Abstract
Often, translation procedures
are applied to ensure that the target language wording is as near as possible to
that in the source language. This results in ignoring emic meaning of concepts
in both the source and the target contexts. This paper investigates the use of
semantics and translation for generating insights into shared, etic meaning and
into emic facets of meaning.
Introduction
A basic issue in cross-cultural research is the determination of
whether the concepts used have similar meaning across the social contexts
surveyed. Concepts used in cross-cultural business research predominately come
from a particular source culture (i.e. US) and were developed in a particular
linguistic context (i. e. American English). Conceptual equivalence with target
contexts should be better assessed before using popular US management
constructs which have often been imported by foreign researchers in
significantly different linguistic and cultural contexts with much enthusiasm
but little care as to their cross-border transportability.
Language is rarely considered in cross-cultural business
research as a resource for the in-depth understanding of conceptual equivalence
issues. This is probably due to the predominance of English and to the obvious
fact that management researchers are not linguists.
Language is viewed rather as a constraint in the research process which can
typically be overcome by appropriate translation procedures. However, language
matters: it has been recently emphasized as being important
for consumer's categorization, judgment and choice (Schmitt and Zhang, 1998);
it has also been shown to be an important indicator of cultural distance (West and Graham,
1998).
This paper investigates the use of translation (back-translation,
parallel translation and combinations) for generating insights into facets of
meaning shared across languages and facets of meaning specific to each culture.
Examples of conceptual equivalence are taken from the cross-cultural management
literature in consumer behaviour, organization, and accounting, in order to
illustrate how meaning differentials can be assessed by using a semantic and
not a lexical approach to the use of dictionaries and translation procedures.
Language-Free International
Research: A Scenario For Cross-Cultural Ignorance
Problems of conceptual
equivalence are frequent when testing the influence of certain constructs
on behaviour. For instance, the hypothesis of the cognitive theory that
consumers do not willingly behave contradictorily may hold true in the United
States while not being applicable to some other countries. The Western belief
in individual, informed, linear, and rational decision-making, is, for
instance, a reason for the lack of conceptual equivalence in the areas of
information search, evaluation, and choice of alternatives. Geertz (1983)
explains that "The Western conception of a person as a bounded, unique,
more or less integrated, motivational and cognitive universe, a dynamic center
of awareness, emotions, judgement and action, organized in a distinctive whole
. . . is, however incorrigible it may seem, a rather peculiar idea, within the
context of world's cultures." (1983, p. 59)
Three basic elements of language may help in assessing conceptual
equivalence: (1) words and their specific meanings, both literal and
figurative; (2) words assembled in sentences and text since languages work as
codes that must be "translated" into other linguistic codes, if
source and target are different; (3) worldviews expressed by speakers of a
particular language. Because language is used daily in local contexts,
it is reflective of local knowledge in the Geertzian sense. Comparing across
cultures, without awareness of language differences may result in biased and
impoverished findings. Faucheux (1976, p. 271) phrases this issue in a
definitive way: "The hope of doing language-free cross-cultural research
is vain".
Basic concepts such as autonomy, leadership, friendliness, motivation,
honesty, satisfaction, authority, well-being, etc. are often used in management
and marketing research questionnaires as if they had universal meaning.
However, words express meaning shared only in the cultural and language group,
especially when they relate to perceptions and to interactions. For instance,
Shenkar and Von Glinow (1994) note that, when interviewing Chinese employees, a
word such as "autonomy", which is a key concept in organizational
research, "cannot be adequately translated into Chinese, or that alternate
Chinese terms, such as "right of self-determination" (zi zhu quan) convey a quite different
meaning." (p. 67).
