Culture and language: the case of cultural dimensions and personal pronoun use.

Journal of Cross-Cultural Psychology

 | May 01, 1998 | Kashima, Emiko S.; Kashima, Yoshihisa | COPYRIGHT 1997 Sage Publications, Inc

he relationship between language and culture has been a major issue of concern for psychology and anthropology since Sapir (1970) and Whorf (1956) advanced their hypothesis that language determines, or at least influences, the way we look at our world. Although a range of studies have challenged the validity of both linguistic determinism and linguistic relativism on empirical and theoretical grounds (Au, 1983, 1992; Brown, 1976; Brown & Lenneberg, 1954; Foss & Hakes, 1978; Rosch, 1987), recent reconceptualizations of the language-culture relationship that emphasize sociocultural context of language and culture acquisition suggest an interactional relationship between them (Gumperz & Levinson, 1991; Hardin & Banaji, 1993; Hill & Mannheim, 1992; Hunt & Agnoli, 1991; Hunt & Banaji, 1988; Lucy, 1992; Lucy & Shweder, 1979; Sherzer, 1987). According to these views, cognitive processes obviously affect our language use, yet language also shapes our higher order cognitive processes, such as social inferences and value judgments, by virtue of its inherent involvement in the process of acquiring cultural practices. Sherzer (1987), for example, considered discourse to be the concrete expression of language-culturerelationship. Gumperz and Levinson (1991) noted that researchers are increasingly interested in how meaning can be rooted in the systematic uses of language. Hill and Mannheim (1992) further suggested that grammatical categories will form a privileged location for transmitting and reproducing culture and social categories.

Inspired by the recent research that emphasizes the sociocultural context of culture learning and language acquisition, we sought to examine the relationship between culture and language by directly testing the correlation between global characteristics of cultures and rules of language use in those cultures. In the last decade and a half, cross-cultural research on human values (e.g., Chinese Culture Connection, 1987; Hofstede, 1980; Schwartz, 1994; Smith, Dugan, & Trompenaars, 1996) and self-construals (e.g., Kashima et al., 1995; Markus & Kitayama, 1991; Triandis, 1989) have identified several major dimensions along which many cultures can be placed. Hofstede (1980, 1991), for example, researched work values, whereas the Chinese Culture Connection (1987) studied Chinese values in a number of Eastern and Western cultures. Schwartz (1992, 1994) examined values collected from various theoretical and cultural perspectives, and Smith and colleagues (1996) analyzed value orientations based on Parsons and Shils's (1951) theoretical analysis. These researchers conducted a cultural level analysis (Leung & Bond, 1989): Individual responses to questionnaire items were averaged to compute cultural mean scores, which were further aggregated into summary variables (often called cultural dimensions) by some data reduction methods such as ecological factor analysis (Hofstede, the Chinese Culture Connection), multidimensional scaling (Schwartz, Smith et al.), and conceptual analysis (Hofstede). The score for each country on a cultural dimension can be interpreted as a measure of the extent to which a certain value system is consensually held by the sample of the country.

Hofstede (1991) identified four dimensions--Individualism, Power Distance, Uncertainty Avoidance, and Masculinity-Femininity--from data that were based on his earlier work (Hofstede, 1980) on work values of employees in a multinational company and he reported country scores on these dimensions in 50 countries and geographic regions. The Chinese Culture Connection (1987) extracted four dimensions--Integration, Human Heartedness, Confucian Work Dynamism, and Moral Discipline--from responses to values generated by Chinese social scientists and reported country scores from 22 nations. Schwartz (1994) distinguished seven value regions--Conservatism, Affective Autonomy, Intellectual Autonomy, Hierarchy, Egalitarian Commitment, Mastery, and Harmony--and reported country scores on them in 38 countries. Smith and colleagues (1996) conducted an analysis on managers' and administrative staff's responses to value orientation questions based on Parsons and Shils's (1951) theory. Smith (personal communication, February 9, 1996) provided us with country scores on four major variables--Achievement, Universalism, Paternalism, and Involvement--from 43 countries used in their analysis. These dimensions are globally concerned with cultural patterns of social relationships (Fiske, 1991) and imply diversity in cultural conceptions of the person across the cultures. Individualism (Hofstede), for example, is primarily concerned with the relationship between the individual and the collective, or lack of such a relationship. Power Distance (Hofstede) and Hierarchy (Schwartz) largely have to do with the relationship between individuals with different status (we will examine these constructs in more detail later).

