig [31], a case study emphasizes what occurs and changes in the processes that take place over time. As this is a case study, we do not mean to generalize our results to all virtual exchanges taking place around the world. However, it will help us identify elements for future studies to explore.
The data derived from a one-month virtual exchange between university students from China and Finland during the COVID-19 pandemic (autumn 2020). Students had been placed into eight groups, with 5-6 Chinese students (CSs) and 2-3 Finnish students (FSs) per group. Most of the FSs were older than CSs since most FSs started their studies after getting some work experience, while CSs went to university straight after high school. The students' majors (education) and language level in English were matching. Under the Common European Framework of Reference for Foreign Languages, their English levels ranged from B2 to C1.
Students met online via Zoom.us to prepare a group presentation on pre-determined intercultural communication tasks (e.g., plans for orientation day for international students). The majority of students used the app on their own; however, several CSs (e.g., Group 2 (G2), G4, and G6) shared a camera in the same room. Each time, one student per group shared his or her screen and edited the PowerPoint (PPT) primarily based on the group's conversation.
The data was collected through "participant observations" [32], in which fieldnotes concerning what they discussed and what the researcher felt were taken. Each discussion of approximately one hour was recorded with the consent of the students. We analyzed 10 moments regarding soothing and code-switching from the 10-h online discussions involving, in total, 11 Finnish students and 14 Chinese students, since soothing and codeswitching frequently occurred among the students. The moments were coded from M1 to M10, and the students were coded as Finnish student 1 (FS1) to FS11 and Chinese student 1 (CS1) to CS14, according to the order of appearance in their group discussion. For example, moment 1 happened at G1's first discussion minute 6 and was coded as M1 G1D1-6 . We have only shown excerpts from these moments due to space constraints, so not all participants are listed in the analysis.

Method
We analyze our data with sociolinguistic tools [19,29], which help us identify the moments of crisis by studying the factors that trigger individuals' different reactions to others' utterances. According to Goffman [19], there are different forms of replies and responses, such as silence, laughter, and blurted vocalizations ("response cries"). These forms of replies and responses "carry all kinds of implication and meaning" [19] (p. 1). In this paper, we focus on these forms of replies and responses to identify the moments of crisis. For example, "response cries" such as "OMG" constitute signals for moments of crisis. This helps us answer our first research question concerning what seems to trigger the moments of crisis. Finally, to describe vivid synchronous online discussions, we employ transcription symbols from Antaki & Widdicombe [33] and Jefferson et al. [34]. For example, (.) A dot in parentheses indicates a short hearable pause (less than half a second) within or between utterances. ( . . . ) three dots in parentheses indicate a long pause (more than half a second but less than 5 s) within or between utterances. Hehh and Hahh represent giggling and laughing, respectively, by showing different pitches. (()) Doubled parentheses contain transcribers' descriptions, such as the translation of utterances. A dash indicates a cut-off. Italics show special turn-taking in the online conversations.

Triggers behind Moments of Crisis
Tables 1 and 2 below summarize the data analysis by describing what triggers the moments of crisis amongst the students and how to deal with these moments. Having looked into the signals for moments of crisis (i.e., response cries and code-switching), we found three types of "moment of crisis" in which students (re)act strategically and purposefully when they feel tension and anxiety: -others' trouble (e.g., spelling errors) or complaints, -work arrangement disputes, -and difficulty in making oneself understood. For the 10 moments of crisis identified in the data, Table 1 shows that spelling errors, someone's absence, and misunderstandings trigger some moments of crisis in the data. When dealing with these moments of crisis, students employ three sub-strategies: offering an explanation for the other; giving an affirmative response; and providing optional answers for an imagined other. As far as soothing is concerned, this strategy is used to comfort others' nerves; soften the tensional atmosphere; or mitigate complaints in interactions. Table 2 demonstrates that situations such as being unable to express oneself and disagreeing with something may also cause some moments of crisis. In these situations, students use code-switching as a strategy to create ambiguity without risking loss of face and to keep the online discussions going. Finally, the identities of mediator, fence-sitter, and facilitator are noted as being (co)constructed when dealing with these moments of crisis in intercultural encounters.
In the following sections, we first discuss when, why, and how individuals employ soothing and code-switching as strategies in the interactions. We then describe the identities related to interculturality, i.e., identity as a mediator by employing soothing in three particular ways. Finally, we analyze how two identities (i.e., fence-sitter and facilitator) are constructed by using code-switching.