Words are seemingly universal when one looks at a dictionary. However,
if concepts are understood differently, there is great chance that findings in
cross-national research are to a large extent flawed by differences in
understanding. Therefore, we should investigate conceptual equivalence with greater
care when designing cross-cultural questionnaire surveys or psychometric
scales. This makes sense for the theoretical concepts used in the grounding of
empirical research as well for the concepts used in interviewing, especially
when interviews are conducted through interpreters. Concepts cover a broad
range of situations (e.g. waiting), qualities (e.g. beauty), institutions (e.g.
board of directors), techniques (e.g. cost accounting), actions (e.g. working),
interactions (e.g. obeying), outcomes (e.g. success, satisfaction), etc. They
suffuse the textual part of research instruments.
There are basically two types of concepts: the first category is that
of general-purpose concepts used by the whole language group; the second
category is that of ad hoc business
concepts used by researchers in the framework of their theories (e.g. hygiene
factors, self-actualization, role ambiguity, etc.). General purpose concepts
are often conceptually equivalent at the broad level but not when understood
presicely, in context. For instance, concepts which have a moral connotation,
such as patronage, can be viewed differently in the Anglo world in contrast to
the Mediterranean world where a system structured around kin-based loyalties,
involving patterns of reciprocal obligations, provides the natural basis for a
patron-client relationship (Chapman, 1992), elsewhere considered somewhat
immoral. Shame and guilt, which are key concepts in intercultural interactions
(Tangney, 1995), are understood differently according to the cultural context
and need to be examined for conceptual equivalence across cultures when one
considers for instance the role played by shame in regulating behaviour in
Asian versus Western cultures (Ha, 1995).
The second category, that of business concepts, is even more threatened
by a loss of its source meaning when transferred into another context since its
national and cultural base is much narrower than that of general concepts. For
example the concept of "consumer ethnocentrism" (Shimp and Sharma,
1987), and the associated psychometric scale (CETSCALE), refer to situations
where consumers associate the purchase of foreign products to the potential
loss of jobs in domestic industry and consequently refrain from buying foreign.
It applies fairly well in developed countries to regions in industrial decline
where people resent foreign-made products because they feel that jobs are
threatened by competitive imports. This holds true in European countries; that
is, probably why Netemeyer et al.
(1991) have successfully replicated the CETSCALE in other developed countries.
But the context underlying the concept of consumer ethnocentrism is not found
in most developing country where there is no local manufactures competing
against imported products. One may wonder, for instance, how item 14 in the
scale ("Foreigners should not be allowed to put their products on our
markets") can be understood in countries where supply for a number of
goods is objectively dependent on imports.
Concepts used in cross-cultural research have a linguistic source
context. Generally, cross-cultural studies start from US-based research
instruments, especially psychometric scales (see for instance Bearden et al.
1993) and interview guides. They use textual elements originally conceived, written
and published in English. English is an ideal language for business: its
grammar is fairly simple; it is precise, action and facts oriented. However,
the lack of diversification of source contexts in terms of language and
culture, creates an effect of pensée
unique, that is, particular cultural assumption such as doing orientation,
future orientation, individualism, rational behavior, etc. suffuse the whole
research process.
The constant recourse to English as lingua
franca tends to blur conceptual differences across linguistic contexts. A
contrast must constantly be made between the worldviews expressed by English
and the worldviews expressed by other languages. The importance of language is
highlighted by Claude Levi-Strauss in the introduction of his Anthropologie Structurale: "When
building this volume, I came up against a difficulty on which I must draw the
attention of the reader. Several of my articles have been written directly in
English, it was necessary therefore to translate them. Yet, while doing this
work, I have been impressed by the difference in tone and structure between the
texts conceived in either language. From this results heterogeneity which, I
fear, jeopardizes the balance and the unity of the work." (1974, pp. 7-8).
Language reveals worldviews in the form of underlying assumptions, inferred key
situations, typical interactions, views of adequate inputs, desirable outcomes,
and appropriate processes which link inputs to outcomes.
