INTEGRATING EMIC and ETIC INSIGHTS ABOUT CULTURE and Lustice JUDGMENT

^ Academy ol Maaagem»nt Review 1999, Vol. 24. No. 1781-796.

VIEWS FROM INSIDE AND OUTSIDE: INTEGRATING EMIC AND ETIC INSIGHTS ABOUT CULTURE AND lUSTICE JUDGMENT

MICHAEL W. MORRIS Stanford University

KWOK LEUNG Chinese University of Hong Kong

DANIEL AMES University of California at Berkeley

BRIAN LICKEL University of California at Santa Barbara

We analyze forms ol synergy between emic and etic approaches to research on culture and cognition. Drawing on the justice judgment literature, we describe dynamics through which the two approaches stimulate each other's progress. Moreover, we delineate ways in which integrative emic/etic Irameworks overcome limitations of narrower frameworks in modeling culture and cognition. Finally, we identify advantages of integrative frameworks in guiding responses to the diverse justice sensitivities in international organiiKations.

In the study of cognition in organizations, and tages of frameworks integrating emic and etic in social science more broadly, there are two accounts—both as middle-range theories of cul- long-standing approaches to understanding the ture and cognition and as applied guides to role of culture: (1) the inside perspective of eth- responding to diverse justice concerns in inter- nographers, who strive to describe a particular national organizations. culture in its own terms, and (2) the outside perspective of comparativist researchers, who at- tempt to describe differences across cultures in EMIC AND ETIC PERSPECTIVES terms of a general, external standard. Pike (1967) The emic and etic perspectives have equally designates these approaches the emic and etic long pedigrees in social science. The emic or perspectives, respectively, by analogy to two inside perspective follows in the tradition of approaches to language: phonemic analysis of psychological studies of folk beliefs (Wundt, the units of meaning, which reveals the unique 1888) and in cultural anthropologists' striving to structure of a particular language, and phonetic understand culture from "the native's point of analysis of units of sound, which affords com- view" (Malinowski, 1922). The etic or outside per- parisons among languages. The emic and etic spective follows in the tradition of behaviorist perspectives are often seen as being at psychology (Skinner, 1938) and anthropological odds—as incommensurable paradigms. In this approaches that link cultural practices to exter- article we argue that these two approaches to nal, antecedent factors, such as economic or culture are complementary. Drawing on the jus- ecological conditions, that may not be salient to tice judgment literature, we delineate forms of cultural insiders (Harris, 1979). synergy between the two research perspectives that go beyond those identified previously (e.g., The divide between these two approaches Berry, 1990; Brett, Tinsley, lanssens, Barsness, & persists in contemporary scholarship on culture: Lytle, 1997). We first analyze ways in which emic in anthropology, between interpretivists (Geertz, and etic research programs have stimulated 1976, 1983) and comparativists (Munroe & Mun- each other's progress. Then we analyze advan- roe, 1991), and in psychology, between cultural psychologists (Shweder, 1991) and cross-cultural 781 782 Academy oi Management Review October psychologists (Smith & Bond, 1998). In the liter- Etic and emic approaches traditionally have ature on international differences in organiza- been associated with differing research meth- tions, the divide is manifest in the contrast be- ods. As Table 1 summarizes, methods in emic tween classic studies based on fieldwork in a research are more likely to involve sustained, single culture (Rohlen, 1974), as opposed to sur- wide-ranging observation of a single cultural veys across many {Hofstede, 1980). Likewise, in group. In classical fieldwork, for example, an the large body of literature on organizational ethnographer immerses him or herself in a set- culture, there is a divide between researchers ting, developing relationships with informants employing ethnographic methods (Gregory, and taking on social roles (e.g., Geertz, 1983; 1983; Van Maanen, 1988) and those who favor Kondo, 1990). Yet, emic description also can be comparative survey research (Schneider, 1990). pursued in more structured programs of inter- The conceptual assumptions with which Pike view and observation (e.g., Goodenough, 1970). (1967) defined the emic and etic dichotomy are Methods in etic research are more likely to summarized in Table 1. Emic accounts describe involve brief, structured observations of several thoughts and actions primarily in terms of the cultural groups. A key feature of etic methods is actors' self-understanding—terms that are often that observations are made in a parallel manner culturally and historically bound. For example, across differing settings. For instance, matched emic studies of justice perceptions in North samples of employees in many different coun- American organizations today might center on tries may be surveyed to uncover dimensions of such constructs as "age-ism" and nondiscrlmi- cross-national variation in values and attitudes nation, whereas studies of Japanese workplaces (e.g., Hofstede, 1980), or they may be assigned to might be couched in qualitatively different con- experimental conditions in order to test the mod- structs, such as amae and gimu (see Kashima & erating influence of cultural setting on the rela- Callan, 1998). In contrast, etic models describe tion among other variables (e.g., Earley, 1989). In phenomena in constructs that apply across cul- sum, although the two perspectives are defined tures. For example, a country's level on the cul- in terms of theory, rather than method, the per- tural dimension of individualism-collectivism spectives lend themselves to differing sets of might be linked to the prevalence with which methods.^ managers reason about justice in terms of the Given the differences between emic and etic equity rule (i.e., rewards received should be pro- approaches to culture, it is not surprising that portional to contributions). researchers taking each perspective have ques- Along with differing constructs, emic and etic tioned the utility of integrating insights from the researchers tend to have differing assumptions other tradition. A common tendency is to dismiss about culture. Emic researchers tend to assume insights from the other perspective based on that a culture is best understood as an intercon- perceived conceptual or methodological weak- nected whole or system, whereas etic research- nesses (see reviews of this tendency in particu- ers are more likely to isolate particular compo- lar research areas by Harris, 1979, and Martin & nents of culture and state hypotheses about Frost, 1998). On one side, emic accounts based their distinct antecedents and consequences. Al- on ethnographic observation are often dis- though, of course, the emic/etic contrast is, in counted on the basis of inconsistency across practice, a continuum, this dichotomy has played a central role in the metatheory debates in many social science disciplines (see Head- 1971). Although there may be a correlation in some research land, Pike, & Harris, 1990).' areas between the emic versus etic perspective and orientations toward control (e.g., in studies of "organizational culture"; Martin & Frost, 1996), there is no necessary link and no strong correlation in the literature on national culture— ' Some scholars have used the terms emic and etic in our focus. ways that depart irom Pike's definitions (see Headland et al.. ^The association between perspectives and methods is 1990). A narrower usage refers to the contrast between cul- not absolute. Sometimes, in emic investigations of indige- ture-specific versus culture-general constructs. This misses nous constructs, data are collected with survey methods and the essence of the distinction, because culture-specific con- analyzed with quantitative techniques (Farh, Earley, & Lin, structs do not necessarily resonate with cultural insiders' 1997; Yang, 1986). Likewise, ethnographic observation and self-understandings, A broader usage refers to the underly- qualitative data are sometimes used to support arguments ing interests of understanding versus control (Habermas, from an etic perspective (Nelsen & Barley, 1997: Sutton, 1994). 1999 Morris, Leung, Ames, and Lickel 783