Soothing as a Strategy: Participants' Identity as Mediators
The identity of a mediator here is based on the concept of the "cultural mediator" described by Zarate et al. [35]. The cultural mediator, as an agent for social cohesion, "regulates the functioning of the social bond and contributes to the cohesion of a society" [35] (p. 230). In other words, individuals identify themselves as mediators when they recognize social bonds in intercultural interaction, when they are willing to take risks in situations of identity stress, and when they use strategies to deal with these situations (e.g., explain, negotiate, and (re)mediate). [35]. In our study, we find that participants sense others' response cries (e.g., My God) and react to them by applying soothing as a strategy for social cohesion in online intercultural interactions.
We have identified the following three sub-strategies used as soothing: offering an explanation/excuse for the other (e.g., M1 and M2); giving an affirmative response (e.g., M3 and M6); and providing an option for an imagined other (e.g., M4 and M10). We discuss these sub-strategies in more detail below.

Sub-Strategy 1: Offering an Explanation/Excuse for the Other
Offering an explanation/excuse for the other is a direct reaction to others' response cries. According to sociolinguistic analysis, for example [19], FS1's semi-words "Oh my; No, no" and self-complaint "My brain is so broken for writing," are signals for moments of crisis. When these occur, his partners, FS2 and CS1, sense his anxiety and react by offering an explanation/excuse to soothe his nerves or embarrassment. These instances show self-other alignment.
In M1 below, FS1 is ashamed when he has difficulties spelling "schedule". His spelling process, which was being watched by others via Zoom, might be the direct cause of his response cries [19]. "Oh My" and "Oh No" as response cries seem to indicate his embarrassment. FS2 then takes the floor, "That's quite hard word for Finnish people", to soothe FS1's embarrassment by providing an insider's perspective on his plight. FS2's interactional occasioned identity [11] as a mediator is constructed by using their knowledge of the Finnish language versus English to comfort FS1.
Excerpt M1 G1D1-6 49" FS1: This is not in our language not even close. There are three consonants behind each other oh, no, no, no, no. FS2: Yeah, that's quite hard word for Finnish people.
The situation is similar in M2 when FS1 misspells "whether" when they are discussing "weather". The first to notice this error, CS1, repeats "whether" instead of directly correcting FS1. FS1 starts to doubt himself, "Why my brain is so broken now?". After spelling "weather" correctly, FS1 comments further on his self-doubt: "My brain is so broken at the moment for writing." His self-doubt arouses CS1's sympathy [26]. To save FS1's face, CS1 offers an explanation/excuse to soothe his embarrassment, "You are doing multitask [ing]." CS1's identity as a mediator is displayed by showing his positive care to soothe FS1's nerves [26]. CS1 reacts to FS1 differently after FS1's response cries: from simply ignoring FS1's spelling error to soothing his nerves. In this instance, CS1 identifies himself as a mediator in a similar way as Zarate et al.'s [35] concept of cultural mediator. He "senses himself involved in constructing the social bond in situations characterized by cultural and linguistic plurality" [35] and adopts strategies to deal with the moments of crisis in which FS1 finds himself in a situation of "identity stress" [35].

Sub-Strategy 2: Giving an Affirmative Reply
From a psychotherapy perspective [26], giving an affirmative reply (e.g., expressing caring words) as a one-way direct reaction to a response cry is regarded as soothing in interaction. M3 and M6 below illustrate how the identity of a mediator emerges by giving an affirmative reply. FS1 tries to spell "achievement" correctly by repeating it again and again in M3. His four "OMGs" (i.e., imprecation) arouse FS2's sympathy [19]. FS2 takes this turn, saying, "It's really quite hard word". He finally spells "achievement" correctly, but he seems to be less confident in his spelling, "It is the correct way because it seems wrong to my eye." His self-doubt is noticed by other students. To cheer FS1's spirits up, FS2/3 and CS1 "self-select" the turn [17] (p. 136) and respond to him supportively again and again to soothe his embarrassment [26]. In M6, soothing is employed to deal with a dispute about the work arrangement, which causes tension between FSs and CSs, who do not seem to understand each other. FS4 suggests checking the spelling on their PPT against certain spelling errors. However, CS5 misunderstands, assuming that they want to modify some content in the PPT. Poor network connection as a factor in their physical surroundings [36] makes the situation even worse. CS5's utterance, "Er because we don't know the game so maybe" implies that she still does not understand. FS4 timely gives an affirmative reply to soothe CS5: "You just relax and wait for that email". This is an affirmative reply to CS5's indirect request "you'd better do this part since we do not know it very well". So, FS4's identity as the mediator is established when she senses CS5's nerves and responds affirmatively to assure CS5 that they will do this part of the job. Considering others' feelings when dealing with interactions in online intercultural interaction (e.g., poor connectivity) could be considered as evidence of, e.g., intercultural competence [13].