Thus language, and especially semantics, are excellent tools for the
study of cross-cultural equivalence. However, mere lexical equivalence is
generally sought, ignoring other dimensions of translation equivalence
(idiomatic, experiential and grammatical-syntaxic equivalence), with the
unfortunate consequence that conceptual equivalence with the target context is
not really investigated. Back-translation is applied to ensure that the target
language wording is as near as possible to meaning in the source language. This
appears as a necessary technicality of cross-cultural research, needed for
academic research to be considered as acceptable when being reviewed. The
technical use of back-translation corresponds to language-free research because
it is based on the universalistic assumption that meaning equivalent to source
can be found in all target languages, an assumption which denies the obvious
capacity of particular languages to create unique meanings. This results in
ignoring emic meaning of seemingly similar concepts both in source (US) and
target (non-US) contexts.
For example the German institution of Aufsichtsrat will be typically translated into ‘supervisory board’.
While this translation is not as such bad (it is even an accepted equivalent in
the European Union) it loses its emic meanings, related to who supervises what for which purposes. In fact, the presence in
the Aufsichtsrat of employees
associated to strategic decision making makes it a German emic concept. Those
not accustomed to employees being close satkeholders in corporate governance
will not see this crucial facet of the concept. Similarly, concepts such as
role ambiguity and role conflict may seem to find easily equivalents in most
languages which have both the concept of "role" (as in playing a role
on a stage) and the concepts of "ambiguity" and "conflict".
However, the compound of role and ambiguity may seem strange in many target
linguistic contexts because people do not think spontaneously of the (lack of)
clarity of assigned tasks.
Respondents are exposed to research instruments which have been
transposed with too basic "linguistic" precautions, namely that
lexical equivalents of the source words are found in the target language and
that the wording is roughly appropriate. However, informants often do not feel
at ease with some of the key concepts used in transposed research instruments
and do not dare to express it openly in the fear of looking ignorant or
unsophisticated. Mail questionnaires or face-to-face interviews with close-end
questions do not allow feedback from the respondents. As a consequence,
seemingly etic meaning is imposed on the respondents with the result that emic
meaning from the source language is not understood and emic meaning from the
target language cannot emerge from the data collection process. Typically this
results in high levels of measurement error and low reliability of measurement
variables in target as compared with source context.
The Use of Translation
Procedures to Assess the Level
of Cross-Cultural Equivalence
This paper adopts an emic-etic hybrid position. The assumption is that
there are both areas of shared meaning and facets of meaning specific to each
culture (see figure 1). Pure emic researchers would criticize such a position
arguing that etic and emic positions are irreconcilable. The existence of
dictionnaries and their actual use in translation processes are, however,
proofs that many meanings are shared across languages. On the other hand, the
limitations of dictionnaries is also proof that much is language-specific. Consequently,
the issue becomes that of disentangling emic and etic meanings in apparently
similar concepts.
Concepts such as opportunism, leadership, motivation, etc. may have
similar facets in different language/cultures but one of these facets may be
emically emphasized in one culture while it is almost ignored in another
culture. For example, when working on the Globe project with a questionnaire on
leadership administered in more than 60 countries, the phrase “charismatic
leadership” could not be translated into Iranian[1]; in fact, this
facet of leadership is so obvious to Iranians that charismatic leadersip, far
from being unknown, is fully assimilated with the concepts of leader and
leadership, thereby leaving in the dark other dimensions of the source concept
linked to role or expert authority (see Dorfman, 1996). The underlying
dimensions of a construct may not be equivalently weighted or articulated in
the total construct across language/cultures.
Comparison across languages enables to generate insights into possible
conceptual differences. Meaning differential can be investigated across
languages for apparently similar words and utterances in the following areas:
(1) multiple meanings of a word; when a concept in the source language has
mutiple meanings in the target language, it means that the target language
offers a richer description (e.g. Spanish is more detailed in describing beauty
than English whereas English is more accurate in describing waiting); (2)
central, most important meaning (modal meaning); (3) frequency of use of
certain words; (4) latent value judgements put on words, positive and/or
negative in particular contexts; pejorative meanings indicating a normative
orientation; (5) meaning subtleties: context of use of words and experiential
aspect (for this, insert individual words in sentences which are culturally
typical); (6) idiomatic expressions; (7) figurative and proper senses: in
English language figurative meanings are rare, whereas other languages such as
German and French make a larger use figurative meanings.