TABLE 1 Assumptions of Emic and Etic Perspectives and Associated Methods

Features Emic/Inside View Etic/Outside View

Defining assumptions and Behavior described as seen from the Behavior described from a vantage external to goals perspective of cultural insiders, in the culture, in constructs that apply equally constructs drawn from their self- well to other cultures understandings Describe the cultural system as a Describe the ways in which cultural variables working whole fit into general causal models of a particular behavior

Typical features of methods Observations recorded in a rich Focus on external, measurable features that associated with this view qualitative form that avoids imposition can be assessed by parallel procedures at of the researchers' constructs different cultural sites Long-standing, wide-ranging observation Brief, narrow observation of more than one of one setting or a few settings setting, often a large number of settings

Examples of typical study Ethnographic fieldwork; participant Multisetting survey; cross-sectional types observation along with interviews comparison of responses to instruments measuring justice perceptions and related variables Content analysis of texts providing a Comparative experiment treating culture as a window into indigenous thinking quasi experimental manipulation to assess about justice whether the impact of particular factors varies across cultures

reports (Kloos, 1988) and for inheriting miscon- proach serves best in testing hypotheses (e.g., ceptions from cultural insiders (Marano, 1982). Greenfield, 1996). On the other side, etic accounts based on survey In a more explicit selectionist proposal. Berry data are often dismissed because researchers (1990) endorses a three-stage sequence. In the remained at a distance from respondents, poten- first stage, initial exploratory research relies on tially insensitive to how respondents were af- "imposed-etic" constructs—theoretical concepts fected by their questions (Geertz, 1983). and measurement methods that are simply ex- Yet, not all arguments against integration are ported from the researcher's home culture, In the staked on critiques of either approach. Separat- second stage, emic insights about the other cul- ism has been defended as a means to protect ture are used to interpret initial findings, with less well-institutionalized traditions from being an eye to possible limitations of the original assimilated by mainstream traditions. Writing constructs, such as details that are unfamiliar or about organizational culture, Martin argues that meaningless outside of the home culture. On "pressures toward assimilation would under- this basis, then, the constructs in the model are mine a perspective's inherently oppositional filtered to eliminate details that cannot be mea- stance ... threatening its conceptual and politi- sured with equivalence across cultural settings. cal integrity" (1992; 187). In sum, both partisan The factors that survive this filter—"derived- and protective agendas have led scholars to ad- etic" constructs—are culture-general dimen- vocate keeping emic and etic insights about a sions of persons, such as value orientations, or phenomenon somewhat separate. of their environments, such as economic or eco- However, not all previous scholars hold that logical factors. In the third and final stage, the emic and etic approaches should be kept apart. researcher tests an explanation constructed Some have suggested that researchers should solely of derived etic constructs. select between approaches, depending on the Brett and colleagues (1997; Lytle, Brett, Bars- stage of a research program. For example, it has ness, Tinsley, & Janssens. 1995) describe another been argued that an emic approach serves best proposal based on a three-stage sequence. in exploratory research, whereas an etic ap- These scholars differ from Berry in sharply dis- 784 Academy of Management Review October tinguishing cultural factors from ecological and much a witnessed reality as a constructed inter- economic factors in analyzing cross-national pretation. differences. Also, they suggest that etic con- Fortunately, frameworks for understanding structs may not always require measurement how observers interpret fairness have been de- equivalence. However, as in Berry's model, the veloped in research on the psychology of justice end state is an explanation drawn in terms of (Sheppard, Lewicki, & Minton, 1992). The primary etic constructs; emic insights guide the initial tradition in this research concerns distributive steps but are not retained in the final explana- justice judgments—that is, responses to partic- tion.^ ular distributions or allocation patterns (Adams, The sequential selection models of Berry 1965; Deutsch, 1985). Although there is increas- (1990) and Brett et al. (1997) have been influential ing evidence that justice perceptions depend in guiding psychological and organizational re- greatly as well on the procedures through which searchers in their approaches to culture. Yet authorities bring about these distributions (Lind these analyses only begin to explore the syner- & Tyler, 1988; Thibaut & Walker, 1975; Tyler & gies between perspectives. Although they ad- Bies, 1990), we restrict our attention, for reasons dress the role of emic insights in refining etic of brevity, to distributive justice. explanations, they say little about how etic in- Judging the fairness of resource allocations in sights stimulate emic investigation. Although organizations involves two major components: they address the interplay between perspec- (1) selecting a rule to serve as the principle or tives within a given research program, they do "scale" of justice and (2) construing the actions not analyze long-term interplay across research of the persons involved. This distinction can be programs within a general research area. To lay illustrated in terms of the traditional metaphor the groundwork for a long-term analysis, we that justice is weighed on a scale (see Figure 1), now introduce the research area of justice judg- The figure represents elements of the stimulus ments. event that spurs a justice judgment (on the left) and the corresponding elements of an observer's subjective interpretation (on the right). When a lUSTICE lUDGMENT resource allocation occurs in a setting, an observer applies a scale that is appropriate to the Judgments of justice occur whenever authori- setting (i.e., a rule is selected from the observer's ties in a group allocate resources or rewards stock of justice principles). At the same time, the among its members. For instance, when a man- observer does interpretive work in order to de- ager gives a larger bonus to an energetic young cide what to place on each side of the scale. The salesperson than to her more senior and expe- observer must construe the meaning or rele- rienced colleague, observers will evaluate this vance of the rewards allocated by the manager, manager positively or negatively, depending on on the one side, and construe the deservingness whether they judge the rewards to be in balance of the employee, on the other. This observer can with the employees' respective contributions. then weigh the manager's actions against the Managers and others who wish to be perceived employees' actions to check for balance (justice) positively need to understand how observers ar- or imbalance (injustice). Although other cogni- rive at justice judgments. This is not a trivial tive steps enter into justice perceptions, this task, however, because it is not always self- framework captures the core components. evident what is fair or balanced; justice is not so Before turning to cultural differences, let us review the key points about each component of justice cognition that have emerged from main- ^ Brett et al. (1997) describe a second form of emic-etic stream research—that is, research in the North interplay in research conducted by a multinational research American and Western European settings, team. Team members rely on emic understandings of their where almost all psychological and organiza- respective local cultural environments when developing in- tional research has been situated. Research on struments and yet rely on etic frameworks when communi- how people select a principle of justice began cating with their collaborators from other cultures. Although with tests of the notion that fairness perceptions less explicitly than in the first sequence model described by these authors, the authors suggest that the end state is an generally follow an equity rule that rewards etic perspective on the phenomenon. should be proportional to contributions (Adams, 1999 Morris. Leung, Ames, and Lickel 785

HGURE 1 Components of Distributive Justice Judgment*^

Stimulus event Observer judges organizational justice

In a particular social Observer selects a context... principle of fairness (e.g., the equity rule).

a manager allocates construes the level of resources, tangible reward conferred by the and intangible,... manager's actions,

to an employee ... and construes the whose attributes and level of deservingness actions contribute to of the employee the organization

" Principle application is like selecting a scale for weighing justice; construing human behavior is like judging what belongs on the scale.