Sub-Strategy 3: Providing a Choice for an Imagined Other
Providing a choice for an imagined other is an indirect reaction to others' response cries. The imagined other here refers to a person whose attributes are imagined by the speakers. In this way, "speakers summon up the presence of others in order to say something to them" and "self-consciously transplant the participation arrangement that is natural in one social situation into an interactional environment" [19] (p. 153).
For example, long conversations between FSs and long pauses during their discussions in M4 indicate the rising tension caused by a work arrangement: who is going to beautify their PPT? The situation becomes tenser after FS1 rejects FS3's willingness to do this job but puts forward CSs to do it. When communication is blocked, FS2 consciously transplants a social situation that not everyone wants to do extra work in this online intercultural interaction [19]. As someone who tries to re-mediate the work dispute [35], she provides a choice for the imagined other who "does not want to speak" at the presentation to ease tensions. Positive replies from CS1/2 to her indicate that soothing works at this moment. M10 also shows when and why FS10 provides a choice for an imagined other to mitigate others' complaints about someone's absence. When FS10 asks, "Did somebody message CS14?", CS11 replies, "I'm calling him but hehehe." FS10 senses the implicit complaint about CS14's absence from CS11's "but hehh". To mitigate the complaint, she constructs an interactional arrangement in which she transplants some conventional excuses of absence for the imagined/absent other (CS14) [19], "Do you know is there a difficulty for CS14 to attend?". She then offers help, "Does he need some help to join us?". This shows how strategically FS10 deals with the issue of CS14's absence: she does not directly ask why he is absent but offers help.
Excerpt M10 G8 D2-2 FS10: Did somebody message CS14? CS11: I'm calling him but Hehh. FS10: OK. CS11: Just wait a minute. FS10: Do you know is there a difficulty for CS14 to attend? Does he need some help to join us? CS12: I think this not difficult but but actually we don't know what he is doing. Hehh. CS11: I think she, Er, I think he is, Er, just have forgotten it. FS10: Oh, oh.
These moments of crisis illustrate that soothing as a positive strategy is employed by participants. We find that FSs use soothing more frequently than CSs. This may be influenced by participants' working experiences according to their demographic characteristics. As such, most of the FSs might have certain experiences as mediators through their work, which the Chinese students do not seem to have. Moreover, language use might be another reason. This might be due to what we could call the social emotional 'automatization' of language use (knowing how to use a foreign language in certain ways to e.g., motivate the other). As such, the living environment (the "Finnish context") of FSs might provide more chances to practice their English in their daily lives, while CSs lack a context where practice of the language is made possible on a regular basis. FSs could thus be more experienced in transplanting a form of language use that is 'natural' in one social situation into this online intercultural interaction [19].

Code-Switching as a Strategy: Identity as Fence-Sitter and Facilitator
Code-switching is both a signal for moments of crisis as well as a strategy that students use to deal with them. The English learning experience of most students (more than 10 years of study and varying use) implies that their English is good enough to communicate with others. However, code-switching happens very often, especially among CSs (e.g., G2, G4, and G6). The reason could be attributed to their physical surroundings, i.e., many CSs sharing the same camera in the same room. The physical surroundings of the speech situation here refer to the objects and events in the co-participants' sensory, particularly visual, reach [36] (p. 17): "Participants may rely on [it] in production and interpretation of future activities" [36] (p. [18][19]. In this section, two identities (i.e., fence-sitter and facilitator) are displayed when the participants apply code-switching to negotiate work-arrangement disputes or to make themselves understood. The definition and identification of the two identities are explained in the following sections.

