Saving and Maintaining Face: Metapragmatic Expressions in BELF Meetings

English as a business lingua franca (BELF) has contributed to an increasing number of intercultural studies that aim to evaluate its complexities and participants’ pragmatic competence in this context. The worldwide use of BELF interactions is subject to multiple contextual factors, with metapragmatic expressions (MPEs) serving as one of the linguistic devices and pragmatic strategies. Based on BELF meetings from the Vienna-Oxford International Corpus of English (VOICE), this paper explores how MPEs are used to save and maintain face in order to promote the effectiveness of communication. It is found that the use of MPEs in BELF meetings can help speakers to accomplish their transactional goals by showing concern for face in this intercultural business context. Also, the use of MPEs dynamically reflects the participants’ interpersonal pragmatic competence in intercultural settings. The present study deepens the understanding of metapragmatic awareness and the mechanism of MPEs in BELF interactions.

Besides providing insights into business English teaching, the present study also helps businessmen to show concern for the face of different parties by using MPEs so as to establish harmonious relationships in BELF meetings.
Another conceptualization proposed by Louhiala-Salminen et al. (2005) is called English as a business lingua franca (BELF), which draws explicitly on ELF research and contextualizes it in international business (Kankaanranta & Planken, 2010;Louhiala-Salminen & Kankaanranta, 2012;Kankaanranta et al., 2015). BELF is perceived as being "highly contextbound and situation-specific", and BELF competence calls for "clarity and accuracy in the presentation of business content, knowledge of business-specific vocabulary and genre conventions, and the ability to connect on the relational level" (Kankaanranta & Planken, 2010;Kankaanranta et al., 2015). While many BELF interactions may be brief, spontaneous, and not repeated, the significance of building a harmonious relationship with business partners implies that a sustained and repeated interaction is the more likely, or perhaps more useful, scenario in BELF research (Komori-Glatz, 2018). Hence, there is still a need for further research into BELF in general, as well as for researchers to engage more across disciplines and coordinate their findings.
BELF is characterized as being constituted and constitutive at two levels. It includes a flexible, variable, and adapted discourse itself, affected by the participants' linguistic repertoires and the demands of the specific context, as well as the level of the interaction, particularly in the context of teamwork and/or repeated interactions, in creating, shaping, and confirming group/team processes (Komori-Glatz, 2018). How English is used as a lingua franca in any given context depends greatly on many factors, including but not limited to the setting and purpose of the interaction, the speakers' (shared) linguistic repertoire(s), their experience in technical and multicultural communication, and the length and power dynamics of the relationship (cf. Ehrenreich, 2009Ehrenreich, , 2010Ehrenreich, , 2016Kankaanranta & Planken 2010;Kankaanranta et al., 2015). For many businesses and businessmen, English is deemed as the only viable option for communicating with speakers having otherwise incompatible linguistic repertoires (Ehrenreich, 2010), and plus, BELF interactions are frequently reported to be successful due to substantial efforts towards cooperation and collaboration from the participants (Kankaanranta & Planken, 2010;Kankaanranta et al., 2015;Ehrenreich, 2016). While research on BELF has grown massively both in quantity and in its conceptual underpinnings (Komori-Glatz, 2018), there is still a need to clarify this concept in a specific context of international business communication. As the relational aspect of business meetings is still under-researched, the present study will focus on facework in BELF meetings.

Face in business meetings
Studies on face and facework are interrelated with politeness research. The notion of face originates in the work of Goffman (1967), and has developed dynamically and comprehensively from a pragmatic perspective (e.g., Matsumoto, 1988;Gu, 1990;Janney & Arndt, 1993;Bargiela-Chiappini, 2003;Watts, 2003). While their approach will not be entirely applied, the notions of "positive" and "negative" face are still relevant to areas of institutional discourse, including politeness, power, conflict, and convergence (Spencer-Oatey, 2000;Handford & Koester, 2010). According to Brown & Levinson's (1987:61) face theory, a social interaction is characterized by the desire of each participant to enable their positive and negative face wants to be met, i.e., the wants or needs for praise and admiration and the desire for freedom from imposition. In general, people will not go out of their way to violate the face of others in the workplace, so when there is a necessity to perform a face-threatening act, people will usually attempt to use mitigation, usually achieved by employing some politeness strategies (Brown & Levinson, 1978, 1987, such as being positively polite by complimenting him/her, allowing him/her plenty of room to reject an imposition (negative politeness), or wording it in a way that appears non-conflictual.
Workplace interactions may potentially develop into conflict and confrontation. Face-threatening acts such as "requests, orders, complaints, and refusals" (Handford, 2010:36) are commonly seen in the business context. Most workplace interactions provide evidence of "mutual respect and concern for the feelings or face needs of others, that is, politeness" (Holmes & Stubbe, 2003:7). Watts (2003) holds that instead of being regarded as a reason for modifying impositions, politeness is seen as a context-sensitive means for softening impositions and may indeed be a socially constrained prerequisite of many types of meetings in particular cultures. Business is, by definition, primarily concerned with transactions (Handford, 2010:36), but it would be incorrect to conclude that workplace interactions are consistently polite, as business people are "inclined to override politeness considerations for the sake of conversational clarity" (Bargiela-Chiappini et al., 2007:195), particularly in international contexts. In Bhatia's (2004:15) observation, a successful business activity "always thrives on building positive relations between various participants". As little attention has been paid to facework in intercultural contexts, the present study will fill this gap by exploring how to save and maintain face via MPEs in BELF meetings.

