Language in International Business: A Review and Agenda for Future Research

A fast growing number of studies demonstrates that language diversity influences almost all management decisions in modern multinational corporations. Whereas no doubt remains about the practical importance of language, the empirical investigation and theoretical conceptualization of its complex and multifaceted effects still presents a substantial challenge. To summarize and evaluate the current state of the literature in a coherent picture informing future research, we systematically review 264 articles on language in international business. We scrutinize the geographic distributions of data, evaluate the field’s achievements to date in terms of theories and methodologies, and summarize core findings by individual, group, firm, and country levels of analysis. For each of these dimensions, we then put forward a future research agenda. We encourage scholars to transcend disciplinary boundaries and to draw on, integrate, and test a variety of theories from disciplines such as psychology, linguistics, and neuroscience to gain a more profound understanding of language in international business. We advocate more multi-level studies and cross-national research collaborations and suggest greater attention to potential new data sources and means of analysis.


Introduction
Exactly 30 years ago, a review of nearly 500 English-language management texts (Holden 1987) demonstrated that only very few authors considered language, and those who did quickly brushed over the topic without considering its complexity. Much has changed since that time. Today's international business scholars treat language as an issue at the heart of their subject area (Brannen et al. 2014;Mughan 2015), as language determines organizational communication, constitutes the foundation of knowledge creation  and is considered essential for the construction of organizational realities (Piekkari and Tietze 2011). Highlighting the theoretical and practical relevance of language in international business, Piekkari et al. (2014, p. 1) stated: ''To say that language permeates every facet of international business would meet with little argument, especially from those involved in global activities in any form''.
As noted by Brannen et al. (2014), scholars approach language issues in business from many different angles. Among the diverse conceptualizations of language they use, three facets feature most prominently: national languages spoken in multinational corporations (MNCs), officially mandated corporate languages, and English as the language of global business. Many scholars focus on the national languages of corporate headquarters and globally dispersed subsidiaries, which are spoken alongside each other in MNCs (Angouri 2014), mingle in employees' speech (Janssens and Steyaert 2014), and thus form ''linguascapes'' (Steyaert et al. 2011), which are constantly subject to negotiation. Others deal with the notion of a common corporate language, mostly defined as an ''administrative managerial tool'' (Latukha et al. 2016) that acts as a facilitator or barrier to internal and external communication . Beyond the frequent, but simplistic understanding of top management mandating that a specific national tongue (mostly English) must always be chosen (Berthoud et al. 2015), scholars have started to recognize the complexities of common corporate languages, which ''often reflect the industry context and the national language environment in the country of origin'' (Brannen et al. 2014, p. 497;Brannen and Doz 2012). The role of English constitutes the third facet of language frequently studied in business. Depending on their disciplinary socialization, international business scholars varyingly conceptualize English as a hegemonic force (Tietze and Dick 2013), which recreates postcolonial power structures (Boussebaa et al. 2014) or as a more neutral communicative tool in the form of business English as a lingua franca 1 (BELF) (Kankaanranta and Planken 2010). Yet other scholars investigate the interplay 1 International business scholars typically conceptualize lingua franca as ''a common language different from the parties' native language, very often English'' (Cuypers et al. 2015, p. 430). Whereas some researchers see a hegemony of English native speakers in a world focused on English (Tietze and Dick 2013), others believe that this hegemony ''is now gradually being replaced, particularly in business contexts, by the use of a neutral form of BELF that neither originates in native speaker models nor is owned or influenced by them'' (Nickerson 2015, p. 392). Their idea is in tune with Brannen et al.'s (2014, p. 496) statement that ''lingua franca was originally conceived as a neutral form of communication without cultural or political bias''. Reinforcing that view, Berthoud et al. (2015, p. 7) emphasize that lingua franca use need not even be monolingual, but may be a ''hybrid code'' drawing on speakers' multilingual repertoires. between national and corporate languages and English (Kuznetsov and Kuznetsova 2014). Language-related research in economics developed largely separate from those bodies of literature. This economic stream analyzes semantic structures of national languages such as future-time reference (Chen 2013) or gender marking (Hicks et al. 2015) and investigates their impact of economic behavior at the country level. Cross-national economic research mostly relies on linguistic distance, i.e. a measure of how difficult speakers of one language find it to learn the other (Hutchinson 2005), or as a predictor of trade patterns and various other outcomes (Sauter 2012;Melitz and Toubal 2014).
But has the proliferation of publications studying international business activities under a language lens made scholars more sophisticated in their conceptualization of language? We review the fast-growing literature on language diversity in international business in order to consolidate and evaluate its achievements to date, identify remaining desiderata, and suggest a research agenda for the years to come. Based on our reading of 264 journal articles on language issues in international business contexts, we show that different streams within the field have developed separately. Whilst economic approaches strive to make the features of specific languages measurable, business studies are divided in their conceptualizations of languages as static and discrete entities versus hybrid, fluid, and situational codes. Whereas some business studies perpetuate the notion of language as an easily accessible instrument or management tool, an increasing number of publications on multilingual business phenomena draws on translation studies, socio-and psycholinguistics to capture language as a multifaceted, complex, and dynamic concept. Revealing patterns in theory, methodology, data, and content within the extant literature, we conclude that international business as a subject area has substantially broadened and deepened its coverage of language issues, but would still benefit from drawing more extensively on language-focused disciplines such as linguistics, in particular applied linguistics, sociolinguistics, and psycholinguistics, as well as translation and communication studies. Only by integrating the concepts and methods from different academic disciplines can the complexity of linguistic influences on international business be adequately understood. Building on this finding, our review aims to provide an inspiring and actionable agenda for future research.
We will start by describing our systematic review methodology and show how we identified, selected, and reviewed relevant publications. Subsequently, we will develop an organizing framework through which we summarize the current status of research in language in international business by research setting, theories, methodologies, and key findings at individual, group, firm, and country levels. On this basis, the second half of our review develops a future research agenda.
2 Methodology: Systematic Literature Review

Data Collection and Analysis
We followed the systematic literature review methodology (Tranfield et al. 2003) using Business Source Premier, JSTOR, and ProQuest to identify language-related research in international business. Following Cantwell and Brannen's (2011) We initially identified 390 articles, of which 264 met the criteria for inclusion outlined above. The ''Appendix'' lists the final set of references, which we coded in an Excel spreadsheet according to a broad range of criteria including, among others, theoretical approaches, levels of analysis, empirical methods (if applicable), major findings, and future research suggestions. Having jointly coded the first ten papers, we noticed very large inter-coder agreement, so we proceeded to code independently with regular crosschecking.

