international student stress due in 20 hours

Denice E. Welch () Professor of International Business, Melbourne Business School, University of Melbourne, Melbourne, Australia.

Lawrence S. Welch () Professor of International Marketing and International Business, Melbourne Business School, University of Melbourne, Melbourne, Australia.

The Importance of Language in International Knowledge Transfer

Denice E. Welch, Lawrence S. Welch

Manuscript received July 2006, revised April 2007, final revision received November 2007.

Abstract and Key Results ● In this article, we examine the effects of language on the transfer of knowledge within

multinational companies. We unbundle language from the culture box and use the basic communication model to show how language affects all stages.

● We also examine a range of influences – cost, transfer medium, teams, networks, trust, staff movements and motivation – on international knowledge transfer to demonstrate how language effects their operation.

● We develop the concept of language as a “reconfiguration agent” to explain its exten- sive, pervasive, on-going and system-altering characteristics.

Key Words Language · Knowledge Transfer · Communication · Reconfiguration Agent

mir vol. 48, 2008/3 339

340 mir vol. 48, 2008/3

Introduction

As companies expand internationally and their foreign operations become more dispersed, language demands increasingly become an issue in inter-unit knowledge transfers. This seems axiomatic, yet it is surprisingly ignored in the broad sweep of international busi- ness research on knowledge transfer. Articles in the 2004 focused issue of the Journal of International Business Studies, and the 2005 special issue of Management International Review – both on knowledge and knowledge transfer in multinationals – are indicative. This neglect is incongruous given the simultaneous rise of research interest in subsidiary management and inter-unit resource and knowledge transfers (e.g., Birkinshaw/Hood 1998, Foss/Pedersen 2002).

This is not to say that language has been ignored. The frequently cited article on inter- national knowledge transfer by Kogut and Zander (1992) recognises that a shared lan- guage enhances communication and exchanges in general. Hedlund (1999, p. 11) pointed out: “For knowledge to be exchanged and combined, there has to be a shared medium of communication. People have to be able to make sense to and of each other. One aspect of this is shared spoken general language”. Nahapiet and Ghoshal (1998, p. 254) consider that a shared language may provide “a common conceptual apparatus for evaluating the likely benefits of exchange and combination”. In this context, language is viewed as a component of corporate identity that enables a multinational to transmit and share knowl- edge (e.g., Phene/Madhok/Liu 2005). Other writers recognise the existence of profes- sional language and company jargon (what Argote, McEvily and Reagans (2003) term ‘short-hand’ and ‘home-made’ language), but this is not taken any further, especially into a multiple foreign language context. Generally speaking, though, in the investigation of international knowledge transfer, language is only hinted at or given fleeting reference. Those citing Kogut and Zander’s work tend to treat language as a given, or conveniently ignore it.

In the broad knowledge management literature, the international aspect receives little coverage, so it is not surprising that language aspects are somewhat ignored. For example, Cohen and Levinthal (1990) include a shared language as part of prior related knowledge when defining their concept of absorptive capacity, but there is scant elaboration on this issue. Even Nonaka’s much cited work that draws on research on Japanese firms is curi- ously quiet on the impact that language issues have on communication and knowledge flows (e.g., Nonaka/Takeuchi 1995).

Recognition of the role of language has been emerging in other fields. Apart from the obvious areas of communication and linguistics, researchers in organization behaviour have shown a renewed interest in language related issues, as exemplified by a 2004 special issue of the Academy of Management Review on language and organization. There is also evidence of a movement of language and communication research into areas of inter- national management (see 2005 special issue of International Studies of Management and Organization).

In this article, we seek to redress the relative neglect of language in the general IB literature through highlighting the various ways language influences the transfer of knowledge throughout the multinational. While we recognise the various layers of lan- guage (that is, company speak, professional jargon and the like; Welch/Welch/Piekkari

mir vol. 48, 2008/3 341

2005), our emphasis is on language as encountered in the day-to-day operations of di- verse multinationals. We begin by noting the links between language and culture but stress the distinctive nature of language that warrants its separation. This discussion is followed by a review of extant work on knowledge transfer. The basic communication model is then used as the context to show how language is more than simply noise or an element of distortion but affects all stages of the international knowledge transfer process. We then examine the way in which language affects a range of international knowledge transfer factors: cost, transfer medium, teams, social networks, trust, staff movements and motivation. Based on this analysis, we draw out conceptual, research and managerial implications. We develop the concept of language as a “reconfiguration agent”: arguing that language continually reconfigures the total international knowledge transfer system: acting as a precursor, contextual influence and even reconstructing basic messages. While our primary focus is the IB research community, managers in multinationals who wrestle with the issues raised in this article should find it of particular relevance.

Language and Culture

In this article, we concentrate on the specific role of language, in a sense taking it out of the ‘culture box’. Language, of course, is a key aspect of culture: it is inherent in a spe- cific culture and also an embodiment of it: “If language influences the way we behave and how we perceive things, it means that culture is also inherent in the language itself” (Claes 1995, p. 99). Despite these well recognised connections, we argue that language is sufficiently important in its own right to warrant a more focused treatment. By treating language as a separate variable, its influence on communication patterns and information and knowledge flows can be more readily identified (Piekkari 2006).

In and of itself, language is a mental model, framing activity and behaviour. It is part of the mindscape. One example is that of international teams comprised of members who speak different mother tongues. Henderson (2005 p.75) regards the challenge for such teams is to recognise that although: “one audible and visible ‘surface’ language is being used to facilitate exchanges, members continue to use diverse expressive and interpretive mechanisms derived from their respective language systems”. Hambrick et al. (1998, p. 198) maintain that language, while interacting with other cultural components – such as values, cognitive schema and demeanour – has a distinct influence on the functioning of international management teams. They conclude that “a low degree of common lan- guage facility [will] generally … impair group functioning, hampering the exchange of information and trust”. Even in instances where there is a high degree of common lan- guage facility, not every team member will exhibit the same degree of confidence or ability to communicate complex ideas in a foreign language. Moreover, “language not only reflects social context, it may also influence social interactions within teams” (Chen/ Geluykens/Choi 2006, p. 688). Thus, culture and language do not always simply overlap.

Language has an importance above and beyond the ‘embeddedness-in-culture’ per- spective. Many staff in diverse multinationals are called upon to operate in the company language, often English, which is not their ‘mother’ tongue, and by definition, this simple act creates some disconnection with their national cultural base. To illustrate, in their

342 mir vol. 48, 2008/3

study of Kone, the Finnish multinational, Marschan, Welch and Welch (1997) found that 65 per cent of employees worldwide, including Finns, were required to operate in a second language – English – which was the company’s common language. For some multinationals, part of the intention in using a company-imposed language is to facilitate the develop- ment of a stronger sense of identification, and thereby corporate culture. This also emerged within Kone, as one manager from the Italian subsidiary explained: “I started to [speak English] … I feel international and I feel Kone because of that” (Marschan/Welch/Welch 1997, p. 595). At the same time, individuals in such situations may still attempt to create their own language-based networks. These attempts tend to be driven not by a need to develop cultural links but are language specific – that is, the need to access critical infor- mation, knowledge and networks in order to facilitate informal transfer processes to cir- cumvent barriers imposed by the requirement to operate in the corporate language. For example, there was concern within Kone’s Spanish subsidiary that, due to its relatively low level of English competence, information and knowledge was not being effectively transferred into the subsidiary. To overcome this, the Spanish subsidiary manager created a personal, Spanish-speaking network across other subsidiary units and at the Finnish headquarters. On his own initiative, he sought out Spanish speakers throughout Kone, and passed this onto his managers so they could access these as well (Marschan/Welch/Welch 1997).