Metapragmatic expressions (MPEs)
Metapragmatics is typically defined as "the pragmatics of actually performed meta-utterances that serve as [a] means of commenting on and interfering with ongoing discourse or text" (Hübler & Bublitz, 2007:6), aiming to explore how to manage discourse "based on common knowledge and reflexivity" (Caffi, 2007:83). Smith & Liang (2007:172) argues that metapragmatic expressions (MPEs) "referred not to the content but to how the audience might understand, use, or orient themselves to it", which means MPEs are used to comment on and interfere with interactions. Structurally, MPEs appear in the form of clauses, sentences, and sentence fragments; semantically, they do not directly deal with the issues or topics under discussion (Liu & Ran, 2016a). Instead, MPEs indicate that "the speaker is both the involved participant and the observer of him/herself and of the interaction" (Caffi, 2007:86); in other words, MPEs show the speakers' reflexive awareness and reveal the speakers' ability to be better involved in the regulation and management of communication.
Extensive research has progressed in different directions from a metapragmatics perspective. The term metapragmatics has been used to analyze reported speech and indexicality (Lucy, 1993;Robinson, 2006). Many approaches have been employed in analyzing metapragmatics in both daily interaction and institutional discourse (Liu & Ran, 2016b), focusing on metapragmatic indicators (Verschueren, 2000), metapragmatic comments (Ciliberti & Anderson, 2007), and metapragmatic utterances (Hübler & Bublitz, 2007). From a functional perspective, Hewitt & Stokes (1975) find that some MPEs function in a similar way to what Caffi (1999) calls "mitigation", a set of strategies rooted in a metapragmatic awareness (Liu & Liu, 2017), contributing to definition and classification of MPEs.
The current study mainly refers to Liu & Ran's (2016a:463) definition of MPEs as "linguistic expressions which explicitly display the speaker's reflexive awareness of language use and his/her intention to manipulate the ongoing interactions to meet particular communicative goals and/or needs". According to Liu & Liu (2017), MPEs reveal specific, goal-oriented motives and mechanisms in BELF meetings. This article argues that the use of MPEs indicates an intention of saving and maintaining face based on the dynamic relationship between participants in BELF meetings.

The Vienna-Oxford International Corpus of English (VOICE)
The data of the present study is selected from the Vienna-Oxford International Corpus of English (VOICE), which was organized and created by Barbara Seidlhofer, professor of English and Applied Linguistics at the University of Vienna in Austria. The corpus contains a collection of ELF users' natural face-to-face communication activities in five different scenarios: educational (ED), leisure (LE), professional business (PB) professional organisational (PO), and professional research (PR). The communicative events included in each scenario include conversations, interviews, meetings, group discussions, panel discussions, and press conferences. The present study selects some representative extracts from PBmtg 3, which is the longest of the eight business meetings.
VOICE provides an account of the events of the meeting, which is elaborated on as follows. All participants in the meeting are non-native speakers of English and have different organizational roles, so the interaction is typical of multicultural business organizations. The meeting lasts 3 hours, 28 minutes, and 6 seconds with a transcription corpus of 24,601 words. Six speakers emerge in the meeting in total, including two Korean logistics managers (S1 and S2) for the distribution company, with Korean as their first language; three Austrian salespeople (S3, S4, and S5), and one Austrian researcher (S6), all of whom speak German as their first language. The power imbalance within and between the five individuals restricts communicative competence and reflects the complexity of communication in business organizations. On the Korean side, S1 ranks higher than S2 though they are both managers; for the Austrian party, S4 is the sales manager and chairs the meeting in the meantime, while the other participants are the sales employee in the company (S3), and the sales assistant (S5) who assists S4 in the meeting. The meeting was held during the visits of S1 and S2. The main task of the Korean side is to advise the Austrian partners on the promotion activities in Korea, while the main task of the Austrian side is to inform the Korean side of the product news. In addition to a business relationship, this is also an opportunity for S4 to establish an ongoing interpersonal cooperating relationship with S1 and S2 since S4 has just taken charge of the Korean market without meeting these two Koreans before.
The reasons for selecting this meeting for analysis are as follows. First of all, contextual constraints allow more room for analysis. Focusing on a particular communicative practice is helpful to reduce the influence of other factors on the research. In addition, this meeting typically belongs to intercultural communication (BELF interactions), which reflects the features of multicultural business institutions. With one side coming from Europe and the other from Asia, there are obvious differences between those non-native English speakers in languages and cultures. Thirdly, a BELF meeting has the dual purpose of being task-oriented and relation-oriented, which provides abundant and diverse data for this study. Last but not least, the power imbalance affects the process of communication. Due to the asymmetrical power distribution among participants, some of them may play a leading role in the interaction, which may trigger more use of MPEs to manage and regulate the interaction.

