INTRODUCTION
‘Working together in teams is hard enough. It doesn't bode well when there are so many languages being spoken around you; it complicates matters even more’. (Interview respondent)
In today's global business environment, multinational, multicultural, and multilingual teams are among the most valuable resources for obtaining an organizational competitive advantage (Tenzer & Pudelko, Reference Tenzer and Pudelko2017). Managing diverse teams has recently become a critical topic in organizational research as it is the key to innovation, better decision-making, problem-solving, and accelerating organizational change. Team members may differ in many ways. Some of them, such as age, gender, and race, are easily noticeable; others, such as personality or culture, are more difficult to recognize and analyze.
Diversity management literature has evolved from looking at superficial diversity characteristics based on simple demographics to debating the role of faultlines, that is, hypothetical dividing lines that split a group into an ingroup and outgroup (Lau & Murnighan, Reference Lau and Murnighan1998). Social identification and self-categorization processes lead to subgroup formations, which can cause conflict, hinder information processing, and negatively affect team performance. Diversity does not always lead to faultline formation as its emergence depends on the relevance of a specific attribute to the team and the task. Since communication is at the core of international business (IB) and language permeates ‘every aspect of organizational life’ (Neeley & Kaplan, Reference Neeley and Kaplan2014: 70), linguistic differences may lead to the formation of language-based faultlines within multinational teams (Dotan-Eliaz, Sommer, & Rubin, Reference Dotan-Eliaz, Sommer and Rubin2009; Vigier & Spencer-Oatey, Reference Vigier and Spencer-Oatey2017).
Whereas prior research acknowledges language as one aspect of demographic faultlines (Barner-Rasmussen & Aarnio, Reference Barner-Rasmussen and Aarnio2011; Kulkarni, Reference Kulkarni2015), multiple faultlines can emerge from language diversity in teams. Some may be more salient than others depending upon the nature of the diversity. There seems to be an implicit assumption that language-based faultlines emerge between native and non-native speakers, without much thought given to a more fine-tuned understanding of language and its local dialects. We address this gap by answering the following research question: How does language diversity influence interpersonal relationships in multinational and national multilingual teams? To address this question, we conducted a qualitative study that consisted of observations and semi-structured interviews with 12 project teams in national and multinational organizations in India.
We intend to make three key contributions. First, we decouple linguistic diversity from the unconscious assumption of linguistic homogeneity of teams consisting of one country's nationals. Language-based faultlines are an explicitly acknowledged challenge in a multinational setting, and yet they are by and large ignored in a domestic, single-nation context. We demonstrate how even subtle linguistic differences in regional dialects can lead to faultline formation. Second, this study extends the work on multilingual boundary spanners by demonstrating that multilingual leaders are more effective as boundary spanners in multinational project teams (MPTs) than in national project teams (NPTs) that consist of people from the same country. Third, we answer the call by IB scholars to incorporate ‘intra-national differences’ in research (Peng & Lebedev, Reference Peng and Lebedev2017: 241; Tung, Reference Tung2008) by studying project teams in a multilingual context of India.
The article develops as follows: first, we review the existing literature on linguistic diversity and analyze how it can lead to faultline formation. Next, we introduce our methods and discuss the benefits of studying language diversity in the Indian multilingual context. We present our results and then discuss the effects of faultlines caused by linguistic diversity and the role of multilingual managers in mitigating those. We conclude the article with a call for further research into the part of regional and ethnic dialects in team collaboration.
THEORETICAL BACKGROUND
Linguistic Diversity and Multilingualism
Language diversity refers to not just the multiplicity of languages spoken by individuals but also the variety of ways in which individuals hear and interpret them (Kassis-Henderson, Reference Kassis-Henderson2005). Acknowledging and understanding the role of language in multinational enterprises (MNEs) is vital, mainly because of the increase in employees who speak and utilize many local languages (Harzing & Feely, Reference Harzing and Feely2008). Although language can create and strengthen interpersonal relationships, prior research shows that it can also serve as an exclusion tool and lead to linguistic ostracism, which results in interpersonal deviance and inhibits interpersonal and intergroup information exchange (Fiset & Bhave, Reference Fiset and Bhave2021). While some scholars such as Aichhorn and Puck (Reference Aichhorn and Puck2017) and Fredriksson, Barner-Rasmussen, and Piekkari (Reference Fredriksson, Barner-Rasmussen and Piekkari2006) suggest that the development of a common corporate language may reduce some of these challenges, others have analyzed the complexities and language-specific barriers associated with it (Brannen & Doz, Reference Brannen and Doz2012; Brannen, Piekkari, & Tietze, Reference Brannen, Piekkari and Tietze2014; Fredriksson et al., Reference Fredriksson, Barner-Rasmussen and Piekkari2006; Tenzer & Pudelko, Reference Tenzer and Pudelko2017; Welch & Welch, Reference Welch and Welch2008). Examples of such barriers include the status loss of non-native speakers of the chosen corporate language (e.g., Gaibrois & Nentwich, Reference Gaibrois and Nentwich2020; Tenzer & Pudelko, Reference Tenzer and Pudelko2017), feelings of exclusion and dilution of individual identity because of low proficiency levels (e.g., Kulkarni, Reference Kulkarni2015), and conflicts in communication processes between the headquarters and subsidiaries and within teams (e.g., Harzing & Feely, Reference Harzing and Feely2008; Tenzer & Pudelko, Reference Tenzer and Pudelko2017). A recent qualitative study by Wang, Clegg, Gajewska-De Mattos, and Buckley (Reference Wang, Clegg, Gajewska-De Mattos and Buckley2020) showed that language standardization could induce anxiety in both native and non-native speakers of the corporate language, thereby affecting knowledge exchange and organizational functioning. Since language allows for knowledge transfer at work, one's inability to communicate with colleagues could be detrimental to the overall strategic objectives of any organization (Fiset & Bhave, Reference Fiset and Bhave2021).