These cultural dimensions provide an opportunity to test the relationship between culture and language use, yet few studies have made use of the opportunity. In one of the rare studies of cultural dimensions and language use, Semin and Rubini (1990) investigated the relationship between individualism-collectivism and verbal abuses. They suggested that individualism and collectivism tap cultural differences in conception of the person (e.g., Kashima, 1987; Miller, 1984; Shweder & Bourne, 1982; Triandis, 1995). In individualist cultures, the person is conceptualized as a decontextualized agent who is solely responsible for his or her own actions. By contrast, in collectivist cultures, the person is considered to be enveloped by a collective and suspended in the network of social relationships with other persons. Consequently, Semin and Rubini hypothesized that verbal abuses in individualist cultures (e.g., Northern Italy) are likely to be directed to the individual (e.g., "stupid," "I hope you will be murdered"), whereas those in collectivist cultures (e.g., Southern Italy) may be directed to the individual and others who are significant for the individual (e.g., "I wish a cancer on you and on all your relatives"). Indeed, they found that insults in Southern Italy (Catania) were more directed toward relations of the individual than those in Central Italy (Bologna), which were in turn more relational than those in Northern Italy (Trieste). More generally, the study implies that language may embed a particular conception about relationships among people, or a conception of the person. Furthermore, cultural dimensions such as individualism and collectivism may tap into such conceptions commonly held by the people in a culture. The research thus implies that cultureand language may be connected through the conception of the person.

A conception of the person is also coded in the use of person-indexing pronouns, or deixis, such as "I" and "you" in English. Deixis are used to indicate extralinguistic entities in discourse: Personal deictic pronouns index the speaker and the addressee within the specific social context. Muhlhausler and Harre (1990) argued that this function of deictic personal pronouns contributes to identity negotiation and to the construction of a social reality in the social situation. For example, whether the speaker refers to the conversational partner as tu or vous in a French conversation, or as anata or omae in a Japanese conversation, has immense implications for the relationship between the speaker and the addressee as well as for the definition of the social situation.

On a slightly different plane, a conception of the person unique to a particular culture may facilitate, and is reinforced by, the specific uses of personal indexicals prescribed in the language spoken in thatculture. Supporting such a possibility, Hanks (1990) argued that deictic systems evolve, to a large extent, through culturally specific, situated practices. Specific uses of personal deixis in everyday discourse may require their users to pay close attention to the relevant, specific aspects of interpersonal relationships over and above others. For example, a speaker addressing another person in a language in which male and female persons are addressed in pronouns of distinct forms may attend automatically to gender-relevant information in the interpersonal situations. In addition, a language that allows personal deixis to be omitted from surface utterances in different situations may require its speakers to attend to the relevant social cues.

In fact, in some languages--including English--the use of subject pronouns is obligatory: I'll, or "you" must be mentioned in an utterance even if the referent is unambiguous. By contrast, other languages do not require the utterance of subject pronouns (or subjects, simply), and these words can be dropped from the surface of a grammatical sentence by the speaker's choice. In some Indo-European languages (e.g., Spanish), personal pronouns are not obligatory, partly because the referents can be recovered from the verb inflections. In some languages such as Chinese and Japanese, however, pronouns can be omitted despite the absence of verb inflections and the grammatical rule of subject-verb agreement (Hinds, 1982; Huang, 1984; Kashima & Kashima, 1997; Li & Thompson, 1976). Explanations for such omission of subject pronouns, or pronoun drop as it is often called, have been extensively debated in linguistic and language acquisition literatures (e.g., L. Berman, 1992; R. Berman, 1990; Bloom, 1990; Valian, 1990, 1991). Although a discussion of such literatures is beyond the scope of the present paper, clearly pronoun use in social interactions involves at least two aspects: the question of "which pronoun to use," as examined by Muhlhausler and Harre (1990), and the question of "whether to use a pronoun." We attempt to relate both to cultural differences in conception of the person.

PRONOUN DROP

We propose that the linguistic practice of pronoun drop, particularly the omission of the first-person singular pronoun (e.g., "I" in English), is linked to the psychological differentiation between the speaker and the context of speech, including the conversational partner.(1) An explicit use of "I" (i.e., first-person singular deictic pronoun) signals that the person is highlighted as a figure against the …

Source: Access my Library 

Image Credit: Appalachian State University