Cross-Cultural Interactions

Abstract

International management studies have been based primarily on the comparison of managerial behavior in countries around the world. Often, these studies have implied that business-people behave similarly with their domestic colleagues as with their foreign counterparts. In questioning that assumption, this study tests whether intra-cultural behavior accurately predicts cross-cultural behavior. Using a negotiation simulation and a sample of 462 Japanese, American, and Canadian business people, behaviors in cross-cultural negotiations were found to differ in some important ways from those in intra-cultural negotiations.

Source: Nancy J. Adler and John L. Graham
View original source.
This work is licensed under a Creative Commons Attribution 4.0 License.

Introduction

Joint ventures, mergers and acquisitions, licensing and distribution agreements, and sales of products and services crucial aspects of all such interorganizational relationships, are face-to-face negotiations. As the proportion of foreign to domestic trade increases, so does the frequency of business negotiations between people from different countries and cultures. Perlmutter estimates that over 50% of an international manager's time is spent negotiating. To successfully manage these negotiations, businesspeople need to know how to influence and communicate with members of cultures other than their own. 

While a growing literature exists documenting international negotiating styles, most studies are descriptive or comparative; that is, they describe the behavior of managers in a particular country or compare behaviors across a range of cultures. For example, there are articles describing the negotiating behavior of French, Russians, Mexicans, Brazilians, Middle Eastern Arabs, Chinese, and Japanese, along with a number of multicountry studies. The worldwide negotiation literature fits within the tradition of research on international organizational behavior. Unfortunately, this tradition has focused on single-culture descriptive studies and multiculture comparative studies, rather than studies investigating cross-cultural interaction. In a survey of articles published in top American management journals over the last decade, only 0.9% focused on cross-cultural interaction. As further exemplified in the most commonly used label for the field, "comparative management," both the field's earlier studies and the more recent research have emphasized within-culture description and comparison. This trend is not surprising given the field's inception during the era of international firms operating multidomestically rather than their current increasingly global scope. While the field of international management has included cross-national interaction studies, they tended to emphasize macro-level organizational, structural, and financial issues, not organizational behavior issues. The interaction of people within and between organizations has remained the domain of comparative management, thus casting most research in the light of single-culture descriptions and multicountry comparisons. 

Implicit in many of the descriptive and comparative studies is the assumption that people behave similarly with their domestic colleagues as they do with their foreign counterparts. That is, these studies suggest that domestic negotiating styles predict international styles. Because most international businesspeople need to know how to negotiate with people from other cultures not simply how foreigners negotiate among themselves the validity of this assumption is critical. If valid, then the comparative literature can directly inform international managerial behavior; if invalid, we need to understand the ways in which within-culture behavior differs from intercultural behavior. The question, therefore, has both theoretical and practical importance. 

To our knowledge, no other researchers have looked at the consistency of negotiator behavior across intra-culture and cross-cultural bargaining. In the present study, the behaviors of businesspeople from four culturally distinct groups (i.e., 190 Americans, 72 Japanese, 100 Francophone Canadians, and 100 Anglophone Canadians) are compared across intra-cultural and cross-cultural situations. 

Theoretical background

In their seminal article regarding international negotiation behaviors, Sawyer and Guetzkow were among the first to posit that negotiators' behaviors and outcomes can be influenced by situational constraints, i.e., cross-cultural negotiations versus intra-cultural negotiations: 

Support for their supposition has come from a broad array of disciplines. The cross-cultural communication and psychology literature suggests that people behave differently with members of their own culture than with members of foreign cultures. Research in nonbusiness contexts has demonstrated that when individuals interact with people from different cultures, the differences between them become salient. Moreover, when people in interpersonal situations confront these actual differences, they tend to exaggerate them. And, when differences become very apparent, some research suggests that relationships among managers deteriorate. Perceived similarity, not difference, has been the important predictor of satisfaction with work relationship. 