Identity as a Fence-Sitter
A fence-sitter is defined in the Cambridge Dictionary [37] (accessed on 29 May 2021) as someone who supports both sides in a disagreement because they do not want to offend either side. Based on this meaning, the identity as a fence-sitter relates to strategies whereby participants use code-switching to create ambiguity by suggesting inferences without risking loss of face [29].
For example, in M5, when CS4 stumbles over her explanation of the game that FSs do not understand, CS5 interrupts her in Chinese: "你就说到时候我们来解决就可以了" (in English: "Just tell them that we will fix this issue"). This utterance implies that there is no need to explain the game to FSs since it is hard to describe it in English. From their online discussions, it appears that CS5 speaks English well and is very active before the moment of crisis. The physical surroundings might be one of the triggers for CS5's code switching [36]: she shares the same camera with CS4. The marked choice (Chinese) made by CS5 indexes her identity as a fence-sitter. By using code-switching, she mitigates the refusal to explain the game to FSs "without actually putting (herself) on record and risking loss of face" [29] (p. 98).
Excerpt of M5 G2D2-2 FS4: The second one is these two truths and one lie. I think it's we all know that. But then this is this one I don't understand. This is different file because I don't understand the meaning of this. So if we are keeping this, can someone explain what is happening in here in this game? Because I don't understand now what they have to do, what they are doing. ( . . . ) CS4: OK, the game of some reasoning of this game and er -. CS5: 你就说到时候我们来解决就可以了。你就说到时候我们会来解决。 ((Just tell them that we will fix this issue. Tell them that we will fix it.)) In M9, participants construct the identity of fence-sitters by using code-switching to distance themselves from their disagreement on work arrangements [28]. CS8/9/10's physical surroundings [36] are the same as CS4/5 in G2: they share the same camera. By switching to Chinese, CS10 distances herself from the disagreement on how to assign their presentation, which is discussed by CS8 and FS9 in English. Instead of disagreeing directly with them in English, she switches to Chinese and suggests revising their presentation first, before deciding who will speak which parts of it. When FS9 is confused with what they have said, CS10 asks CS8 to interpret, "让他们，你让他们听一下" (in English: "Let them hear what we have discussed"). CS9 agrees with CS10's opinion, asking CS8 to retell FSs about revising the PPT first. CS9 and CS10 use code-switching as an avoidance strategy to distance themselves from the disagreement on work arrangements [28].

Identity as a Facilitator
Identity as a facilitator in this subsection has the same characteristics as the social identity of a "facilitator" in Cashman's [27] study. In her study, "the facilitator" negotiates with different groups by switching languages, demonstrating three characteristics: neutrality, fairness, and objectivity [27] (p. 309). As was discussed for M9, CS8's identity must be "answered" as a facilitator [16] since she is required to switch languages to facilitate the online discussion. According to Holland et al. [16], identities remain dependent on relations and conditions. If these relations and conditions change, "identities must be "answered", and old "answers" about who one is may be undone" [16] (p. 189). Identities "answered" here means that identities are re-constructed and re-negotiated in different contexts.
Code-switching in M7 and M8 serves as a membership categorization device to "ascribe, accept or reject membership in groups" [27] (p. 306). By actively employing codeswitching in M7, FS7's identity as a "fair" facilitator is highlighted. She facilitates online interactions by using Finnish ("we" code) and English ("they" code). When FS8 shifts from English ("they" code) to Finnish ("we" code) in an attempt to gain support from ingroup members, FS7 rejects the ingroup membership by using English ("they" code), encouraging FS8 to speak English in turn 2. FS7 only switches to Finnish ("we" code) in turn 6 after FS8 continues communicating with her in Finnish ("we" code). FS7 rejects becoming an ingroup member of FS8 by using the "they code" (English) to facilitate online discussions [27]. This moment demonstrates that FS7's attitude toward and eagerness for interculturality could be developed by rejecting "ingroupness" while remaining aware of the intercultural context to continue the online conversation [27]. In excerpt 8, CS6, as a facilitator, switches to Chinese with her Chinese classmates for assistance when she does not understand FS7. After not receiving any responses, she switches to English and asks the Chinese researcher/assistant for help. She now communicates in English with the researcher rather than conversing in Chinese, in order to position the researcher as a teacher in the online intercultural interaction. Chinese here becomes "we" code for in-group and informal interactions with CSs, while English becomes "they" code for more formal out-group relations with the Chinese researcher [29] (p. 66). This moment shows that code-switching as a membership categorization device does not include/exclude individuals by their nation-state roles [14], but accepts/rejects membership with consideration for different purposes in different contexts [27].
Excerpt 8 G4D2-7 FS7: The one the one you showed with the pictures. ( . . . ) Hehh. Where you have the for six and for two and the other slides you showed. FS8: That PowerPoint you have made. FS7 talks about that. CS6: 没听懂，啥意思啊? ((Whispering to other CSs who were in the same room with her.)) ((I don't understand. What did they mean?)) Teacher, can you help me translate their words. This network is not well. Hehh.
The result shows that code-switching as a strategy is employed by students to align themselves with others and build interactional arrangements. For instance, they employ code-switching as fence-sitters to create ambiguity by suggesting inferences without risking loss of face [29]. It also serves as a membership categorization device to include or exclude others as intragroup or outgroup [27].