Overview of Our Sample
Since the earliest articles were published in 1987, language-related research in international business has grown exponentially. There were only 14 articles published from 1987-1999, 73 published 2000-2009, and 177 published 2010present. We visualize this development in Fig. 1. Whereas prior studies frequently emphasized the ''infancy'' of language-related international business research (see e.g., Feely and Harzing 2002;Neeley 2013), there has been a dramatic increase in research output over the past decade.
For each of the 264 publications in our sample, we verified its number of Google Scholar citations. 3 The field's slow start is reflected in the low number of citations most of the earliest publications have garnered to date (Holden 1987: 29;San Antonio 1987: 71;Fixman 1990: 129;Swift 1991: 81;Tsalikis et al. 1992: 37; Sims

Current Status of Language Research in International Business
As most of the journal articles in our review follow a conventional sequence of presentation-i.e. background, theory, methods, and research findings-we organize our literature overview into similar categories. Our structure also mirrors the choices of other recent systematic reviews (see e.g., Aguinis and Glavas 2012;Terjesen et al. 2016). We will start by reporting on the geographic settings of language-related international business research reflecting the fact that most papers open with presenting their studies' background. Based on a review of theoretical framework sections, we will then discuss key theories used in the field. Drawing on the methods sections of our sample papers, we will go on to discuss frequently used methods and data sources in our focal field. Finally, we will mirror the results sections of empirical papers by providing an overview of their findings. As it is difficult to cluster the highly fragmented content around ''big'' research questions, we will build on Brannen et al.'s (2014) characterization of language as a ''multilevel construct'' and categorize findings according to the corresponding levels of analysis. Table 1 summarizes the aspects covered in our review.

Research Setting
Much of the early research originates from outside the US. Although ten out of seventeen authors of the 14 articles published between 1987 and 1999 were from US business schools, this is only due to their large author teams. Language scholars from Finnish and Norwegian institutions published more prolifically, producing several papers per author. The United Kingdom is the most frequently studied country in these early works. This strong representation of European countries is rather atypical for the otherwise very US-centered international business research. Harzing and Feely (2008, p. 51) explain this pattern with the fact that ''American researchers (…), because of the dominance of the English language, have a reduced perception of the importance of language''. Although US scholars have caught on to the topic in recent years, author origins and target regions for empirical studies on language are still more diverse than other fields within the broader subject area of international business. In the overall sample ranging from 1987-2016, the number of countries examined ranges from 1 to 224, with a mean of 8, a median of 2, and a mode of 1. The most common countries examined to date are the UK, USA, Finland, Germany, Japan, and Sweden. Compared to China, there is a paucity of language research related to the other BRIC and emerging economies.

Theory
Having discussed the development in the geographic centers of language-sensitive international business research, we now turn to the theoretical background of publications. Depending on their disciplinary socialization, international business scholars with an interest in language draw on a variety of theories from organizational behavior, economics, and strategy. Organizational behavior and cross-cultural management scholars approach language with theories on culture (e.g., Harzing et al. 2002;Kassis Henderson 2005), social identity (e.g., Groot 2012; Reiche et al. 2015), power relations (e.g., Neeley 2013;Hinds et al. 2014), emotions (e.g., Neeley et al. 2012;Tenzer and Pudelko 2015), and a range of other phenomena. Those with a background in economics apply, among others, the gravity model of trade (e.g., Melitz and Toubal 2014;Sauter 2012), transaction cost economics (e.g., Selmier and Oh 2013), or linguistic relativity (e.g., Chen 2013). Strategy researchers focus predominantly on resource-based explanations for the internationalization of MNCs and new ventures (e.g., Fernandez-Ortiz and Lombardo 2009; Hurmerinta et al. 2015). Despite their common goal-to explain the impact of language on international business and economic activities-these bodies of literature have hitherto only spoken to each other to a very limited extent. To broaden international business scholars' view beyond their respective home disciplines, we will now summarize the key contributions of the most utilized theories in language-related international business research in order of their frequency of use: culture, the gravity model, internationalization, linguistic relativity, and social identity.

Culture
Ever since language first emerged as a topic in international business, the relationship between language and culture has challenged international business researchers. No one has doubted the tight link between the two concepts, but their relationship has been conceptualized disparately. Early international business research often conflated language with culture (Kassis Henderson 2005) or implied that cultural modeling based on value systems substituted for specific language studies, a stance that may have delayed the recognition of language as a separate construct (Brannen and Mughan 2016). Gradually, however, the mutual relationship between language and culture came to the forefront, with some authors considering language to be ''inherent in a specific culture and also an embodiment of it'' (Welch and Welch 2008, p. 341) and others positioning it at the center of culture .
In recent years, management scholars (e.g., Harzing et al. 2005;Akkermans et al. 2010) have applied the psychological concept of cultural accommodation to capture the link between language and culture in a business context. Showing that language priming induces individuals to adapt their thoughts and behaviors to the cultural norms associated with the language they are currently speaking, those authors demonstrate that language use activates what the neuroscience literature identifies as the neural pathways resulting from engagement in cultural practices. Along these lines, Dutch students were found to behave more competitively when playing a price-setting game in English compared to their native language (Akkermans et al. 2010), especially if they had spent time in an Anglophone culture.
International business researchers taking inspiration from sociolinguistics have approached the culture-specific elements of language use from a different angle. Building on cross-cultural pragmatics, they analyze the culture-specific rhetoric patterns in speech acts such as requesting, refusing, and thanking to understand how speakers of different cultures use language in interactive contexts to create specific meaning (Kassis Henderson 2005). As this implied meaning was found to create frequent misunderstandings in global business communication (Chen et al. 2006), an increasing number of scholars recognized the ''transformative power of translation'' (Brannen et al. 2014, p. 501). Analyzing the difficulty of translating Western management terms such as ''knowledge sharing'' into Russian, Holden and Michailova (2014) caution against simplistic attempts to replace terms from one tongue with those of another. Following their call, international business researchers have begun to understand translation as a process of interaction across cultures (Brannen and Mughan 2016;Chidlow et al. 2014), where meaning may be found in the space between cultures.

Gravity Model of Trade
The second most frequently applied theory is based on the gravity model of trade, which correlates the direction and size of trade between political entities with the size and geographic distance between these trading partners. Largely separated from other streams covered in this review, economists started in the early 2000s to extend this work to consider the role of language variation as a barrier to bilateral trade. Initial work relied on binary variables to indicate whether or not countries share an official language, whereas later studies have considered the distance between language families, the role of minority languages (Sauter 2012), and differences in translation or direct communication (Melitz 2008;Melitz and Toubal 2014). Controlling for the stock of immigrants and other factors, a consistent finding around the world is that greater distance between/amongst languages is associated with less trade across these nations. As summarized by Sauter (2012), countries with a common language trade 1.5 times more and the language barrier amounts to a tax equivalent of about 7%, while Egger and Lassman's (2015) meta-analysis suggests that a common language increases trade flows by 44%. Related research demonstrates that language is a barrier to trade across Canadian provinces (Sauter 2012), 36 countries (Hutchinson 2005), and a 19 language, 195-country dataset (Melitz and Toubal 2014).