Language as a Source of Division in Teams
The complex and multifaceted construct of language is one of the determinants of team success (e.g., Hinds, Neeley, & Cramton, Reference Hinds, Neeley and Cramton2014; Kassis-Henderson, Reference Kassis-Henderson2005; Tenzer & Pudelko, Reference Tenzer and Pudelko2017). In their study, Tenzer, Pudelko, and Zellmer-Bruhn (Reference Tenzer, Pudelko and Zellmer-Bruhn2021) demonstrated that language diversity could reduce individuals’ participation in team communication and thereby inhibit knowledge processing in multinational teams. They found that evident language barriers, that is, barriers related to lexical and syntactical proficiency, and hidden language barriers, that is, barriers related to pragmatic and prosodic cues, can both lead to ostracization in teams. Language differences can become a potential faultline, that is, a hypothetical dividing line that splits a group into an ingroup and outgroup (Lau & Murnighan, Reference Lau and Murnighan1998; Vigier & Spencer-Oatey, Reference Vigier and Spencer-Oatey2017). Contrary to other diversity measures (e.g., demographics, age), faultlines tend to involve one or more of the diversity characteristics. The same individuals may be aligned with other teammates on one dimension (making them a member of a given subgroup) and crossed on another dimension (making them members of another subgroup). These underlying patterns of group member characteristics lead to subgroup formation through social identification and self-categorization processes (Lau & Murnighan, Reference Lau and Murnighan1998). Research shows that team faultline creation can lead to conflict, inhibit information processing, and have detrimental effects on key team outcomes such as decision quality, accuracy, and overall performance (Antino, Rico, & Thatcher, Reference Antino, Rico and Thatcher2019; Thatcher & Patel, Reference Thatcher and Patel2012). It is important to note that team diversity does not always lead to faultline creation; the emergence of a faultline depends on the relevance of a specific attribute (e.g., native language or a regional dialect) to the team and tasks at hand.
Dotan-Eliaz et al. (Reference Dotan-Eliaz, Sommer and Rubin2009) built on the idea of language-based faultlines within teams and examined the effects of linguistic ostracism. As expected, teams in which some members talk in a language that other members cannot understand reported lower perceived team potency. Other scholars such as Voss, Albert, and Ferring (Reference Voss, Albert and Ferring2014) have studied the relation between the comfort of using a foreign language and the perception of conflicts in multinational teams and stated that the presence of subgroups creates a dynamic that leads to relationship and/or task conflicts. Similar findings have been reported by Tenzer, Pudelko, and Harzing (Reference Tenzer, Pudelko and Harzing2014) and Tenzer and Pudelko (Reference Tenzer and Pudelko2017) who found that linguistic differences often contribute to the perception of high team diversity but create mistrust among team members. Using an inductive approach, Aichhorn and Puck (Reference Aichhorn and Puck2017) demonstrated that the use of accommodation approaches, such as translation, by certain bridging individuals within teams could reduce the negative consequences of linguistic differences and lead to clear communication; however, this field of study is still in a nascent stage.
Research has identified the problems related to communication and knowledge transfers that arise due to the formation of subgroups along the faultlines of a shared native tongue and nationality. With emerging economies being home to a vast multilingual workforce (Jiménez & Bayraktar, Reference Jiménez and Bayraktar2021), it is possible for employees to form subgroups around regional languages; however, this has not been analyzed in depth.
The Role of Language in Project Teams
This study was focused on understanding the impact of language diversity in project teams – temporary teams that are assembled to accomplish short-term projects with a duration of 6–12 months. Short-term project teams are prevalent in all spheres of the global business environment, and their members are required to collaborate and complete tasks at a fast pace (Vigier & Spencer-Oatey, Reference Vigier and Spencer-Oatey2017); thus, effective and efficient interactions among team members are critical for success (Kozlowski & Ilgen, Reference Kozlowski and Ilgen2006; Oosterhof, Van der Vegt, Van de Vliert, & Sanders, Reference Oosterhof, Van der Vegt, Van de Vliert and Sanders2009). Tenzer et al. (Reference Tenzer, Pudelko and Harzing2014) found language diversity to affect the levels of intra-team trust; therefore, communication and intra-team trust building are of paramount importance, especially in project teams where members may not have previous experience of working with each other (Langfred, Reference Langfred2007). Interestingly enough, while a significant body of research has dealt with cultural diversity (e.g., Mach & Baruch, Reference Mach and Baruch2015), language diversity has received much less scholarly attention (for more detail, see Shemla, Meyer, Greer, & Jehn, Reference Shemla, Meyer, Greer and Jehn2016). This literature gap can be attributed to two reasons: (1) language being considered an element of the larger construct of culture and the assumption that cultural characteristics play a greater role within a multicultural (multinational) team (Wang et al., Reference Wang, Clegg, Gajewska-De Mattos and Buckley2020) and (2) the assumption that team members who come from the same country all speak the same native language. In the context of linguistically diverse economies such as India, these assumptions do not hold, as team members may, by and large, share similar cultural characteristics and yet differ in native language (Jiménez & Bayraktar, Reference Jiménez and Bayraktar2021). Faultlines may be less likely to emerge in situations with high levels of linguistic diversity, since subgroup divisions are unclear (because of the shared cultural background), yet they tend to be deeper when subgroup members share similar (linguistic) attributes (Lau & Murnighan, Reference Lau and Murnighan1998; Vigier & Spencer-Oatey, Reference Vigier and Spencer-Oatey2017). Such a research context allows for the examination of language diversity and the confounding effects of multicultural interactions. Accordingly, in the current study, we aimed to determine the similarities and differences in the impact of language diversity on NPTs and MPTs in India (Table 1 highlights the differences between the present study and previous studies on language diversity in the Indian context). In doing so, this study incorporates both inter-lingual (i.e., national languages) and intra-lingual (i.e., regional languages and dialects) conceptualizations of language (Tenzer et al., Reference Tenzer, Pudelko and Harzing2014).