1965). For example, when a piece-rate compen- quantitative performance measures may not be sation system seems fair, it is because the ob- the only attributes of employees that are rele- server applies the equity rule and sees the man- vant. Interpreting the overall contributions may ager's allocation of rewards as balanced with require a detailed knowledge of the setting. De- the employees' contributions. pending on the local norms and practices, fac- Research then progressed to incorporate other tors such as effort, attitude, or seniority may principles. Deutsch (1975) argued that principle figure prominently in assessments of contribu- selection depends on the primary goal in the tions. context: productivity goals, primary in the work- Moreover, in construing the rewards provided place, are linked with the equity rule; interper- to employees by a manager, an observer must sonal harmony goals, predominant in friendship first interpret the ways that the actions of the groups, call for the equality rule; and personal manager benefit employees. This may involve a welfare goals, as in the family, are linked with great deal of concrete, specific knowledge, such the principle of need-based distribution. On the as cultural scripts, roles, and symbols. Addition- whole. Western researchers have assumed that ally, the meaning of some resource allocations equity is the primary principle in work organi- may be affected by the context of other alloca- zations, and other principles are applied only in tions. For instance, Martin and Harder (1994) contexts where the resource exchange has to do found that, in the United States, a manager's specifically with cultivating employee relation- unequal distribution of financial resources, such ships (e.g., invitations to an office party) or with as salary, is tolerated when accompanied by ensuring welfare (e.g., health insurance plans). equal distribution of intangible, socioemotional In addition to selection of a principle, justice rewards, such as friendliness. Evaluating the judgments also require interpretation of the rel- meaning of a manager's action, then, requires evant contributions and rewards. In construing more complex interpretative processes than a what an employee contributes, for example, an simple tallying across the tangible resources observer must determine which attributes and provided. performances of the employee are relevant. A In sum, mainstream Western research on the salesperson who generates no sales for her com- psychology of justice over the past two decades pany may not have contributed as much as the has identified increasingly subtle relationships company's top performer, yet sales and other among contexts, resource allocations, and ob- 786 Academy of Management Review October servers' reactions; these, in tum, have spawned uncovered a strong theme of harmony in Chi- applied frameworks to help managers antici- nese cultural discourse concerning groups, rela- pate justice sensitivities (Sheppard et al., 1992). tionships, and justice. Nevertheless, the impact Yet, at the same time, researchers in non- on researchers of justice was slight for several Western settings have begun to provide increas- reasons. First, ethnographic descriptions of har- ingly compelling evidence that culture influ- mony remained somewhat vague. Also, some ences the process of justice judgment. Some ethnographic evidence, such as that from histor- findings indicate differences in selecting princi- ical studies of Chinese negotiations with other ples of justice (e.g., Leung & Bond, 1984). Other countries, showed a seemingly deliberate lack findings indicate differences in the interpreta- of concern for harmony (Pye, 1982), challenging tion of actions by authorities and employees the generalization. (e.g., Redding & Wong, 1986). We now review There is also a long tradition of etic perspec- how this research has progressed. tive research examining cultural differences in general values related to principles ot justice. The first wave of such studies involved compar- RESEARCH ON CULTURAL INFIUENCE isons with translated survey instruments that To illustrate the kinds of cultural differences had originally been developed to measure work that arise in organizations and to lay the foun- values in Western settings, such as individual dation for our analysis of forms of synergy be- freedom, equality, and the welfare of the group. tween etic and emic approaches, we briefly re- In several studies scholars found that Chinese view findings concerning differences in how respondents give more weight to group-oriented justice judgments are made in East Asian cul- values than do North Americans (e.g., Singh, tural settings, as opposed to Western settings. Huang, & Thompson, 1962). Although certainly the cultures within the gen- Although these studies produced sharply de- eral area of East Asia vary greatly, there is a fined and replicable differences, a limitation of common heritage of Confucian values and insti- the approach is that inappropriate Western con- tutions that makes some general comparisons structs may have been imposed onto other cul- and contrasts meaningful. We proceed by exam- tures. A step forward was Hofstede's (1980) study ining selected results from the two key compo- of 40 countries, in which he distilled value di- nents of distributive justice perception: select- mensions, such as individualism-collectivism, ing principles and construing behavior. from a factor analysis of country means. Rather than a general orientation toward other people, the individualism-collectivism dimension taps Selecting Principles the extent to which individuals conceive of The idea that the justice of a given event may themselves as embedded in particular ingroups be judged by different principles in different and follow a norm of sacrificing personal bene- cultures is a theme raised in emic studies of fit for ingroup others. Hofstede (1980) placed 40 justice in East Asian settings. For instance, eth- countries on several major dimensions of value nographers in East Asian cultures have sug- orientation and linked these positions to ante- gested that the principle of harmony is salient in cedents in economic and ecological conditions Confucist cultural settings (Hsu, 1953). Whereas and to consequences in social behavior. Western common sense and theory (Deutsch, Triandis, Hui, and others developed derived- 1985) distinguish the goal of harmony from that etic instruments to measure individualism- of productivity, descriptions of Chinese concep- collectivism at the individual level. In these tions of groups suggest that harmony is central studies the researchers consistently found that to social organization and productivity (Hsu, individuals socialized into Chinese and other 1971). This idea that harmony is a means to Confucian-influenced cultures held more coUec- productivity, rather than an opposing goal, is tivist social values and beliefs than individuals expressed in a proverb by the Confucian scholar in Western settings and placed more emphasis Mencius; "Weather is less important than a on the ingroup/outgroup distinction (see Trian- fertile field, and a fertile field is less important dis, 1995, for a review). than human harmony." In sum, early emic Although both emic and etic studies sug- scholarship—fieldwork and textual study— gested that Chinese culture may be associated 1999 Moiris, Leung, Ames, and Lickel 787