While not denying the importance of the cultural link, these situations illustrate how language imposes its own demands upon those working in cross-border situations, with potentially far-reaching consequences for communication and knowledge transfer proc- esses. As such, language is deserving of investigation in its own right.

Knowledge Transfer

There are many definitions of knowledge, but for the purposes of this paper we use that offered by Roberts (2000, p. 430) who separates knowledge, information and data, as a way of sharpening the definition of knowledge. Data is “a series of observations, meas- urements, or facts”. Information is “data that have been arranged into a meaningful pattern”, whereas knowledge is “the application and productive use of information… it involves an awareness or understanding gained through experience, familiarity or learn- ing”. This is not to say that there is a simple linear hierarchy of process between these three, rather, in many instances, knowledge is a prerequisite for the generation and utilisa- tion of data (Alavi/Leidner 2001).

The two main categories or properties of knowledge are now generally accepted as being tacit and explicit (or codified) knowledge (Nonaka/Takeuchi 1995). For our pur- poses, while there is not a simple delineation between these two categories – they operate in tandem and interactively – we shall differentiate between tacit and explicit knowledge in a transfer context. For example, it is generally argued that codified/explicit knowledge is easier to transfer, although there is a recognised dependence on the possession of rele- vant tacit knowledge in order to decode explicit knowledge (Roberts 2000). Other writers point to the embedded nature of knowledge: that it cannot easily be separated from its social and organization context, thus making it ‘sticky’ for transfer purposes (Kogut/

mir vol. 48, 2008/3 343

Zander 1994, Martin/Salomon 2003, Inkpen/Tsang 2005). Lam (1997) has noted that some knowledge can become so embedded in the social and team context within organi- zations that it is simply not ‘migratory’.

Knowledge transfer is said to occur when knowledge is disseminated or diffused, that is communicated from one entity to another. Such transfer can be deliberate or as a con- sequence of unintended processes such as social interaction and chance meetings. For knowledge to be transferred there must be some form of articulation, making the knowl- edge explicit. However this articulation occurs, language is absolutely basic. Articulation can involve non-verbal communication. For example, demonstration – the show-how aspect of knowledge transfer – is normally accompanied by some verbal instruction. Ex- plicit knowledge is transferred in codified form, such as words, technical drawings, and graphics. One of the aspects of knowledge transfer stressed in the literature is that codi- fied knowledge, to be useful, inevitably calls upon tacit knowledge. That is, useful codi- fied knowledge is rarely disembodied from the tacit.

Knowledge is resident in and transferred between individuals, groups, and organiza- tions, although individuals are the key. As Howells and Roberts (2000, pp. 22 et seq.) point out, while knowledge can radiate out from the individual, “becoming more codified, more information-like, more transferable and more global in its reach, it still requires in- terpretation at the individual level … its [knowledge] interpretation, comprehension and absorption back into a knowledge state remains at the individual level”. This underlies much of the thinking about absorptive capacity: that is, the ability to recognise, accept and apply new knowledge that has been transferred into an organization from elsewhere. It is not just the sum of the absorptive capacity of individual employees, but is contained in organizational routines and the like (Cohen/Levinthal 1990). Language is part of ab- sorptive capacity but it is primarily reliant on individual competence. A recently reported study by Buckley et al. (2005, p. 56), found language as a constraint in the three Chinese IJVs investigated. Once managers attempted to move knowledge to the shop-floor level, foreign language competence was at a very basic level: “As a result, all documents relat- ing to such workers needed to be translated into Chinese, which delayed the process of knowledge transfer and increased transfer costs”.

Communication and Knowledge Transfer

In order to appreciate the role of language in knowledge transfer, it is useful to begin with the basic communication model, where the sender encodes the message, uses an appropri- ate medium/channel to transmit the message, and the receiver decodes the message (Hollensen 2001). The effectiveness of the transfer depends on the ability of the sender to accurately encode a meaningful, complete message; the selection of the correct medium or channel of transfer; and the ability of the receiver to decode and understand the mes- sage as it was intended. Communication scholars have produced a wealth of evidence to demonstrate how, despite its seeming simplicity, the communication process is fraught with barriers, noise, distortion and leakage. In other words, it does not occur in a seamless, automatic, enclosed ‘vacuum tube’, but is subject to many influences that determine its successful transfer. Feedback is important. Face-to-face communication generally is the

344 mir vol. 48, 2008/3

preferred mode of communication as it allows sender and receiver to check for meaning- ful transfer, through verbal clarification as well as non-verbal interpretation. As Buckley et al. (2005, p. 48) note: “Knowledge transfer requires the use of language and communi- cation to enable articulation in order to promote assimilation”. Thus, we superimpose knowledge transfer onto the basic communication model. Figure 1 illustrates how lan- guage affects both sides of the transfer process, in situations involving individuals (I), groups (G) and/or organizations (O); including the reverse feedback loop.

At the receiving end, as part of decoding, translation may be necessary. The receiver may not be the ultimate end user, such as when an intermediary acts as a translator. The initial recipient may even decide not to translate at all, thus stopping the transfer process. The final recipient is then dependent on the intermediary’s accuracy and preparedness to share the knowledge as sent.

Figure 1. International Knowledge Transfer Model

Intermediaries act as gate-keepers, particularly in situations where possession of rel- evant language fluency delivers power to the intermediary. Marschan-Piekkari, Welch and Welch (1999a, p. 389) relate that, in the case of the Finnish multinational Kone, gate-keeping activity in some subsidiaries (for example Italy) was very deliberate, in- cluding the restriction of subsidiary staff training in English, the corporate language. Similarly, gate-keeping was evident at the subsidiary level in another Finnish multina- tional, Wartsila, where expatriates in the Italian subsidiary controlled the flow of infor- mation and knowledge through language competence (Finnish and the corporate lan- guage, English), inhibiting the extent of knowledge transfer. In a third example, Kone’s Taiwanese subsidiary manager was the only person within the subsidiary with an ade- quate competence in English and therefore was responsible for all communication be- tween Taiwan and other Kone units including headquarters. He undertook the primary translation role and is an example of the ultimate gatekeeper. Research also has revealed how the translation gate-keeping role can affect international technology transfer (Fai/

mir vol. 48, 2008/3 345

Marschan 2003). As Janssens, Lambert and Steyaert (2004, p. 424) stress, translators always “deform the original text” in the process of translation. Inevitably, knowledge is altered in some way.

Obviously, there is an equivalent process at the sender’s end leading to a final encoded form for transmission. In the course of this process, there may be deliberate or unintended distortion of the original message, translation errors, codification problems and gate- keeping (see Figure 1). Again, language can be an important part of the distortion or blockage within this process. For example, Lam (1997, p. 992) contrasts the sender be- haviour on the two sides of a Japanese-British international joint venture. One of the British interviewees commented: “I think we have supplied more information than J-firm has, but in most cases, there is no reluctance to supply information. The only barrier is technical language, translations and that kind of thing.” Another British interviewee summed the transfer process as: “Little coming from Japan, a lot going to Japan”. This asymmetry was confirmed by a Japanese interviewee:

We are definitely getting more [information] out of B-firm, than the other way around … It is fairly well-documented and can be passed over just like that. Whereas, in our company, most of the things are not documented. Even if they were, they are not in English in the first place (p. 992).