Mishler reports that in international exchanges: "The greater the cultural differences, the more likely barriers to communication and misunderstandings become". Some researchers have even questioned whether "managers from significantly different cultures such as Japan and the United States can ever completely understand each other". Studies in the following five research areas are particularly relevant. 

Interpersonal orientation

Most of the literature summarized in later sections suggests that negotiators will adjust their behavior from one situation to another. However, Rubin and Brown imply that people with a low interpersonal orientation (IO) will behave consistently across intra-and cross-cultural situations. They suggest that a high (IO) person is "responsive to the interpersonal aspects of his relationship with others. He is both interested in, and reactive to, variation in the other's behavior". Alternatively, a low IO is "characterized, first and foremost, by a nonresponsiveness to the interpersonal aspects of his relationship with the other..." .

Thus, one might conclude that some people will behave in the same way no matter who is on the other side of the negotiation table someone from the same culture or someone from a different culture. Graham and Herberger carry this idea one step further when they suggest that American negotiators naturally tend to be low IOs: 

So an explanation for ethnocentricity and obstinacy at the international negotiation table is offered. However, most of the rest of the pertinent literature argues that behavior differences will occur across the two settings, and for a variety of reasons. 

Negotiator similarity

The present study provides an excellent opportunity to test Evans' "similarity hypothesis". Evans' ideas "the more similar the parties in a dyad are, the more likely a favorable outcome, a sale" have stimulated a series of studies investigating relationships between similarity and a variety of negotiation outcomes. Weitz (1979), in his excellent critical review of this stream of research, concludes that support for Evans' similarity hypotheses is weak, and in some cases, flawed by confounds. However, the previous work provides an important background for the issues to be considered here. 

McGuire cites a "considerable body of evidence" and posits the mechanism underlying the influence of similarity: 

Mathews, Wilson and Monoky conclude that "perceived" similarity leads to more cooperative behaviors during negotiations. Despite the threats to internal and external validities of the Mathews et al. study identified by Weitz, their reasoning is consistent with Evans. That is, similarity affects negotiation processes, such as the degree of cooperation. So, negotiators in same culture dyads can be expected to behave more cooperatively than negotiators in cross-cultural dyads. Graham provides empirical support for greater cooperation in intra-cultural negotiations compared to cross-cultural. 

Evans, Davis and Silk, and Bagozzi all discuss at some length the relationship among similarity, attraction, and outcomes. Implied in Evans' work is a causal relation among the three constructs, with attraction intervening: similarity → attraction → outcomes. Thus, negotiators in same culture dyads might be expected to be more attracted to partners and achieve higher negotiation outcomes profits and satisfaction. 

Communication problems

Closely related to the issue of negotiator similarity are cross-cultural communication problems. Everyone writing in the area of international negotiations reports substantial communication problems at the negotiation table which often lead to undesirable outcomes for one or both parties. Condon's views are most insightful he classifies cross-cultural communication problems into four categories: 

1. Language and language behavior; 
2. Nonverbal behavior; 
3. Values; 
4. Patterns of thought. 

Condon adds that these categories might be considered in order of ascending perplexity. That is, misunderstandings at the level of language are often obvious and most easily corrected. Misunderstandings at the lower levels are seldom obvious to the participants in an interaction. 

Language is a system which can be studied, described and taught. Our understanding of nonverbal behaviors and communication is, by comparison, disorganized and incomplete. Unlike the first two categories, cultural values are not directly observable. Values, attitudes, and other comparable terms are abstractions, broad concepts which provide a basis for understanding otherwise apparently unrelated behaviors. As such, values and their relation to communication are even more difficult to investigate in a systematic way. Regarding the fourth category, patterns of thought, Condon adds that cultural differences at this level

may result from a combination of differences in language and values. Thus, "while one may come to understand or at least appreciate the vital differences in cultural values, many have given up trying to understand how another person 'reasons'". In cross-cultural negotiations, we might expect problems of communication caused not only by what is said, but also by how what is said is interpreted. 