Table 1. Key studies focusing on language diversity in India
METHODS
A qualitative research design was adopted for this exploratory study (Plakoyiannaki, Wei, & Prashantham, Reference Plakoyiannaki, Wei and Prashantham2019), as it is suitable when the core concepts of the study are immature due to a clear lack of theory and previous research (Creswell, Reference Creswell2007; Pratt, Reference Pratt2009). Specifically, a case study design (Stake, Reference Stake1995) consisting of in-depth interviews (level of analysis = individual) and observations (level of analysis = team) was employed, with a focus on theory elaboration, that is, to fill gaps in the existing research on language diversity in teams (Lee, Mitchell, & Sablynski, Reference Lee, Mitchell and Sablynski1999; Pratt, Reference Pratt2009). Utilizing two methods of data collection also facilitated methodological triangulation (Stake, Reference Stake1995).
Rationale for Selecting India as the Country Context
India is rapidly becoming an investment hub for large global organizations such as The Alphabet Inc., Apple, and Foxconn (S&P Market Intelligence, 2020). As India is ranked fourth in the world in linguistic diversity (Skutnabb-Kangas, Reference Skutnabb-Kangas2000), it provides a fascinating and rich empirical context for studying the implications of language diversity (Kulkarni, Reference Kulkarni2015). Twenty-two official languages and 30 languages are spoken by more than one million people in the country. According to the data on bilingualism and trilingualism in India, 255 million people speak at least two languages, and 90 million people speak three or more languages (Registrar General & Census Commissioner, India, 2011). It has been argued that multilingualism in India is a ‘natural phenomenon’ and the multiplicity of languages and linguistic identities is a ‘defining (feature) of Indian bilingualism’ (Bhatia & Ritchie, Reference Bhatia, Ritchie, Bhatia and Ritchie2004: 795). India is thus an interesting, yet understudied, context for analyzing the effects of language diversity.
Case Selection Criteria
When selecting project teams for this study, intentional, theoretical sampling (Stake, Reference Stake1995) was employed; data sources (e.g., organizations and multilingual project teams) were purposefully selected based on their relevance to the research question at hand. The case was selected based on its ability to contribute to the existing literature by filling important conceptual categories. Specifically, a domestic and a multinational organization, wherein most tasks are accomplished via the division of employees into teams, was chosen so that the internal workings, that is, the languages spoken during team meetings and patterns of interactions between team members, could be observed.
Multilingual teams were intentionally selected for the study, which was not a difficult task as many of the teams in the chosen organizations were multilingual. Furthermore, the research setting was limited to one industry (management consulting) so that other macro environmental factors could be kept constant while the impact of language diversity was studied. All the MPTs considered for this study were located in India and consisted of individuals from different countries, including Brazil, China, India, Denmark, France, USA, and UK. NPTs consisted of individuals from various regions of India. In addition to English, an average of two regional languages were spoken within each team. Notably, in both multinational and national teams, team members vary in terms of their age (24–46 years), experience (2–16 years), and managerial responsibilities as well as employment duration in their current organization (1–15 years). This variation helped make the sampling more purposeful (Lee et al., Reference Lee, Mitchell and Sablynski1999). Tables 2 and 3 provide information related to the project teams and interview respondents, respectively.
Table 2. Project team information
Table 3. Interview respondent information
Notes: n, native; NT, member of NPT; MT, member of MPT.
Data Collection Approach
Semi-structured interviews and direct observations were used for the purpose of this study, that is, to understand the complexities at play within project teams.
Direct observations
All 12 project teams (6 multinational and 6 national) were observed to experience and understand the nature of the work and social interactions (e.g., team meetings, managerial interactions with team members), and the reactions and behaviors of those team members who could not be interviewed due to organizational restrictions were noted down. The primary goal was to observe the pattern of interactions among members: who was the contact person for clarifications, which languages were being used, when individuals switched between languages, whether there were any language-based subgroups, and what was the role of team managers/leaders during the meetings as well as whether clarification questions were asked in the corporate language (English) or a different language, whether the manager answered questions in English or in a local language, and the manager's reaction when team-members interacted in their native languages. The intention was to understand the role of language diversity at the team level. Overall, 35 hours of observation data were collected. Direct observations were conducted before individual interviews for two reasons: to identify individuals for the interviews and to follow up on specific events/interactions during the semi-structured interviews.
Semi-structured interviews
Thirty-six interviews were conducted with three persons from each project team[Footnote 1], and each interview ranged from 45 to 60 minutes. Twenty interviews were conducted in person, and 16 were conducted via Skype. The interviews helped us elaborate on existing theory by focusing on the specific experiences of respondents in linguistically diverse teams. Semi-structured interviews assure a certain extent of consistency in the questions so that the responses of interviewees can be compared; at the same time, this approach is flexible enough to enable the discussion of important, unanticipated topics. For instance, when we learned about the role of boundary spanners, we edited our interview structure to include questions related to this issue.
All participants were given the option to communicate in their native language during the interviews, however they opted to do the interviews in English. They explicitly stated that they were comfortable speaking in English, and the authors found them to have excellent command of the language (lexicon, syntax, and sentence structure), negating the need for translators. Third parties, such as translators, have the potential to create disturbances and hinder the flow of communication during interviews (Usunier, Reference Usunier1998), due to which we did not want the presence of a third party to inhibit the process of trust formation between the interviewer and interviewee. The interviews focused on the informants’ experiences within their current teams and thoughts concerning language diversity. The interview structure consisted of questions related to themes discussed in prior research – specifically, research on language diversity and identities was used to form questions related to language-based subgroups and faultlines within organizations. Additionally, the notes taken during direct observations also guided the framing of questions. The initial set of interview questions was modified after a pilot study with five respondents. The results of the pilot study were used to refine the research design and ensure that the respondents were able to understand and answer the questions clearly (Kim, Reference Kim2011). The pilot study also helped enhance the credibility of our qualitative study.
