psych questions

Cultural Differences in Children's Emotional Reactions to Difficult Situations

Author(s): Pamela M. Cole, Carole J. Bruschi and Babu L. Tamang

Source: Child Development , May - Jun., 2002, Vol. 73, No. 3 (May - Jun., 2002), pp. 983-996

Published by: Wiley on behalf of the Society for Research in Child Development

Stable URL: https://www.jstor.org/stable/3696263

REFERENCES Linked references are available on JSTOR for this article: https://www.jstor.org/stable/3696263?seq=1&cid=pdf- reference#references_tab_contents You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Society for Research in Child Development and Wiley are collaborating with JSTOR to digitize, preserve and extend access to Child Development

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

Child Development, May/June 2002, Volume 73, Number 3, Pages 983-996

Cultural Differences in Children's Emotional Reactions

to Difficult Situations

Pamela M. Cole, Carole J. Bruschi, and Babu L. Tamang

Although cultures vary in terms of how their members appraise situations, communicate emotions, and act on them, little is known about how culture influences children's emotional reactions. This study examined beliefs about revealing emotion in 223 second-, fourth-, and fifth-grade children from three cultures (Brahman, Tamang, and the United States). Interviews yielded descriptions of how children would feel, whether they would want others to know their feelings, why they would or would not, and what they would do in difficult interpersonal situations. Findings revealed three distinct cultural patterns. Tamang were more likely to ap- praise difficult situations in terms of shame than were Brahman and U.S. children, who endorsed anger. Brah- man children, however, were more likely to not communicate negative emotion than were Tamang and U.S. children. The responses of U.S. children appeared to be more problem focused and action oriented than those of Brahman and Tamang children. Age influenced the degree to which children used emotion-focused coping, and affected decisions about communicating anger in Tamang and U.S. children. Features of cultural contexts that influence children's sense of appropriate emotional behavior are discussed.

INTRODUCTION

The capacity to be emotional is a basic attribute of all human beings. Emotions are adaptive processes that involve appraisals of situations in terms of one's well- being and readiness to act to preserve well-being (e.g., Barrett & Campos, 1987; Frijda, 1986; Lazarus, 1991). Moreover, the appropriate regulation of emotion, particularly in terms of communicating and acting on emotion, is crucial to social competence (Eisenberg et al., 1994; Halberstadt, Denham, & Dunsmore, 2001; Hubbard & Coie, 1994; Jones, Abbey, & Cumberland, 1998; Rubin, Bukowski, & Parker, 1998; Saarni, 1999; Selman & Demorest, 1984). Which emotions and be- haviors are appropriate, however, may be culturally variable.

Culture and Emotion

Views differ on the degree to which emotions, and which aspects of emotions, are universal or socially constructed (e.g., Ekman, 1994; Hochschild, 1983; Russell, 1994; Wierzbicka, 1994). Evidence indicates that there are both variations and similarities among cultures in the emotional significance given to situa- tions (appraisal), the manner in which emotions are conveyed from one person to another (communica- tion), and the manner in which people deal with situ- ations that elicit emotion (coping; for a review, see Mesquita & Frijda, 1992). Unfortunately, that body of research has tended to focus on adult emotion. A full

understanding of the role of culture requires the study of cultural variations in children's emotional understanding and behavior as well.

Generally, explicit knowledge of display rules and conflict solutions are thought to be present by middle childhood (e.g., Chung & Asher, 1996; Fuchs & Thelen, 1988; Gnepp & Hess, 1986; Meerum Terwogt & Stegge, 1995; Saarni, 1979; Zelko, Duncan, Barden, Garber, & Masters, 1986; Zeman & Garber, 1996; Zeman & Shipman, 1996). Socialization pressures, however, which vary across cultures, may influence implicit awareness and therefore the age of onset of emotional understanding. In one study, Indian girls were shown to appreciate the distinction between felt and ex- pressed emotion as preschoolers, earlier than Indian boys and British boys and girls, suggesting that spe- cific cultural patterns of socialization affect the age at which understanding emerges (Joshi & MacLean, 1994). In a different study, another sample of young South Asians (Brahman first graders) understood that one can feel angry but should not communicate it on one's face (Cole & Tamang, 1998), suggesting that the study of culture in children's emotion understanding may affect theories of emotional development.

