How knowledge may be successfully developed across cultural boundaries

    16. Mai 2011 von Dipl.-Ing. Michael Fegerl, Univ.-Prof. Dr. Wilfried Wieden

    Increasingly people have to develop knowledge across cultural and language boundaries. Even though recent technologies offer powerful communication facilities people often feel confronted with problems which clearly reduce their chances for making their efforts a success. Concrete evidence concerning such problems is in this article derived from an EU-project in which both authors are currently involved. This contribution intends to describe selectively observed problems and experiences and interim results from the application of procedures and tools to remediate the observed problems.

    This articel was also published in Open Journal of Knowledge Management, Issue III/2011.

    1. Introduction

    In the course of the EU-project SILMAS (Sustainable Instruments for Lake Management in the Alpine Space; for details cf., in which both authors are involved, it appeared in the very beginning that the partners from 5 different countries could present their topics fairly well in the selected corporate language English. However, when it came to negotiating specific details of their topics considerable misunderstandings emerged, often only after the specific negotiation process had been ended. In follow-up interactions requests for clarification demanded an increasing (and irritating) share of time.

    The objective of this contribution is to make more explicit on the basis of available evidence why cross-cultural knowledge development may be inhibited and how observed problems may be remediated. The approach is data-driven rather than theory-driven: Selected actions which were initiated for this purpose will be described with respect to the achieved positive and negative effects which resulted from the application of these actions. Knowledge management procedures and tools will be described with respect to how they motivated the actions taken.

    2. Specification of problem

    2.1. Selected meta-data

    The following evidence derives from observations which were mainly made during face-to-face interactions at project meetings, in part also from online interactions (skype conferences). Project partners represent a fairly heterogeneous group of people: they derive from 5 different countries (France, Italy, Germany, Austria and Slovenia), different educational backgrounds (biology, physics, business administration, tourism, law, teaching), and professional affiliations  (free-lancers, full-time employees, managers). All are highly motivated because of the dire need for knowledge standards. The language of interaction was and still is (mainly) English, a language which is nobody's mother tongue. The criteria of analysis derive from a theoretical basis which is discussed in more detail in section 4.

    2.2. Evidence

    Case (a)
    Participants made use of one and the same English term, but apparently related it to different concepts (meanings).

    Persons from France made use of the term lake contract to refer to 'general agreements about the use of lakes', whereas persons from Italy used the term to refer to 'a specific legal document which applies to one lake only'.

    Case (b)
    There was agreement across partners to set up new (cross-cultural) types of organizational units, but faced the problem of finding an appropriate designation.

    The new unit is designed to develop new ideas about lake management, implement and test these ideas and appropriate the results for dissemination to the public. French and Italian participants suggested the term laboratory for this unit, which the German-speaking colleagues declined, because they had connotations with this term which related to a small room in which only experiments were conducted. The latter group in turn suggested the term think tank, which the former group declined because of negative connotations (political jargon). Only the Slowene partners said that they could live with both terms.

    Case (c)
    After the first stage of project work, which was marked by mutual information about local situations and personal expertise, the coordinators of specified work packages (topics) agreed to collect further data from the rest of the consortium. But feedback of data and information remained scarce and negotiated knowledge largely "insular".

    One partner dispatched a questionnaire concerning 'harbor management', how well regulations were regarded, how much the authorities controlled these regulations, which were the observable negative effects of harbors on water quality. Addressees were highly motivated to participate in the exchange of data, but often communicated that the kind of questions was difficult to answer for them because they did not have comparable harbors at their lakes, had different or no regulations.

    Case (d)
    Experts from specific domains such as biology or measurement technology did not face serious problems in negotiating knowledge standards within their domain across cultural boundaries, but often had difficulty to make their expertise accessible to experts from other domains (non-experts from their point of view). Participation in discussion therefore often tended to remain domain-specific.

    Biologists addressed topics like 'macrophyte growth' which for partners from other domains was clearly remote.

    Case (e)
    To negotiate contingencies between different domains, which is one of the core project objectives, partners preferred workgroups within their own cultural and language community, to later report results to the plenum. However, it often turned out that the culture-specific groups discussed different aspects of an assigned topic, which were not immediately compatible with the results reported by others.

    Sample topic:
    Which are the contingencies between the commercial use of lakes, environmental conditions, etc.  and water quality? As indicated, contexts were hardly comparable across different lake instances.

    2.3. Interpretation of evidence

    Case (a)
    It is assumed that

    • for lack of corporate (project) knowledge the terms of the selected corporate language are still related to culture-specific knowledge (e.g. variant legal systems
    • compound designations such as lake contract are reduced representations of a concept and thus potentially ambiguous, because they do not make explicit the relationship between the two word components (cf. the German example Babyöl vs. Olivenöl)
    • corporate language designations are tacitly translated into the mother tongue to exploit their meaning.