FIGURE 1
Facets of Meaning Across
Languages/Cultures
The main limitation to the lexical use of back-translation is that the
source language "closes" the possible area of meanings covered.
Therefore, back translation does not enable to discover emic meaning in the
target language because this meaning does not exist in the source language.
Parallel translation combined with back-translation may, by chance, let emic
meaning emerge because of different insights among independent translaters.
A good example of how
back-translation can be used as an investigation instrument rather than a mere
technical device is given by Levine's (1988) comments about research into
Brazilian versus US concepts of punctuality. He takes the example of the translation
into Portuguese of a questionnaire containing the verb "to wait":
“Several of our questions were concerned with how long the respondent would wait for someone to arrive versus when
they hoped the person would arrive
versus when they actually expected the
person would come. Unfortunately for us, it turns out that the term to wait, to hope and to expect are all typically translated
as the single verb esperar in
Portuguese. In many ways our translation difficulties taught us more about
Brazilian-Anglo differences in time conception than did the subjects' answers
to the questions.” (Levine, 1988, pp. 48-9). There is a sort of continuum
across languages in the accuracy of description of the waiting phenomenon. The
French language, which lies somewhere between English and Portuguese in terms
of temporal accuracy, uses two words: attendre
(to wait) and espérer (to hope).
To expect has no direct equivalent in French and must be translated by a
lengthy circumlocution (compter sur
l'arrivée de).
However, the best way to generate insights as to conceptual equivalence
is for a multilingual/multicultural team of researchers to use multiple source
and target languages and to allow for decentering in the translation process
(Campbell and Werner, 1970). For this, researchers should start from a rather
broad conceptual area which is almost etic because it corresponds to common
problems in all languages/cultures. The procedure then consists in asking
native speakers of the languages and cultures concerned to elicit key-words
related to the broad conceptual area (e.g. authority in figure 2) as in
cognitive mapping (Eden, 1988). A collaborative team of cross-cultural
researchers then chooses the most frequently cited terms related to the concept
in each language/culture and back-translates them; the team may then debrief to
generate etic and emic conceptual dimensions. The frequency of citation of
certain facets enables to assess the degree of emicity. Figure 2 shows how to
generate insights into conceptual equivalence of authority and autorité.
FIGURE 2
Double Back-translation to
Generate Insights into Conceptual Equivalence
Generating a Cross-Cultural
Inventory of a Concept's Facets: The Case of ‘Trust’
A way for understanding which facets can be used jointly in different
language versions of a common research instrument is to generate a
cross-cultural inventory of a concept's facets. The concept of trust is now
more and more widely applied, especially in the areas where intercultural
interaction are frequent such as conflict resolution in joint ventures
(Sullivan and Peterson, 1981, 1988) or contract enforcement (Choi, 1994). Trust
is considered conceptually equivalent across languages and cultures and the
linguistic equivalence of the concept of trust across cultures is rarely
questioned. Fukuyama, contrasting high-trust (Germany, Japan) and low-trust
societies (China, France), defines trust as "the expectation that arises
within a community of regular, honest, and cooperative behavior, based on
commonly shared norms, on the part of other members of that community"
(1994, p. 26).
By looking at how trust is expressed in four languages (English,
French, German, and Japanese), we can derive some insights on which aspects of
the concept are put in the forefront by the corresponding cultures. The English
concept of Trust is the reliance on
and confidence in the truth, worth, reliability of a person or thing. Reliance
is central in the Anglo-Saxon concept of trust, that is why the legal
institution of trust has been highly developed in the Common law tradition
whereas it is non existent in the Roman-Germanic tradition. The German concept
is based on two verbs: trauen and vertrauen, both of them meaning
literally to "trust". However, Germans use the first form, trauen, mostly in the negative sense,
"Ich traue Dir nicht"
("I do not trust you"), and the second in the positive sense "Ich vertraue Dir" ("I trust
you"). The prefix Ver indicates
a transformation; this documents the worldview which lies behind the German
concept of trust: (1) the initial position is distrust (2) only after a
favourable change has occurred can trust be established. The German Vertrauen evokes the process of changing
an initial trustless situation into one in which the parties have built
confidence into the relationship.