Culture has been conceptualized as a developmen- tal niche in which children acquire knowledge and skills. The developmental niche includes the physical and social features of the child's setting, the commu- nity's child-care and childrearing customs, and care- givers' beliefs about child development (Super & Harkness, 1993). Although this cultural framework has not focused on emotional development, it follows that the developmental niche is a context for acquir-

? 2002 by the Society for Research in Child Development, Inc. All rights reserved. 0009-3920 / 2002 / 7303-0021

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

984 Child Development

ing culturally specific emotional reactions. A first step in exploring cultural variations in children's emotional development is to determine whether children from distinctly different cultural settings describe different ways of reacting emotionally in public situations.

Comparison of Values in Asia and the United States

Cultural priorities appear to affect values with re- gard to specific emotions. For example, Asian values- which are often described as collectivistic, interdepen- dent, or self-accommodating (Baer, Curtis, Grabb, & Johnston, 1996; Markus & Kitayama, 1991; Roth- baum, Pott, Azuma, Miyake, & Weisz, 2000; Triandis, 1994); and place high priority on relationship har- mony and respect for authority-discourage anger expression and value shame (Kitayama & Markus, 1995). Anger involves the appraisal of a situation as blocking one's goals or treating one unjustly, as well as the associated readiness to act to overcome these

obstacles and wrongs (Barrett & Campos, 1987). Shame involves different appraisal and action readi- ness patterns; it involves viewing the self as having behaved wrongly and being ready to avoid or hide from others (Barrett & Campos, 1987). Thus, anger and shame serve different social functions: anger fa- cilitates individual domination whereas shame facili-

tates self-submission (Barrett & Campos, 1987). Anger is discouraged in Asian societies because it

threatens authority and relationship harmony, whereas shame is valued because it involves regarding the self in relation to others (Barrett, 1995; Lebra, 1983; Miller,

1996: Miyake & Yamazaki, 1995; Tangney, 1990). In the United States, greater emphasis is placed on the development of individuality, autonomy, and self- expression in children than in Asian societies (Chao, 1995; Zahn-Waxler, Friedman, Cole, Mizuta, & Hiruma, 1996) and other Western societies (Harkness, Super, & van Tijen, 2000). American society tolerates anger in the interest of self-assertion and protection of individ- ual rights and freedoms, provided it is expressed in socially acceptable ways (Stearns & Stearns, 1986). Shame, on the other hand, is often seen by Americans as harmful to children's self-esteem (e.g., Ferguson, Stegge, Miller, & Olsen, 1999). Thus, because the de- velopmental niches of Asian children include values and practices aimed at relationship harmony and re- spect for authority, it follows that Asian children should develop a culturally specific sense that one must not communicate anger or act angrily. On the other hand, U.S. children, who develop in a cultural setting that values self-expression, self-esteem, and self-assertion, should believe that shame is undesir-

able and that anger can be used to defend oneself. In sum, children's reports of their emotion-related behavior-how they appraise situations, whether they communicate felt emotion, and how they choose to act-should reflect their culture's values.

Although there is little research on cultural differ- ences and emotion, there is evidence that children's social understanding reflects cultural values. For ex- ample, Chinese and Indian children refer more to re- lationship concerns and less to self-interest in discuss- ing conflicts than do Western children (Keller, Edelstein, Schmid, Fang, & Fang, 1998; Miller & Ber- soff, 1992). Social understanding can also vary within societies and within groups (Kagitcibasi & Poortinga, 2000). Korean sixth graders who held relation- oriented, collectivistic ideals solved peer conflicts dif- ferently than did those children who held self- oriented, individualistic ideals (Han & Park, 1995). In addition to individual variation, subgroups within a culture (e.g., different ethnic groups) may hold differ- ent values. Within-nation comparisons may be partic- ularly useful in identifying specific features of cul- tural settings that influence social understanding.