Linguistic Influences on MNC and New Venture Internationalization
Strategy scholars concerned with internationalization theories composed the third most prominent theoretical approach to language in business. Inspired by Johanson and Vahlne's (1977) seminal model of firms' internationalization process, which positions language diversity as an important element of psychic distance, scholars have examined how corporate decision makers' foreign language skills influence their international opportunity recognition (Hurmerinta et al. 2015). As a consequence, small to medium-sized enterprises (SMEs)' directors' knowledge of a foreign language and international experience is significantly and positively correlated to SMEs' international diversification strategies (Fernandez-Ortiz and Lombardo 2009). Compared to native-born, monolingual Americans, immigrant and transnational entrepreneurs are more likely to start export-oriented businesses; however, language does not affect global imports (Light et al. 2002). More than a decade later, a study of immigrant entrepreneurs reports similar findings: Canada's French and Allophone speakers are more likely to start ventures that export to global markets (Sui et al. 2015).

Linguistic Relativity
Linguistic relativity theory, the fourth most frequently applied approach in our sample, rests on the idea that different languages shape different worlds, a premise attributed to linguists and anthropologists (Sapir 1921(Sapir , 1951Whorf 1956; von Humboldt 1836) who examine how different languages' semantic structures shape human cognition. After being virtually discarded in the 1970s, this theory has recently attracted renewed interest from researchers. An emerging body of research examines the Sapir/Whorf hypothesis in relation to consumer behavior (Puntoni et al. 2009) and economic activity (Chen 2013). For example, the presence of gender-differentiated pronouns is correlated with attitudes towards gender-based discrimination. Hicks et al.'s (2015) study of US immigrants show that households where members come from countries with gender-intensive languages are more likely to allocate household tasks by sex, whereas countries with a lack of gender markers in their language have higher female board representation (Santacreu-Vasut et al. 2014). Malul et al. (2016) demonstrate that the linguistic gender marking gap between an MNC's home and host country influences the success of female expatriates. Chen (2013) examines the linguistic structure of future tense, finding that native speakers of languages that grammatically associate the future and the present (e.g., French, English, Czech) are more likely than weak future language speakers to display future-oriented behavior such as greater savings, more wealth at retirement, less smoking, greater safe sex, and less obesity.

Language-Based Social Identity Formation
The fifth most frequently applied theoretical approach to language in international business draws on early research in organizational psychology. Leveraging social identity and self-categorization theories (Tajfel and Turner 1979), international organizational behavior scholars explain why language diversity can separate employees into groups based on a shared language and thus give rise to language boundaries in MNCs (Born and Peltokorpi 2010). As the use of specific language nuances signals a sense of familiarity (Chong et al. 2010), language-based clusters form within the MNC based on homophily, a tendency to interact with similar others (Mäkelä et al. 2007). These clusters unite employees sharing a common mother tongue who can easily create and maintain interpersonal relationships and exchange knowledge (Fredriksson et al. 2006). In contrast, language differences separate expatriates as out-group members from host country nationals, thus diminishing social support, interactions, and network building (Zhang and Peltokorpi 2015). Consequently, language emerges as a key factor for selfcategorization and the categorization of others (Feely and Harzing 2003). These language-based intergroup boundaries can have detrimental effects of decreasing MNCs' organizational identity, knowledge transfer, control, coordination, and communication (Born and Peltokorpi 2010).

Methodology
In the following, we will examine which methods were most frequently used for studying language in international business. We will also review the data sources empirical studies have been drawing on.

Methods
The heritage of the pioneering qualitative work by Piekkari (Marschan et al. 1997;Marschan-Piekkari et al. 1999a, b) is still reflected today in a large proportion of qualitative case-study research-certainly much higher than in other fields within international business. Figure 1 provides the breakdown of qualitative, quantitative, and theoretical/conceptual research over time. In our sample, 127 (47.9%) studies are qualitative in nature, while 113 (42.6%) use quantitative methods, 11 (4.2%) use both methodologies, and 13 (4.9%) are theoretical/conceptual. Among the qualitative studies, researchers employ a variety of methods, from organizational ethnography to grounded theory and discourse analysis. Of the quantitative articles, a large majority use purely descriptive statistics. The next most utilized methods are regression analyses.

Data Sources
Scholars have utilized a variety of data sources, from interviews and observations to survey data, from multilingual managers from a single or multiple countries to multi-country/language studies. Of the studies that provide a time frame for data collection, the majority are cross-sectional. The few longitudinal studies are a relatively recent development.
Of the studies with data, surveys and questionnaires represent the most common data source, followed by interviews and interviews with supplemental data. Other popular data collection means include online search and other options such as firms' e-mails, internal documents, and website content, institutional archives and databases, in-house/laboratory experiments/tests, press/census, and participant observations.

Findings by Levels of Analysis
Over the last three decades, the number of topics covered by language-related research in international business has proliferated along with the fast growth in publications. Following Brannen et al.'s (2014) portrayal of language as a ''multilevel construct'', we organize our review of research findings according to their levels of analysis. The most common level in our sample is individual, followed by firm, and then group and country levels. Approximately 17% of studies include multiple levels of analysis, most commonly the combination of individual and firm levels. Below we summarize major research topics at each level of analysis. Table 2 lists some representative recent publications in these categories and provides some examples of theories, phenomena, and research questions, which language-sensitive international business studies have addressed at each level.