The French notion of confiance,
similarly as in other Romance languages, is
based on the Latin confidentia , a
compound of cum (with, shared) and fides (faith, belief): the notion of
sharing common beliefs, religion or group membership, is central to the Latin
concept of trust. Confiance assumes
similarity as a pre-requisite for a trustful relationship.There is only one
word in French for "trust" and "confidence" suggesting the
relative unsophistication of the French low-trust culture in comparison to
English and German languages (see figure 3). Luhmann (1988), a German writing
in English, makes a sharp distinction between "confidence", that is, a
broad feeling that our expectations will not be disappointed, and
"trust" which requires engagement of the self and rational
evaluations of risks, a difference which does not exist in French. Finally, the
Japanese word for trust is shin-yô meaning
literally: sincere business; it is based on a compound of shin, a character for "sincerity" and yô which means literally "something
to do, a business," (Sakade, 1982). The Japanese shin-yô insists on the orientation of trust towards the future, a
common enterprise and the sincere expectations of the parties.
FIGURE 3
Facets of the Concept of
Trust
The core meaning of the concept of trust is roughly the same
cross-culturally, but different languages favour different facets of the
concept. Concepts of trust largely overlap across culture but the dominant
emphasis may be revealed by linguistic investigation, at least as insights to
be verified. Insights from the four language/cultures reviewed enable to derive
the facets of the concept of trust: (1) trust is reliance on and confidence in
people, words and things; (2) trust is inseparable from distrust: since obvious
showing of distrust is detrimental to the establishment of trust, every culture
has to deal with the paradox of their inseparability; (3) trust is about
sharing common faith, beliefs, possibly education or group membership; (4)
trust is directed to common and future achievements.
Searching for Meaning through
‘Untranslation’: Illustrations
Untranslation is the use of back-translation to generate insights into
conceptual equivalence. In addition to the assessment of emic and etic meanings
as in figure 1, it allows to identify faux
amis (deceptive cognates) and hidden “emicity” in seemingly etic concepts. Faux amis are words that look the same
and have in fact different meanings: a fully different meaning may be
attributed to a lexically equivalent concept. For instance, the concept of
"household", widely used in market research, is subject to possible
inequivalence: Mytton (1996) cites the case of Northern Nigeria where people
often live in large extended family compounds or gida which are difficult to compare with the prevalent concept of
household which reflects the living unit of a nuclear family.
Rather than to force source meaning into the target context, back
translation can be used to assess the degree of divergence in interpretation of
basic concepts across linguistic contexts. It is then easier to base choices as
to whether source concepts can be transposed cross-culturally on actual meaning
equivalence rather than spurious lexical equivalence. For instance, hair seems
an etic equivalent to the French cheveu;
however, cheveu is only the hair on
the top part of the head, on the skull (hidden emicity); thus it excludes beard
hair which are in the category "poil"
(poil de barbe) and hair on the body. Poil
has a masculine connotation while cheveu is
rather neutral. The French, product universes of hair care (cheveu) and body care (poil) are rather strictly separated. It
is difficult to conceive of a combined body- and hair-care product. Figure 4
presents a case where English has one concept for two concepts in French and, a
contrasting case where English has two concepts for only one in French where faire both means "to do" and
"to make".