Discussion and Conclusions
Based on a case study of educational virtual exchanges between Chinese and Finnish students, this paper drew on an understanding of interculturality as "critical" to examine how identities are (co)constructed by employing soothing and code-switching as strategies to deal with moments of crisis during the students' interactions. Such moments are central to understanding what individuals "do" when they meet interculturally. As a case study, this study is not meant to generalize for the thousands of intercultural virtual exchanges taking place around the world but to open up our eyes to an under-researched aspect of interculturality in education: how to deal with moments of crisis? We described 10 moments of 10-h online synchronous oral discussions between Chinese and Finnish students who had not been trained specifically about intercultural communication. Our data was examined based on critical interculturality [30]-beyond "culture" with a focus on the processes of co-construction and negotiations of meanings, identities, and ideas when people meet across borders-and sociolinguistic [19,29] perspectives.
Going back to our research questions, this is what we find for the question concerning what triggers moments of crisis. Looking into "response cries" and code-switching, which are signals for moments of crisis here, we find five specific issues that trigger them: someone's spelling errors; someone's absence; misunderstandings; difficulty expressing oneself; and disagreement. The participants use soothing and code-switching in these moments of crisis to achieve interactional arrangement (e.g., moving the discussion forward) and self-other alignments (e.g., saving self's and/or other's face). In other words, identities are co-constructed by aligning oneself with the other [19].
Regarding research question two: how do they deal with moments of crisis? We find that soothing and code-switching are employed as strategies which can be considered as double-edged in "doing" interculturality. On the one hand, soothing as a positive strategy is employed to make people feel comfortable when individuals react to others' "response cries" [19]. On the other hand, code-switching seems to correspond to an avoidance strategy used to shun getting involved in interactional incidents by asking others to mediate.
Regarding research question three (what does the use of these two strategies reveal about identities?) our results support Tracy and Robles' [17] rhetorical perspective on identities: individuals act strategically and purposefully, demonstrating that they have agency/ability to deal with crises in interaction [16]. Individuals employ soothing and code-switching for different purposes in different situations. When and why they use different strategies projects their different identities. Identity as a mediator, for example, emerges by employing soothing as a positive strategy to comfort others' nerves, soften tensional atmospheres, and mitigate complaints [26]. For a fence-sitter, code-switching protects them from losing face in online intercultural interactions [28,29]. Finally, facilitators use code-switching as a means of accepting/rejecting members as ingroups/outgroups for different purposes [27].
Our study also has pedagogical implications for offline/online intercultural education. Teachers can train students to deal with incidents in interactions by including work on the strategies of soothing and code-switching in their classes. Our study advises teachers to pay close attention to students' (re)actions and use of strategies in virtual exchanges since they can reveal more about their attitudes or emotions in their engagement with interculturality. As Goffman [18] (p. 33) maintains, "evidence of social worth and of mutual evaluations will be conveyed by very minor things". It is also beneficial to encourage students to evaluate the strategies they themselves use when faced with moments of crisis, as this could develop, e.g., their intercultural competence, especially their self-observation and self-evaluation in interactions. Intercultural virtual exchanges in education already have a long history, and we hope that the COVID-19 pandemic will urge more researchers and educators to explore their richness and intricacies and to help them expand their take on the central position that interculturality plays in them.
Author Contributions: Both authors J.P. and F.D. contributed equally to conceptualization, methodology, formal analysis, data treatment, writing-original draft preparation, writing-review and editing, and project administration. Jun Peng collected and transcribed the data. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the University of Helsinki.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: Data are available upon request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.