In a general sense, there was more emphasis by the British partner on codification as a basis for knowledge sharing. For the Japanese, however, there was greater emphasis on tacit knowledge and interpersonal exchanges, to a certain extent locking them into the use of the Japanese language rather than codifying knowledge into English to facilitate knowl- edge transfer. The small number of those with requisite language competence restricted the extent of personal exchange between the parties, thus limiting sending and receiving capacity on both sides (see also Buckley/Clegg/Tan 2003). Of course, as Figure 1 shows, if the receiver responds to any message, inevitably language intrudes during the feedback process with potentially equivalent encoding and decoding issues.

Faced with the language issue in internal inter-unit information and knowledge flows, over time, multinationals tend to move towards the use of a common corporate language, and sometimes more than one. Introducing a common language into the internal commu- nication process of the multinational is evidence of the managerial concern to achieve effective modes of knowledge, information and data transfer. It is sometimes seen as a panacea for the problem of language differences. While language standardisation through a common corporate language aims to facilitate inter-unit communication, it often intro- duces new forms of barriers and distortion. As senders, top management may not recognise the unintended consequences of language standardisation upon subsidiary staff. Welch, Welch and Piekkari (2005, p. 14) note that: “Having adopted a common corporate lan- guage for internal reporting and communication purposes, and as top management oper- ates predominately in the adopted common corporate language, such executives often lose sight of the way in which language effects remain pertinent for others within the multinational”. In effect, language standardisation pushes the translation question down the line, becoming an absorptive capacity issue for subsidiary management.

The above discussion highlights how language in cross-cultural communication is more than ‘noise’ that occurs in message transmission. It has effects at all stages of knowl- edge transfer – on both sender and receiver sides, as well as in transmission and feedback.

346 mir vol. 48, 2008/3

However, the impact of language extends beyond the immediate transfer process, as the following section demonstrates.

Influences on International Knowledge Transfer Examined Through a Language Lens

To fully appreciate the broader impact of language, we analyse seven of the most com- monly identified influencing factors that prior research has identified as driving the inter- national knowledge transfer process. The seven factors are: cost, transfer medium, teams, social networks, trust, staff movements and motivation. These are examined through the prism of language to demonstrate the extent to which language may facilitate, filter, distort and/or act as a barrier to effective international knowledge transfer. While each of the seven influences is examined separately, we also stress their interrelatedness, as shown in Figure 2. Exploration of the language impact on this broad array of influences is con- sidered an important step towards conceptualising the role of language as a change agent that is capable of reconfiguring the system or set of interacting influences that is part of, as well as surrounding, the basic knowledge transfer model. Further, inserting language automatically highlights the role of individuals who are key players in the accumulation, dissemination and absorption of knowledge. Our analysis illustrates how knowledge transfer depends on the way in which individuals use their language competence within the organisational context.

Figure 2. Language and Influences on International Knowledge Transfer

mir vol. 48, 2008/3 347

Cost

As much of the research points out, international knowledge transfer is an important, yet costly process, and therefore of considerable concern from a company perspective (Carlson 1975, Kogut/Zander 1993, Foss/Pedersen 2002). The literature seems to concur that tacit knowledge transfer tends to be high cost, whereas explicit or codified knowledge transfer is relatively low cost (Roberts 2000). However, as Lamberton (1997, p. 77) observes, “… codification, because it is not a purely technical matter, but involves human and organiza- tional capabilities, runs up against limits of affordability”. Language costs are part of this picture. For example, Fixman (1990) notes the time (six months to a year in some cases) and cost involved in translation of major documents consequent upon changes, such as in the telecommunications sector, in European Union policy. Indeed, the expanded EU lan- guage context illustrates the challenge and cost to governments and firms of transferring information and knowledge when multiple languages have to be used.

Obviously, firms can decide not to undertake translation or only on a limited basis, thus saving upfront expenses but there is clearly a range of hidden costs associated with such an approach – for example, staff denied access to new knowledge, or misapplying new technology. For the individual, there is an opportunity cost of sharing knowledge. Time spent explaining or showing the knowledge content to others can be considerable when that time could be used on their own work for which they are judged in performance terms. This is even more pronounced when knowledge has to be translated, thereby ac- centuating the time and cost involved.

In the codification process, electronic transfer of even the simplest of documents may be costly because of the translation demands involved with multiple languages, either at the sender’s or receiver’s end. In a case known to the authors, the Australian subsidiary HR director of a Swedish multinational consistently checked all communication from the parent company written in English (the company language) even though the subsidiary was in an English speaking country. This action was deemed necessary, though costly and time consuming, to ensure that the ‘Swenglish’ did not obscure the intended meaning. Costs associated with language codification are not insubstantial and are exacerbated when there is a concern to check for equivalence and cultural sensitivity that requires back-translation as an additional step in the transfer process (Janssens/Lambert/Stayaert 2004). Thus, to say that codified knowledge is relatively cheaper to transfer between MNC units can be misleading. As Buckley, Clegg and Tan (2003) report, the Chinese IJV, Shanghai Bell, was forced to establish a special centre for documentation and language translation because of the perceived low absorptive capacity of the joint venture that had inhibited the transfer of technology. Materials were translated into Chinese for distribu- tion to relevant departments.

Transfer Medium

The form or medium of knowledge transfer has long been exposed as critical. It is not our intention to examine all aspects of the various mediums or channels. Instead, we focus on face-to-face interaction and the growing use of information and communication technolo- gies, such as the intranet and Internet.

348 mir vol. 48, 2008/3

Face-to-face interaction

Face-to-face interaction still tends to be viewed as the most powerful medium of knowl- edge transfer, despite advances in modern telecommunications. De Meyer’s (1991) study of R & D centres in 14 large multinationals found that among six broad mechanisms to improve inter-unit communication, socialisation was the most important: the firms used training, temporary assignments and constant travelling as socialisation tools. A further gauge of its importance is the continuing and even growing use of international business travel. Multinationals continue to place great emphasis on the need to gather staff together in formats such as meetings, training sessions and the like as a basis for knowledge sharing. Participation in such gatherings is, naturally, language dependent. Finnish research demonstrates how subsidiary staff participation in training programs was restricted by the need to be fluent in English, the language of instruction. As well, language competence is fundamental to comprehension and understanding, as illustrated by the experience of Wartsila’s Italian subsidiary staff. When Finnish expatriates ran a quality control course for Italians in English, many participants did not understand what they were being told. At meetings, Italians were not able to raise important questions because of their lack of corporate language skills (Marschan-Piekkari/Welch/Welch 1999a).

In transferring tacit knowledge, face-to-face interaction is imperative. Hedlund (1999, p. 12) has argued that: “The fact that much individual as well as organizational knowl- edge is tacit requires interaction to be intensive and extended in time for demerging, merging, and remerging of knowledge elements to be possible”. Research stresses that effective interaction as a basis for tacit knowledge transfer depends on the relationship context and of relationship development as a precursor to the transfer and absorption of knowledge (Johannessen/Olaisen/Olsen 2001). Further, Roberts (2000, p. 434) notes: “Given the importance of tacit knowledge as a factor enabling the assimilation of codified knowledge, face-to-face contact may well be a prerequisite for the transfer of much codi- fied knowledge. In addition, the establishment of a level of trust to facilitate the exchange of knowledge also favours co-presence and co-location.”