Cultural Variations in Rural Nepal

Rural Nepal is a natural setting for studying cul- tural variations. Nepal and its people can be char- acterized by terms such as collectivistic, interde- pendent, and self-accommodating. Nepal has rich cultural diversity, however; its 1994 census identified 60 distinct groups (Central Bureau of Statistics, 1994). The major ethnic groups retain fairly distinct social customs (e.g., language and dress) due to barriers in commerce, communication, and transportation in re- mote areas. Some of the variations in social organization and customs in Nepali ethnic groups may influence the appraisal of situations and how one communicates and acts on emotion. The Brahman and Tamang cultures of Nepal are two such distinct groups.

There were many common features between the rural Brahman and Tamang villages that participated in this research in the 1990s. The villages were mid- sized (about 1,000 people) and located in Nepal's middle hills, at least a day's walk from a motor-able road. Homes were earthen houses with thatched

grass roofs. Households were composed of parents, their married sons and their wives and children, and unmarried children, averaging seven to eight persons per household. Family members worked, ate, and slept together, except during the hours that children attended school. Adults and older siblings shared childrearing, household, and farming duties. Corn was the main crop and there was rarely surplus. Most

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

Cole, Bruschi, and Tamang 985

children were barefoot and family members received new clothes once a year, if fortune permitted, during the annual autumn festival.

In these villages, everyone knew each other and there were few secrets. Private information was

known widely but not discussed openly (cf. Berre- man, 1997; Cameron, 1998). Children were expected to get along with and know their places in relation to others-values that appeared to be more important than children's individual achievements and unique- ness. For example, everyone was addressed by kin- ship terms (e.g., younger sister), not by given names. These kinship terms were used frequently in any sin- gle conversation. Thus, an individual was constantly addressed differently (e.g., older sister by one person, daughter by another) with terms that were shared by all members of the same gender. In school, children recited lessons in unison, and were not called on to speak individually. A child who was out of step with the group or failed to repeat what the teacher said was punished.

Despite these similarities, Brahman and Tamang societies differed in ways that were expected to influ- ence children's emotional appraisals and actions. Brahmans are Aryan people who migrated to Nepal from northern India. They are high-caste Hindus who speak Nepali, a Sanskritized language. The caste sys- tem organizes their social interactions and personal behavior as a means of preserving high-caste spiritual purity and avoiding spiritual pollution. For example, lower castes may not serve certain foods to Brah- mans, Brahman women must remove themselves from the hearth and home when they are menstruat- ing, interactions are monitored carefully in terms of where one walks and what one touches, and inter- caste violations are penalized (Bennett, 1983; Berre- man, 1997; Cameron, 1998; Gray, 1995; Skinner, Pach, & Holland, 1998). In 1854, the Hindu caste system was first codified into law in Nepal and penalties for vio- lating caste were legalized (H6fer, 1979). Although Himalayan Brahmans are regarded as more lax than those of the Indian plains (Berreman, 1997), within Nepal, Brahmans are regarded as the highest caste. Thus, the village Brahmans have ethnic pride and sta- tus, despite their poverty and lack of formal educa- tion (Berreman, 1997; Cameron, 1998).

The Tamang are regarded as an indigenous or "tribal" people of Nepal (Gautam & Thapa-Magar, 1994). They have no recorded history. They appear to be a Mongol people and are presumed to have emi- grated from China. They practice Tibetan Buddhism, and speak an unwritten Tibeto-Burman language, called Tamang. Tamang society is noted for its egali- tarianism, tolerance, and minimization of psychic dis-

tress (Fricke, 1986; Holmberg, 1989; March, 1986; Tamang, V.S. 2051). For example, Tamang mute differ- ences in fortune or success by sharing what they ac- quire and engaging in community problem solving that takes into account all members' views. The use of