with different conceptions of justice, there were has revealed subtle distinctions in the kinds of no striking findings that challenged main- relationships that trigger particular principles stream research on the psychology of justice. of justice in Chinese societies. Of course, these The first findings concerning Chinese culture to newly noticed emic distinctions are candidates have a major impact on mainstream justice for new derived-etic constructs. For instance, scholarship came in studies that brought to- Leung (1997) has drawn on emic insights about gether the emic insight concerning the central- harmony motives and has distinguished the ity of harmony concerns with the etic insight concern for maintaining a tie to a peripheral that the ingroup/outgroup distinction matters ingroup member, which triggers an equality more. Leung and Bond (1984) tested the hypoth- rule, from the concern for enhancing harmony in esis that, for Chinese individuals, harmony con- a close relationship, which triggers a generosity cerns will shape the justice principles applied rule. Although this proposal about harmony in interactions with ingroup members. The ex- goals in different kinds of relationships derives periments varied whether participants com- from emic studies of Chinese culture, Leung of- pleted a shared project with ingroup versus out- fers it as a potential etic hypothesis about group members and varied the contribution highly coUectivist societies. level of the members. Results showed that Chi- Thus, in summary, research on the principle- nese participants, like U.S. participants, apply selection component of justice judgment has in- the equity rule with outgroup members, but, un- volved an active interplay between emic and like U.S. participants, apply a more complex rule etic insights. Although most of the research on with ingroup members. With ingroup members, this component has been from the etic perspec- their pattern is one of generosity: allocating by tive, findings from the emic perspective have the equality principle when the other has con- spurred insights. Emic research has revealed tributed less than oneself but allocating by the novel constructs (e.g., generosity as a means to equity principle when the other has contributed harmony), has challenged etic constructs (the more than oneself. A virtue of these findings, notion that individual's adherence to individu- relative to early survey findings, is that they alist and collectivist values is captured by a reveal the dynamics of cultural influence by un- unitary dimension), and has suggested new so- covering how culture interacts with contextual lutions (e.g., distinguishing types of ingroup re- variables (e.g., ingroup versus outgroup). These lations). Before delineating these forms of emic/ findings were also bolstered by replications in etic interplay, let us similarly review cultural other East Asian settings (Kim, Park, & Suzuki, research on the other component of justice judg- 1990; Leung & Iwawaki, 1988), and extensions ment. have uncovered further relevant contextual factors (Chen, 1995). In recent years the strategy of linking the Construing Behavior choice of justice principles to broad cultural Insights concerning cultural influence on the value dimensions has come under critique. interpretation of behavior relevant to justice Anomalous findings within the etic tradition comes mostly from emic studies. A recurrent have resonated with long-standing conceptual theme in field studies and ethnographies of or- challenges by emic researchers (for a review see ganizations in East Asian settings (e.g.. Redding Earley & Gibson, 1998). The controversy has & Wong, 1986; Rohlen, 1974) is that, compared sparked a wave of emic studies in which re- with Western settings, assessment of employ- searchers have taken a closer look at indige- ees' contributions is based less strictly on task- nous conceptions: Chiu (1991) conducted a fine- relevant performances. In analyses of cultural grained content analysis of popular Chinese differences in policies and practices of firms, sayings about injustice, and Ho and Chiu (1994) researchers have noted related tendencies, such uncovered relations among the many concep- as the relatively large weight placed by Japa- tual components of individualism-collectivism nese and Korean organizations on an employ- through a similar content analysis. ee's seniority when assessing the employee's To a greater extent than in early exploratory contribution (Pascale & Athos, 1981) or the con- studies or in studies with derived-etic concern in Chinese societies for the value of a per- structs, this new wave of focused emic analyses son's social connections (Redding & Wong, 1986). 788 Academy ot Management Review October

Such emic analyses have suggested the possi- ever, little direct evidence has accumulated for bility that even when there is agreement across etic hypotheses concerning how interpretation cultures on the principle of justice that applies of deservingness hinges on collectivism and in- (e.g., the equity rule), there may be disagree- groups. ment in justice judgments that arise from dif- Perhaps the closest development is research ferences in what observers count as appropri- comparing how an individual's network is eval- ate contributions (e.g., performance versus uated across cultures, exploring parallels to seniority). guanxi in other collectivist cultures. For exam- Emic analysis of how culture shapes judg- ple, a form of relation known as compadies is an ment of deservingness has gone furthest in stud- important link in business interactions in some ies of indigenous Chinese constructs. Although Latin America cultures (Stephens & Greer, 1995). there can be no doubt that an employee's social In a step toward an etic analysis, Xin and Pearce connections enter into appraisals in many West- (1994) have argued that emphasis on guanxi ern settings, the role of an employee's connec- functions to protect against fraud and, hence, tions in an evaluation of his or her worth gener- that it arises in settings lacking institutional ally is left implicit and unarticulated by protections (i.e., strong courts). Although these Western observers. In Chinese culture, in con- authors have presented supportive evidence for trast, there is a rich lexicon of lay constructs for this functional relationship within China, cross- articulating how an individual's network con- cultural evidence has yet to emerge. Morris, fers worth or value (Hsu, 1971). Guanxi (which Podolny, and Ariel (1998) have used a survey of translates literally to "connections" but has employee networks in an intemational organi- more positive connotations) is a construct that zation to compare how an employee's network guides the ascription of credit to people with made others obligated to that employee. These extensive networks (King, 1991). Within tradi- authors did not observe a general tendency for tional Chinese commercial settings, an individ- relationships to be valued in collectivist societ- ual's guanxi plays a key role in that person's ies; rather, different kinds of relationships in- social face or mianzi (Hu, 1944). Although guanxi curred value in different collectivist settings has received particular attention, other con- (e.g., dependence relations in Hong Kong, as structs, such as renging, which concerns how opposed to friendship relations in Spain). In much one person owes in a relationship, have sum, although etic analyses of the role of rela- also been analyzed (Gabreyena & Hwang, 1996). tions in perceptions of employees' deserving- Hwang (1987) presented a general model of Chi- ness have exposed interesting questions about nese social and organizational behavior as a the function, they appear to miss some impor- function of these relational constructs. tant details. Although the bulk of emic research on inter- Overall, research on how observers interpret preting action involves qualitative data from actions in terms of justice-relevant rewards and ethnographic observation and study of texts, contributions has moved forward through the some emic analyses have applied quantitative mutual influence of researchers working in the methods while still taking the emic approach. emic and etic traditions. Interestingly, however, For example, predictions about justice behavior the majority of cultural research relevant to this from Hwang's (1987) model have been tested ex- component has been from the emic perspec- perimentally. Emic predictions about relation- tive—the opposite of what we observed concern- ships have also been approached with quanti- ing the abstract component of principle selec- tative survey tools; for instance, guanxi has tion. This may reflect that in a more concrete, been modeled in terms of relational demogra- knowledge-based cognition, there are fewer rel- phy (Tsui & Fahr, 1995). evant etic constructs—constructs with an equiv- Some scholars have proposed etic hypotheses alent meaning across cultures. It may be that about construal of employee deservingness. For construals of deservingness vary across cul- instance, Hofstede argues that in collectivist so- tures in so many specific ways that generalities cieties "promotion decisions take employees' in- in terms of etic constructs, such as the ingroup, group into account," whereas in individualistic fail to capture the variance. societies they "are supposed to be based on Having reviewed major cultural findings re- skills and rules only" (Hofstede, 1991: 67). How- lated to justice perception, we now are in the 1999 Morris, Leung, Ames, and Uckel 789 position to draw some conclusions about how with those from the other tradition. When pre- emic and etic insights complement each other. liminary exploratory studies from either perspective suggest the possibility of cultural influ- FORMS OF SYNERGY ence on a judgment, this often spurs a second generation of studies, which often come at the Drawing on our example of the justice judg- problem from the opposite perspective in order ment literature, we now analyze several general to critique or challenge the initial claims. In forms of synergy that can occur between emic providing apt challenges to the limitations of and etic research programs within a topic area. initial claims, second-generation studies in one We start by analyzing how emic and etic contri- tradition often evoke new formulations that syn- butions stimulate each other's progress. We thesize the original claim with the critique. This then suggest that the emic and etic approaches ongoing, mutually enriching relationship be- are partly able to counteract one another's the- tween the two kinds of research is illustrated in oretical weaknesses in describing culture. Fi- Figure 2. A general theme is that different nally, we argue that a full psychological model strengths of the two approaches create comple- of judgment may only be possible and meaning- mentarities. Let us now describe these specific ful when etic and emic insights are combined. paths of influence between perspectives in more detail. Mutual Stimulation First, we examine the different merits of exploratory studies in the emic and etic traditions. The primary form of synergy is the stimulation Emic exploration proceeds through open-ended that emic and etic insights can provide to each and long-standing observations of ethnogra- other. Differences between the perspectives phers who immerse themselves in a particular mean that there are lessons from exploratory culture. The strength of this method is in the studies in one tradition that are not redundant wealth of detail conveyed in "thick description"