Empirical support for Condon's views is broad. Of particular interest is a recent article by Graham and Andrews which describes in depth how communication problems, at all four levels, result in undesirable outcomes for Americans negotiating with Japanese businesspeople. It follows then that negotiation outcomes will be less favorable in crosscultural negotiations because communication problems are much more likely to occur. 

Reciprocity and interactional synchrony

A series of studies by social psychologists and sociolinguists suggests that negotiators in a dyad tend to imitate one another's behaviors and balance individual negotiation outcomes. Gouldner explains that a "reciprocity norm" establishes a stable set of mutual rewards that guides interactions such as negotiations. Putnam and Jones report that reciprocity is more evident in integrative message patterns than in distributive strategies. Walton and McKersie, Rubin and Brown, and Pruitt all describe a tendency of negotiators to match one another's bargaining strategies. 

Even deeper than Gouldner's reciprocity norm are the unconscious influences of the sociolinguistic concept of interactional synchrony. Condon and others have reported that a speaker's body movements are coordinated with one another and coordinated with the articulation of speech. Moreover, these movements manifest a hierarchical organization parallel to that of speech. That is, minor body movements may be associated with phrase transition within sentences, whereas grosser body movements may be associated with thematic transition within a conversation. More significant is the finding that listener's back-channel verbal responses and body movements (e.g., the use of the word "yes" or head-nodding to indicate comprehension) are also coordinated in the same hierarchical way with the articulation of another's speech. Interactional synchrony has been defined as the isomorphism of verbal and nonverbal behaviors between speaker and listeners (i.e., "the precise 'dance-like' sharing of micro-body-motion patterns of change between speakers and listeners").

Therefore, based on these concepts of reciprocity and synchrony, negotiators in cross-cultural interactions might be expected to adapt their usual intra-cultural behaviors to more closely reflect those of their foreign counterparts. Likewise, outcomes of cross-cultural negotiations may reflect a compromise between results typical of the differing intracultural styles. 

Acculturation theory

Acculturation theory suggests what might happen at the point of culture contact. That is, what will result from the mix of negotiation and communication styles? 

Acculturation theory is a "mature" paradigm in anthropology. It received the most attention during the 1930s and 1940s. This attention was primarily a response to problems with Indian peoples in the Americas and problems of British colonial rule. The questions were: To what extent can indigenous peoples be assimilated into "advanced" cultures, and how might this process of assimilation be facilitated? The most widely accepted definition of acculturation is that of Redfield, Linton, and Herskovits: "Acculturation comprehends those phenomena which result when groups of individuals having different cultures come into continuous first-hand contact, with subsequent changes in the original cultural patterns of either or both groups". 

In most cases, including the above, the emphasis was on "what" happens when cultures meet. Secondarily (in keeping with the tradition of anthropology), the "why" of the phenomena was investigated and hypotheses formed. Consequently, most research into acculturation is descriptive in nature. However, implicit in most reports is a causal model: the dependent variable in most studies might be described as some degree of acceptance of aspects of another culture. At one end of such a "scale of acceptance" is the complete assimilation of behaviors, traits, or values of another culture. At the other end of the scale is "reaction," that is, rejection of the aspects of the other culture. Somewhere in the middle of the scale is adaption or syncretism the mixing of the two cultures, or as Beals puts it: "While the totality of the emergent culture may be regarded as a syncretism, large areas of the social structure may be essentially new". More complex conceptions of the dependent variable have also been suggested (e.g., Bateson). 

Various explanatory variables have been posited, including types of contact situations (e.g., friendly vs. hostile; inequalities in size, complexity or power of groups; existence of force), processes of contact (e.g., order and manner of presentation or cultural traits, resistances, time constraints), and individual characteristics (e.g., class, role, status, and personality differences). The degree of acceptance was hypothesized to be functionally related to three classes of variables situation, process and individual characteristics; i.e., A = f(S,P,I). It should be noted that power (or force) was the most commonly reported important influence on acculturation. Such a paradigm bears more than coincidental resemblance to attitude change models and the social exchange theory (described in a subsequent section of the paper). Only the dependent variable is different acceptance versus attitude change or specific behaviors such as agreements. 