the term Tamang to classify them became official when the Hindu caste system was codified into law (Hofer, 1979). Because Tamang and other indigenous groups engage in behaviors that pollute according to caste law (e.g., eat beef), Tamang are regarded as a lower, but not untouchable, caste. Along with other indigenous groups, the Tamang are also a minority group; until the 1950s they were restricted from edu- cation and, in some areas, from owning land (Hoftun, Raeper, & Whelpton, 1999). Thus, in addition to being a group that tries to minimize individual differences among its members, Tamang may also have an identity of being lower status within the larger Nepali society. Minority status might constrain self-assertion, promote submissive social behavior, and affect one's model of appropriate emotional appraisal and action (Clark, Anderson, Clark, & Williams, 1999; Ogbu, 1981).

In sum, the niche in which Brahman children de- velop achieves social coherence through caste hierar- chy. This form of sociality has been characterized as vertical collectivism, power distance, and authority ranking (Billings, 1989; Fiske, 1992; Hofstede, 1980). Brahman heritage promotes a sense of relative indi- vidual achievement, particularly in contrast to other groups, in terms of both the sense of a heritage as the educated elite, and their higher caste status. Thus Brahman social prestige, pride, and authority over lower castes may support an openness to the experi- ence of anger-an emotion that supports pride and dominance-while understanding that managing anger is required in intracaste relationships. The niche in which Tamang children develop achieves social co- herence through the minimization of individual dif- ferences and egalitarianism, a form that has been called communal sharing or horizontal collectivism. The Tibetan Buddhist perspective is consistent with these values and adds an obligation to make oneself selfless and calm. In intercaste relations, Tamang are a minority group of lower status. Thus, there may be multiple cultural pressures that deter Tamang chil- dren from experiencing and communicating anger. These various factors may create a niche for the devel- opment of culturally specific ideas about appropriate emotional appraisal, communication, and action.

Purpose of the Study

We predicted that culture would influence chil- dren's descriptions of their emotional reactions in dif-

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

986 Child Development

ficult interpersonal situations in the following ways. First, we predicted that U.S. children, by virtue of a cultural emphasis on self-assertion and individual achievement, would endorse anger more often than would Nepali children. As a corollary prediction, based on cultural values on maintaining group har- mony and respecting authority, we expected that Ne- pali children would be more likely to appraise situa- tions in terms of shame than would U.S. children.

Because previous research found a tendency for Brah- man first graders to endorse anger more than did Tamang children, we predicted this difference in older children. Moreover, we expected that various influ- ences of Tamang culture, religion, and relative social status differences might lead Tamang children to en- dorse shame more than would Brahmans children.-

Culture was also expected to influence whether anger or shame was communicated. United States children were expected to reveal anger more than would Brahman children, because anger supports self-assertion. Brahman children, who may feel en- titled to endorse anger, would nonetheless have a social consciousness that one can and should act dif-

ferently than one feels. They were expected to say that one does not communicate anger more than would Tamang children. Shame was more difficult to pre- dict: we expected that U.S. children would report not showing shame but that Tamang children would re- port showing shame. As a check that cultural values were organizing these emotional reactions, children's associated justifications and actions were studied in hypothetical difficult situations. We predicted that U.S. children would justify their reactions from a self- assertive, problem-focused point of view, and de- scribe efforts to alter difficult situations. Nepali chil- dren would rely on internal, emotion-focused rea- sons, and would accept, rather than act to alter, difficult situations. These predictions are based, in part, on the fact that Asian youth appear to engage in emotion-focused or secondary control coping more than Western youth (Essau & Trommsdorff, 1996; Mc- Carty et al., 1999).