FIGURE 2 Interplay Between Emic and Etic Research'

Emic perspective Etic perspective

Initial exploration Initial exploration

Ethnographic Translated survey study of one setting across many settings

Later refinements Later refinements

Structured study of Tests of an account in distinctions emphasized derived-etic constructs by cultural insiders

Dual-perspective account of cultural influence on justice judgment

Applied framework to guide responses to justice concerns in culturally diverse organizations

° Paths of mutual influence between the two research perspectives, leading to an integrative explanatory framework. 790 Academy of Management Review October

(Geertz, 1983). The ethnographer aims to under- enced by second-generation etic research, be- stand the culture on its own terms, rather than cause they are directed toward the goal of through imposition of prior theories. This en- uncovering the contextual variables omitted in ables the discovery of novel features. Yet, a these etic studies. For example, studies repre- weakness is that subjectivity in what the re- senting individualistic and coUectivistic values searcher notices and how much he or she on a single dimension gave rise to focused emic chooses to generalize can easily distort the por- studies of the structure of these values in Chi- trait. However, flawed and conflicting ethnogra- nese culture (Chiu, 1991; Ho & Chiu, 1994). These phies can serve a valuable function in provok- studies, in tum, provided a strong challenge to ing further research. For instance, Hsu's (1953) etic researchers, in that they were based on sys- descriptions of all-pervading harmony concerns tematic data collection and were operation- and the conflicting descriptions of others (Pye, alized quantitatively. Indeed, these studies 1982) sparked Leung and Bond's (1984) etic inves- have been incorporated into new etic proposals tigation of the social contexts that moderate cul- (e.g., Leung, 1997). In sum, the emic critiques of tural differences in the justice principles. etic work drew attention to important concerns underemphasized in the original etic scholar- Exploratory studies from the etic perspective ship and suggested tools for a subsequent wave often take the form of imposed-etic surveys. The of etic accounts. dubious equivalence of measuring instruments and the lack of sustained first-person observation mean that nuances of meaning can be lost Richness in Models of Culture | in translation. Nevertheless, there is often a re- liable signal through the noise, and the broad Whereas our first point concerned how find- outlines of cultural differences and their associ- ings from the two perspectives challenge each ations to other variables (such as ecological and other and stimulate each other's new questions, economic factors) are identified. The images oi our second point concerns how the two kinds of culture produced by exploratory etic studies are explanation complement each other in contrib- like the crude maps of world geography uting to rich accounts of culture. A fault line runs sketched centuries ago, based on the reports of through the disciplines concerning culture. On returning sea navigators. Such maps simplify one side are disciplines like history or cultural the terrain, but in doing so provide a guide to anthropology, rooted in a historicist logic of which places might be interesting for a closer seeking local regularities within a bounded mi- look. lieu. On the other are disciplines like econom- An example of how etic maps spur closer, ics, driven by a functionalist logic of seeking emic study can be seen in the aftermath of Hof- transhistorical generalizations. Organizational stede's (1980) findings that even the most highly behavior involves both of these logics (e.g., Bur- industrialized East Asian societies differ rell & Morgan, 1979). Yet, the emic and etic per- sharply from the West. Until this study, many spectives each provide only half of the story. social scientists believed that, regardless of tra- Because emic studies tap into the explanations ditional cultures, wealthy industrialized societ- held by cultural insiders, the emic perspective ies were fast converging in their social values inherently leads to an emphasis on the causes (Inkeles, 1980). Emic studies generally focused of phenomena that are internal and local to the on less industrialized societies. Hofstede's find- cultures and organizations being studied. Be- ings provoked a renewed interest in emic stud- cause etic perspectives attune one to relation- ies of Japan, Korea, Hong Kong, and other indus- ships between external structural variables and trialized East Asian societies. behaviors, a functionalist story is more likely to We now examine the interplay between sec- result. ond-generation studies in each tradition. In sec- However, a richer account of culture can re- ond-generation emic studies, as we have seen, sult when an integrative explanatory frame- researchers retain an interest in understanding work arises. One way this happens is when a culture through the constructs of cultural in- emic findings from several cultures reveal siders, yet they focus more narrowly on one part parallel patterns. Etic researchers often re- of a culture than do researchers conducting ex- spond by attempting to capture the pattern in ploratory studies. Often, these studies are influ- terms of more general factors that can be as- 1999 Morris, Leung, Ames, and Lickel 791

sessed through parallel constructs and mea- The tendency of etic perspective research to sures. For example, emic studies of employee omit all but the most abstract components of a relationships in Chinese societies and other cognitive process is, in fact, rooted in the very societies have led to more general functional- procedure for identifying derived-etic con- ist accounts of when relationships are valued structs. Derived-etic constructs (Berry, 1990) are (e.g., Xin & Poarce, 1994). These attempts supposed to be the common denominators of the rarely succeed in full, but they raise the valu- variables involved in a psychological process able question of why a given historically across cultures. Identifying behavioral events rooted practice persists in its contemporary with equivalent meanings across cultures is form. These accounts, in turn, spur emic stud- easier to do if the behavior and meaning are ies highlighting unique details not captured conceptualized abstractly. As a result, all of the by functionalist generalizations. In this exam- concrete details that are not equivalent across ple we see that looking at the same phenom- cultures drop out of the description. In other enon from two perspectives adds depth and words, the search for constructs with measure- richness to the explanatory framework. The ment equivalence across cultures creates an up- emic account of guanxi highlights the embed- ward pressure toward abstract descriptions. dedness of Chinese conceptions of distributive justice in other aspects of Chinese culture. The Concrete details differ qualitatively in ways etic account illustrates that these constructs that prevent comparison with parallel method- and this culture exemplify a diffuse but recog- ological procedures. nizable syndrome of collectivism that has par- Let us consider why an etic approach is able ticular ecological and economic antecedents. to illuminate the abstract component of apply- An account of culture that acknowledges both ing justice principles but not the concrete com- historicist and functionalist logics is important ponent of interpreting actions. There seems to in organizational behavior, in that many re- be a finite list of basic justice principles across searchers have agendas that are both intellec- cultures; a workable derived-etic model is pos- tual and practical. For instance, many organiza- sible. Event construal, however, involves a large tional researchers are aligned with the critical number of overlapping constructs concerning interest of exposing power relations, which re- social roles, relationships, symbols, and so quires understanding informants on their own forth. Endless variation seems possible—not terms but also going beyond their reports to merely in the frequency with which constructs describe economic and other conditions that are evoked but in the qualitative content of the "envelop" the informants' world (Jermier, 1998). constructs. As we have seen, etic hypotheses Accounts of culture serving a critical agenda or about construal of deservingness have not other change agendas require the richness of found empirical support. In sum, an etic ap- integrating emic and etic insights. proach leaves us with a highly abstract and incomplete view of culture and cognition that fails to offer predictions about any particular case. Comprehensiveness in Models of Judgment Turning to the other side of the problem, we Another way in which an integrative, emic- can see that a purely emic approach is no better. etic framework serves better than a single- With purely emic studies, like early ethnogra- perspective explanation is in capturing the phies of Chinese harmony, researchers have a kinds of cognition involved in justice and other difficult time distilling the key principles. It may judgments. A pattern revealed in our review of be that seeing the behavior in two different cul- research on culture and justice was that most tures makes it easier to spot the abstraction that etic research activity has focused on the more unites them. In sum, although the two perspec- abstract component of applying a justice princi- tives often bring researchers' attention to the ple. Conversely, most emic research activity has same phenomenon, it is also true that they are focused on the more concrete component of con- complementary, in that they draw researchers' struing the deservingness of the people in- attention to different components of justice judg- volved. Quite likely, these biases of etic and ments, making it more likely that all of the im- emic perspectives are inherent ones. portant aspects of cognition will be recognized. 792 Academy of Management Review October