Individual Level Perspectives
Language research at the individual level incorporates multiple perspectives and covers a variety of topics. From an economic perspective, Gary Becker's (1992) notion of the importance of human capital is apparent in the large body of research that consistently indicates that one's language abilities (when one is operating in a 'host'/non-native environment, i.e., as an immigrant) condition access to information and labor market opportunities. In a multinational firm setting, language skills influence the cross-cultural adjustment of expatriates (Selmer and Lauring 2015;Zhang and Peltokorpi 2015) and individual employees' career mobility (Itani et al. 2015;Latukha et al. 2016). Moreover, multilingual employees find it easier to create social capital (Barner-Rasmussen et al. 2014), enabling them to function as boundary spanners, language nodes, and information gatekeepers (Heikkilä and Smale 2011;Peltokorpi and Vaara 2012). Furthermore, individuals who are bilingual and bicultural in their ability to navigate institutional environments are more likely to pursue entrepreneurial activity, often as transnational entrepreneurs (Light et al. 2002). A small but growing body of research (e.g., Luna et al. 2008;Brown and Sachdev 2009;Alvarez et al. 2017) examines the impact of bilingualism Do foreign language settings foster or diminish individual free-riding behaviors? (Urbig et al. 2016) How does the differential use of task indicative cues in a shared language influence the performance of multinational work teams? (Butler 2011) How does organization members' fluency in the adopted lingua franca influence their reactions to a cross-border merger? (Kroon et al. 2015) How does linguistic diversity within the home and host country influence the equity stake taken by acquirers? (Dow et al. 2016) Which contextual factors influence corporate language-based communication avoidance? (Lauring and Klitmøller 2015) How does expatriates' willingness to learn the host country language influence their relationship with host country employees (Zhang and Harzing 2016)? Does language use influence perceived team conflicts and team atmosphere? (Voss et al. 2014) How can MNCs manage language as a strategic resource ( (2015) on individual outcomes, such as self-sufficiency or perceived vitality. Other research studies the adverse impact of the lack of native English-language skills. For example, Hosoda et al. (2012) examined the discrimination against potential job applicants with Spanish-accented English in hiring and promotion processes. Scholars have also established that a lack of understanding due to foreign language use creates uncertainty with resulting feelings of uncertainty, anxiety, and tension (Neeley et al. 2012;Tenzer and Pudelko 2015) which can spill over to other contexts and lead to general feelings of negativity and a fear of exploitation. Again others look at the impact of language choice in bilingual advertising on individual consumers (e.g., Ying-Ching and Wang 2016; Kubat and Swaminathan 2015).

Group Level Perspectives
As early research viewed language exclusively as a skill residing in individuals, international business studies only gradually recognized it as ''a constitutive, collective force contained in the MNC'' (Brannen et al. 2014, p. 499) and started to explore language at higher levels of analysis. At the group level, existing work investigates a diversity of settings from co-located teams (Tenzer et al. 2014) to global virtual work groups  and corporate boards . For example, recent work examines the impact of linguistic diversity on team processes and emergent states such as group cohesiveness (Lauring and Selmer 2010), social categorization , power relations (Tenzer and Pudelko 2017) (Klitmøller and Lauring 2013;Klitmøller et al. 2015) and language-related status evaluations (Butler 2011;Neeley 2013). At the corporate board level, language diversity can lead to impoverished and silenced discussions, particularly if employee representatives lack sufficient proficiency in the board's working language .

Firm Level Perspectives
Firm-level research focuses on language competencies, policies, and practices within MNCs (e.g., Harzing and Pudelko 2013), including HQ-subsidiary relationships (Harzing et al. 2011;Harzing and Pudelko 2014) and mergers, acquisitions, and alliances (Joshi and Lahiri 2014;Cuypers et al. 2015). These studies investigate the impact of linguistic diversity on social identity formation (e.g. Mäkelä et al. 2007;Harzing and Feely 2008), subgroup dynamics (Steyaert et al. 2011), and knowledge sharing . The latter study, for instance, finds that the positive relationship between a shared language and knowledge transfer is mediated by how much subsidiary managers share the goals and visions of HQ. Cuypers et al. (2015) find that linguistic distance is linked negatively and lingua franca proficiency is linked positively with higher stakes in acquisition targets.

Country Level Perspectives
Research at the country level highlights the role of language as an institution that shapes behavior and activity. Intra-country research frequently characterizes countries according to their official language(s); some detailed work investigates the languages a country's citizens speak, for example using the World Values Survey data (WVS 2014). Luiz (2015), for example, draws on the South African context to develop a new measure, ethno-linguistic fractionalization, of a nation's ethnic and/or linguistic diversity. Much of the research carried out in Switzerland (e.g., Steyaert et al. 2011, Berthoud et al. 2013 and Finland (e.g., Vaara et al. 2005; Barner-Rasmussen and Aarnio 2011) explicitly engages with language dynamics in countries with more than one official language. Inter-country research examines the linguistic distance between national languages or between English as a global language and specific countries' official languages to determine the language costs of economic transactions  and their effect on bilateral trade (Hutchinson 2005) or the choice of target countries for foreign direct investment (Lien et al. 2012). A recent paper develops a measure to capture the aggregate impact of common native language, common spoken language, common official language, and linguistic proximity on bilateral trade, disentangles ease of communication from other trade enabling factors and additionally considers translators and interpreters' roles (Melitz and Toubal 2014).

Multiple Level Perspectives
Whereas the large majority of language-related studies in international business focus on a single level of analysis, a growing body of research recognizes that language ''is a multi-level issue'' (Piekkari et al. 2014, p. 244;Brannen et al. 2014). Equally split between qualitative and quantitative approaches, this stream investigates the impact of individual experiences with language on group dynamics and firm performance or explores influences of language issues at higher levels on individual cognitions, emotions, and behavior. However, only a few studies have implemented genuinely multi-level designs, which integrate data collection and analysis at several levels with theory building or testing spanning the same levels (Hitt et al. 2007). Studying bottom-up influences, Hinds et al. (2014) apply an exemplary multilevel approach, which combines individual-level interviews with team-level observation in multinational work groups. The authors analyze these datasets separately and on this basis demonstrate that asymmetries in individual team members' language proficiency levels lead to subgroup formation and teamlevel power contests. Research spanning the individual and firm levels explores the implications of individual and corporate translation behavior on an organization's absorptive capacity (Piekkari et al. 2013) or shows how the language capital of individual employees interacts with organizational resources to shape a corporations language operative capacity (Welch and Welch 2015). Exploring top-down effects, Boussebaa et al. (2014) demonstrate how corporate mandates to use English created a transnational language-based hierarchy between different employee groups. Other studies look at the emotional experiences (Neeley et al. 2012) or knowledge sharing activities (Mäkelä et al. 2007) of employees under a language mandate.