This study also examined age as indexed by grade to determine the stability or variability of cultural patterns during the transition to middle childhood. The paucity of cross-cultural developmental work limited the ability to make predictions. Brahman first graders understood that one might feel angry but should not communicate or act on the anger (Cole & Tamang, 1998), lending support to the interpretation that Hindu socialization pressures influence the age at which appropriate cultural knowledge emerges (Joshi & MacLean, 1994). Thus, it may be that Brah- man children acquire culturally specific display rules

earlier than other children and that age-related differ- ences may depend, in part, upon a child's cultural membership. We predicted, therefore, that changes in emotion and communication would exist only for non-Brahman groups. Changes in social expectations and cognitive skills that accompany development in middle childhood and allow children to reflect on in-

ternal state processes and increase their repertoire of coping skills might, however, influence all children's responses to communicating negative emotions and their justifications of why they would or would not show certain emotions.

In addition to including two age groups and exam- ining children's explanations for their responses, the present study built on earlier work in two other ways. First, shame, a highly valued negative emotion in Ne- pali society, was included. Thus, it was possible to de- termine whether cultural differences influenced emo-

tional appraisals as well as communication of emotion. It must be noted, however, that the term for shame in

Nepali and Tamang languages is also the term for em- barrassment and shyness. It should be interpreted as meaning a feeling that the self is not worthy of the in- terest, respect, or admiration of others. Second, a group of U.S. children was included to aid in articu- lating the cultural differences and similarities be- tween Nepali ethnic groups and U.S. society. Such a comparative design risks losing the essence of the phenomenon of study in trying to maintain relevance across cultures (Poortinga, 1989). In the present study, this risk was minimized by collaborating with a Ne- pali (third author) and field-testing procedures in each locale.

METHOD

Participants

Two groups of children (younger and older) from three cultural groups (Brahman, Tamang, and the United States) participated in the study. The younger children attended second grade in each community. The older children attended fifth grade and, because there were not enough fifth graders in the Nepali vil- lages, Nepali fourth graders were also included. De- spite this, there were no significant age differences be- tween cultural groups at either age level, both ts < 1. The Brahman children (n = 62) lived in a farming vil- lage in western Nepal (younger age: M = 8.91 years, SD = 1.04; older age: M = 11.45, SD = .82). The Tamang children (n = 62) lived in a similar commu- nity in eastern Nepal (younger age: M = 9.44 years, SD = 1.21; older age: M = 11.46, SD = 1.45). The U.S. group (n = 99) lived in a rural northeastern farming

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

Cole, Bruschi, and Tamang 987

community (younger age: M = 8.96, SD = .40; older age: M = 11.82, SD = .66). The 223 children were rel- atively evenly divided in terms of child gender (110 boys, 113 girls) and age (102 younger, 121 older).

Procedure

Materials

Children's beliefs about emotion were assessed in

interviews administered individually to each child during the day in a private location at the school. The interview was conducted in the child's mother tongue by a native adult speaker. The interview involved nine vignettes that described emotional situations (see the Appendix), which were read one at a time to the child. Graphic depictions of each vignette were used to aid the child's understanding. The procedure took approximately 30 min.

Because of the differences between life in U.S. and

Nepali farming communities, existing procedures for assessing emotion knowledge could not be used. For example, a story about a birthday gift (Saarni, 1979) is not emotionally significant in rural Nepal because birthdays are not celebrated. Therefore, vignettes were developed for this study that followed the for- mat of stories in the literature on children's emotion

knowledge. The stories were composed in English and translated and backtranslated by Nepali native speakers who edited them to reflect everyday speech while preserving meaning. The stories were then piloted in each culture.

The resulting nine stories consisted of six difficult and three pleasant situations. These were presented to each child in one of four orders. In each order, two difficult situations were followed by a pleasant situa- tion to provide relief from negative content. (Data generated in response to pleasant situations were not included in the data analyses.) Difficult situations were counterbalanced for (1) the potential of the situ- ation to elicit shame versus anger appraisals (i.e., in- justice and thwarted goals, which might be more likely to evoke anger; and public mistakes or mis- haps, which might be more likely to evoke shame), and (2) who elicited the child's emotion (see the Ap- pendix). The vignettes varied with regard to whether a friend or parent elicited the emotion but a fully crossed model, in which anger and shame varied sys- tematically with interpersonal context, required more stories than the young participants could tolerate.