After the pilot study, questions related to the role of team managers in multilingual teams were added. The following are some of the categories of questions that were asked: the notion of linguistic identity and subgroups (e.g., describe the relationship between team members who speak the same language(s), describe the relationship between people who speak different native languages); the relationship between language diversity and team interactions (e.g., describe an instance where you encountered a situation in your team where people spoke multiple languages; what were the challenges/opportunities associated with that?); positive and/or negative outcomes of such diversity (e.g., can you elaborate on a situation where the language differences positively/negatively impacted the team interactions and functioning?); and the role of team manager and leader (e.g., how can one ensure that the subgroup formation does not hinder the team functioning? How can the manager help run the process smoothly?).
It should be noted that the questions asked of each interviewee varied based on their responses as well as behaviors during the observations. For instance, if an interviewee stated that language diversity created opportunities in their team, they were then asked to elaborate on its positive impacts on team interactions. On the other hand, if someone noted that language diversity created challenges, they were asked to elaborate on the negative effects. Many of the questions were purposefully created to be open-ended and thereby provide insights into the participants’ opinions on the current workings of their respective teams as well as their involvement in other activities within their organizations. The aim of the study was to understand the role of language diversity at the individual level.
Data Analysis
Thematic analysis was used to identify key initial themes/first-order codes in the data based on their recurrence, repetition, and forcefulness (Owen, Reference Owen1984). We intended for the themes to emerge from the data rather than any a priori assumptions. Initial themes or first-order codes were noted when three criteria were present: (1) recurrence, (2) repetition, and (3) forcefulness (Owen, Reference Owen1984: 275). Recurrence is observed when two or more parts of a response have a similar meaning, even when different words are used. Repetition, on the other hand, refers to the use of key terms multiple times, and forcefulness refers to ‘vocal inflection, volume, or dramatic pauses which stress and subordinate some utterances from locutions in oral reports’ (Owen, Reference Owen1984: 275). The data collection and analysis followed an iterative process of cycling between the data, existing literature, and emergent theory until saturation was reached (Locke, Reference Locke2001; Tenzer & Pudelko, Reference Tenzer and Pudelko2017). Through this process, relationships were found between the data gathered, which were finally integrated into a set of three core findings after theoretical saturation was reached. Tables 4 and 5 provide an example of the data analysis process. Themes were determined based on the theoretical categories indicated by prior research as well as emergent categories. Categories based on prior research include exclusion, power imbalance, and faultlines (e.g., Harzing & Feely, Reference Harzing and Feely2008; Marschan-Piekkari, Welch, & Welch, Reference Marschan-Piekkari, Welch and Welch1999). Other categories, such as the role of multilingual managers in gaining trust and reducing language-based subgroups, were recognized as emergent. The generated data was repeatedly compared with emerging categories and themes until no new information emerged (Creswell, Reference Creswell2007), and only findings that were triangulated through both interviews and observations were finally considered.
Table 4. Deduction of first-order categories from data
Table 5. Deduction of final conceptual dimensions from first-order themes
RESULTS
The three broad themes (language-based faultlines, dialect-based faultlines, and multinational managers as better boundary spanners) that emerged from data analysis were grouped into two main findings; these are discussed below.
Dual Faultlines in NPTs Compared with MPTs
Language-based faultlines were found to exist in both NPTs and MPTs. However, evidence of regional dialect-based faultlines was found only in NPTs. Conversations within both teams frequently drifted from English to the employees’ native languages. During one direct observation session of team NPT#3, some members were constantly switching between English and their local language, and no one was translating for other team members. Three of the team members were visibly uncomfortable, and there were smirks on their faces. We inferred from this observation that there were two different language subgroups conversing among themselves.
When probed further during an interview, a participant (NT18) conveyed the following:
‘Not a single conversation happens entirely in English, which is the official language here. I do not understand much Hindi, so I don't understand these insider jokes. I feel like an outsider, and sometimes, I just pretend to laugh at those jokes to avoid feeling embarrassed’.
Another interviewee (MT5) who worked in a MPT conveyed the following:
‘Of course, people interact more with those who share their language – isn't language like a part of culture? We were told this in our MBA, but I think we never talk about such stuff here at the office. I find it easier to work with people who speak my language; I don't have to explain everything to them in English … plus I am more comfortable with them’.
Thus, the above interviewees indicated that language is a strong part of their identity, and that language diversity can often lead to feelings of exclusion, discomfort, and anger among the team members, which leads to the creation of language-based subgroups and divisions (i.e., faultlines). Furthermore, language is closely related to one's emotions, and individuals feel more comfortable sharing information with those who speak their language. The interviewees from both types of teams further mentioned that such divisions could delay project completion by hindering team-level communication, particularly because employees might withhold essential task-related knowledge from those who are perceived as outsiders or members of other language subgroups.
Additionally, regional dialect-based divisions were seen within the native language-based subgroups in NPTs, as faultlines and language-based groups can disintegrate into smaller regional dialect-based subgroups due to the emergence of dialect faultlines (refer to Table 6 for the differences between language- and dialect-based divisions and faultlines). A regional dialect refers to ‘a variety of language which differs grammatically, phonologically and lexically from other varieties, and which is associated with a particular geographical area’ (Trudgill, Reference Trudgill2003: 35). This implies that some of the vocabulary used by an individual speaking in their regional dialect is unique to their region or state. For example, Hindi is the native language of people living in several Indian states; however, the vocabulary, grammar, and sentence structures employed by Hindi speakers can vary depending upon their geographical locations. A few examples of dialects of Hindi include Haryanvi (associated with the state of Haryana), Marwari (associated with the state of Rajasthan), and Bhojpuri (associated with certain cities in the state of Bihar). Similar variations can be found in other languages such as Tamil, Punjabi, and Telugu. During the observations, it was noted that not all individuals who speak Hindi were comfortable with communicating with each other – it was clear that faultlines existed within those groups as well. For instance, during an observation of team NPT#6, we noted that the conversation had abruptly shifted from English to Telugu and members were speaking in two different dialects of Telugu. We learned that A and S were talking in a dialect that H and D could not completely understand, which made the latter visibly uncomfortable and frustrated.