    Case (b)
    It is assumed that

    • new concepts are formed, which obviously do not yet have a verbal address
    • if established terms are adduced, on the one hand access to the new concept may be facilitated (this process is called metaphoric extension, cf. the trivial case of mouse in the IT-domain)
    • if established terms are adduced, on the other hand unsuitable or even negative connotations are likely to emerge if used across language boundaries (which e.g. was why Mitsubishi failed to sell a car named Pajero in Spanish-speaking countries)

    Case (c)
    It is assumed that

    • the questions from a culture-specific island of knowledge rather than a draft of corporate project knowledge
    • the format of questionnaires did not leave sufficient room for expanding the scope of questions
    • the questions did not offer bridges to other domains of expertise

    Case (d)
    It is assumed that

    • project partners in this case were dealing with more or less global types of knowledge, which is subject to certain standardization processes in the respective scientific community
    • in expert-expert communication the use of highly condensed verbal representations (terms, acronyms, implicit use of language) on the one hand makes much sense, because being more economical (see Terminology in Fig. 1)
    • in communication between experts and non-experts condensed verbal representations obviously make expert knowledge more or less inaccessible  (see “shorezone functionality index SFI” in Fig.1)
    • without explicit reference to more familiar concepts expert knowledge cannot be embedded (sufficiently understood) (see Definition in Fig.1).

    Case (e)
    It is assumed that

    • making contingencies between different knowledge domains across cultural and language boundaries is a complex process which involves a number of different parameters
    • the parameters include those of quality (e.g. explicit relationships between knowledge types, domains, cultural contexts, see Fig.2), time (sequential constraints in developing corporate knowledge), as well as knowledge representation (e.g. multilingual knowledge representation)
    • without adequate procedural guidelines (e.g. identifying commonalities and culture-specific differences between observations) and technical tools the process of cross-cultural knowledge development is unlikely to be successful.


    Fig. 1: elements of knowledge - a topic to manage the different languages: designation, definition, documentation and terminology as well as structural information (find topic relations) and bridges between different domains of knowledge (Association)

    3. Specification of approach

    On the basis of conclusions of this type, which do not radically differ from conclusions obtained in other projects which the authors have accompanied, the procedure knowledge refinement was designed. This procedure is strongly tied to adequate tool support (multilingual concept mapping facilities) and is being applied and further developed in a growing range of knowledge domains, including the educational, service and production industry sector.

    The approach rests mainly on three basic pillars, a practical, a technical, and a theoretical one, with a number of subdivisions. The practical pillar consists of the following processes and involves the following technical facilities:

    • Raise awareness relative to expectable problems, e.g. by making explicit that
      • transmission of documents (including questionnaires) is no guarantee that knowledge has been successfully transmitted, or that the task can be fulfilled
      • cultural differences among communication partners are a potential source of conflict, but need not necessarily lead to conflict (often people see cultural differences as a resource)
      • language differences among communication partners are a potential source of conflict, but need not necessarily lead to conflict (it often needs differences in language use to make people aware of differences in knowledge).
    • Facilitate access to knowledge, e.g. by
      • defining conceptual knowledge units in a way that they can be sufficiently understood by involved persons, across cultural boundaries
      • representing conceptual knowledge units verbally in a way that they can be searched or communicated unambiguously across language boundaries
      • providing visual support to show knowledge structures and thus help involved persons gain orientation (to be satisfied by knowledge map)
      • providing visual support to make bridges (relationships) between different domains of knowledge (to be satisfied by knowledge map)
      • providing facilities to a knowledge resource which allow users to search for or contribute to knowledge resources by using their mother tongues or a corporate language (to be satisfied by a web application which allows to combine knowledge management, document management, as well as multilingual management).


    4. Specification of theoretical basis

    The theoretical basis rests on pillars of different fields of study, including linguistics, cross-cultural studies, and knowledge management. Beyond consideration of input from these fields, coordinated alignment of relevant input from these studies into a holistic model appeared indispensable. In a highly selective form resources like the following were adduced from the given domains.

    4.1. Linguistics

    Model building rested heavily on findings from linguistic semantics. What seems worth mentioning in this context is the clear distinction between language-driven (semasiological) and knowledge-driven (onomasiological) approaches to semantics (for details cf. Lyons 1995, Busse 2009). Even though linguistic semantics is predominantly oriented at the language-based approach, model procedures in multilingual contexts were found to require a knowledge-based approach. Only under this approach was it possible to design facilities for the machine-based generation of a multilingual glossary for selected knowledge domains, or for explaining project partners why translation is in many cases undesirable. Since a clear distinction between words and conceptual units is indispensable to provide such facilities and since in German-speaking countries Wort and Begriff are often confused, the given approach may however be difficult to understand in this community.