Data identification
As a typical business discourse study situated in one particular intercultural setting, this paper seeks to explore the role of MPEs in saving and maintaining face in BELF meetings. Given the institutional features of BELF interactions (cf. Ehrenreich, 2009Ehrenreich, , 2010Ehrenreich, , 2011Ehrenreich, , 2016Kankaanranta & Planken, 2010;Louhiala-Salminen & Kankaanranta, 2012;Kankaanranta et al., 2015) and categories of MPEs in previous studies (Liu & Ran, 2016a, 2016bLiu & Liu, 2017, 2021, MPEs are identified based on their roles in the business meeting. The present study will not expound on specific types of MPEs, but regard them as a general category of pragmatic strategies.
The procedures for transcribing the data are as follows. First, two coders read the corpus and relevant literature for the context and working definitions of MPEs in the present study. Second, the two coders respectively coded the meeting strictly, with examples and analyses of previous studies serving as a reference. Third, upon the completion of data coding, the coders shared findings and exchanged ideas by figuring out differences in the corpus. Controversial identification was discussed carefully, and coders sought advice from other researchers if necessary until a final agreement was reached.

Meaning and profession negotiations in BELF meetings
Despite sharing some business knowledge and expertise, participants in BELF meetings are generally lacking in core common ground. MPEs used in the meeting serve as adjusting devices to help create common ground when there is insufficient specific knowledge of some issues in the interaction (Kecskes, 2013). This is associated with what Kankaanranta & Planken (2010) and Ehrenreich (2016) point out regarding the characteristics of BELF: mutual understanding of business-related issues and interpersonal relationships. In this process, self-face should be taken into account via MPEs for a better display of mutual understanding.

Meaning negotiation for the addresser
Participants in BELF meetings negotiate word meanings since some terms are culturally-specific. In extract 1, the participants are working out the meaning of "gullible". During the negotiation, MPEs are employed to show the addresser's concern for the positive self-face in order to establish a shared sense and mutual understanding of this unfamiliar word.
Extract 1: They are discussing what "gullible" means together.
2278 S1: okay er now er talking about target er kids are no more gullible than adults perhaps even less so they're The opening expression in the extract "okay er now er talking about" (line 2278) signals a new topic during the conversation, which shows S1's desire to drive the participants' focus to the product presentation. That indicates S1's application of MPEs concerning his own positive face, wishing to be accepted and gain attention from the other participants. In the presentation material comes an unfamiliar word "gullible", which is hard for the Korean party to understand. Realizing S4's confusion "i've never heard" (line 2280), S1 tries to answer the question with different expressions but they fail to come into play (see "a bit confused" in line 2291). This situation points to the lack of common ground in BELF meetings regarding terms of different expressions. Then S4 suggests referring to the dictionary for help to figure out the exact meaning of "gullible". Two MPEs "does not matter" (line 2296) and "i'm just curious because i've never heard it" (line 2298) reveal S4's positive attitude toward the meaning negotiation of this new word, which signals his appreciation for S1's work of turning to the dictionary and displays his concern for S1's positive face. What is more typical is three MPEs "i'm not sure" (line 2300), "if i'm correct" (line 2302), and "if i'm right" (line 2307), which clearly disclose S1's uncertainty about his utterance. In order to actively carry on the meeting, he employs these three MPEs, hoping that the addressee can accept his explanation for the unfamiliar word. Positive self-face is evidently manifested as S1 wishes his opinions to be approved of and adopted via MPEs.
In this extract, the opening expression activates the topic for discussion. The main conversation consists of several MPEs to present different attitudes towards meaning negotiation. The interaction is dynamic as is seen in interpersonal relationships in these meetings. That is, in order to achieve relational goals, the initial consideration of the positive selfface of the addresser will involve the negative face of the addresser in the following. For example, what is discussed above is that "i'm not sure", "if i'm correct', and "if i'm right" reflect S1's desire to be accepted, which allows the other party to better take S1's point of view without being too imposing, therefore making the view more acceptable.