FIGURE 4
Different Conceptual
Divisions Across Languages
Another use of untranslation is to identify emic meaning which contains
none the less some universal meaning, e.g. goodwill,
trust as a legal institution, ringi, or kaizen. Certain concepts remain untranslated such as guanxi because of the belief that they
are purely Chinese-emic. While this may be true to a certain extent,
"social networking" is a proximate term. Broadly considered business
networks are a fairly universal reality. The very notion of guanxi can to a large extent be
considered as a Chinese, and more broadly an East Asian form of networking,
consisting in maintaining relationships with the appropriate organizations and
individuals within these organizations. The Chinese guanxi corresponds to Kankei
in Japan and Kwankye in Korea, that
is, after-hours socialization which become important forums for meeting and
convincing key decision makers in a socially more comfortable atmosphere (Tung,
1996). Guanxi mixes social behaviour
and business practices in a complex set of disinterested and interested
personal interactions. It is not necessarily directed at short term results and
consists in an investment in relationships which may or may not be called upon
in the future. Firms engaged in a connected set of companies, called guanxihu, do their best to avoid
embarrassing a business partner which experiences temporary financial problems.
At a broad level, the Chinese concept of Guanxi shares some common traits with
the Western concept of networking, especially its emphasis on the continuity of
business relationships between firms engaged in cooperative rather than
competitive behaviour. There are, however, some significant differences which
Luo and Chen (1997, pp. 3-4) explain as follows: ". . . guanxi primarily relates to personal,
not to corporate, relations, and exchanges that take place amongst members of
the guanxi network are not solely
commercial, but also social, involving the exchange of renqing (social or humanized obligation) and the giving of mianzi (face in the society), or social
status. This feature often leads guanxi to
be named "social capital". In contrast, networking in Western
marketing and management literature is the term primarily associated with
commercially based corporate-to-corporate relations. Because of this
difference, many Western business people are often in danger of overemphasizing
the gift-giving and wining-and-dining components of a guanxi relationships, thereby coming dangerously close to crass
bribery or to be perceived as "meat and wine friends" which is a
Chinese metaphor for mistrust."
Another example of an emic concept which is used world-wide is that of
"goodwill". In southern European languages, the term
"goodwill" is used directly in English because there is no equivalent
for so intangible an asset as goodwill. In countries where accounting is mostly
framed by law and directed at the fiscal authorities, concepts which emphasize
book value for fully intangible assets seem relatively strange. The conceptual
equivalence of "goodwill accounting" has been discussed by Brunovs
and Kirsch (1991) in the framework of the harmonization of international
accounting standards. They conclude that the most significant conceptual
difference is that between the United Kingdom and Ireland on the one hand and
the other countries examined on the other hand. "The UK. standard
advocates that goodwill be eliminated immediately on acquisition by write-off
directly against reserves, whereas the other countries require goodwill to be
carried forward in the balance sheet and systematically amortized against
income over the estimated useful life of the goodwill. This difference is not
just an inconsistency, but rather represents the chasm of conceptual division
as to the underlying question of what goodwill represents." (1991, p. 155)
Conclusion
Any cross-cultural research research in business should include a
preliminary phase of conceptual equivalence assessment. Core etic meanings and
significant emic meanings should be assessed before the cross-border transfer
of research instruments. Variation is obviously not unlimited. I do not pretend
that all should be checked following the rather heavy approaches suggested in
this paper, but at least the conceptual equivalence of some key concepts should
be better investigated across major linguistic contexts and the findings shared
among researchers. A personal suggestion is that part of the cross-cultural
research agenda in business studies should focus in the future on the
progressive emergence of a corpus of etic and emic meanings for about 100 key
business concepts across major linguistic contexts.
References
Bearden, William O., Richard G. Netemeyer and Mary F. Mobley (1993) Handbook
of Marketing Scales, Newbury Park, CA: Sage Publications.
Brunovs, Rudolf, and Robert J. Kirsch (1991), "Goodwill accounting
in selected countries and the harmonization of international accounting
standards", Abacus, 27(2), 135-161.
Campbell, D.T. and O. Werner (1970), "Translating, working through
interpreters and the problem of decentering", in R. Naroll and R. Cohen
(eds.), A Handbook of Method in
Cultural Anthropology, The Natural History Press: New York, 398-420.