Information and Communication Technology

Proponents of the use of information and communication technology (ICT) point to the ways in which ICT can enhance the knowledge transfer process: from knowledge creation through to its storage, retrieval and application. For example, Alavi and Leidner (2001, p. 118) consider that company intranets expose staff to greater amounts of on-line or- ganizational information and “support individual learning (conversion of explicit knowl- edge to personal tacit knowledge) through provision of capabilities such as computer simulation (to support learning-by-doing) and smart software tutors”. Improved ICT has undoubtedly enhanced international contact at all levels of the multinational, and enables more rapid transfer of information and codified knowledge. The volume of knowledge that resides in some companies’ centralised knowledge banks though can be overwhelm- ing (Nielsen/Ciabuschi 2003, Garud/Kumaraswamy 2005). Further, as McDermott (1999) recognises, cultural aspects surrounding ICT are important. There is a danger that the

mir vol. 48, 2008/3 349

centralisation of knowledge codified into the company language will disconnect those who lack requisite language competence.

As noted above, while ICT facilitates contacts, it does not replace the need for direct, face-to-face contact, particularly in terms of its “richness, variety and scope, [which] can- not yet be fully reproduced by ICTs” (Roberts 2000, p. 439). Whether face-to-face or electronic, the need for a shared language in communication is self-evident.

Teams

Research on knowledge transfer emphasises the importance of teams and groups within multinationals. Advocates consider that global teams have the capacity to overcome the normal constraints on organizational transfers of resources, knowledge and information (for a review, see Zárraga/Bonache 2003). They can be a highly focused solution to a range of issues and problems as well as powerful mechanisms of international knowledge transfer, given the varied contributions of team members, their diverse networks, and sometimes multiple languages (Harvey/Novicevic 2002). However, on closer examina- tion, language diversity can inhibit effective team performance. In a recent study of com- munication within international teams operating within MNCs in France, Henderson (2005) found that, even when teams operated in a common language (English), language diversity affected interpretations and socialization processes. These were “hidden obsta- cles” in team performance and knowledge sharing.

Teams operating within the subsidiary context may become isolated from the rest of the multinational, with negative consequences for access to and transfer of relevant, useful knowledge, thus reducing the subsidiary’s absorptive capacity (Hedlund 1994, Zellmer- Bruhn 2003, Mahnke/Pedersen/Venzin 2005). Knowledge can easily become locked within a subsidiary-based team thereby restricting sending capacity. Language is one fac- tor that can cause subsidiary team isolation. This is why expatriates with the required language competences (local and corporate) can play such an important role in breaking down subsidiary team isolation and thereby ensuring knowledge transfer in both direc- tions.

From another perspective, informal global teams may emerge due to staff interaction in the normal course of international operations. For example, Goodall and Roberts (2003), in their study of an European-based oil multinational, found that expatriates de- veloped into a highly socialised, networked team when operating in project teams in various countries. However, while the global team supported knowledge sharing between expatriates, it acted as a barrier for locals who did not possess the appropriate languages – “English and corporate speak” – and this restricted knowledge sharing with the locals.

Social Networks and Relationship Development

The knowledge management literature recognises the important role of social networks and inter-personal relationships in facilitating knowledge transfer – the know-who aspect (Hansen 2002). This is also stressed in the literature on social capital (e.g., Nahapiet/ Ghoshal 1998). Social networks are predominantly informal. As Reagans and McEvily (2003, p. 240) stress: “Informal interpersonal networks are thought to play a critical role

350 mir vol. 48, 2008/3

in the knowledge transfer process”. Informal networks are not always obvious, but are used extensively – in some cases contrary to the formal direction of top management (Marschan/Welch/Welch 1996). Informal networks act like a powerful underground movement ensuring that knowledge and information reach those who require it.

Language is not just a pre-requisite for informal social networks at the inter-unit level but also can play a role in structuring patterns of connection, generating language-based clusters within an MNC that form ‘shadow structures’ (Marschan-Piekkari/Welch/Welch 1999b). Language-based networks within the MNC facilitate knowledge transfer by providing members with quick access to a large pool of potential knowledge. However, cultural and language similarity does not necessarily ensure effective communication (Zander 2005). Further, research shows that it is not just direct personal connections that are important. The wider indirect connections in a social network may also be brought into play. For example, an associate in another unit could provide direction to the source of required knowledge that is known to this associate’s broader network (Hansen 2002). The intricate web of contacts and links in a network and their importance in transferring information and knowledge has been identified in the vast body of literature generated by the Industrial Marketing and Purchasing (IMP) group (Håkansson/Snehota 1995).

Trust

Trust has emerged as an important factor in the knowledge literature, somewhat inevitably given the emphasis on the importance of trust in relationships and networks. Again, re- search by the IMP group and others stresses the importance of trust (e.g., Young/Wilkin- son 1989, Henderson 2005). There is an inherent perception of reliability and veracity of knowledge disseminated from sources that are seen as trustworthy. Trust also mediates or alleviates concerns about how knowledge and information is to be used, and encourages disclosure and sharing of knowledge. In other words, there is trust in the person and how the knowledge is to be used (McEvily/Peronne/Zaheer 2003). Preparedness to identify members of a personal network for others to utilise for knowledge transfer relies on the ability to trust that access will not be open to abuse by that person (Michailova/Husted 2003). Research indicates that trust is particularly important in the transfer of tacit knowl- edge (Dhanaraj et al. 2004). For trust to be established between the sender and the receiver of knowledge, a relationship has to be built. Roberts (2000, p. 434) stresses face-to-face contact and socialisation as important in this process especially in the context of com- munication difficulties that arise from “differences in culture and language”.

If trust, which is relationship based, plays such a critical role in tacit knowledge transfer, then language becomes one of the keys that unlocks the potential contribution of trust. It is very difficult to establish a trusting relationship without an adequate basis of commu- nication. Abrams et al. (2003) provide an example of how a new product development team, comprising two Americans and an Englishman, descended into mistrust and dys- functional interaction in part because of the way in which members applied different meanings to the same English words. Cross-cultural communication and marketing lit- eratures are replete with similar examples.

Within the multinational, employee perceptions and expectations may generate a situ- ation of hostility and mistrust – such as expatriates from headquarters being regarded as

mir vol. 48, 2008/3 351

‘spies from headquarters’. Expatriate interaction with local staff is important in creating an environment of trust, and competence in the local language is often seen as an aspect of this. For example, Goodall and Roberts (2003), in their study of a European oil com- pany’s operation in Colombia, report on the reaction of local staff to expatriates. They felt a lack of trust from the expatriate community, as evidenced by the failure to develop local staff. Locals were simply not trusted. One interviewee commented: “Language makes a big difference. They [the local staff] can’t speak English and it’s a huge barrier”. The feel- ings of hostility towards expatriates and the perceived lack of trust had implications for knowledge transfer. There was a lack of technology transfer between expatriates and local Colombians: “During the last two years, nothing is learned. The expat goes and then they bring in another expat”. Lack of language skills and mistrust reinforced each other as major barriers to knowledge transfer. Expatriates had their own network, but locals were excluded from network access, thus further accentuating problems of current and future knowledge transfer possibilities. The importance of language for the development of trust has been confirmed in a recent study of 310 inter-unit relationships of multinational subsidiaries. “Results show that language fluency related significantly to shared vision and perceived trustworthiness in both the Chinese and Finnish subsidiaries” (Barner- Rasmussen/Björkman 2007, p. 105).

Teamwork is enhanced by the development of trust, but this depends on a degree of face-to-face interaction (transfer medium) over time. However, these factors are con- nected back to having a shared language as a basis for effective inter-unit communication that enhances information and knowledge flow and knowledge sharing.