Professional negotiation for the addressee
In BELF meetings, participants focus on the professional aspects of business negotiation as the primary goal is to "get their work done" (Bargiela-Chiappini et al., 2007:3). In extract 2, the participants are talking about the packaging in promotion activities. During the meeting, MPEs are used to show different ideas when the addresser would like to deliver his idea to the addressee.
Extract 2: They are discussing how the product will be packaged.
3339 S1: pluto is also angry huh 3340 S2: @@ 3341 S1: mickey is angry 3342 S3: mickey is mickey is angry 3343 S5: haeh The topic of the extract is the picture on the package. When S1 says "goofy looks dull" (line 3344) and "goofy looks stupid" (line 3347), S5 holds an opposite view by saying "he looks nice'' (line 3349). Noticing S5's reaction, S1 then adds an MPE "yeah i'm just kidding" (line 3351). The MPE used in the interaction indicates that S1 does not have the intention to make poor comments on the package by negating his previous opinion so as to mitigate potential conflict. Literally, S1's reply suggests that he agrees with S5 that the goofy looks nice, but from a relational perspective, it can be seen that S1 attempts to manage rapport (Spencer-Oatey, 2008) in the interaction, that is, to maintain a harmonious relationship during the negotiation when different views emerge. The MPE "yeah i'm just kidding" is directly related to reducing the threat of S5's negative face; that is, he does not want his opinions to be unimpeded by others. More evidence can be seen in the next two MPEs, "well actually we're we're well we we have different opinions" (line 3354) and "we all have a different erm idea" (lines 3357), which reveal that S1 essentially thinks poorly of the package by acknowledging that they have different opinions. Opposite opinions are demonstrated in the maker "well actually" and the following S5's MPE "but they would give some refreshment for the" (line 3358).
The extract shows different ideas of different parties about the package, which is definitely a critical issue that deserves great attention. From the very beginning, S1 uses the MPE "yeah i'm just kidding" to show his concern about the idea of the counterparty, attempting to alleviate conflict and maintain a harmonious interpersonal relationship (S5's negative face). Next, "well actually" reflects that the addresser further puts forward his own point of view by employing a mitigation strategy, so as to make his own point of view more acceptable to the other party (S1's positive face). The dynamic interpersonal factors start from the consideration of the negative face of the addressee to the positive face of the addresser.

Interlocutors' linguistic and identity resources in BELF interactions
BELF participants are less strict with the use of English as their primary goal is to promote effective business communication. This means that BELF participants may not speak English idiomatically, so some deviation from or violation of English will not be considered as communicative failure or lack of linguistic capability. In their interactions, MPEs can be used to enable speakers to find out the relevance and limitations of communicative tasks, behaviors, and discourse. They accommodate each other to ensure intelligibility and to display rapport-building (Spencer-Oatey, 2008;Ehrenreich, 2016). The utilization of linguistic resources and identity resources play a crucial role (cf. Spencer-Oatey, 2007), which is achieved via MPEs, including relevant professional background knowledge, language level, language expression, and strategy usability, as well as differences in the speaker's mother tongue and cultural background.