Another difference between the acculturation paradigm and more recent social-psychological models is the units of analysis. Acculturation theory has really been applied in a macro sense, the units of analysis being entire cultures. The units of analysis in social psychology have been the individual or, at most, small groups.

Acculturation theory fits the specific situation of cross-cultural negotiations very well. Acculturation theory is particularly useful if process measures are selected as the dependent variable. That is, what factors will determine which parties will adopt which negotiation and communication styles, given that these styles are culture specific?

Certainly, the most obvious example is language. What language will be spoken during cross-cultural negotiations? Will one party adopt the language of the other party? Given that the parties to the negotiation are from cultures X and Y, there exist several possibilities or operational definitions of the dependent variable, adoption of language:

1. Language X used; 
2. Language Y used; 
3. Language X used part of the time, language Y used the rest, by both parties; 
4. Interpreters used for translations;
5. A third language, Z, used; and 
6. Combinations of the above. 

As mentioned earlier, acculturation theory suggests three classes of determinants situation, interaction process, and individual characteristics. Certainly one individual characteristic will be critical linguistic capability. The results of previous research also suggest that power difference will be another important determinant. 

The circumstance of Japanese and American cross-cultural negotiations seems to fit this theory. Most often, English is the language spoken during the negotiations between Japanese and Americans. Part of the explanation is that the Japanese possess greater linguistic abilities than Americans. Japanese schools teach and emphasize English. However, in the long term, the most important explanation is the power-differential, both economic and military. The Japanese emphasis on learning English can be attributed in large part to the American occupation following World War II. Additionally, until recent times, Japan has been economically dependent on the United States. However, there are exceptions. For example, it is common practice for high-level Japanese executives to use interpreters, even though they may speak and understand English. Here the use of interpreters is expressive of the person's power. Further, with the increasing economic interdependence of the 1980s, changes are taking place. Japanese businessmen now complain about Americans' ignorance of Japanese business customs. 

In the case of negotiations between Canadians, one might also predict a greater use of the English language, and for similar reasons as those described regarding Japanese/American negotiations. That is, Francophones (French speakers) make up only about 25% of the Canadian population, and their per capita income is also lower than the Canadian average, suggesting concomitant power differences. Acculturation theory not only suggests how the process of negotiation might vary acceptance, syncretism, or reaction; but also the theory provides clues to the determinants and associated mechanisms of the variation. 

Acculturation theory can serve as a useful guide for the investigation of the relation of cultural variation of the parties and the process of cross-cultural negotiations. 

Dependent constructs

Negotiation theory

Theory suggests that outcomes of business negotiations will be influenced by three classes of constructs bargainer characteristics, situational constraints, and process-related measures. The first two constructs are exogenous and are determined before negotiations begin. Process-related measures are endogenous constructs which may be influenced by bargainer characteristics and situational constraints, and which in turn may influence the outcomes of negotiations. The present study investigates the influence of a situational constraint (intracultural versus cross-cultural negotiations) on the processes (strategies, interpersonal attraction, and duration), and outcomes (satisfaction and profits) of business negotiations. 

Negotiation process constructs

Problem-solving approach

The problem-solving approach (hereafter PSA) to negotiations involves first an emphasis on questions and getting information from clients about their needs and preferences. Second, once the client's requirements and circumstances are fully understood, then the negotiator accommodates the offering to the client's needs. The focus is on cooperation and an integrative approach, wherein the needs of both parties are honestly discussed and eventually satisfied. For a discussion of the place of honesty in negotiations, see Murray. A PSA then can be concisely defined as a set of negotiation behaviors which are cooperative, integrative, and information-exchange oriented. Such strategies tend to maximize the number of alternative solutions considered, thus allowing negotiators to optimize outcomes. 

The relationship between a problem-solving approach and negotiation outcomes has been frequently investigated during the last twenty years. Different researchers have used various labels for the PSA concept, but findings have been relatively consistent. Generally, PSA has been found to positively influence negotiation outcomes. 