Table 6. Differences between language- and regional dialect-based faultlines
Subsequently, we followed up on this finding during the interviews. Of the 18 interviewees who work in national organizations, 14 noted that the regional dialects of their colleagues constitute an important distinguishing factor that can influence communication. Our interviewees from NPTs mentioned that regional dialect is a strong attribute of their identity and that further divisions along regional dialects often formed within the linguistic subgroups. For instance, one interviewee (NT8) stated the following:
‘The same language can be spoken in numerous different ways, and sometimes, the grammar or vocabulary is so different that it is difficult to understand all the words. To me, it signifies that the other person has a different background, and I don't always feel at ease with them’.
This interviewee hails from Chennai and speaks an urban (city) dialect of Tamil; he experiences emotional and processing difficulties when interacting with those who use a different lexicon. The other interviews revealed that processing difficulties are generally experienced by speakers of both urban and rural dialects. Furthermore, being part of dialect-based subgroups is a source of support for NPT members. Notably, regional dialects are closely associated with one's regional identity and caste in the Indian context (Chen, Chittoor, & Vissa, Reference Chen, Chittoor and Vissa2015), resulting in social stigma being attached to certain dialects. One's verbal speech can easily indicate one's (presumed) status and hierarchy in the social environment, which is one reason why dialect-based faultlines and conflicts emerge in NPTs.
Dialects did not emerge as an important communication factor in MPTs during the direct observations; to verify this result, the participants from MPTs were probed regarding the role of dialects during their semi-structured interviews. As the multinational teams consist of people from different countries, the participants did not distinguish between people based on their regions within India. Respondent MT4, who is originally from the UK, noted,
‘I am comfortable as long as someone speaks English properly. I have good relationships with people who keep the interactions limited to English. It does not matter where they are from’.
In multinational teams, there are many visible artifacts – such as easily identifiable nationality, race, or ethnic accents – that make it relatively easy to observe social categorizations. However, in domestic teams, nationality is not a differentiator because all team members hail from the same country, resulting in a greater emphasis on dialect-based categorizations for creating social categorizations and faultlines. Prior research notes that the presence of faultlines can negatively impact group processes, such as by reducing cohesion and creating conflicts, which can further reduce overall group and organizational performance (Choi & Sy, Reference Choi and Sy2010; Lau & Murnighan, Reference Lau and Murnighan2005; Thatcher & Patel, Reference Thatcher and Patel2012). Therefore, organizations need to consider the impact of both language and dialect while developing language management strategies (Tang, Zheng, & Chen, Reference Tang, Zheng and Chen2017). Kulkarni (Reference Kulkarni2015) studied the influence of linguistic diversity and identities on routine activities in organizations in India and found that language is often used as a tool to define social boundaries; however, the author only followed the inter-lingual approach to language conceptualization (Tenzer et al., Reference Tenzer, Pudelko and Harzing2014). The present study's findings highlight the importance of both inter-lingual and intra-lingual conceptualization of language by incorporating both national-level languages and regional dialects. There is a key gap in the current literature on language diversity, with scholarly work assuming the presence of only a few language-based faultlines in multilingual teams, whereas our results suggest that a larger number of faultlines based on both native language and regional dialect differences are active (Meyer & Glenz, Reference Meyer and Glenz2013; Thatcher & Patel, Reference Thatcher and Patel2012) in NPTs. Certain regional dialects carry a negative connotation and signal a status divide between the speakers (e.g., Khazzoom, Reference Khazzoom2003), and such a division could deepen the existing faultlines created by national language differences. This rift could be hazardous for the long-term success of teams not only because it is considered trivial by current scholarly work but also because most team managers are unaware of its existence. Furthermore, prior studies have criticized the research on faultlines for its limited focus on inter- and intra-subgroup processes and for discounting the role of management and leadership (e.g., Meyer, Shemla, Li, & Wegge, Reference Meyer, Shemla, Li and Wegge2015). Therefore, the present study contributes to the literature by exploring the role of multilingual boundary spanners. This is discussed in the next section.
Effectiveness of Multilingual Managers as Boundary Spanners: National Team Managers Versus Multinational Team Members
Boundary spanners can be defined as those ‘individuals who are perceived by other members of both their own ingroup and/or relevant outgroups to engage in and facilitate significant interactions between the two groups’ (Barner-Rasmussen, Ehrnrooth, Koveshnikov, & Mäkelä, Reference Barner-Rasmussen, Ehrnrooth, Koveshnikov and Mäkelä2014: 887). Essentially, boundary spanners possess a particular set of characteristics, abilities, and skills that enables them to associate and communicate with multiple (linguistic) groups/subgroups and act as gatekeepers of information transfers between ingroup and outgroup members, both within teams and across organizational units/boundaries (Kane & Levina, Reference Kane and Levina2017). Overall, 28 participants (15 from multinational and 13 from national teams) emphasized the importance of multilingual managers in facilitating boundary spanning activities in multilingual teams and reducing the faultlines that frequently emerge in such teams; however, our analysis shows that multilingual managers were able to perform boundary spanning roles effectively only in MPTs. During observations of NPTs, the team members frequently shifted to their native/regional language while sharing information with their colleagues, which created confusion and frustration among those who did not understand that language. We noted that several NPT managers were oblivious to the conflicts caused by such behaviors. For instance, during one of the observations of NPT, two of the five team members switched to Hindi while explaining tasks to each other, and the manager did not pay attention to this switch. A, a non-native Hindi speaker, was noticeably angry as she was continuously shaking her head, but the manager did not notice this. We probed the team manager regarding the above observation during an interview, and he said,
‘There are absolutely no problems in our team. Everyone gets along and most work-related tasks are discussed in English. If anyone had problems, I am sure they would let myself or the management know’.
When probed further on the details of phrase ‘most work-related tasks’, the interviewee said, ‘by most, I mean 70–75%’. Another manager (interviewee NT17) stated the following:
‘Look, we have this rule that everyone has to speak in English here. And they can always ask me questions and clarifications in Hindi; Hindi is our national language, and everyone understands it. I have never seen any problems—language is not a big deal’.