Staff Movements

There is agreement in the knowledge transfer literature that moving staff is one way to assist in knowledge transfer. For example, Alvai and Leider (2001) regard personnel transfers as more effective for highly context specific knowledge distribution as it allows access to the routines and tacit knowledge of others. The knowledge transfer and sharing role of staff movement is reflected in the MNC literature (for a review see Dowling and Welch 2004). Motorola (China) used “extensive short term movements internationally in order to personalise knowledge” (Buckley/Clegg/Tan 2003, p. 79). This involved mainly expatriates working in China for short periods. At the same time, in order to ensure that such knowledge was transferred, the Chinese subsidiary invested considerably in lan- guage training in English, the operating language, and in recruiting English-speaking Chinese employees. Know-how development – assisting making tacit knowledge explicit through demonstration and explanation of routines – was a common aim of staff transfer, though movement also exposed staff to new ideas and ways of working. Thus, staff move- ments are not just restricted to expatriates from headquarters into subsidiaries, but between subsidiary units, and into headquarters. Shanghai Bell moved Chinese staff into the Bel- gian headquarters as a way of building a learning network.

However, there are cases where staff movements have not resulted in effective transfer. Research on the use of Japanese expatriates in Chinese subsidiaries found that orientations to work, along with language and cultural differences, impeded knowledge transfer. Japa- nese expatriates tended to “focus on their jobs rather than seek to cross-fertilize manage-

352 mir vol. 48, 2008/3

ment knowledge and practice” (Taylor 1999, p. 864). A similar finding is reported by Marschan-Piekkari, Welch and Welch (1999a) where technical staff assigned to interna- tional subsidiaries of the Finnish multinational, Kone, concentrated on the project, stayed for a short time, and had relatively limited interaction with local staff. This restricted the transfer of knowledge to local staff. In this case, the lack of local language fluency was the major constraint.

Staff movements are considered to support the development of organizational networks and relationships. Such networks include access to a broader range of knowledge and contacts. Research shows that informal networks may be deliberately built on the basis of language commonality. As the Goodall and Roberts (2003) example above illustrates, expatriates can form a strong personal network, but they also may act as language nodes. If expatriates have both the local language and the common corporate language, they become a key knowledge interface between subsidiary staff and headquarters – the more so, of course, when competence in the corporate language within the subsidiary is weak. This language node role can persist after repatriation, with the person becoming the ‘know-who’: a short-cut in various exchanges such as problem-solving (Marschan- Piekkari/Welch/Welch 1999a).

Traditional expatriates are costly and demographic changes, such as the dual career couple, have affected global staff mobility. To counter these constraining factors, multi- nationals have increased their use of short-term assignments and other forms such as commuting and rotational assignments. Aside from postings to foreign operations, multi- nationals will have numerous staff travelling internationally to support their various ac- tivities. Emerging research is revealing how language has a constraining effect on the knowledge transfer role of short-term assignees and international business travellers. Un- like traditional expatriates, these groups of multinational employees are often seen as “just passing through”, and usually lack local contacts and support mechanisms provided to longer-term assignees, such as language training. Welch, Welch and Piekkari (2005) relate a number of examples of language related problems encountered by international business travellers, such as having to work through interpreters, who modify the mes- sages for local audience sensitivities or, in cases where the corporate language is used, a varied level of local competence constrains receptivity.

Motivation

Research shows that motivation is a consideration in knowledge transfer. Szulanski (1996) refers to the potential reluctance, even hostility, of individuals or groups to share crucial knowledge, given that the act of transferring involves a loss of ownership and the ceding of some position of privilege. In a study of Russian and Western companies oper- ating in Russia, Michailova and Husted (2003, p. 63) found an environment of knowledge sharing hostility which generated a prevailing attitude among the Russians that there was little to be gained through sharing knowledge: “Russian managers strongly expressed the view that they should always be more knowledgeable than their employees”. The per- ceived opportunity cost of sharing was high. In addition, as Nielsen and Ciabuschi (2003, p. 36) explain: “In times of business restructuring and resulting job reductions, people tend to protect their intellectual capital (knowledge)”. Uncertainty increases knowledge

mir vol. 48, 2008/3 353

sharing hostility and impacts on motivation to share. Likewise, Lazarova and Tarique (2005) discuss the importance of repatriates’ willingness to share knowledge upon their return as a factor in international knowledge transfer. The two-step process that language introduces into the equation thus reduces the individual’s motivation to share knowl- edge.

These motivational aspects are not surprising as knowledge confers power, as does language competence. The Kone study referred to earlier is indicative. According to one of Kone’s global HR staff, in some subsidiaries, management were deliberately restrict- ing communication and controlling inter-unit contacts through language. One tactic was to limit subordinate access to English language courses even though Kone subsidiary budgets allowed for this expenditure (Marschan-Piekkari/Welch/Welch 1999a). In a gen- eral sense, individuals are more prepared to share knowledge with known and trusted counterparts who also share a language.

The above analysis reveals that, behind the scenes, the role played by language is more substantial and long-lasting than has been previously recognized. It is evident that language has the potential to interpose in myriad ways and at different stages of the knowledge transfer process, far beyond the basic communication model. As Brannen (2007) points out, the basic communication model is deficient as an explanatory tool as it omits the process of comprehension: knowledge transfer involves more than the decoding of a linguistic signal. By identifying the various ways in which language impinges on a variety of influences on international knowledge transfer (as depicted in Figure 2), we show how language has an impact well beyond the encoding and decoding process of cross-cultural communication. Further, we highlight the role of individuals as enactors of language influence, with the capacity to introduce personal concerns and connections, thus widening the potential range of knowledge transfer issues and factors; for example, with respect to tacit knowledge and motivation to share.

Implications of Inserting Language

Based on the above, we now delineate and discuss the conceptual, research and manage- rial implications of the interplay between language effects and the various influences on the transfer process.

Language as a Reconfiguration Agent

By commencing with the fundamental communication model, we have shown how lan- guage impinges in a direct fashion on the basic international knowledge transfer act. However, the impact of language goes beyond that. It determines aspects such as who has the information and knowledge, whether and how it is articulated, when and if it is shared, and in what form. We have demonstrated how language may affect sender transfer capacity, recipient absorptive capacity, and operative influences (such as cost and transfer medium, identified in Figure 2). Through reconfiguring these various elements, language generates ongoing impacts beyond a simple one-off knowledge transfer act – for exam- ple, through its contribution to the development of social capital. That is, language is

354 mir vol. 48, 2008/3

simultaneously an active agent in the knowledge transfer process itself, as well as influ- encing the background set of determinants. In this section, we develop the concept of “reconfiguration agent” to convey the sense of how language affects the total system within which knowledge transfer takes place. We draw out the interactions between lan- guage and social capital; language and subsidiary clustering and inter-unit communica- tion; and language and human resource policies to illustrate the way that language acts as a reconfiguration agent.

As mentioned earlier, social networks provide the context within which interaction occurs between individuals and groups. Interaction is based on connections that are heavily influenced by similarity between people, an element of which is a shared language. The tendency to interact with similar others has been termed homophily, based on research that shows how interpersonal similarity breeds connections, and this has implications for the information individuals receive, how this is interpreted, and the nature and quality of the interaction (Makela/Kalla/Piekkari 2007). Thus, interpersonal similarity supports the development of social networks and a key part of this homophily is language. It facilitates the formation and operation of social networks through which knowledge is transferred, acting as a bridge or conduit between various organizational, functional and national groups. At the same time, language similarity can act as a barrier, determining who is inside the network and who is outside. Through this mechanism, the social network is constantly being reconfigured with continuing effects on international knowledge transfer.