Stories were told in the second person to help the child imagine being in the situation. Minor variations in the stories were used to ensure cultural relevance

(e.g., the tea was iced in the United States and hot in

Nepal; the child's chore was raking leaves in the United States and cutting grass in Nepal). The graphic depictions of the situations were also adapted for each culture. A U.S. artist used photographs of Nepali village life to guide his portrayal of physi- cal qualities of Nepali villages and their residents (e.g., facial physiognomies, clothing, buildings, and landscape).

To assist each child in endorsing emotions, sche- matic drawings of the four targeted emotional expres- sions were provided. The drawings were based on re- search on the appearance of the eyes and mouth and the posture of the head for the emotions ashamed, angry, happy, and okay (e.g., Barrett, Zahn-Waxler, & Cole, 1993; Ekman & Friesen, 1978; Izard, 1979). The hair and eye shape were adapted for each culture, but the facial configurations and head positions were identical. Male and female versions of the situations, their pictorial depictions, and the facial expressions were used for boys and girls. (A copy of all graphic materials is available on request from the correspond- ing author.)

Interview

After each story was read to the child, the inter- viewer asked five questions:

1. When [the situation] happens to you, how do you feel? This question assessed the child's emotional re- actions in the situations. The child first answered

freely and then was asked to select one of four emo- tions (happy, ashamed, angry, okay) that was closest to the spontaneous reply. The frequency with which the child endorsed each of the emotions based on the

forced-choice procedure was used in the statistical analyses.

2. Would you want [parent/friend] to know that you were feeling [child's emotion term]? This question as- sessed the child's preference for communicating or not communicating each selected emotion. The fre- quency with which children responded yes and no was used in the analyses.

3. How would your face look? Each child used the drawn facial expressions to indicate the face that best represented how they would look in the situation. These replies were used to check the child's under- standing of the situation but were not used in the data analyses. The responses were redundant with chil- dren's responses to the second question.

4. Why would you show/not show your [parent/friend] that you felt [child's emotion term]? This question as- sessed the child's explanations for his or her choices, providing a window into whether cultural values might organize the selected responses. The frequency

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

988 Child Development

with which self-evident, emotion-focused, and problem- focused justifications were given was used in the analyses. The coding for justification based on these open-ended responses is described below.

5. So when you felt [child's emotion term] and wanted to show/not show that you felt that way, then what would you do? This question assessed what the child thought he or she would do in the circumstances. The fre-

quency with which the child described tolerating or accepting the difficult situation and trying to change it was used in the analyses. The coding for behaviors is also described below.

Coding of Justifications for Revealing Emotion

The coding system for categorizing children's rea- sons for why they would or would not communicate their emotions was guided by research on coping, control orientation, and emotion display rule knowl- edge (Heckhausen & Schulz, 1995; Lazarus, 1991; Saarni, 1979, 1997). Every response was coded using one of the following four categories:

1. Self-evident justifications (Saarni, 1997). The child's justification referred to the facts of the situa- tion (e.g., "I showed I was ashamed because I fell." or "I would show that I was happy because mother is happy."). These justifications seem to capture a child's sense that their emotions would be expected given the situation, with no indication that the child was exerting control.

2. Emotion-focused justifications (Lazarus, 1991). The child's justification referred to a reappraisal of the situation or a general rule or principle for emotional behavior that involved the child's state (e.g., "I wouldn't show I'm angry because I don't need the eraser that much anyway." or "It's not polite to look too happy when eating someone else's food."). That is, these justifications were aimed at altering the per- sonal experience of the situation rather than at modi- fying the external circumstances. This has also been referred to as a secondary control process in which change is directed at the self, a pattern that would be consistent with cultural values involving self- accommodation in relationships (Heckhausen & Schulz, 1995; Rothbaum et al., 2000).