Interestingly, India does not have a national language. The interviews led to the inference that NPT managers were aware of the diversity of their team members but not the resultant divisions. In the abovementioned instance, the interviewee spoke with an urban dialect of Hindi, and on probing further, we found that the manager had previously worked in a small organization where Hindi was the only spoken language, with no diversity, and thus had not experienced such divisions.
Interestingly, managers from MPTs were aware of language-based faultlines and actively worked to resolve the resultant problems, thereby performing boundary spanning activities effectively. During an observation of an MPT, the manager was actively engaging with everyone in the team and reminding everyone to be respectful of the diversity and differences. When K tried to initiate conversations in Hindi, the manager immediately called her out and asked her to speak English. He also reiterated that a constant back and forth between native languages and English could cause communication problems and delay meaningful discussions. The direct observations revealed that the multilingual managers in the MPTs actively engaged with their multilingual team members and talked to them about the role of language in accomplishing projects on time.
During an interview, interviewee MT6, who is a team leader, noted,
‘I have lived in multiple countries as an expatriate; thus, I am aware of the issues caused by differences in cultures and languages. I keep them in mind while working with my team here’.
Interviewee MT9 mentioned the following:
‘I am fortunate that R is our team lead. He obviously knows that communicating in a multitude of languages can create confusion and delays in project deliveries. In my other team, these communication problems and frustrations are very common, but in this one, it's different. R has a good rapport with all of us, speaks both Hindi and English well, and we joke around in our native languages. But he always makes it a point to remind us to speak English during meetings, and he can be quite firm’.
The interviewees also mentioned the importance of organizational culture in creating such awareness. Multinational managers often interact with colleagues from other parts of the world, and are, therefore, required to undergo sensitivity training to ensure their knowledge of the problems associated with managing a diverse workforce. Several interviewees also highlighted that acculturation activities and social events initiated by team leaders/managers during initial meetings helped build trust within their multinational teams. During one observation of MPT6, we learned that the team lead frequently planned drinks after work and we heard some of the team members discussing food options and joking about their time at the bar from a previous get-together two weeks prior. We inferred that such get-togethers had helped build comradery in MPT6.
Barner-Rasmussen et al. (Reference Barner-Rasmussen, Ehrnrooth, Koveshnikov and Mäkelä2014) have demonstrated that multilingual individuals can play boundary spanning roles and reduce the faultlines created by language diversity by understanding all the sides of a story and acting as a bridge that helps the members of different linguistic groups and subgroups appreciate each other's perspectives and concerns. The present research replicates these findings in the Indian context. Additionally, a key point that emerged during the interviews was that multilingual managers need to be aware of the faultlines caused by language diversity in order to be effective boundary spanners. This finding is significant because current research on multilingual boundary spanners is sparse, and to the extent of our knowledge, assumes that multilingual individuals are aware of the conflicts caused by language diversity and can, therefore, rely on their bridging skills or use certain accommodation strategies to reduce the associated negative effects (e.g., Aichhorn & Puck, Reference Aichhorn and Puck2017). Furthermore, prior research suggests that a multilingual manager who speaks the languages of both ingroup and outgroup members within a workgroup can cultivate a sense of belonging among the team members (Woo & Giles, Reference Woo and Giles2017), whereas our findings suggest that such assumptions cannot be generalized to heterogeneous country contexts, such as India, as they do not hold true within national teams.
DISCUSSION
The Existence of Faultlines
There is a large body of literature on language diversity and multilingualism that discusses the role of language in creating faultlines in teams and organizations (e.g., Kassis-Henderson, Reference Kassis-Henderson2005). However, the role of dialects in the creation of subgroups has not been discussed in depth in the literature on language in management and organization studies, mainly because language research is still a nascent field and attention has mostly been given to the differences between the national languages of employees working in multinational organizations (Karhunen, Kankaanranta, Louhiala-Salminen, & Piekkari, Reference Karhunen, Kankaanranta, Louhiala-Salminen and Piekkari2018). Our study contributes to this domain by showing that dialects trigger faultlines and create subgroups in NPTs but not in MPTs. Within the Indian context, dialects are closely tied to one's social class and status, which can significantly impact business relationships (Chen et al., Reference Chen, Chittoor and Vissa2015). Thus, team members might be biased toward individuals or groups that are perceived to be inferior or different, and stigmas are attached to people who speak in a certain dialect, even when they are proficient in the corporate language of their organization (i.e., English). Prior research recognizes that a shared dialect can be used as a common ground for establishing personal relationships and that individuals can use their regional dialect to identify their fit with an organization; this can further impact their decision to stay or leave an organization, thus affecting turnover (Gong, Chow, & Ahlstrom, Reference Gong, Chow and Ahlstrom2011).
This study shows that dialects are a strong source of one's identity and are associated with one's status within a team. Based on the present findings as well as arguments from prior research, it is clear that status differences create divisions and trigger faultlines in NPTs, that is, creating a situation that could turn a formerly dormant dialect faultline into an active one (Chrobot-Mason, Ruderman, Weber, & Ernst, Reference Chrobot-Mason, Ruderman, Weber and Ernst2009). Theoretically, there are two prerequisites/assumptions for the formation of strong faultlines: (1) a moderate level of diversity (Lau & Murnighan, Reference Lau and Murnighan1998, Reference Lau and Murnighan2005) and (2) a small number of subgroups divided along a specific number of attributes. Our findings align with the former but contradict the latter. Prior studies have argued that a higher number of subgroups within a team is likely to lead to weaker faultlines (e.g., Nishii & Goncalo, Reference Nishii and Goncalo2008). In fact, Polzer, Crisp, Jarvenpaa, and Kim (Reference Polzer, Crisp, Jarvenpaa and Kim2006) found that groups divided across two geographic locations tend to have the highest levels of conflict; groups divided across three locations have moderate levels of conflict; and groups divided across six locations display the lowest levels of conflict. In the present study, MPTs demonstrated a moderate level of diversity whereas the number of subgroups was higher in NPTs, yet they were able to sustain faultlines and maintain conflict. It is plausible that such strong faultlines were sustained in NPTs because of weak leadership compared with MPTs. We elaborate on this in the following section.