In the organizational context, the reconfiguration of social networks may be deliber- ately pursued by management as part of a proactive strategy, or it may develop as a result of fortuitous activities by individual staff and the connections they develop. Deliberate network creation is illustrated by our earlier example of the Spanish subsidiary manager who developed an “information and knowledge language path” through his identification of Spanish-speakers in other subsidiary units. Though informal, this initiative had positive outcomes for the multinational as the language path enhanced distributive and absorptive capacity, providing a more assured basis for the transfer of information and knowledge across language boundaries, even though in the process it was reconfiguring the commu- nication channels that were based on the common corporate language (that is, English).

Alternatively, language-based social networks may evolve in a fortuitous manner as a result of individual appointments, staff movements, and corporate activities such as train- ing programs and group meetings. For example, a Finnish expatriate (Mr. X) working in a Latin American subsidiary of the Finnish multinational, Kone, voluntarily learnt Spanish while on assignment. His motivation was to improve his own ability to perform his job. Upon repatriation back to Finland, he discovered that he had become known as “the per- son at headquarters who spoke Spanish”. For 20 years, he was a key formal and informal gatekeeper and boundary spanner, providing technical information and knowledge in Spanish to Latin American and Spanish subsidiary staff to support official and informal, codified and tacit knowledge transfer in both directions (Marschan-Piekkari/Welch/Welch 1999a). Through his acquired language skills, Mr X enhanced headquarters’ knowledge sender capacity and improved the knowledge transfer process in general. This case demonstrates the potential, long-lasting influence of a language-based social network. Eventually, Mr X’s position as a language node was formalised through promotion to a relevant position that reflected his language capability and network role.

mir vol. 48, 2008/3 355

These examples indicate how the movement of knowledge that resides in one part of the multinational to another unit is often informal, unplanned, even inadvertent, as a result of individual interaction between parties who share a common language. As discussed earlier, language nodes are important agents of knowledge transfer, in and of themselves as well as through their language-based social networks. As such, they become powerful reconfiguration agents based on their language capabilities.

Language also may reconfigure the knowledge transfer system by generating informal subsidiary clusters – creating so-called shadow structures that lie behind the formal or- ganizational structure (Marschan-Piekkari/Welch/Welch 1999b). These clusters are an attempt to overcome a lack of corporate language competence within the relevant sub- sidiaries through a range of joint activities (such as, combined training programs), sharing new product information, and staff transfers. A language-based cluster can be regarded as a form of aggregated social network, though often with an organizational imprimatur, given the endorsement of subsidiary managers and formal nature of many of the activities.

In a general sense, a multinational’s ability to transfer knowledge is related to its stock of language capital, reflected in the aggregate possession of relevant foreign language skills by its employees. In this respect, company policies such as the hiring of language competent staff, maintaining language abilities through training, and strategic movement of staff are important. Peltokorpi (2007, p. 78), in his study of Nordic subsidiaries in Japan, found that a lack of English skills among Japanese employees created communica- tion problems. Responses included recruitment of staff with English language skills, en- couraging staff to attend company-sponsored language classes, and sending Japanese employees to Europe to gain language skills. In addition, some firms deliberately sought to employ other Nordics resident in Japan who had Japanese language competence. These individuals acted as language intermediaries between the expatriate Nordics and their Japanese subsidiary employees, as they were able to “transfer information closer to its original meaning. They were further told to facilitate subsidiary-headquarter connec- tions”. The fact that these companies were undertaking such measures is indicative of the importance attached to language in improving the flow of knowledge between the Nordic headquarters and their Japanese subsidiaries.

The knowledge transfer literature tends to analyse processes at the organization level, such as unit-to-unit transfer, while much of what is transferred occurs at the individual level. As noted earlier, language competence may deliver power to certain individuals within the company but there is no certainty about how this power will be used. An indi- vidual may be constructive and use the relevant language competence to facilitate knowl- edge transfer. Alternatively, the actions may be that of a powerful gatekeeper who modi- fies and frustrates the transfer process. Key individuals can also facilitate or hamper the development of language capital by controlling access to language training, staff relocation and promotion, and the like. Thus, the consideration of language at the individual level captures some of the subtle, behind-the-scenes, reconfiguring effects it has on interna- tional knowledge transfer.

Thus, the conceptual contribution we make in this paper is to move the treatment of language and international knowledge transfer beyond the basic communication model. The concept of “language as a reconfiguration agent” demonstrates how language has far wider and longer-term systemic consequences than current IB research into the process of

356 mir vol. 48, 2008/3

international transfer has generally recognised. We now draw attention to the research and managerial implications of giving language a more substantial role.

Research Implications

Apart from its contribution in conceptual development discussed above, the more sub- stantive inclusion of language in IB research into knowledge transfer processes has meth- odological value. First, our analysis demonstrates the value of unbundling language from the concept of culture and treating it as a variable in its own right. Addressing language explicitly should also enhance the veracity of conclusions drawn from empirical research as it helps in avoiding the use of culture as a convenient catch-all – the ‘villain of the piece’ – to explain research findings regarding international knowledge transfer processes and outcomes.

Second, there is a need for studies of international knowledge transfer to cover lan- guage aspects in their conduct and their reporting. Many of the research studies referred to above leave major questions about how language has been treated during the course of the research itself, given that such studies, by their very nature, typically cross multiple language boundaries and contexts. These questions include:

What language was used in data collection? Often, one is left to assume that the lan- guage was English but readers cannot make an informed judgement about the reliability and validity of the resulting data that might have been affected by language considera- tions.

What was the respondent’s language competence? We are not informed of the steps taken, if any, in many studies to ensure that those sent the questionnaires had the requisite language fluency to respond in a meaningful manner.

Who is the respondent? It is not always clear from published results how the researchers have ensured that the intended informant has not passed the questionnaire onto someone who has the requisite language fluency but lacks the knowledge to respond meaningfully.

How did the researchers control for the possibility that translators may be used to overcome the language barrier? Literal translation is not always meaningful, even when the researcher has carefully followed protocols such as back translation.

To what extent did a lack of requisite language capability influence the response rate, and pattern of responses? For example, expatriates within a subsidiary are generally the persons with corporate language fluency. Are they the ones who are responding thereby biasing the results?

Third, treating language in its own right in international knowledge transfer, particu- larly the subtleties and process aspects over time, as argued above, would seem to call for research methodologies that focus on longitudinal and in-depth case analysis.

Managerial Implications

Given the above, it is evident that language should be taken into account in a more sub- stantive way through its incorporation in the design and implementation of international

mir vol. 48, 2008/3 357

knowledge transfer policies and practices. As a starting point, managers should recognise that language standardisation – the adoption of a common corporate language for infor- mation flows, reporting and general inter-unit communication – despite the benefits, does not necessarily remove potential communication problems posed by language diversity. In some respects, language standardisation recasts the language problem by pushing it further down the hierarchy, and creating new centres of language power. As senior man- agers within multinationals tend to be fluent in the common corporate language, their own competence may mask recipient difficulties due to lack of ability on the part of those lower down the hierarchy and in some subsidiaries to decode the knowledge being trans- ferred in the common corporate language. As (Simonin 2004) points out, part of the sender’s capacity is the recognition of the recipient’s needs and abilities, including com- parative language differences between sender and receiver.