Future Research Agenda
Despite the fast growth of language research in international business over the past few years, the field is still far from achieving a holistic understanding of the multidimensional role of language in business. Only a few years ago, Brannen, et al. (2012, p. 1) remarked that ''IB research remains unsophisticated in appreciating the multiple forms, facets, and features of language and its impact on MNCs and on the way in which we study IB phenomena''. Taking stock of recent developments, our literature review has shown some progress in this regard, but also revealed that large gaps remain. A juxtaposition of the earliest and latest papers included in our sample illustrates this noticeable, but slow growth. In the years following Holden's (1987, p. 236) critique of the ''naive, misinformed and unconsciously (or unashamedly) chauvinistic'' pronouncements on the nature and functions of language in international business, most authors viewed language as ''mechanical and manageable'' (Fixman 1990, p. 25), simplistically focused on grammatical errors of nonnative speakers (Sims and Guice 1992), or characterized accented speech as a fixed personal characteristic (Tsalikis et al. 1992). However, others already considered the social and cognitive dimensions of language (Swift 1991) and the unintended consequences of corporate language policies (San Antonio 1987) in those early days. We still see a varied picture today. On the one hand, scholars increasingly acknowledge the manifold languages spoken in MNCs (Tenzer and Pudelko 2017), study instances of language mixing (Schau et al. 2017), explore speakers' linguistic positioning behavior (Millot 2017), and generally recognize the contextually conditioned, co-constructed, and culturally created nature of language (Du-Babcock and Tanaka 2017). On the other hand, natural languages are still often used as categorical variables, suggesting they are self-contained (Bell and Puzakova 2017;Touchstone et al. 2017). Given the multidisciplinary nature of international business as a subject area, we hope that future studies will integrate concepts and methods from different academic disciplines to gain a deeper understanding of the complex linguistic influences on globalized business environments. To stimulate the field's further development in this direction, we will now point out untapped opportunities for future research. Consistent with the structure of our review above, which follows the conventional sequence of presentation in empirical papers, we highlight promising future research directions for (1) the geographic settings of language research, (2) theoretical approaches from different disciplines, (3) methods and data, and (4) findings on different levels of analysis.

Research Settings: Future Directions
Having shown an imbalance in authors' target regions of research, our review suggests that the field would benefit from extending the scope of investigated regions, countries, and languages. Whereas many general effects of language diversity were confirmed across contexts, others may be subject to regional variation. Considering that Harzing and Pudelko (2013) reported considerable variations in corporate language policies across countries, a more comprehensive coverage is needed, for example of emerging market multinationals. Corporations and individual employees may also face different situations in host countries with multiple official languages (such as Serbia or Singapore), where speakers may mobilize a broad array of linguistic resources to express voice (Janssens and Steyaert 2014). Harzing (2016) tentatively suggests that people of Nordic and Germanic countries tend to perceive language more mechanically as a means of communication, whereas the inhabitants of Latin and East Asian countries consider it to be at the very core of their culture. The latter are hitherto underrepresented, as the current research focuses on Finland, Germany, the UK, and the US. This may be one reason why many international business scholars have understood translation as the mere search for ''equivalence'' rather than a process of interaction across cultures (Chidlow et al. 2014).
Given the varying linguistic distance of local tongues to English (Hutchinson 2005) as the language of global business, our review also encourages a more comprehensive coverage of influential languages in global business. In a recent study, Ly et al. (2013) list Arabic, English, French, German, Hindi, Japanese, Mandarin Chinese, Portuguese, Russian, and Spanish as the ten most influential languages on a global scale. Considering the growing importance of BRIC countries, languages such as Chinese, Russian, and Portuguese are now significant in the global arena. Whereas our review uncovered a growing number of studies on the use of Chinese in business, there is much less research on the languages of the other emerging BRIC countries. Researchers speaking Portuguese, Russian, Hindi, or Tamil as their mother tongues could enrich the field with an intimate understanding of these languages in their native context. Resting on empirical data collected in Finland, China and Russia by native speakers of Finnish, Swedish, English, Russian, Mandarin and Cantonese, Barner-Rasmussen et al. (2014)'s study on multinational boundary spanning demonstrates the enormous potential of such endeavors.
In parallel, European and North American international business scholars should go beyond the dominant domestic collaborations and aim for more international and cross-lingual cooperation with colleagues in emerging economies. They may activate collaborative relationships of this kind through conferences or mailing lists and develop them using virtual collaboration technologies such as Skype, Lync, or WebEx. Selecting languages, countries, and regions based on theoretical considerations, these international research teams could juxtapose different language combinations in one study, thus extending recent comparisons of language issues in MNCs in Nordic, Anglo, continental European, and Asian language clusters (Harzing and Pudelko 2013). Empirical projects of this scale and scope are particularly challenging to plan and carry out, as they require scholars to carefully reflect upon their methodological practices (Piekkari and Tietze 2011). Along these lines, Welch and Piekkari (2006) illustrate the difficulties of reaching shared understanding with interviewees when using foreign languages in qualitative interviewing. Chidlow et al. (2014) provide helpful guidance how international business researchers can responsibly account for their translation decisions when managing multilingual datasets.

Theory: Future Directions
As we have shown above, theoretical perspectives such as culture, the gravity model from economics, theories of firm internationalization, linguistic relativity, and social identity currently prevail among language research in international business. Consistent with Brannen et al.'s (2012) call for a ''reexamination of current international business models and frameworks'' under a linguistic lens, we believe that research on language diversity in international business should build on extant achievements, but also extend its theoretical scope beyond the approaches used by previous studies. To this end, organizational behavior researchers, strategy scholars or economic theorists need to look beyond the boundaries of their academic socialization. Whereas individual researchers can gain inspiration from other disciplines by way of cross-disciplinary pollination, we see the largest potential for advancement in inter-disciplinary collaboration by representatives with different academic socialization. The innovation resulting from this creative recombination of theoretical angles will help the field overcome lingering simplistic uses of the language concept and approach its focal phenomenon in ways that are more sophisticated.

Building on Existing Theories
There are many promising opportunities to enrich the dominant theoretical angles in the field. In terms of culture, we second Pudelko et al.'s (2015) call for a better clarification of the relationship between culture and language. Researchers can build on the pioneering publications about cross-cultural speech pragmatics in international business settings (Chen et al. 2006;Kassis Henderson 2005). Going beyond the readily detected issues with lexical and syntactical understanding across language barriers, sociolinguistic speech act theory (e.g., Pütz and Neff-van Aertselear 2008) may help to examine the impact of culturally conditioned language use (House 1996;Wierzbicka 2003) on international business communication. Differences in the use of language for particular purposes such as informing, demanding, or promising, and diverse conversation styles, e.g. in turn-taking or intonation, merit particular attention, as these forms of language barriers ''often go unnoticed and are all the more pernicious for that reason'' (Kassis Henderson 2005, p. 70).
Building on Egger and Toubal's (2016) suggestions to refine research on language and trade, economic perspectives using the gravity model could be extended to examine the effects of immigration, transnational entrepreneurship, and a country's foreign language education on trade activity. Such endeavors may follow up works by Genc et al. (2012), Drori et al. (2009), and Byram (2008), respectively. The related research into linguistic influences on MNC internationalization could examine the role of language policy on firm outcomes, for example investigating how policies to use English as the corporate language affect firm growth and international expansion. This line of work could answer calls (e.g., Allen et al. 2015) to better understand how MNCs manage human capital across borders.
Linguistic relativity theory can be applied to new topics such as cognitive theories of decision-making (Wood and Bandura 1989) and the related crosscultural differences (Mann et al. 1998) or gendered structures of the workplace (Holmes 2008) such as the persistent gender gap in entrepreneurial activity. Social identity research could examine the development of linguistic identities over time and the congruence or divergence of MNC employees' language-related identities with their national, functional, or location-based identities. Theories of intersectionality (Anthias 2008;Harper 2011) may help to conceptualize this complex interplay of multiple identities.