3. Problem-focused justifications (Lazarus, 1991). The child's justification referred to an action or a gen- eral rule or principle for behavior involving an ex- plicit intention to alter the situation at the time or in the future (e.g., "If I show I'm happy, then he would help me again." or "If I don't look angry, they'll stop teasing me."). This has also been referred to as a pri- mary control process, involving change directed at external conditions, and is a pattern that is associated

with individualistic societies (Essau & Trommsdorff, 1996; Heckhausen & Schulz, 1995).

4. Uncodable justifications. The child did not reply to the question or the child's explanation could not be categorized in any of the above categories because the justification was illogical, vague, or insufficient in the explanation.

Coding of Ensuing Behavior

Children's open-ended descriptions of what they would do in the situations were categorized as pre- dominantly reflecting self-assertion (i.e., trying to change the situation) to preserve individual needs and rights, or predominantly reflecting tolerance or accep- tance of the situation to preserve social harmony.

1. Acting to change the situation. The child described acting on the situation either physically or socially in such a way as to preserve his or her own rights, needs, and intentions. For the purposes of this study, no dis- tinction was made between appropriate and poten- tially inappropriate behaviors. Examples of chil- dren's responses included verbal assertions ("I would say, 'It's not nice to laugh at me."'), challenging or questioning the behavior of the other ("I would say, 'Why did you hit me?"'), and fixing the problem (re- doing or correcting homework). As in the case of problem-focused, or primary control strategies, this category of behavioral choice reflects a sense that one can and should act to change undesirable events and situations.

2. Accepting the situation. The child described not acting on the situation but either staying quietly or moving away from the person who provoked the emotion. Examples included leaving ("I would go somewhere else."), sitting quietly and saying noth- ing, and altering one's thoughts ("Why be upset? The homework has to be done again anyway."). Descrip- tions of talking to others in the situation were not in- cluded in this category. This category of behavior re- flects values that one should not, or cannot, change situations.

3. No reply or uncodable response. The child did not answer or gave a description that could not be coded as trying to change or accept the situation.

Nepali children's responses were translated by the first and third authors into English for coding. The translations were then checked by an independent translator fluent in all three languages. The translated emotion display justifications and behavioral choices were coded by two U.S. undergraduate research as- sistants who were unaware of the study predictions. The second and third authors served as reliability checks for the coding. Reliability assessments were

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

Cole, Bruschi, and Tamang 989

conducted for a randomly selected 25% of the proto- cols for each cultural group. There was 89% agree- ment between raters for emotion display justifica- tions and 92% agreement for behavior choices. The average Cohen's K calculated for each category ranged from substantial to almost perfect (M = .84, range = .66-1.00).

RESULTS

Effects of Culture and Age on Emotion and Communication

Three mixed-factorial MANOVA models were used

to examine the effects of culture (Brahman, Tamang, and the United States) and age (younger, older). The first MANOVA examined whether the emotions that

children endorsed (angry, ashamed, happy, okay) and their responses about communicating each emotion (revealing, not revealing) were affected by the child's cultural group and age. The second MANOVA exam- ined the same group differences in children's reason- ing (self-evident, emotion focused, problem focused), and the third examined actions they believed they would take in difficult situations (accept situation, try to alter situation). Because the regulation of negative emotion was the focus of the study, these analyses were limited to data from the six difficult situations.

The MANOVA results are presented in Table 1 and significant findings are described in the text. Signifi- cant Fs were followed by planned (t test) comparisons for predicted effects and Scheffe's tests for a posteriori comparisons.

The literature on culture and children's emotion

understanding provides mixed evidence of gender differences. Gender was not featured in the study's hypotheses but it was necessary to determine whether gender should be considered. Therefore gender was entered into the full mixed model, which yielded one significant effect, an interaction of Emotion X Culture X Gender, F(6, 633) = 2.21, p < .05. United States girls endorsed anger more than did Nepali girls. Given that gender did not add considerably to the overall find- ings, it was omitted from further analyses.