Regular language audits are required in order to have a clearer picture of how extensive the organization’s language resources are and where they are located. This is a necessary starting point to any policy development in the language area. An audit should also identify how effective company-sponsored language classes are, particularly in terms of numbers of employees accessing this opportunity; whether all units are hiring employees with requisite language skills; and that language competent staff are being effectively utilised. Language should also be part of any due diligence conducted prior to making a foreign acquisition. Emerging research has shown how language can be a disruptive force in mergers across national boundaries and constrain the integration process (Vaara et al. 2005).

Alongside the language audit, it would be useful to follow “the language trail” within the company. That is, what happens in a language sense when knowledge is transferred across language boundaries within the company? Company studies would reveal the extent to which knowledge is diffused and distributed and how language is facilitating, filtering, blocking, and/or distorting the knowledge flow.

Further, managers also need to consider language abilities and competences when for- mulating personnel policies, particularly appointment and promotion decisions; staff deployment associated with international assignments and transfers; international team composition; and allocation to and monitoring of training budgets for language acquisi- tion and maintenance. Finally, developments in IT, and their application – including translation software – are likely to help with language differences but, as with language standardisation, should not been seen as a panacea, as they do not remove the language barrier in person-to-person verbal interaction, nor facilitate tacit knowledge transfer.

Conclusion

The purpose of this article has been to draw attention to the various ways language affects the process of international knowledge transfer, thus redressing what we consider has been a deficiency in much of the research into how knowledge is transferred around the modern multinational organization. In the course of our analysis, we have exposed the pervasive and dynamic influence of language on the whole knowledge transfer system, acting as a powerful reconfiguration agent. It is a precursor to transfer processes. It affects the sender’s ability to transmit knowledge; the receiver’s absorptive capacity; the transfer

358 mir vol. 48, 2008/3

medium; feedback mechanisms; and the organization’s distributive capacity. Language also impinges on the operative influences within the system; and reconfigures the context in which international knowledge transfer takes place: creating, driving and reconstruct- ing social networks and informal structural clusters through which knowledge circulates.

By examining language, we also refocus attention on the role of individual actors in the international knowledge transfer process. For example, we show how relevant lan- guage competence delivers power to certain individuals who may exercise this power to control information and knowledge flows in a positive or negative way.

Some may take comfort in the importance of English and its widespread global use as seemingly overcoming the language barrier. The rise of China as a global economic power points to the likelihood that language issues will become more, rather than less, promi- nent in the future. As one of our Chinese MBA students in Beijing recently commented: “English is difficult. It will be nice when Chinese companies become so global that you will be forced to use Chinese”.

References

Abrams, L. C./Cross, R./Lesser, E./Levin, D. Z., Nurturing Interpersonal Trust in Knowledge-Sharing Net- works, Academy of Management Executive, 17, 4, 2003, pp. 64–77.

Alavi, M./Leidner, D. E., Knowledge Management and Knowledge Management Systems: Conceptual Founda- tions and Research Issues, MIS Quarterly, 25, 1, 2001, pp. 107–136.

Argote, L./McEvily, B./Reagans, R., Managing Knowledge in Organizations: An Integrative Framework and Review of Emerging Themes, Management Science, 49, 4, 2003, pp. 571–582.

Barner-Rasmussen, W./Björkman, I., Language Fluency, Socialization and Inter-Unit Relationships in Chinese and Finnish Subsidiaries, Management and Organization Review, 3,1, 2007, pp. 105–128.

Birkinshaw, J./Hood, N. (eds.), Multinational Corporate Evolution and Subsidiary Development, London: McMillan Press 1998.

Borgatti, S. P./Cross, R., A Relational View of Information Seeking and Learning in Social Networks, Manage- ment Science, 49, 4, 2003, pp. 432–445.

Brannen, M. Y, When Mickey Loses Face: Recontextualization, Semantic Fit, and the Semiotics of Foreignness, Academy of Management Review, 29, 4, 2004, pp. 593–616.

Buckley P. J./Carter, M. J./Clegg, J./Tan, H., Language and Social Knowledge in Foreign-Knowledge Transfer to China, International Studies of Management and Organization, 35, 1, 2005, pp. 47–65.

Buckley, P. J./Clegg, J./Tan, H., The Art of Knowledge Transfer: Secondary and Reverse Transfer in China’s Telecommunications Manufacturing Industry, Management International Review, 43, Special Issue 2, 2003, pp. 67–93.

Carlson, S., How Foreign is Foreign Trade?, Acta Universitatis, Upsaliensis, Studia Oeconomiae Negotiorum 11, Uppsala 1975.

Casson, M., The Organisation and Evolution of the Multinational Enterprise, Management International Review, 39, Special Issue 1, 1999, pp.77–121.

Chen, S./Geluykens, R./Choi, C. J., The Importance of Language in Global Teams: A Linguistic Perspective, Management International Review, 46, 6, 2006, pp. 679–695.

Claes, M.-T., A Case Study of Language Overlap and Cross-Cultural Communication Problems, Interface Jour- nal of Applied Linguistics, 10, 1, 1995, pp. 99–111.

Cohen, W./Levinthal, D., Absorptive Capacity: A New Perspective on Learning and Innovation, Administrative Science Quarterly, 35, 1, 1990, pp. 128–152.

De Meyer, A., Tech Talk: How Managers are Stimulating Global R&D Communication, Sloan Management Review, 33, 3, 1991, pp. 49–58.

Dhanaraj, C./Lyles, M. A./Steensma, K./Tihanyi, L., Managing Tacit and Explicit Knowledge Transfer in IJVs: The Role of Relational Embeddedness and the Impact on Performance, Journal of International Business Studies, 35, 5, 2004, pp. 428–443.

mir vol. 48, 2008/3 359

Dowling, P. J./Welch, D. E., International Human Resource Management: Managing People in a Multinational Context, London: Thomson Learning 2004.

Fai, F./Marschan-Piekkari, R., Language Issues in Cross-border Strategic Alliances: An Investigation of Tech- nological Knowledge Transfers, Proceedings of the 29th Conference of the European International Business Academy, Copenhagen December 11-13 2003, CD-ROM.

Fixman, C. S., The Foreign Language Needs of U.S.-Based Corporations. Annals, 511, September, 1990, pp. 25–46.

Foss, N./Pedersen, T., Sources of Subsidiary Knowledge and Knowledge Transfer in MNCs, in Lundan, S. (ed.), Network Knowledge in International Business, Cheltenham: Edward Elgar 2002, pp. 91–114.

Garud, R./Kumaraswamy, A., Vicious and Virtuous Circles in the Management of Knowledge: The Case of Infosys Technologies, MIS Quarterly, 29, 1, 2005, pp. 9–33.

Goodall, K./Roberts, J., Only Connect: Teamwork in the Multinational, Journal of World Business, 38, 2, 2003, pp. 150–164.

Griffith, D. A., The Role of Communication Competencies in International Business Relationship Develop- ment, Journal of World Business, 37, 4, 2002, pp. 256–265.

Harvey, M./Novicevic, M.M., The Co-ordination of Strategic Initiatives within Global Organizations: The Role of Global Teams, International Journal of Human Resource Management, 13, 4, 2002, pp. 666–676.

Hansen, M. T., Knowledge Networks: Explaining Effective Knowledge Sharing in Multiunit Companies, Organization Science, 13, 3, 2002, pp. 232–248.

Hedlund, G., A Model of Knowledge Management and the N-Form Corporation, Strategic Management Jour- nal, 15, Special Issue Summer, 1994, pp. 73–90.

Hedlund, G., The Multinational Corporation as a Nearly Recomposable System (NRS), Management Interna- tional Review, 39, Special Issue 1, 1999, pp. 5–44.