The Promise of Theories from Other Academic Disciplines
Having outlined fruitful contributions from different disciplines to the investigation of language effects in business, we reinforce Brannen et al.'s (2012) view that insights gained from disciplines such as linguistics, political science, and psychology can create frames of reference helping to understand the role of language in international business more profoundly.
Having seen the successful application of linguistic theories (e.g., Chen et al. 2006;Virkkula-Räisänen 2010) to business settings and economic phenomena, we support Pudelko et al.'s (2015, p. 90) view that linguistics is an ''obvious candidate'' for cross-disciplinary pollination in this field. Recent work on the economic repercussions of linguistic gender-marking and future-time reference suggests that researchers should consider how other elements of language structure may correspond to labor market allocations as well as preferences for entrepreneurship. Moreover, the usefulness of these theories suggests that other theoretical lenses such as semiotics (Smith and Anderson 2007), evolutionary linguistics (Croft 2008), or socio-linguistics (Wardhaugh and Fuller 2015) could also contribute to the investigation of language diversity in business settings. Semiotics, for example, is the study of signs and how they are used to communicate with others (Chandler 2007). Besides considering pragmatic conventions of culturally conditioned language use (Wierzbicka 2003), researchers may examine how differences in prosodic conventions, i.e. acoustic cues like loudness of the voice, intonation, speaking rhythm, and speed (Sporer and Schwandt 2006) influence mutual understanding between employees speaking different mother tongues.
Our review also highlighted the enormous cognitive challenges employees are facing in today's multilingual organizations. International business settings already entail high cognitive demands due to their dynamic and complex nature (Volk et al. 2014;Hadjichristidis et al. 2016), but these are substantially exacerbated by the burden of foreign language processing. According to Takano and Noda (1995), activities such as conversation or negotiation require both linguistic (i.e., communication) and non-linguistic information processing (i.e., thinking and deciding). Employees speaking a foreign language at work use a larger part of their working memory to allow for the linguistic processes, thereby sacrificing resources available for thinking and decision tasks (Baddeley 2003). Tenzer and Pudelko (2016) recently connected language-based cognitive load to the choice of communication media in virtual teams. However, it remains unclear how this mechanism generally affects decision making in business. Whereas some studies find that foreign language processing causes psychological distance and therefore triggers deliberate and reflective thinking (Keysar et al. 2012), others demonstrate the opposite, namely that decision making and behavior becomes more intuitive, automatic, emotional, and less analytic when people are cognitively distracted . Neurolinguistic research on the processing of language in the human brain (for a review see Leikin 2016) should aim to resolve this puzzle, for example by using functional MRI technology (Cabeza and Nyberg 2000) to measure individuals' brain activity while performing foreign language tasks.

Looking Beyond Disciplinary Boundaries
Given the multidimensional influence of language on international business activities, we encourage scholars to look beyond their respective mother disciplines, as interdisciplinary perspectives allow addressing ''more complex questions than those which are typically formulated when relying on the standard assumptions and the narrowing focus usually found within disciplines'' (Cantwell and Brannen 2011, p. 3). Whereas the subject area of international business has already assembled an array of approaches from different disciplines, these mostly occurred independently from each other in separate publications. To proceed from this multidisciplinary setup to truly interdisciplinary research, scholars need to synthesize and interrelate arguments taken from different disciplinary perspectives (Cantwell and Brannen 2011). Among many promising combinations, an integration of theories from international strategy research, organizational behavior, and international economics could advance our focal field with novel approaches. Strategy research could harness psychological insights from organizational behavior to gain a deeper understanding of individual employees' reactions to corporate language policies and the ensuing dynamics on the group-level. This may reveal the contested and negotiated nature of language practices, show how tongues are mingling in daily business communication and, consequently, facilitate the design of more sophisticated language strategies. We also encourage strategy researchers with a language interest to take inspiration from the concepts and theoretical angles applied in economics. For example, they could expand current work on linguistic distance as an independent variable to explain entry modes or analyze language structures such as gender marking and future-time reference to analyze particular features of national institutions and policies. Finally, experimental work in behavioral economics on how language choice influences thoughts, feelings, and behavior could complement the psychological perspective of language-related organizational behavior research.
According to Harzing and Feely (2008, p. 51), such synergies have not been fully realized so far, since scholars have been ''deterred by the cross-disciplinary nature of the subject''. The slow progress may also be due to the enormous difficulties of achieving genuine interdisciplinarity, which Kockelmans (1979) already cautioned against almost four decades ago. Interdisciplinary research requires that specialists combine their expertise into an integrated response to the problem (Piso 2015), but there is a lack of clarity concerning how exactly this may be achieved (Repko 2007). Rogers et al. (2005, p. 268) point to the ''incommensurability of concepts, different units of analysis, differences in world views, expectations, criteria, and value judgments'' between academic disciplines as obstacles to integration. Scholars have captured these challenges in different metaphors. Whereas Horn (2015) likens disciplines to cultures which require scholars involved in interdisciplinary work to undergo adjustment processes, dominant images come from the realm of languages. Since each discipline has its own conceptual vocabulary (Newell 2001) and scholars ''speak in dialects that are specialized to their disciplines'' (Wear 1999, p. 299), the central barrier comes down to the difficulty of communicating concepts, theories, and methods across disciplinary boundaries (Stone 2013;Piso 2015). Disciplinary institutions such as academic journals, funding agencies, or university management furthermore discourage interdisciplinary integration, as they tend to evaluate individual scholars according to their capacity to adhere to idiosyncratic disciplinary conventions (Horn 2015).
Researchers aiming to capture the role of language in international business through interdisciplinary collaborations therefore need to prepare for setbacks (Horn 2015). However, if cross-disciplinary teams strive to explicate basic premises to each other (Wear 1999), communicate extensively about conceptual differences and engage in constant self-reflection and -evaluation (Szostak 2013), they can broaden their horizons and achieve theoretical innovation (Cantwell and Brannen 2016). If scholars overcome the related obstacles, interdisciplinary research endeavors promise to resolve complex issues which transcend the scope of a single research expertise (Piso 2015).

Methodology: Future Directions
Our review uncovered a number of patterns in methodology and data sourcing. Specifically, we found a slightly higher proportion of qualitative studies than quantitative work in the field. In terms of data sources, most research is crosssectional and interview-based. Based on these findings, we offer recommendations for extending the field's methodological toolbox and substantiating its empirical basis with new forms of data collection.