ADVANTAGES OF INTEGRATIVE Identifying and Choosing Policy Options FRAMEWORKS IN APPLICATION One of the implications of the preceding dis- In addition to the benefits of an integrative cussion is that managers who have anticipated framework in basic theory, there are also bene- cultural differences in justice sensitivities may fits in applied problems. The practical need for a need to adjust their policies from country to framework to guide managers in coping with country in order to respond to diverse justice cultural differences in justice judgments has sensitivities. Yet, this localization of policies never been greater. Organizations increasingly works against another goal of many organiza- span many cultures, both because of the diver- tions, which is to standardize policies across the sification of the workforce in many nations and globe. Increasing numbers of organizations fol- because of the globalization of organizations low the model of Citicorp, Philips, Sony, and themselves. other global firms by maintaining coordination There are advantages of an integrative emic- through the rotation of managers from one coun- etic framework at several points in managing try to another. Standardizing or globalizing pol- across cultures: (1) anticipating the justice sen- icies greatly facilitates the process of employ- sitivities of employees in other cultures, (2) iden- ees moving from one unit to another. Hence, tifying and choosing policy options, and (3) suc- firms face a tradeoff. cessfully implementing a policy option. When considering this tradeoff in policies relevant to justice, one needs a framework that incorporates cultural influences on each compo- Anticipating Cultural Differences in Justice nent of justice judgment. That is, firms have to Sensitivities contend with differences not only in the principles or rationales that employees endorse but in Etic research has had a high stature in organ- the concrete beliefs that guide how employees izational research in part because it seems prac- construe behavior of management and of their tical and efficient. The hope has been that the peers. The four possible combinations are rep- myriad cultural traditions in an international resented in Figure 3. Although research re- organization might be reducible to a few simple garding when and how to develop global or- cultural dimensions. Interestingly, however, the ganizational standardization is advancing, popular literature of cultural training guides for understanding of this issue (particularly how to managers has not been influenced much by the implement standardization) remains limited etic research found in management journals. (Shenkar & Zeira, 1987: Sullivan, 1992; Taylor, Managers still tend to learn about cultural sen- Beechler, & Napier, 1996). Let us see how an sitivities in a country-by-country fashion (e.g., encompassing framework for conceptualizing Cushner & Brislin, 1996). cultural influence helps clarify the options. Training guides convey potentially relevant points by referring to recurrent roles and situa- Using our framework, we can think of differ- tional scripts. These guides include little in the ences in justice perception (and many other or- way of etic generalizations, such as that collec- ganizational domains) as being one of three tivist cultures promote a harmony-preserving types: (1) a difference mainly in concrete beliefs distribution rule in ingroup interactions. Al- related to event construal, (2) a difference though an etic generalization may allow a man- mainly in applying abstract principles, or (3) a ager to make general predictions about many difference in both. Firms deciding to apply a cultural groups, it does not allow a manager to policy across cultural boundaries will face dif- make precise predictions about any of them. ferent issues, depending on which kind of differ- That is, to apply etic predictions in such terms ences exist. When no differences exist, of course, as ingroup, these terms must be fleshed out or the firm can "go global" in its policies. instantiated into the concrete details of a local A difference only in construal-related beliefs cultural setting. Without the cultural specifics may be the least difficult to manage toward a identified through emic analysis, the abstract global policy since employees already share a principles identified by etic methods are un- common framework of principles. For example, likely to be sufficient for developing organiza- employees in both cultures might use equity tional policies. principles for distribution but in one culture re- 1999 Morris, Leung. Ames, and Lickel 793 HGURE 3 Framework for Decision Making About Cross-Cultural Standardization"

Construal oi Deservingness Same across cultures Diflerent across cultures

e.g., weight placed on tenure Same versus shorf-ferm per/ormance across - Easy to standardize policies - Difiicult to standardize cultures - Stress principle similarity and construal definitions

e.g., equity versus equality lule Different - Very difficult to standardize - Most difficult to standardize across cultures - Stress that some resources • To standardize may require are distributed by the explicit programs; worthwhile culturally favored principle only if policy is core to firm