Interpersonal attraction

In addition to the bargaining strategy itself, interpersonal attraction (e.g., like/dislike, friendly/unfriendly feelings) can strongly influence current negotiation outcomes and the success of future transactions. Rubin and Brown, in their review of the negotiation literature, conclude that interpersonal attraction generally enhances bargaining outcomes. McGuire explains that when people are attracted to each other they will make sacrifices (i.e., concessions in a negotiation) to preserve the gratifying personal relationship. Thus, an individual bargainer may give up economic rewards for the social rewards of a relationship with an attractive partner. To the extent that one receives rewards from a relationship with an attractive partner, the person will be more satisfied with the relationship (or in this case, with the negotiated agreement).

It should be noted that interpersonal attraction might be conceived as an exogenous construct determined before negotiations begin, as a part of the combination of the negotiators' characteristics. It may also be argued that attraction is a consequence of the negotiation, an outcome construct. However, in this study, attractiveness is considered a process-related construct. This is consistent with the views of Evans and Zunin and Zunin. Evans suggests that similarity of negotiators leads to more favorable negotiation outcomes. Zunin and Zunin suggest that during the first few minutes of conversation "decisions" are made about interpersonal attractiveness and whether to continue the interaction. 

Time

In the cross-cultural negotiation literature, duration of the negotiations is described as a key aspect of the process. For example, Tung and Van Zandt report that negotiations with Chinese and Japanese are exasperatingly long from the perspective of most American managers. Pruitt discusses at great length the pervasive influence of time on negotiations. That is, time limits affect the qualities of the aspirations, concession making, and negotiation satisfaction. Although time limits per se are not varied in this study, negotiators from different cultures may have different expectations about "appropriate" durations, which may in turn influence behaviors. 

Time can also be thought of as an outcome construct. Indeed, Green et al. considered duration of negotiations as such. This takes into account the economic value of a negotiator's time. However, Hall has suggested that such a "time is money" view is peculiarly American.

Negotiation outcomes

Profits and satisfaction

Researchers often find outcomes of business negotiations difficult to measure and to compare. Various studies have used sale versus no sale, an obvious measure of bargaining effectiveness, profits obtained by bargainers, and a combination of individual and joint profits. Beyond profits, negotiator satisfaction is an important measure of success, especially if partners desire a continued relationship. Given the dual importance of task accomplishment (profit) and relationship building (satisfaction), especially in international negotiations, the present study uses both as outcomes. 

Hypotheses

The literature suggests a series of hypotheses to be tested, several of which are conflicting. At the more global level, we find implied in many comparative studies that negotiator behavior will not vary between cross-cultural and intra-cultural situations. Rubin and Brown and Graham and Herberger provide explanations for such obstinacy or the null hypothesis. 

HO   Processes and outcomes in cross-cultural negotiations will be no different from those in intra-cultural negotiations. That is, a person from culture X will negotiate with a person from culture Y in the same way as a person from culture X will negotiate with a person from culture X (i.e., XY = XX). 

Alternatively, the preponderance of research suggests that negotiators adjust behaviors in cross-cultural negotiations, thus leading to variations in processes and outcomes. 

HI     Processes and outcomes in cross-cultural negotiations will be different from those in intra-cultural negotiations (i.e., XY ≠ XX). 

The research regarding negotiator similarity and cross-cultural communication problems suggests behaviors and outcomes will vary across cultural situations in the following manner: 

Hla    Cooperativeness (PSA), interpersonal attraction, satisfaction, and profits will be lower and duration (time) longer in cross-cultural negotiations than in intra-cultural negotiations (i.e., when XY ≠ XX, then XY < XX). 

The literatures on reciprocity and interactional synchrony also suggest that variations in behaviors and outcomes will occur across situations. However, the direction of variation will differ: 

Hlb    When cooperativeness (PSA), interpersonal attraction, time, satisfaction, and profits differ from crossto intra-cultural situations, the cross-cultural behaviors and outcomes will be similar within the dyads (i.e., when XY ≠ XX, then XY = YX). 