Chapman, Malcolm (1992), "Social anthropology and international
business - Some suggestions", Paper presented at the Conference of the
Academy of International Business, November 20-22.
Choi, C.J. (1994), "Contract enforcement across cultures", Organization
Studies, 15, 673-82.
Dorfman, Peter W. (1996), "International and cross-cultural
leadership", in Betty Jane Punnett and Oded Shenkar editors, Handbook
for International Management Research, Cambridge, Mass.: Blackwell,
267-349.
Eden, Colin (1988), Cognitive Mapping as a Visionary Tool : Strategy
Embedded in Issue Management, Alfred Houle seminar, Université Laval, Québec
March.
Faucheux, Claude (1976), "Cross-cultural research in experimental
social psychology", European Journal of Social Psychology, vol. 6,
no. 3, 269-322.
Fukuyama, Francis (1994), Trust: The Social Virtues and the Creation
of Prosperity, New York: Free Press.
Geertz, Clifford. (1983), Local Knowledge, New York: Basic
Books.
Ha, Francis Inki (1995), "Shame in Asian and Western
cultures", American, Behavioral Scientist, 38(8), August,
1114-1131.
Levi-Strauss, Claude (1974), Anthropologie Structurale, Paris:
Plon.
Levine, Robert V. (1988), "The pace of life across cultures",
in Joseph E. McGrath ed., The Social Psychology of Time, Newbury Park,
CA.: Sage Publications, 39-60.
Luhmann, Niklas (1988), "Familiarity, Confidence, and Trust"
in Diego Gambetta(ed.), Trust: Making and Breaking Cooperative Relationships,
Oxford: Blackwell, 94-107.
Luo, Yadong and Min Chen (1997), "Does guanxi influence firm
performance?", Asia Pacific Journal of Management, 14, 1–16.
Mytton, Graham (1996), "Research in new fields", Journal
of the Market Research Society, 38(1), pp. 19–32.
Netemeyer, Richard G., Srinivas Durvasula, and Donald R. Lichtenstein
(1991), "A cross-national assessment of the reliability and validity of
the CETSCALE", Journal of Marketing Research, 28 (August), 320-327.
Sakade, Florence, general editor (1982), A Guide to Reading and
Writing Japanese, Tokyo: Charles E. Tuttle.
Schmitt, Bernd H. and Shi Zhang (1998),
"Language Structure and Categorization: A Study of Classifiers in Consumer
Cognition, Judgment, and Choice", Journal of Consumer Research, 25
(September), 108-122.
Shenkar, Oded and Mary Ann Von Glinow (1994), "Paradoxes of
organizational theory and research: Using the case of China to illustrate
national contingency", Management Science, 40(1), January, 56-71.
Shimp, T.A. and S. Sharma (1987), "Consumer ethnocentrism :
Construction and validation of the CETSCALE", Journal of Marketing
Research, 26 (August), 280-289.
Sullivan, Jeremiah, and Richard B. Peterson (1981), "The
relationship between conflict resolution approaches and trust: A cross-cultural
study", Academy of Management Journal, 24(4), 803-15.
Sullivan, Jeremiah, and Richard B. Peterson (1988), "Factors
associated with trust in Japanese-American joint ventures", Management
International Review, 22(2), 30-40.
Tangney, June Price (1995), "Recent advances in the empirical
study of shame and guilt", American, Behavioral Scientist, 38(8),
August, 1132-1145.
Tung, Rosalie L. (1996),"Negotiating with East Asians", in
Pervez N. Ghauri and Jean-Claude Usunier (eds), International Business
Negotiations, Oxford: Pergamon/Elsevier Science, 369-81.
West, Joel and John L. Graham (1998), "Language's Consequences: A
Test of Linguistic-Based Measures of Culture Using Hofstede's Dimensions",
Academy of Management Best Paper Proceedings, San Diego.