° Distinguishing types of cultural influence on justice judgment clarifies a firm's options in the tradeoff between globally standard versus locally sensitive policies. gard seniority as the key contribution, whereas cross-training (rather than individual special- in the other culture count performance as the ization), continually increasing (rather than contribution. Firms would be advised to stress fixed by contract) performance standards, and the similarity in the shared principles as the security of employment (Florida & Kenney, 1991; basis of fairness and develop clear definitions Fucini & Fucini, 1990; Wilms, Hardcastle, & Zell, of what counts as a contribution or input. In sum, 1994; Young, 1992). However, North American the firm can "think globally" about the princi- and Japanese workers favor quite different prin- ples of fairness in its organization, yet "act lo- ciples and possess different beliefs about how cally" in explaining how these principles are to construe workplace events, which made instantiated into concrete practices. transferring these policies difficult. As shown by A difference mainly in principles makes it the experience of Japanese auto companies, more difficult to employ a globally standardized standardization in such instances entails exten- policy and achieve uniform perceptions of fair- sive cultural training and efforts to make clear ness. An example would be a case in which one the reasons for organizational policies (Florida culture favors an equity principle for salary and & Kenney, 1991; Wilms et al, 1994), to avoid cre- the other favors an equality principle. One pos- ating feelings of injustice (Fucini & Fucini, 1990). sible strategy here would be to stress other re- When developing such training programs, it is sources that fit the culture's preferred principle. necessary for organizations to review emic anal- If global salary policy were moved toward being yses of local cultural beliefs in order to develop equity based, other resources still distributed training programs that properly translate con- based on equality, such as respect and benefits, cepts into terms that local workers find accept- could be stressed in the equality-oriented cul- able. ture. If strong differences exist in both construal and principle selection, substantial investments Successfully Implementing a Chosen Policy are likely to be required for successful standard- Having chosen to either globalize or localize ization. For example, Japanese auto manufac- compensation policies, a firm still has to imple- turers have implemented many Japanese organ- ment the policy. A key idea in the implementa- izational policies in their North American tion of a policy is assuring the "buy in" of the factories, including an extremely small number employees affected and of the managers over- of job classifications and intensive functional seeing the policy. A plan that reflects assump- 794 Academy of Management Review October tions held in mutual agreement by all relevant Berry, J. W. 1990. Imposed etics, emics, derived etics: Their parties has the greatest chance of success. conceptual and operational status in cross-cultural psychology. In T, N. Headland, K. L. Pike, & M, Harris (Eds.), When a policy based on an outsider understand- Emics and etics: The insider/outsider debate: 28-47. ing is imposed from on high, insiders affected by Newbury Park, CA: Sage. the policy often put up resistance (Drazin & San- Brett, J. M., Tinsley. C. H., Janssens, M,, Barsness, Z. I,, & Lytle, delands, 1992; Dunbar & Ahlstrom, 1995). Manag- A. L. 1997. New approaches to the study of culture in ers working with a solely etic perspective, industrial/organizational psychology. In P. C, Eariey & hence, are likely to meet resistance from below M. Erez (Eds.), New perspectives on intemational indus- in the cultural group. Yet, local managers work- trial/organizational psychology; 75-129. San Francisco: ing with a solely emic (local) perspective are New Lexington Press. likely to meet resistance from above. That is, the Burrell, G., & Morgan, G. 1979. Sociological paradigms and cultural group whose plight is explained solely organizational analysis. London: Heinemann. in their own emic terms may be dismissed and Chen, C. C. 1995. New trends in reward allocation prefer- marginalized by outsiders who do not under- ences: A Sino-U.S. comparison. Academy oi Manage- stand the emic description of their justice sensi- ment Journal. 38: 408-428. tivity. Hence, only a plan that is expressed in the Chiu, C. Y. 1991. Responses to injustice in popular Chinese dual language of emic and etic is likely to res- sayings among Hong Kong Chinese students. Journal ot onate with all of the stakeholders involved in a Social Psychology, 131: 655-665. policy that responds to cultural differences in Cushner, K., & Brislin, R. W, 1996. Intercultural interactions: A justice sensitivities. practical guide. Thousand Oaks, CA: Sage. Deutsch, M. 1975. Equity, equality, and need: What deter- mines which value will be used as the basis of distributive justice? Journal oi Social Issues, 31: 137-149. CONCLUSION Deutsch, M. 1985. Distributive justice. New Haven, CT: Yale In this article we have extended previous ar- University Press. guments about the complementarity of emic and Drazin, R., & Sandelands, L. 1992. Autogenesis: A perspective etic insights about culture. In contrast to previ- on the process of organizing. Organization Science. 3: ous work by Berry (1990) and Brett et al. (1997), we 230-249. have focused on synergy over the long term Dunbar, R. L., & Ahlstrom, D. 1995. Seeking the institutional within a general research area. We have de- balance of power: Avoiding the power of a balanced view. Academy oi Management Review. 20: 171-192. scribed several forms of stimulation in which developments within each research tradition Earley, P. C. 1989. Social loafing and collectivism: A comparison of United States and People's Republic of China. are provoked and challenged by findings in the Adminis(ra(ive Science Quarterly. 34: 565-581. other tradition. Moreover, we have argued that an integrative explanatory framework incorpo- Earley. P. C. & Gibson, C. B. 1998. Taking stock in our progress on individualism-collectivism: 100 years of sol- rating insights from both traditions avoids lim- idarity and community. Journal oi Management, 24: 265- itations of purely etic and purely emic findings 304. in conceptualizing culture and in capturing its Farh, J. L., Earley, P. C, & Lin, S. C. 1997. Impetus for actions: various influences on cognition. Finally, we A cultural analysis of justice and organizational citizen- have argued that such integrative frameworks ship behavior in Chinese society. Administrative Sci- have several advantages as guides to solving ence Quarterly. 42: 421-444. | the applied problem of managing justice per- Florida, R.. & Kenney, M. 1991. Transplanted organizations: ceptions in international organizations. That is, The transfer oi Japanese industrial organization into the an integrative framework enables better antici- U. S. American Sociologicai Review, 56: 381-398. pation of employees' justice sensitivities, better Fucini, J. J., & Fucini, S. 1990. Working ior the Japanese- decision making about a firm's policy options, inside Mazda's American auto piant. New York: Free and, once a policy is chosen, better implemen- Press. tation. Gabrenya, W. K., & Hwang, K. K. 1996. Chinese social inter- action: Harmony and hierarchy on the good Earth. In M. H. Bond (Ed.), Handbooi oi Chinese psychology: 309- 321. Hong Kong: Oxford University Press. REFERENCES Geertz, C. 1976. From the native's point oi view: On the Adams, I. S. 1965. Inequity in social exchange. In L. Berkowitz nature of anthropological understanding. In K. Basso & (Ed), Advances in experimental social psychology: 267- H. Selby (Eds.), Meaning in anthropology: 221-237. Albu- 299. New York: Academic Press. querque: University of New Mexico Press. 1999 Morris, Leung, Ames, and Lickel 795