The reader will note that hypotheses Hla and Hlb are not mutually exclusive. 

Acculturation theory suggests that Japanese and Francophone negotiators will adapt in cross-cultural settings to a greater degree than their American and Anglophone counterparts. This last "hypothesis" cannot be formally tested using this research design. However, it will be worthwhile to consider the results from this perspective. 

Research methods

How the Study Was Conducted

Negotiation simulation

The simulation, developed by Kelley and used by Pruitt, and Clopton, involves negotiating for the prices of three commodities. Each bargainer receives an instruction sheet, including a price list with associated profits for each price level. Participants were given fifteen minutes to read the instructions and plan their bargaining strategies, and up to one hour to negotiate. The simulation has both competitive and cooperative characteristics; that is, negotiators can attempt to maximize individual or joint profits. While simple enough to learn quickly, the simulation usually provides enough complexity for substantive interaction. Of the other negotiation simulations considered, Kelley's appeared to simulate best the essential elements of actual commercial negotiations as observed in preliminary research. Please see Appendices 3A and 3B for more details.

Following the bargaining session, participants completed a questionnaire. To assure equivalence, the French and Japanese translations of both the simulation instructions and the questionnaire were backtranslated into English by second translators; the original and backtranslated English versions were compared and discrepancies resolved. While participants conducted within-culture negotiations in their native language (English, Japanese, or French), the language of intercultural negotiations was chosen by the pair negotiating.

Since this study sought and measured negotiators' "natural" behaviors, the methods described above depart significantly from traditional experimental manipulations and associated checks. This departure not only allows for a claim to greater external validity than the traditional design, but more important, allows both negotiators to adapt their behavior during the interaction. The present approach has proven useful in simulating actual negotiator behavior. 

Participants

One-hundred-ninety American, 72 Japanese, 100 Canadian Francophones, and 100 Canadian Anglophones participated in the negotiation simulation. Since students and businesspeople bargain differently, the sample was limited to experienced businesspeople. All had at least two years of business experience in their respective countries and were volunteers from executive education or graduate business programs. Their average work experience was 9.3 years and the average percent of inter-organizational contact was 48.1. See Table 3.1 for more detail regarding each group. Participants were randomly paired and randomly assigned to play the role of the buyer or seller with either a same culture partner (80 American/American pairs, 21 Japanese/Japanese pairs, 37 Anglophone/Anglophone pairs and 37 Francophone/Francophone pairs) or different culture partners (30 Japanese/American pairs and 26 Anglophone/Francophone pairs). 

Mean age, years of work experience, and percentage of work involving contact outside the company, with standard deviations in parentheses, for the four participant groups: the United States, Japan, and Canada (split into Francophone and Anglophone samples).

Data collection instruments

Three process-related measures have been included in the analyses. Each negotiation was timed. Cooperativeness (PSA) and interpersonal attraction were measured using 5and 3-item scales respectively. These are included in Appendix 3C. 

The study considered three negotiation outcomes individual profits, joint profits, and satisfaction. Profits are derived directly from the negotiated agreement. Satisfaction with the negotiation was measured using a 4-item scale on the post-simulation questionnaire. See Table 3.2 and Appendix 3C for a detailed explanation. 

The Japanese negotiatiors participating in intra-cultural interactions completed a shorter questionnaire. Thus, only a 3-item measure of cooperativeness and a 1-item measure of satisfaction were used for that group. More detail is provided in Table 3.2 and Appendix 3C. 

Results

Study Results: Cross-Cultural vs. Intra-Cultural Negotiation

Qualities of measures

As can be seen from the Cronbach α coefficients in Table 3.2, the reliability of each measure used in the study is adequate (α > 0.65). Also, the high correlations between the longer and the shorter scales (the latter used by some of the Japanese) suggest convergence. 