    Model building is also rested on findings in sociolinguistics and pragmatic linguistcs. These disciplines are mainly dedicated to relating different forms (varieties) of language to non-linguistic variables like social groups, professional groups, ethnic groups, types of knowledge,  intention, tasks  (for further details cf. Wardhaugh 2002, Stockwell 2002, Austin 2005). As experience has shown people without special linguistic training are often incapable of choosing an appropriate register or form of representation to assure that recipients can safely interpret the coded contents, which often seriously impairs the success of knowledge transfer or knowledge development, even if conducted within their native language.

    4.2. Cross-cultural studies

    Sufficient evidence is available on this topic with diverse types of focus, e.g. cross-cultural management (e.g. Apfelthaler 1999, or Holden 2002), cross-cultural communication (e.g. Bolten 2007, or Gudykunst & Mody 2002), or the specific form of knowledge communication (e.g. Reinhardt & Eppler 2004). Attempts were made to link findings from these resources with findings from linguistic semantics, sociolinguistcs and knowledge management to develop more effective solutions e.g.

    • for settings where experts need to communicate knowledge to non-experts across language, knowledge, and cultural boundaries
    • for making distinctions between culture-specific and global knowledge types which is crucially relevant for term and text translation: e.g. with culture-specific concepts terms should by no means ever be translated, neither should texts if addressed to different target groups (because needing different details of explanation)
    • for designing machine-based facilities like a multilingual glossary (which has turned out unfeasible if based on cross-linguistic associations of terms).

    4.3. Knowledge management

    In the given context basically two different resources were adduced for procedure and tool development: On the one hand expertise in knowledge representation (e.g. Sowa 2000, Reichenberger 2010); which were further developed e.g. with respect to multilingual representations, as well as expertise in cognitive psychology (e.g. Eysenck 2006). Insights from both domains of study were adduced for model development (e.g. Wieden 2006, 2010), which will be described separately.

    5. Specification of procedure

    To make culture-specific forms of knowledge accessible across cultural boundaries the authors in cooperation with software developers have implemented the procedure knowledge refinement in a new created software (syneris®). The procedure has been derived from the findings indicated above in coordination with the model of conceptual graphs (Sowa 2000). It was further developed into a semantic network design which is heavily indebted to insights obtained from application in various educational and business contexts. The resultant design does not generally follow conventional displays, which rely on principles of empty space (cf. Reichenberger 2010), but on systematic exploitation of the four dimensions 'up, down, left, right' for the purpose of super- and subordination, passive  and active associative links. The procedure has the following (recommended) sequential order, but is in part reversible according to the insights quoted above:



    (a) specify core topics   
    (b) assign specific topics   
    (c) assign composite topics   
    (d) append representations  
    (a) – (d)    

    - relevant conceptual categories
    - relevant conceptual structures
    - relevant conceptual network
    - (multilingually) documented knowledge
    - refined knowledge

    Representations can be liberally appended upon demand, including terms, comments, explanations, authentic documents, all of these liberally across an infinite number of languages. Technical support should include facilities for concept mapping and multilingual term and document management, to yield a knowledge map as an accessible product or forum for further discussion and to give rise to add-ons like web-based services or dynamic multilingual glossaries.

    Process (c) is typically the area where issues of concern are made explicit and process (d) is typically the area where appropriate responses are developed, ideally with respect to posed questions. like why? when? when not? how? ... In the given case questions and respective responses like the following may be formulated:

    • Why is it that knowledge refinement may facilitate successful knowledge communication across cultural boundaries? E.g. because
      • knowledge structures are invariant across languages,
      • individual topics can (ideally) be accessed in different languages
      • culture-specific topics can be defined intrinsically (specifying characteristics) and extrinsically (with respect to superordinate, subordinate, and coordinate topics),
      • composite topics can be explained on different levels of elaboration and in different varieties of language for different target groups, if needed.
    • When is it that knowledge refinement may not facilitate successful knowledge communication across cultural boundaries? E.g. if communication partners
      • are not fully aware of having particular problems
      • do not sufficiently share particular interests in a project


    6. Sample evidence from application of knowledge refinement in the SILMAS project

    Experience from project work shows that partners involved, all of them expert of some sort, did not start with the core concepts, but chose speficic topics (compounds) as entry points, and that core topics in a later stage had to be derived from the set of compounds and super-imposed.

    'Specific topics' include for instance:

    • conflict solving governance
    • causes (water quality problems)
    • instruments (water quality problems)

    As soon as relevant compound of specified topics had been sufficiently sub-categorized, attempts at relating them across hierarchy boundaries have been made. An example is indicated in the Fig. 2.

    Fig. 2: Semantic network for immediate neighbors of the topic 'WP5.3 sustainable leisure and professional fishing' 

    Fig. 2: Semantic network for immediate neighbors of the topic 'WP5.3 sustainable leisure and professional fishing'. The pop-up-windows shows the designation, definition (Erläuterung) and terminology (Fachwort) of the topic (work in progress status).