The Role of Multilingual Managers as Boundary Spanners
Scholars in language research have used terminologies such as ‘bridging individuals’ or ‘linguistic intermediaries’ to highlight the ability of multilingual individuals as boundary spanners (e.g., Aichhorn & Puck, Reference Aichhorn and Puck2017; Feely & Harzing, Reference Feely and Harzing2003). However, the contexts that are conducive to honing their bridging activities remain unclear. There seems to be a universal assumption about the success of such persons as it is widely believed that they can neutralize the detrimental effects of language diversity by gaining the trust of their team members (Barner-Rasmussen et al., Reference Barner-Rasmussen, Ehrnrooth, Koveshnikov and Mäkelä2014). However, to bridge language barriers, such persons must have not only the capability but also the willingness to do so (Kane & Levina, Reference Kane and Levina2017). We argue that the willingness will not occur if multilingual managers are not cognizant of the impacts of language diversity and the issues related to subgroup formation and faultlines. Organizational culture and training programs could generate cognizance among employees. For example, in the multinational organizations considered for the present research, sensitivity training helped generate more empathetic responses from MPT managers, whereas no such training was mandated in the national organizations. Consequently, some managers’ ignorance and encouragement of the use of native languages in NPTs triggered the division of individuals into subgroups.
Table 7 compares the MPT and NPT managers’ characteristics. The literature suggests that sharing a common attribute with members of other subgroups may lead to cross-categorization and influence perceived faultline strength, and scholars have analyzed the effects of leadership-based moderators such as leadership role and task structure (e.g., Benet-Martínez, Reference Benet-Martínez, Deaux and Snyder2012; Gratton, Voigt, & Erickson, Reference Gratton, Voigt and Erickson2007). The present study, in contrast, reveals that attributes such as awareness and acknowledgement of differences, which are often considered trivial, can help reduce faultline strength and subgroup formation. Awareness of language differences and the willingness to actively address linguistic faultlines could motivate multilingual managers to employ ice-breaking strategies. In the present study, some managers actively encouraged social contact and dinners to improve the levels of trust among their team and divert attention away from language (and dialect) differences. In teams led by such boundary spanners, the mistrust and conflict between subgroups was replaced with trust stimulated by a positive perception of language diversity.
Table 7. Effectiveness of multilingual managers as boundary spanners in linguistically diverse teams
Theoretical Implications
One of the most significant contributions of this study is the focus on individual-level multilingualism along with team-level language diversity. Individual-level data was collected through interviews and team-level data through observations of team interactions and meetings, and the findings highlight a more nuanced view of individual-level multilingualism. Furthermore, by targeting organizations in India, this study brings attention to the phenomenon of language diversity in an emerging economy and moves beyond the traditional focus of linguistic diversity research on the US and Western Europe. Thus, this study answers the call to study language diversity in emerging economies (Tenzer, Terjesen, & Harzing, Reference Tenzer, Terjesen and Harzing2017). Dialects can enhance the communication and interactions between individuals, enabling the formation of groups (Gong, Chow & Ahlstrom, Reference Gong, Chow and Ahlstrom2011; Du, Reference Du2019). Prior research has also found that dialects play a strong role in emerging countries such as China (e.g., Gong et al., Reference Gong, Chow and Ahlstrom2011). Regional dialects can affect the choice of foreign location during internationalization processes, and dialect differences can have negative effects on intra-national trade (Wang & Ruan, Reference Wang and Ruan2019). Du (Reference Du2019) found that CEO–auditor dialect sharing can create a positive bias and impact the quality of audits. The findings of these studies signify the importance of dialects in managerial decision making in other contexts such as China. However, the impact of dialects on interactions and trust building within an organization's internal environments requires further study. Researchers could test the findings of the present study in the Chinese context to further the understanding of this phenomenon.
The current study further adds to the research on faultlines and team interactions in four ways. First, the interview and observation data show that the ‘us versus them’ dynamic (Hinds et al., Reference Hinds, Neeley and Cramton2014) exists not only in globally distributed teams but also in local teams that are thought to be linguistically homogenous. Language and dialect differences can both affect intragroup communications and cause strained interpersonal relationships. The present study's findings suggest that perceived inequalities related to regional dialects create a need for NPT members to form subgroups that can support the ingroup members. Second, prior research suggests that a higher number of subgroups in a team would lead to weaker faultlines (e.g., Lau & Murnighan, Reference Lau and Murnighan1998), whereas the current study shows that faultline strength is not directly associated with the subgroup numbers. NPTs have a larger number of subgroups than MPTs because of the activation of both language and dialect-related faultlines; however, there is no evidence of weaker faultlines or higher levels of inter-subgroup trust in NPTs. Third, research on faultline theory suggests that group history can influence the strength and number of faultlines in workgroups as the salience of various attributes can change over time (Meister, Thatcher, Park, & Maltarich, Reference Meister, Thatcher, Park and Maltarich2020). However, project teams are formed for a relatively short duration, and they are dismissed once the project is completed, making it difficult to analyze the temporal element in project teams. One way to manage the diversity in such teams is by following the suggestions of linguists Woo and Giles (Reference Woo and Giles2017), who noted the importance of speech accommodation training. Organizations could offer training programs on intergroup communication practices, thereby reducing anxiety and frustration among employees and facilitating quality contact between multilingual individuals and managers. Finally, individuals who speak multiple languages can act as linguistic boundary spanners to facilitate smooth working between the subgroups of a team. We know that leaders of teams tend to have higher power and influence over their team members and, therefore, play a significant role of activation or deactivation of faultlines (Meyer et al., Reference Meyer, Shemla, Li and Wegge2015), and research has suggested that multilingual boundary spanning individuals can be valuable employees in organizations due to their skills and abilities in translation and accommodation (Aichhorn & Puck, Reference Aichhorn and Puck2017). Similarly, research on faultline theory indicates that shared attributes can help leaders reduce subgroup bias and enhance levels of intergroup trust (Benet-Martínez, Reference Benet-Martínez, Deaux and Snyder2012; Meyer et al., Reference Meyer, Shemla, Li and Wegge2015). However, based on the present study's results, we argue that a cross-categorization of attributes with team-members is not sufficient and that multilingual managers need to (1) understand the diversity and differences that lead to the creation of faultlines and (2) actively employ team diversity management strategies such as ice-breaking and acculturation to take employees’ attention away from language and dialect differences, thereby deterring the emergence of faultlines. This finding contributes to the work of van der Kamp, Tjemkes, and Jehn (Reference van der Kamp, Tjemkes, Jehn, Wildman and Griffith2015), who introduced the concept of faultline deactivation, and Schölmerich, Schermuly, and Deller (Reference Schölmerich, Schermuly and Deller2016), who argued that leaders’ diversity beliefs can moderate the activation and functioning of faultlines. As noted in our findings, multilingual boundary spanners have the potential to increase team efficiency by improving communications, promoting intragroup relationships, building rapport, and reducing faultlines. Nevertheless, the role of boundary spanning is not limited to managers; bilingual or multilingual individuals within teams can also act as boundary spanners. This warrants the attention of language diversity researchers.