Methods
Our systematic review reveals that language-related research in international business has evolved considerably, both in terms of qualitative and quantitative methods. This methodological diversification bears the potential for promising complementarities. Qualitative approaches dominated the field in its early days and are certainly well suited to build robust middle range theory in previously unexplored areas (Eisenhardt and Graebner 2007), and therefore highly suitable for investigating still unchartered effects of language diversity on international 123 business. The growing number of quantitative studies can test the propositions generated by exploratory case study research (Creswell 2013). We also encourage a broader application of experimental studies such as prisoner dilemma games in the field (see e.g., Akkermans et al. 2010;Volk et al. 2014). The use of experimental studies has the advantage that the use of a particular language use can be manipulated between treatments and thus the effect of language can be isolated. Experimental games also allow us to measure actual behavior rather than relying on self-reported surveys or interview responses, which in turn reduces the effects of social desirability and self-presentation. Another research method could involve textual analysis of concepts using software such as Diction (Ridge and Ingram 2014). Multi-method studies combining qualitative, quantitative, and experimental approaches to language effects in international business are still rare (see e.g., Angouri 2013 andBarner-Rasmussen et al. 2014 for exceptions), yet they would be invaluable to enhance the robustness of emerging theories in the field. Parallel to our encouragement of interdisciplinary theorizing, we urge scholars to broaden their methodological repertoire by tapping into the toolboxes of neighboring academic disciplines.

Data Sources
We urge qualitative researchers to go beyond the dominant interview methodology, complementing their datasets with observations of naturally occurring linguistic misunderstandings among employees of multinational corporations. Whereas interviews may be biased by social desirability and only reflect consciously processed information, observations capture actual behavior and pick up effects which the interacting parties may not be aware of. Future studies may also extend the pioneering efforts in multi-sited organizational ethnography and introduce approaches from neighboring disciplines such as discourse analysis for sociolinguistics (Schiffrin et al. 2008) or life histories from ethnography (Musson 2004) in order to comprehensively understand the complex influence of linguistic diversity. These techniques would also provide longitudinal data, which could meaningfully advance the field by examining the development of language policies over time. In the realm of quantitative studies, our review highlights the need for more large-scale studies covering MNCs in a wide variety of country contexts in order to probe the generalizability of the impact of foreign language use.

Findings by Levels of Analysis: Future Directions
Our systematic review of language research in international business demonstrates that this fast-growing field captures language-related phenomena on all major levels of analysis. These findings suggest a series of promising future research avenues for examining language at each level, which we will outline below. Table 3 indicates additional theories, phenomena, and research questions on different levels of analysis, thus generating a general framework for future language-related research in international business.

.1 Individual Level Perspectives
A more profound understanding of how language influences individual outcomes could be promoted through fundamental research in behavioral economics. An example is Akkermans et al.'s (2010) experimental study on how language priming influences individuals' thoughts and behaviors. Scholars could for instance explore how individuals' associations with key management concepts from the Anglophone world differ depending on the language in which they voice these thoughts. Individual-level research on language in international business could also generate a deeper understanding of the role bilingual professionals can play in MNCs. Whereas previous research mostly focused on coordinate bilinguals (those who acquire their second language very early in life, usually in the same context), future studies could extend the investigation to compound bilinguals (those who acquire their second language later in life, often in another context; see Larsen et al. 1994), a situation more typical for bilingual professionals (Day and Wagner 2007) and migrant workers (Roberts 2007). As immigrant entrepreneurs play a key role in growing their host economies (Wadhwa et al. 2007), they constitute a particularly promising target group for studies on compound bilinguals. Existing work on language use in polyglossic urban areas and multilingual regions (Lüdi et al. 2010) and on internal migration (Lüdi 1992) can provide useful starting points here. International business scholars may also draw on the work of Berthoud et al. (2015) in studying how individuals draw on multiple linguistic repertoires to construct, transmit and apply knowledge. Linguistic policy research on bilingual education (see e.g., Riagáin and Lüdi 2003) can meaningfully inform studies of linguistic capital in modern multinationals. Furthermore, we encourage the field to investigate the behavioral effects of language diversity in business contexts. For example, language-based cognitive load and anxiety through foreign language use have been largely ignored as a cause of health issues.

Group Level Perspectives
Regarding the group level of analysis, existing studies reveal that language barriers substantially influence team communication, knowledge sharing, and other processes. We therefore suggest that future researchers examine new group phenomenon such as co-located and virtual teams, and the roles of bilingual group members as boundary spanners and bridge-makers. Within these groups, future research could test theories of the consequences of linguistic ostracism (e.g., Robinson et al. 2012). More specifically, future studies could extend recent investigations on the language-based choice of communication media (Tenzer and Pudelko 2016) to probe the suitability of established frameworks like media richness theory (Daft and Lengel 1986) or media naturalness theory (Kock 2004) in multilingual settings. Finally, researchers could examine the interplay between linguistic identities and national, cultural, functional, location-based, gender-driven, age-related, or other identities to explore the disruptive potential of language-based faultlines (Thatcher and Patel 2012;Hinds et al. 2014) within and across multilingual groups.

Firm Level Perspectives
Concerning the firm level, our review demonstrates that the majority of international business scholars interested in linguistic diversity investigate effects in large MNCs. We argue that it might be interesting to study language effects in other firms, particularly small and medium enterprises, new ventures, and NGOs. Considering that business researchers form a transnational community working with English as a lingua franca (Tietze and Dick 2013), the impact of linguistic diversity on business schools (see Lauring and Selmer 2012;Śliwa andJohansson 2014, 2015) also offers a worthwhile avenue for exploration, as does the increasingly interdisciplinary university research environment. Comparing language policies, practices, and effects between these different contexts can assist in understanding the boundary conditions for theories of language diversity in international business. To gain a more comprehensive understanding of corporations' ''transnational business communication capital'' (Tietze et al. 2016), firm-level research should furthermore study occupational vocabulary and sociolects in addition to the commonly investigated diversity in national languages.

Country Level Perspectives
With respect to research on the country level, our systematic literature review identified a substantial upward trend in research on the economic implications of linguistic relativity, determinism, and grammatical structures. This line of ''Whorfian economics'' (Fabb 2016) research could be further extended by examining whether women's occupational choices or the gender pay gap correlate with the intensity of linguistic gender marking in a country's dominant language. Considering that a recent experiment demonstrates significant differences in children's intertemporal choices depending on their mother tongue (Sutter et al. 2015), it would be interesting to investigate the effects of obligatory or optional future-time reference in a country's language on citizens' preferences for long-versus short-term investments. An experiment capturing divergent behaviors between the speakers of minority and majority languages (Cappelletti et al. 2015) furthermore highlights the need to study language effects in countries with several official languages. Related themes involve the impact of countries' colonial past on language use or the influence of government initiatives trying to counteract the ''excessive'' use of English words (for the Chinese case see Economist 2014) on communication.