Henderson, J. K., Language Diversity in International Management Teams, International Studies of Manage- ment and Organization, 35, 1, 2005, pp. 66–82.

Hollensen, S., Global Marketing, Harlow: Pearson Education 2001. Howells, J./Roberts, J., From Innovation Systems to Knowledge Systems, Prometheus, 18, 1, 2000, pp. 17–31. Håkansson, H./Snehota, I. (eds.), Developing Relationships in Business Networks, London: Routledge 1995. Håkanson, L., Knowledge Transfer in Multinational Corporations: An Evolving Research Agenda, Management

International Review, 45, Special Issue 2, 2005, pp. 3–7. Inkpen A. C./Tsang, E. W. K, Social Capital, Networks, and Knowledge Transfer, Academy of Management

Review, 30, 1, 2005, pp. 146–165. Janssens, M./Lambert, J./Steyaert, C., Developing Language Strategies for International Companies: The

Contribution of Translation Studies, Journal of World Business, 39, 4, 2004, pp. 414–430. Johannessen, J.-A./Olaisen, J./Olsen, B., Mismanagement of Tacit Knowledge: The Importance of Tacit Knowl-

edge, the Danger of Information Technology, and What To Do About It, International Journal of Informa- tion Management, 21, 1, 2001, pp. 3–20.

Kogut, B./Zander, U., Knowledge of the Firm, Combinative Capabilities, and the Replication of Technology, Organization Science, 3, 3, 1992, pp. 383–397.

Kogut, B./Zander, U., Knowledge of the Firm and Evolutionary Theory of the Multinational Corporation, Jour- nal of International Business Studies, 24, 1, 1993, pp. 625–645.

Lam, A., Embedded Firms, Embedded Knowledge: Problems of Collaboration and Knowledge Transfer in Global Cooperative Ventures, Organization Studies, 18, 6, 1997, pp. 973–996.

Lamberton, D., The Knowledge-Based Economy: A Sisyphus Model, Prometheus, 15, 1, 1997, pp. 73–81. Lazarova, M./Tarique, I., Knowledge Transfer upon Repatriation, Journal of World Business, 40, 4, 2005,

pp. 361–373. McDermott, R., Why Information Technology Inspired but Cannot Deliver, California Management Review,

41, 4, 1999, pp. 103–117. McEvily, B./Peronne, V./Zaheer, A., Trust as an Organizing Principle, Organization Science, 14, 1, 2003,

pp. 91–103. Madsen, T. L./Mosakowski, E./Zaheer, S., Knowledge Retention and Personnel Mobility: The Nondisruptive

Effects of Inflows of Experience, Organization Science, 14, 2, 2003, pp. 173–191. Mahnke, V./Pedersen, T./Venzin, M., The Impact of Knowledge Management on MNC Subsidiary Performance:

The Role of Absorptive Capacity, Management International Review, 43, Special Issue 2, 2005, pp. 101– 119.

Makela, K./Kalla, H. K./Piekkari, R., Interpersonal Similarity as a Driver of Knowledge Sharing within Multi- national Corporations, International Business Review, 16, 1, 2007, pp. 1–22.

360 mir vol. 48, 2008/3

Marschan, R./Welch, D./Welch, L. Control in Less-hierarchical MNC Structures: The Role of Personal Net- works and Informal Communication, International Business Review, l.5, 2, 1996, pp.137–150.

Marschan-Piekkari, R./Welch, D./Welch, L., Adopting a Common Corporate Language: IHRM Implications, International Journal of Human Resource Management, 10, 3, 1999a, pp. 377–390.

Marschan-Piekkari, R./Welch, D./Welch, L., In the Shadow: The Impact of Language on Structure, Power and Communication in the Multinational, International Business Review, 8, 4, 1999b, pp. 421–440.

Martin, X./Salomon, R., Tacitness, Learning and International Expansion: A Study of Foreign Direct Investment in a Knowledge-Intensive Industry, Organization Science, 14, 3, 2003, pp. 297–311.

Michailova, S./Husted, K., Knowledge-Sharing Hostility in Russian Firms, California Management Review, 45, 3, 2003, pp. 59–77.

Nielsen, B. B./Ciabuschi, F., Siemens ShareNet: Knowledge Management in Practice, Business Strategy Review, 14, 2, 2003, pp. 33–40.

Nonaka, I./Takeuchi, H., The Knowledge-Creating Company, New York: Oxford University Press 1995. Peltokorpi, V., Intercultural Communication Patterns and Tactics: Nordic Expatriates in Japan, International

Business Review, 16, 1, 2007, pp 68–82. Piekkari, R. Language Effects in Multinational Corporations: A Review from an International Human Resource

Management Perspective, in Stahl, G./Bjorkman, I. (eds.), Handbook of Research in International Human Resource Management, Cheltenham: Edward Elgar, 2006, pp. 536–550.

Phene, A./Madhok, A./Liu, K., Knowledge Transfer Within the Multinational Firm, Management International Review, 45, Special Issue 2, 2005, pp. 53–74.

Reagans, R./McEvily, B., Network Structure and Knowledge Transfer: The Effects of Cohesion and Range, Administrative Science Quarterly, 48, 2, 2003, pp. 240–276.

Roberts, J., From Know-How to Show How? Questioning the Role of Information and Communication Tech- nologies in Knowledge, Technology Analysis and Strategic Management, 12, 4, 2000, pp. 429–443.

Schmid, S./Schurig, A./Kutschker, M., The MNC as a Network: A Closer Look at Intra-organizational Flows, in Lundan, S., (ed.), Network Knowledge in International Business, Cheltenham: Edward Elgar 2002, pp. 45–72.

Simonin, B. L., An Empirical Investigation of the Process of Knowledge Transfer in International Strategic Alliances, Journal of International Business Studies, 35, 5, 2004, pp. 407–427.

Song, J./Almeida, P./Wu, G., Learning-by-Hiring: When is Mobility More Likely to Facilitate Inter-Firm Knowledge Transfer?, Management Science, 49, 4, 2003, pp. 351–365.

Szulanski, G., Exploring Internal Stickiness: Impediments to the Transfer of Best Practice within the Firm, Strategic Management Journal, 17, Special Issue Winter, 1996, pp. 27–43.

Taylor, B., Patterns of Control within Japanese Manufacturing Plants in China: Doubts about Japanization in China, Journal of Management Studies, 36, 6, 1999, pp. 853–863.

Vaara, E./Tienari, J./Piekkari, R./Säntii, R., Language and the Circuits of Power in a Merging Multinational Corporation, Journal of Management Studies, 41, 3, 2005, pp. 595–623.

Welch, D./Welch, L./Piekkari, R., Speaking In Tongues: The Importance of Language in International Manage- ment Processes, International Studies of Management and Organization, 35, 1, 2005, pp. 10–27.

Young, L. C./Wilkinson, I. F., The Role of Trust and Cooperation in Marketing Channels: A Preliminary Study, European Journal of Marketing, 23, 2, 1989, pp. 109–122.

Zárraga, C./Bonache, J., Assessing the Team Environment for Knowledge Sharing: An Empirical Analysis, International Journal of Human Resource Management, 14, 7, 2003, pp. 1227–1245.

Zander, L., Communication and Country Clusters: A Study of Language and Leadership Preferences, Interna- tional Studies of Management and Organization, 35, 1, 2005, pp. 83–108.

Zellmer-Bruhn, M. E., Interruptive Events and Team Knowledge Acquisition. Management Science, 49, 4, 2003, pp. 514–528.