Hypotheses tests

Analysis of variance was used to test for statistically significant differences between groups. As indicated in Table 3.3, in the majority of cases, no differences were found between intra-cultural and crosscultural situations. However, a few key differences are also evident, particularly supporting H1. Americans reported higher satisfaction (p < .05) in cross-cultural negotiations than in intra-cultural negotiations. Japanese reported higher levels of interpersonal attraction (p <.05) and achieved lower individual (p < .05) and joint (p < .10) profits in crosscultural negotiations. The Francophone Canadians used more cooperative (PSA) strategies (p < .05) in cross-cultural negotiations. The Anglophone Canadians spent more time (p < .05) and achieved lower joint profits (p < .05) in cross-cultural negotiations. 

Recall that Hla stated that less effective processes and worse outcomes would be associated with cross-cultural negotiations than with intra-cultural negotiations (i.e., XY ≠ XX, XY <  XX). This hypothesis is supported in that the Japanese achieved lower profits (both individual and joint) and the Anglophone Canadians took more time and achieved lower joint profits in cross-cultural negotiations. 

Similarities in processes and outcomes within cross-cultural dyads were predicted in Hlb (i.e., XY ≠ XX, XY = YX). With Canadian negotiators, the hypothesis is partially supported no statistically significant difference was found between Francophone and Anglophone PSA strategies in cross-cultural situations. Likewise, no statistically significant difference was found between satisfaction levels of Japanese and Americans in cross-cultural interactions (using the 4-item measure of satisfaction, the mean for Japanese in cross-cultural negotiations = 15.7).

Each row is a measure used in the study, showing its mean with standard deviation in parentheses and the Pearson correlations among all variables (groups pooled). The matrix is lower-triangular, so each pair appears once and blank cells are intentional; ** marks correlations significant at p < .05.

Analysis-of-variance results comparing intra-cultural and cross-cultural negotiations for four groups (United States, Japan, Francophone Canadians, Anglophone Canadians). Each cell shows the group mean with its standard deviation in parentheses; superscript markers indicate statistical significance and scale notes, explained beneath the table.

Discussion

Concisely stated, the central question of the study has been: Are negotiation processes and outcomes the same in cross-cultural interactions as in intra-cultural interactions? The study has considered six important negotiation constructs and tested for variation using four distinct cultural groups. Changes in behavior and outcomes have been observed in seven out of the twenty-four possible instances. The overall conclusion of the study therefore must be that negotiators do indeed adapt their behaviors in cross-cultural interactions. 

Negotiatiors in each of the four cultural groups made changes. Changes were also made in each of the six variables processes and outcomes. American negotiators were more satisfied in cross-cultural interactions. The Japanese were more attracted to American negotiators than their fellow Japanese, even though their profits were reduced when bargaining with Americans. Francophone Canadians behaved much more cooperatively with Anglophone Canadians. The Anglophone Canadians spent more time and achieved lower joint profits in crosscultural interactions. 

Consistent with Graham and Herberger's comments, the American negotiators appear to be the most obstinate. Their behavior remained consistent across situations, only their post hoc expressions of satisfaction changed from one circumstance to the next. Rubin and Brown's low interpersonal orientation (IO) label seems to fit the Americans in our study, suggesting they did not heed their negotiation partner's behavior and made few adjustments to their own. 

The lack of American adaptation also is consistent with acculturation theory the Japanese should tend to make more adjustments due to their history of dependence on the American economy. Indeed, the Japanese-American negotiations were conducted entirely in English.

While it is true that the data were collected in Los Angeles and none of the Americans spoke Japanese, neither factor mitigates the basic finding that the Japanese tend to adapt more. Indeed, Americans' general weaknesses in foreign language reflects a societal level problem which manifests itself at the individual level in our study. Although not specifically focused upon in this research, all the conversations between Francophones and Anglophones were conducted in English (i.e., as viewed on the subset that were videotaped), even though data were collected in Montreal and many of the Anglophones were bilingual. However, the results (as reported in Table 3.3) show that both Canadian groups adapted in other ways. Obviously, these issues raised by acculturation theory deserve further attention, especially as global economic relationships continue to shift. 