    The given selective display makes clear that the specific topic 'WP5.3 sustainable leisure and professional fishing'

    • is a type of 'WP5 Conflict Solving Governance' (up dimension)
    • has subtypes, like 'individual fishiung', or 'leishure activities' (down dimension)
    • has active associates, like 'has pilote sites with' (right dimension)
    • has passive associates, like 'is managed by SILA Annecy' (left dimension).

    Once the structures has been negotiated and converted into what may be called corporate project knowledge a number of representational processes were initiated, including complementation of topic designations or terms to familiarize the non-expert public with specific language uses or authentic documents for specific target groups.

    Remarkable efforts in working effectivness have been achieved since the common knowledge map, as shown in Fig.1, is accessible for project partners during negotiations via Skype or personal meetings.

    It should be noted, that comments, documents and other types of contents, but not terms and definitions, can also be appended to composite units (e.g. documents with results of measurements are added to the composite unit “macrophyte density - can influence  - fish stock”). For successful working with knowledge refinement in a cross-cultural project the authors have used a specific software bundle for knowledge design and information management (syneris®) because it is necessary for understanding to make new corporate knowledge structures visible particularly in a multilingual environment.

    Using syneris® it was much easier to explain all project partners involved how the given procedures support the development of complex knowledge and how the tool and its use may assures sustainability even after the project has ended. At the current state of development the first results have clearly emerged, e.g.

    • a common knowledge base as web application
    • which is open to further discussion and development with a clear conceptual structure (e.g. during skype conferences)
    • which specifically invites comments to complex (composite) knowledge units
    • which is increasingly enriched with new contents in different languages

    and thus increasingly allows dissemination of standard knowledge to non-project stakeholders (e.g. political decision makers across the Alpine space).

    7. Conclusion

    Available experiences suggest that in the context of cross-cultural knowledge development problems of a specific type emerge to which single-source solutions hardly provide adequate responses. But even multi-source systemic solutions, in which findings from different domains are included, do not appear to be easily applicable, at least not without considerable effort on the part of the recipients, since such approaches are yet little familiar (and still think that translation will solve the problem of cross-cultural knowledge communication and development) and since broadly available tools are hardly geared for such demands. In this context one should also note official recommendations of the educational sector of the EU, which strongly recommends multilingual competences as an educational goal for all European citizens, for good reasons. In this project, such competences would have helped a lot. But what motivates continuation of the respective efforts is that partners in this (and other) projects are increasingly prepared to enter into complex mental activities of this type, if they get adequate IT support, and because the outcome promises considerable added value – cost-benefit relations seem to be well in balance.


    • Apfelthaler, G. (1999) Interkulturelles Management. Wien, Manz.
    • Austin, J. (2005) How to do things with words. Cambridge (Mass.): Harvard University Press,
    • Bolten, J. (2007) Interkulturelle Kompetenz. Erfurt 2007 (Landeszentrale für Politische Bildung).
    • Busse, D. (2009) Semantik. Paderborn, Wilhelm Fink.
    • Eysenck, M. (2006): Fundamentals of Cognition. Hove, Psychology Press.
    • Gudykunst, W., Mody, B. (eds) (2001) Handbook of International and Intercultural Communication. London: Sage Publications.
    • Holden, N. (2002): Cross-cultural management. London: Prentice Hall.
    • Lyons, J. (1995) Linguistic semantics. Cambridge, Cambridge University Press.
    • Reichenberger, K. (2010) Kompendium Semantische Netze. Heidelberg, Springer
    • Reinhardt, R.; Eppler, M. (eds.) (2004). Wissenskommunikation in Organisationen. Berlin, Springer.
    • Sowa, J.(2000) Knowledge representation. Pacific Grove: Brooks/Cole.
    • Stockwell, P. (2002): Sociolinguistics. A Resource Book for Students. London, Routledge.
    • Wardhaugh, R. (2002) An introduction to sociolinguistics. Oxford, Blackwell.
    • Wieden, W. (2006) Corporate linguistics: A knowledge management approach to language. Arbeiten aus Anglistik und Amerikanistik 31:2, 185-207.
    • Wieden, W. (2010): Wissensmanagement und Terminologie. In: eDITion, Fachzeitschrift für Terminologie 7/1, 8-15.
    [Standard] Namensnennung 3.0 Deutschland - Weitergabe unter gleichen Bedingungen 3.0 Deutschland
    Lizenziert unter einer Creative-Commmons Lizenz


Das Kommentarsystem ist zurzeit deaktiviert.


Dieser Beitrag ist den folgenden Themengruppen zugeordnet


Dieser Beitrag ist den folgenden Schlagworten zugeordnet