Geertz, C. 1983. Local knowledge: Further essays in interpre- Kondo, D. K. 1990. Crafting selves: Power, gender, and dis- tive anthropology. New York: Basic Books. courses of identity in Japanese workplace. Chicago: Uni- Goodenough, W. 1970. Descripfion and comparison in cul- versity of Chicago Press. tural anthropology. Chicago: Aldine. Leung. K. 1997. Negotiation and reward allocations across Greenfield, P. 1996. Culture as process: Empiricai methodol- cultures. In P. C. Earley & M. Erez (Eds.), New perspec- ogy for cultural psychology. In J. W. Berry, Y. H. Poort- tives on international industrial/organizational psychol' inga, & J. Pandey (Eds.), HandbooJt of cross-cultural psy- ogy: 640-675. San Francisco: Jossey-Bass. chology, vol. I. (revised ed.): 301-346. Needham Heights, Leung, K., & Bond, M. H. 1984. The impact oi cultural collec- MA: AUyn and Bacon. tivism on reward allocation. Journal of Personality and Gregory. K. 1983. Native-view paradigms: Multiple cultures Social Psychology, 47: 793-804. and culture conflicts in organizations. Adminis(ra(ive Leung, K., & Iwawaki, S. 1988. Cultural collectivism and Science Quarterly. 28: 359-376. distributive behavior: A cross-national study. Journal ot Habermas. J. 1971. Knowledge and human interests. Boston: Cross-Cultural Psychology, 19(1): 35-49. Beacon. Lind, E. A., & Tyler, T. R. 1988. The social psychology oi Harris, M. 1979. Cultural materialism: The struggle for a procedural justice. New York: Plenum. science of culture. New York: Vintage. Lytle, A. L, Brett, J. M., Barsness, Z. L, Tinsley, C. H., & Headland, T., Pike, K., & Harris, M. 1990. Emics and etics: The lanssens. M. 1995. A paradigm ior confirmatory cross- insider/outsider debate. Newbury Park. CA: Sage. cultural research in organizational behavior. In L. L. Cummings & B. M. Staw (Eds.), Research in organization- Ho, D. Y.. & Chiu, C.-Y. 1994. Component ideas oi individual behavior, vol. 17: 167-214. Greenwich, CT: JAI Press. alism, collectivism, and social organization: An application in the study of Chinese culture. In U. Kim. H. C. Malinowski, B. 1922. Argonauts oi the western Pacific. Lon- Triandis, C. Kagitcibasi, S.-C. Choi, & G. Yoon (Eds.), don: Routledge. Individualism and collectivism: Theory, method, and ap- Marano, L. 1982. Windigo psychosis: The anatomy of an plications: 200-212. Newbury Park, CA: Sage. emic-etic confusion. Current Anthropology. 23: 385-412. Hofstede, G. 1980. Culture's consequences: International dif- Martin, J. 1992. Cultures in organizations: Three perspectives. ferences in work-related values. Beverly Hills, CA: Sage. New York: Oxford University Press. Hofstede, G. 1991. Cu/(ure and organ izafions. Beverly Hills, Martin, J., & Frost, P. 1996. The organizational culture war CA: Sage. games: A struggle for intellectual dominance. In S. R. Hsu, F. 1953. Americans and Chinese: Two ways of life. New Clegg, C. Hardy, & W. R. Nord (Eds.), Handbooi of organ- York: Abelard-Schuman. ization studies: 598-621. Thousand Oaks. CA: Sage. Hsu, F. 1971. Psychological homeostasis and jen: Conceptual Martin, J., 8c Harder, J. W. 1994. Bread and roses: Justice and tools Jor advancing psychological anthropology, Amer- the distribution of financial and socio-emotional re- ican Anthropologist. 73(1): 23-44. wards in organizations. Social Justice Research, 7: 241- 264. Hu, H. C. 1944. The Chinese concepts of face. American An(hropoiogis(, 46{2): 45-64, Morris, M. W., Podolny, J., & Ariel. S. 1998. Missing relations: Incorporating network analysis into models oi cultural Hwang, K. K. 1987. Face and favor: The Chinese power game. American Journal of Sociology. 92: 944-974. influence on behavior. Research Paper 1527, Stanford, CA: Stanford Graduate School of Business. Inkeles, A. 1960. Industrial man: The relation ol status to experience, perception, and value. American Journal of Nelsen, B. J., & Barley, S. R. 1997. For love or money? Com- Sociology, 66(1): 1-31. modification and the construction of an occupational mandate. AdminisfraJive Science Quarterly. 42: 619-653. Jermier, ]. M. 1998. Introduction: Critical perspectives on organizational control. Administrative Science Quarterly. Pascale, R. T., & Athos, A. G. 1981. The art of Japanese man- 43: 235-2S6. agement. New York: Simon and Schuster. Kashima, Y., & Callan, V. J. 1998. The Japanese work group. Pike, K. L. 1967. Language in relation to a unified theory oi In M. Dunnett (Ed.), Handbook of industrial and organi- the structure of human behavior. The Hague: Mouton. zational psychology, vol. 4: 609-646. Palo Alto, CA: Con- Pye, L. 1982. Chinese commercial negotiation style. Cam- sulting Psychologists Press. bridge, MA: Oelgeschlager, Gunn, & Hain. Kim, K. I., Park, H. J., & Suzuki, N. 1990. Reward allocations in Redding, G., & Wong, G. Y. 1986. The psychology of Chinese the United Stotes, Japan, and Korea; A comparison of organizational behavior. In M. H. Bond (Ed.), The psy- individualistic and coUectivistic cultures. Academy of chology of the Chinese people. New York: Oxford Uni- Management Journal. 33: 188-198. versity Press. King, A. Y. C. 1991. Kuan-hsi and network building: A socio- Rohlen, T. 1974. For harmony and strength: Japanese white- logical interpretation. DaedaJus, 120(2): 63-84. collar organization in anthropologicai perspective. Kloos, P. 1988. Door het oog van anthropoloog (Through the Berkeley: University of California Press. eyes of the anthropologist). Muiderberg, the Nether- Schneider, B. 1990. Organizational climate and culture. San lands: Coutinho. Francisco: Jossey-Bass. 736 Academy of Management fleview October

Shenkar, O,, & Zeira, Y. 1987. Human resource management Thibaut. J., & Walker, L. 1975. Procedural justice: A psycholog- in international joint ventures: Directions for research. ical analysis. Hillsdale. NJ: Lawrence Eribaum Associates. Academy of Management Review, 12: 546-557. Triandis, H. C. 1995. individualism and coJIecfivism. Boulder, Sheppard, B. H., Lewicki, R. I., & Minton, J. W. 1992. Organi- CO: Westview Press. zational justice: The search for fairness in the workplace. Tsui, A., & Fahr, L. 1995. A relational demography analysis New York: Macmillan. of guanxi. Paper presented at the annual meeting of Shweder, R. A. 1991. Thinking through cultures: Expeditions the Academy of Management, Vancouver, British Co- in cultural psychology. Cambridge, MA: Harvard Univer- lumbia. sity Press. Tyler, T. R., & Bies, R, J. 1990. Beyond formal procedures: The Singh, P. N., Huang, S. C, & Thompson. G. G. 1962. A com- interpersonal context of procedural justice. In J. S. Car- parative study of selected attitudes, values, and person- roll (Ed.), Applied social psychology and organizational ality characteristics of American, Chinese, and Indian settings.' Hillsdale, H]: Lawrence Eribaum Associates. students, /ournal of Social Psychology, 57(1): 123-132. Wilms. W. W., Hardcastle, A. ].. & Zell, D. M. 1994. Cultural Skinner, B. F. 1938. The behavior of organisms: An experimental analysis. Englewood Cliffs, NJ: Prentice-Hall. transformation at NUMML SJoan Managemenf fieview, 36(1): 99-113. Smith, P. B., & Bond, M. H. 1998. Social psychology: Across I cultures (2nd ed.). Boston: Allyn and Bacon. Wundt, W. 1888. Uber Ziele und Wege der Volfierpsychologie. Stephens, G. K., & Greer, C. R. 1995. Doing business in Mex- Philosphische Sludien, 4. ico: Understanding cultural differences. Organizafionai Xin, K.. & Pearce, J. L. 1994. Guanxi: Good connecdons as Dynamics, 24: 39-55. substitutes for institutional support. Paper presented at the annual meeting of the Academy of Management, Sullivan. D. 1992. Organization in American MNCs: The per- Dallas. TX. spective of the European regional headquarters. Man- agement International Review, 32: 237-250. Yang. K.-S. 1986. Chinese personality and change. In M. H. Sutton, R. 1994. The virtues of closet qualitative research. Un- Bond (Ed.). The psychology of the Chinese people: 106- published manuscript, Stanford University, Stanford, CA. 170. Hong Kong: Oxford University Press. Taylor, S., Beechler, S,, & Napier, N. 1996. Towards an inte- Young. S. M. 1992. A framework for successful adoption and grative model of strategic international human resource performance of Japanese manufacturing practices in the management. Academy of Management Review, 21: United States. Academy of Management Review, 17: 959-985. 677-700.

Michael W. Morris is an associate professor of organizational behavior at Stanford University's Graduate School of Business and a research affiliate of the Institute for Social and Personality Research at the University of California at Berkeley. His current research topics include cultural influence on social cognition and action, interpersonal and intergroup conlUct resolution, and judgment and decision making. Kwok Leung is professor ol management at the City University of Hong Kong and president of the Asian Association of Social Psychology. His research interests include justice, conflict, joint ventures in China, cross-cultural psychology, and research methodology. Daniel Ames is completing his Ph.D. in social psychology at the University of Cali- fornia at Berkeley, where he earned his M.A. His research interests include culture and social cognition. Brian Lickel is a graduate student in the Social Psychology Program at the University of California at Santa Barbara. He conducts research on social cognition, groups, and perceptions of justice. He was formerly an associate consultant at the Boston Con- sulting Group.