The present research also highlights the role of language-based diversity in an understudied context. Indian society is highly multilingual, with over 121 languages spoken across the country (Registrar General & Census Commissioner, India, 2011). Language is intimately connected to regional culture, traditions, and ethnic identities. The different languages can be further divided into numerous dialects, both rural and urban, with class and status distinctions (Chen et al., Reference Chen, Chittoor and Vissa2015; Finegan & Rickford, Reference Finegan and Rickford2004). Our study shows that distinct regional languages can create communication barriers between subgroups by creating faultlines and inhibiting interpersonal relationships, and, contrary to popular belief, the use of English as a corporate language does not lower these barriers. Our findings in the Indian context not only highlight the influence of social status and class of dialect speakers but also provide a foundation for future studies on language diversity in other multilingual countries, such as the USA, where one's dialect is often associated with their class (Finegan & Rickford, Reference Finegan and Rickford2004). It is plausible that individuals make value judgment of others based on their dialect and associated social class, thereby creating faultlines that are invisible to the management because they speak the corporate language fluently.
A practical implication of our findings is the acknowledgement of linguistic differences within one-nation teams that outwardly seem linguistically homogeneous. Even when all teammates have a strong grasp of the common language, managers must address the potential faultlines caused by various native tongues and/or distinct regional dialects, which may potentially lead to mistrust and bias and hinder efficient team functioning. It is imperative for managers to understand and acknowledge the distinction and relationship between language diversity and language/dialect-based faultlines (Lau & Murnighan, Reference Lau and Murnighan2005; Thatcher & Patel, Reference Thatcher and Patel2012) to address the conflicts and tensions brewing among the various subgroups.
Limitations and Future Research
Although it significantly contributes to the field of language diversity and multilingualism, this study is by no means free of limitations. First, due to restrictions related to data access, the study sample consisted of teams from two different organizations, one domestic and one multinational, and macro-organizational factors, such as organizational culture, that influence the emergence of faultlines in national and MPTs could not be controlled. To address this gap, future research could analyze the intragroup dynamics of the two types of teams within the same organization. Second, our study did not take into consideration the role of other demographic attributes of the team members, such as gender, age, and experience, while exploring the role of language diversity. Further research could study the intersection between these attributes and one's native language. Third, we focused on only one country, and it is possible that the relationship between diversity, faultlines, and subgroup formation is more complex or different in other emerging economies, such as China, where regional dialects could strongly affect interpersonal relationships. Therefore, future research can test the present findings in other country contexts. Finally, we did not examine longitudinal effects of language diversity. The interviewees shared their experiences with functioning in multilingual project teams, but little was learned about the long-term effects of such diversity.
Despite its limitations, the present work provides a starting point for the further examination of numerous language-related issues in organizations. First, the study findings stem from an analysis of interviews with individuals who work in organizations that have a formal corporate language policy of using English as the lingua franca; future research can test the models on organizations that do not have a formal corporate language policy but, rather, a de-facto lingua franca that is spoken by majority of employees. Furthermore, we call for more research on multilingualism and language-based faultlines. Vigier and Spencer-Oatey (Reference Vigier and Spencer-Oatey2017) showed that the number and strength of cultural faultlines in organizations determines the mechanisms of rule development and implementation and, in turn, team attitude and atmosphere. Therefore, we suppose analogous mechanisms might exist when language diversity is analyzed. Research on the same could potentially be conducted in other multilingual countries such as Canada, China, or Switzerland, where fewer languages are widely spoken, leading to fewer faultlines. Furthermore, in this study, we controlled for certain respondent-related macro factors, such as individual's education background by only interviewing individuals who received their MBAs from top-tier schools in India or their respective countries. However, it is possible that the process of subgroup formation is different in teams wherein individuals are not only from different socio-economic status and speak in various regional dialects but also have different educational backgrounds. Therefore, future research could analyze the interaction, if any, between the dialects and educational backgrounds of the respondents.
CONCLUSION
Language diversity matters because language is a significant element of one's social identity; individuals tend to identify with their native language and give social significance to the groups they associate with. Such associations and beliefs related to social categorizations can have significant effects on the outcomes of work groups and organizations. This is especially noteworthy as workforces become more diverse with each passing year due to factors such as mass migration and the increased representation of minorities in the workplace. However, the contexts in which such diversity research is conducted are not so diverse; most scholarly research has focused on the global North or Western cultures. The present study's data showed that language diversity can be as much of an issue within a domestic or a national project team as it is in multinational teams and that multilingual managers in national teams are less aware of the detrimental effects of the resultant faultline and subgroup formations. Furthermore, contrary to current scholarly assumptions, not all multilingual managers are successful as boundary spanners in multilingual teams, and leaders’ awareness and willingness to reduce tensions and divisions and improve team camaraderie can help them navigate language faultlines and address the associated conflicts. Researchers can engage with the results and limitations of this study to further the understanding of the causes and consequences of language diversity in organizations across diverse national contexts.
Open access