Multi-Level Perspectives
Our review also yields a number of recent studies connecting the perspective of individual speakers with language effects on their teams or organizations. We consider these multi-level approaches highly promising, as multilingual collaborations can only create synergies by integrating the strengths of individual contributors into an outcome greater than the sum of what each employee could have achieved individually (Katzenbach and Smith 1993). To understand how these synergies arise, scholars need to capture so-called ''emergent processes'' (Kozlowski and Klein 2000;Kozlowski et al. 2013), which transform intra-personal thoughts, feelings, and behaviors through interaction and communication into higher-level collective phenomena at the dyadic, team, or organizational level. We argue that emergent processes such as cohesion, confidence, conflict, learning, adaptation, and organizational climate could be fruitfully studied under a language lens. Besides this bottom-up emergence, scholars may also study the top-down influences of a country's linguistic context or organizational language strategies on team dynamics or individual cognitions, emotions, and behaviors. Whereas many studies have tried to tackle such processes with data collection at a single level of analysis, we encourage future research could gain a more holistic understanding by building theory, and collecting and analyzing data at all involved levels (Hitt et al. 2007). To be successful in such complex research designs, scholars have to articulate the theoretical bases of their work carefully (Hitt et al. 2007) and consider the limited capabilities of existing software packages in multilevel modelling (Kozlowski et al. 2013). As the transformation of individual language-related cognitions and emotions into collective processes takes time, they need to conduct more longitudinal research, which are more time-consuming and expensive. Given that each academic discipline tends to favor specific levels of analysis, multilevel research often requires an interdisciplinary mindset. To the extent that these challenges are mastered, multi-level research will break new ground in language-related international business studies.

Limitations
The scope and focus of our study entails some limitations, which indicate possible extensions in future research. First, we had to omit monographs and book chapters from our systematic review, because the major online databases do not list them. This might have excluded relevant contributions by linguists, translation scholars, and members of other disciplines who occasionally touch on the subject area of international business, but rely more on book publications than academics in business studies do. Whereas pragmatic constraints did not allow us to systematically review the theories, methods, and empirical findings of these contributions, we have drawn many suggestions for future research from this body of literature. Interested readers may continue their studies with the Handbook of Multilingualism and Multilingual Communication (Auer and Wei 2007), the language section within the Routledge Companion to Cross-Cultural Management (Holden et al. 2015) or the recent Palgrave Handbook of Economics and Language (Ginsburgh and Weber 2016). Second, we only included English-language publications in our review. To check for bias, we also entered the equivalents of our search terms in French, Spanish, and German as there are established business journals in these national languages (Venard 2007). We found that publications in these languages did not yield substantial insights beyond the English-language literature. Publications in Portuguese, Russian, Hindi, and Mandarin might have yielded more insights on the BRIC countries, but we decided to limit our review English-language material, as no research team would be able to read all major world languages. Most importantly, 75% of articles in the social sciences are written in English and the hegemony of English as a language of science is rising (Enrique Hamel 2007).
Nevertheless, we would hope that scholars with language capabilities in English as well as one of the above languages could act as bridge individuals.
Third, the scope of our study did not allow us to include all forms of language diversity in business settings. Brannen et al. (2014) name technical or electronic language as potentially insightful avenues of research, whereas a large stream of discourse, rhetoric, and narrative analysis by organization theorists investigates how top managers recontextualize content through language, thus shaping sensemaking, organizational identities, and strategic orientations (Boje et al. 2004;Phillips et al. 2004). Future research could fruitfully connect the ''linguistic turn in organizational research'' (Alvesson and Kärreman 2000) focusing on rhetorical and metaphorical language with the linguistic turn in international business targeting on the multilingual realities in global corporations.
Fourth, if we had reviewed a larger number of publications from communication studies, sociolinguistics, and psycholinguistics, our review could have portrayed the complex concept of language in greater depth and could have shown a broader array of methods to empirically capture it. However, it was not feasible to cover these disciplines in their entirety within the space constraints of this article, so we limited our scope to papers studying how language plays out in international business settings. Whereas most publications dealing with this context were written by business scholars and economists, they are informed by many different concepts of and approaches to language. Future studies could build on our efforts by drawing more comprehensively on the achievements of linguistics, the leading discipline studying languages in form, meaning, and context and under a variety of aspects (Akmajian et al. 2001).
Fifth, we need to balance our many suggestions for future research by acknowledging a certain danger of further proliferation in research themes. If the diverse approaches to language in international business evolve in parallel and independently from each other, the field may become even more fragmented than it is today. An active dialogue between approaches and a synthesis between dominant themes will be needed to reach a holistic understanding of language in international business. To this end, we invite scholars from different mother disciplines to collectively define a set of ''big'' research questions, which can unite their efforts for the years to come.

Conclusion
Reinforcing the frequent calls for more conceptual innovation and empirical investigations on the impact of language on international management (see e.g., Holden 2008;Piekkari and Zander 2005), our systematic and thorough review of 264 publications on language in international business identifies some progress in understanding the ''multifaceted role of language in international business'' (Brannen et al. 2014). As we note a growing body of research drawing on concepts from a variety of disciplines and employing diverse methodologies, many international business scholars have gained a better understanding of the function and role of language within their subject area. However, our review also reveals that the field remains fragmented, with serious knowledge gaps in theory, data, methodology, and content. Reflecting on Holden's (1987, p. 234) statement that ''linguists who aspire to an integration of linguistics into the management sciences face a herculean task'', we conclude that part of this task has been fulfilled to date, but much remains to be done. To motivate future research in this direction, we offered multiple opportunities for advancing the investigation of language diversity in international business research. We particularly encourage the integration of insights from different academic disciplines as an opportunity to gain a deeper understanding of language complexity in international business. Extending Piekkari et al.'s (2014, p. 244) recent conclusion that ''the pervasive effects of language need to be taken more fully into account in explanations of international business activity'', we argue that a more profound understanding of its effects will have a very positive impact on business and management studies as a whole. Barner-Rasmussen et al. (2014) Bell and Puzakova (2016) Lüdi (2013, 2015) Björkman and Piekkari ( Kroon, Cornelissen, and Vaara (2015) Ku and Zussman (2010) Kubat and Swaminathan (2015) Kulkarni (2015) Kuznetsov and Kuznetsova (2014) Lagerström and Andersson (2003) Latukha et al. (2016) Lauring (2007 Klitmøller (2014, 2015) Lauring and Selmer (2010and Selmer ( , 2011and Selmer ( , 2012aand Selmer ( , b, 2013 Nickerson (1998Nickerson ( , 2005Nickerson ( , 2015 Oh, Selmier, and Lien