The literature in cross-cultural communication and Evan's similarity hypothesis indicate that cross-cultural negotiations will be more difficult than intra-cultural negotiations, that both processes and outcomes would be affected in a negative way. These views, summarized in Hypothesis la, are supported in four of the seven change situations discovered. The profits (both individual and joint) were reduced for Japanese in cross-cultural negotiations. Cross-cultural negotiations took longer for the Anglophone Canadians, and their joint profits were lower when working with the Francophones. 

The literature on reciprocity and interactional synchrony suggests that negotiators will imitate or reflect one another's behaviors, and thereby adapt to differing cultural situations. In two cases, such views (i.e., Hypothesis lb) are supported. The Americans raised their level of satisfaction to that of the Japanese in the cross-cultural interactions. And the Francophones increased their cooperativeness (PSA) to the level of their Anglophone counterparts in the cross-cultural negotiations. 

The only finding that cannot be explained by established theory is that Japanese negotiators were more attracted to Americans than they were to their fellow Japanese. We can think of two possible post hoc explanations. First, perhaps the Americans were rated more attractive because of a novelty factor. The questions used in the attractiveness measure (see Appendix 3C) include the terms "interest" and "comfort". And, if the novelty factor was important, then the Japanese might rate the Americans higher on the "interest" items and lower on the "comfort" items. However, an examination of the means of the individual items fails to support such a conclusion. 

The second explanation regards differences in behaviors and attitudes between Japanese buyers versus Japanese sellers. Graham et al. describe an unusual hierarchical relationship between Japanese buyers and sellers, contrasted with a more egalitarian relationship between American buyers and sellers. Thus, one might guess that Japanese sellers facing American buyers (and the latter's more egalitarian approach) might express higher levels of interpersonal attraction. And this appears to be the case Japanese sellers rated American buyers more attractive (X = 13.7) than Japanese buyers rated American sellers (X = 12.4), and the difference was statistically significant (p < .05). 

This latter finding, indeed all the results, indicate that much more work needs to be done in this area. All we have been able to do in this initial study is scratch the surface of an increasingly important phenomenon. We now have some indication that negotiators do change their behavior in cross-cultural negotiations. And we therefore must be quite careful when we interpret and suggest implications of comparative studies. Moreover, some of the theories we have interpersonal orientation, similarity, communication problems, reciprocity, interactional synchrony, and acculturation theory seem to be applicable. But the phenomena also appears to be more complex than our simple approach warrants.

Limitations of the study

It is important to be aware of the limitations and shortcomings of the research design. There are several such issues involved in this laboratory simulation.

Perhaps the most important consideration is the validity of the principal outcome measure, individual profits. Kelley's negotiation simulation and similar measures have been used in other studies, but how well the simulation represents actual business negotiations remains problematic. Any laboratory experiment is open to criticism regarding external validity this research is no exception.

Additionally, much of the evidence supplied for accepting or rejecting hypotheses derives from participants' self-reports and judgments. In particular, the reliability and validity of the process measures depend entirely upon the participants' immediate post-negotiation memory and impressions of the actual negotiation. Future studies should include improvements in both these areas.

Finally, the most profound limitations of this and similar studies may be the theories and methods themselves developed by American behavioral scientists. Perhaps the American behavioral consistency across intraand inter-cultural negotiation situations reflects a similar obstinacy in American theory building, or an artifact of the simulation and measures used. Graham and Gronhaug discuss such issues at length. Hofstede and Bond provide further evidence supporting this final caveat in their comparisons of Chinese and several Western cultures' managerial styles. The latter report that three of the four primary dimensions that explain variations in Western managerial behavior (power distance, individualism-collectivism, masculinity-femininity) are salient to the Chinese while one Western dimension (uncertainty avoidance) is irrelevant and one Eastern dimension (Confucian dynamism) must be added. This "Eastern dimension," Confucian dynamism, was invisible to researchers using Western paradigms: it only emerged when theory and measures were derived from empirical studies of the Chinese. The help of foreign researchers and the use of more inductive methods in foreign cultures will be needed to mitigate such limitations in future studies.