Linguistic resources for the addresser
In BELF meetings, MPEs reflect the intention of the choice and utilization of linguistic resources in the exchange of ideas. In extract 3, the participants are explaining some terms during the introduction of a business plan. MPEs are used to allow the addressee to better comprehend the explanation.
Extract 3: They are introducing the business plan by explaining some terms.
2422 S1: well er [S4] i i see clearly er your your er concern but er last year was our our er performance was really er unsatisfactory yeah In the meeting, S1 plays a leading role in the introduction of promotional activities on the Korean market. The MPE "so let's look at er" (line 2424) clearly indicates the intention of initiating a new topic and directs attention to the business plan for the subsequent discussion. The MPE "probably you can give us some er comment" (line 2424) selects the Austrian side as the next speaker, specifically asking for an opinion on the business plan. From the perspective of business communication, the employment of these MPEs reveals the obligation of S1 as the seller, who is responsible for planning the sales activities of the product in Korea, but they need to get recognition from the other party during the communication.
In addition, as S1 dominates in the meeting about the sales situation in Korea, he figures out there are some technical terms in the report while discussing the business plan. The expression "and they created some some er jargon" (line 2428) suggests that S1 realizes that the word "jargon" may be incomprehensible. Then, the use of the MPE "do you know the word" (line 2428) is an explicit attempt to negotiate the meaning of the word. It displays an interactive and dynamic competence to apply domain knowledge, and on the other hand, discloses that S1 has noticed the possibility of a facethreatening act during negotiation. They continue the discussion after making it clear that both parties understand the meaning of the word, which is proved based on the following three MPEs "quite look-looks something huh" (line 2438), "quite impressive" (line 2441), and "quite im-" (line 2442).
The extract signals that the use of MPEs reflects speakers' metapragmatic awareness and pragmatic competence to actively accommodate the other party with proper use of linguistic resources in BELF meetings. The MPEs "probably you can give us some er comment" and "do you know the word" dynamically reveal that the addresser (S1) sacrifices his own negative face to show concern for the addressees' (S4 and S5) positive face. In this way, S1 guides the other side to express their opinions through the use of MPEs and supports the positive face of the Austrian side.

Identity resources for the addressee
Identity resources in BELF meetings not only indicate different roles of speakers, meaning negotiation, decision making, topic discussion, etc., but also disclose the features of BELF contexts, the common ground of business knowledge, and interpersonal relationships. In except 4, the addresser uses MPEs to signal topics to be discussed and to inquire about suggestions for the following business plans, which reflect the utilization of the identity resources of these speakers. MPEs are utilized to further express their rights, responsibilities, and obligations under the restriction of different contextual factors. In this extract, S1 acknowledges the criticism of the counterparty with the MPE "but it it is very good er good er er good advice" (line 2119), showing a positive attitude towards cooperation. This mitigation strategy avoids the direct threat to the positive face of the addressee. After communicating with S4 about the license issue, S1 uses the MPE "a bit unprofessional from our side" (line 2128) to self-reflect and make negative comments. The use of MPE again alleviates the threat to face, making the expressions of viewpoints nor too direct or pushy, so that the negative face of the addressee will not be threatened. To this, S4's response displays his concern for interpersonal relationships. The marker "n-yeah" in the MPE "n-yeah i mean i i see the good intention but" (line 2130) dynamically suggests that faced with S1's selfcriticism of being unprofessional, S4 disagrees at first and then agrees, but he immediately realizes that it would be inappropriate to confirm that S1 is not professional. Two markers, "n-yeah" and "i mean", indicate S4's attempt to find an appropriate stance and adjust his opinions out of concern for S1's negative face. The MPE "n-yeah i mean i i see the good intention but" serve as the preceding sequence, which helps to lead in the subsequent information "i also see the risk of" (line 2130). This strategy reveals the metapragmatic awareness of both parties in maintaining a harmonious interpersonal relationship.
The MPEs in the extract show the utilization of identity resources to avoid potential problems and solve conflicts. On the one hand, the addresser uses "positive" MPEs to affirm the suggestions of the addressee; whereas, on the other hand, they use "moderative" MPEs to mitigate potential negative effects caused by criticism, so that facework is fully achieved in the interaction. In particular, the use of MPEs "a bit unprofessional from our side" and "n-yeah i mean i i see the good intention but" reflects the dynamic face negotiation between BELF participants.

Concluding remarks
Compared with intracultural interactions, BELF interactions reflect the choice of MPEs by speakers from different linguistic and cultural backgrounds, which reveal their metapragmatic awareness to coordinate and manage their participation in the interaction and different resources in multiple contexts, aiming to achieve a win-win interactional goal. This paper sheds light on face and facework in BELF meetings to highlight the dynamics of interpersonal interactions via MPEs, and to explore the interpersonal pragmatic competence in the utilization of various resources and the knowledge construction in intercultural business settings. Basically, participants often use simple and clear ways of expression to promote negotiation and make innovative use of resources available to manage the interactions. That is, in the intercultural business context, appropriate linguistic devices and pragmatic strategies contribute to effective communication by creating common ground. More significantly in the present study, in terms of the dynamism of relations, speakers choose MPEs to relate to the facework of both parties and further establish and maintain a harmonious business relationship by saving and maintaining face in the interaction. Grounding one of the relational elements, this paper touches on face and facework to investigate the role of MPEs in BELF meetings, and how particular pragmatic strategies are used in BELF interactions still need to be explored. Moreover, research on other relational factors, such as identity and emotion, is suggested to be conducted to further evaluate the interpersonal functions of MPEs in BELF interactions, either from an interpersonal perspective or an intercultural perspective.