Emotions and Communication

A significant main effect of emotion indicated that children endorsed feeling angry (M = 3.10, SD 1.47) more than they endorsed feeling ashamed (M = 2.28, SD = 1.32), and they endorsed both of these neg- ative emotions more than they endorsed the positive emotions of happy or okay (Ms = .21, .37, SDs = .57, .83, respectively). Each of these means differed signif-

Table 1 Summary of Analyses of Variance Results for Culture and Age Effects for Emotion and Communication, Justification, and Behavior

Effect Source df F

Emotion 3, 651 280.41*** Communication 1, 217 .08 Culture 2, 217 .34 Age 1, 217 1.82 Emotion X Communication 3, 651 10.78*** Emotion X Culture 6, 651 8.57*** Communication X Culture 6, 651 42.79*** Emotion X Age 3, 651 3.08* Communication X Age 1, 217 1.49 Emotion X Communication X Culture 6, 651 24.30*** Emotion X Culture X Age 6, 651 1.86 Communication X Culture X Age 2, 217 6.73** Emotion X Communication X Culture X Age 6, 651 6.85*** Justification 2, 434 20.07*** Justification X Culture 4, 434 63.96*** Justification x Age 2, 434 3.76* Justification X Culture X Age 4, 434 2.24 Behavior 2, 434 53.12*** Behavior x Culture 4, 434 7.38**** Behavior X Age 2, 434 .31 Behavior x Culture X Age 4, 434 1.46

Note: *p < .05; **p < .01; ***p < .005; ****p < .001.

icantly from all the others, all ps < .01. A significant interaction of emotion and communication indicated

that children preferred to reveal positive emotions, but there was no general preference for revealing or not revealing anger and shame.

Effects of Culture and Emotion and Communication

The predictions for cultural differences in endors- ing emotions were confirmed, with one exception, by a significant two-way interaction of culture and emo- tion (see Table 1). Tamang children endorsed feeling shame (M = 2.76, SD = 1.42) more than did U.S. chil- dren (M = 2.16, SD = 1.13), t(159) = 2.80, p < .01, and Brahman children (M = 1.98, SD = 1.40), t(122) = 3.06, p < .005. Brahman and U.S. children did not dif- fer in appraising situations in terms of shame, t < 1. Also, as predicted, U.S. children endorsed anger (M = 3.06, SD = 2.23) more than did Tamang chil- dren (M = 2.06, SD = 1.53), t(159) = 3.06, p < .005, but, contrary to prediction, U.S. and Brahman children (M = 3.23, SD = 3.16) did not differ in this regard.

The prediction that cultural setting would also in- fluence children's decisions about communication

was confirmed by a significant Culture x Communi- cation interaction (see Table 1). Brahman children pre- ferred not to communicate emotions (M = 4.55, SD = 1.47) more than did Tamang children (M = 2.37, SD =

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

990 Child Development

1.95), t(122) = 7.02, p < .001, and U.S. children (M = 2.04, SD = 1.78), t(159) = 9.28, p < .001. Tamang and U.S. children did not differ in their responses about com- municating emotions, t < 1.

An unexpected significant interaction of culture, emotion, and communication, involving positive emotions, was also found. Specifically, Brahman chil- dren said they would show happiness in difficult sit- uations more than did Tamang children, p < .05, and U.S. children, p < .005. Tamang children said they would show how they felt okay more than did Brah- man children, p < .005 but not U.S. children.

Effects of Age and Emotion and Communication

There was a significant interaction between age and emotion (see Table 1). Specifically, older children reported feeling happy twice as often (M = .28, SD = .70) as did younger children (M = .13, SD = .34), t(222) = 4.13, p < .05. There were no age differences for feeling angry, ashamed, or okay, all ps > .10.

Age was examined to understand whether there were patterns of stability or change in the communi- cation of particular emotions. We expected Brahman children to show a stable pattern of not revealing neg- ative emotion but examined whether cognitive ad- vances of middle childhood were associated with a

greater tendency to hide negative emotions among U.S. and Tamang children. A complex and significant four-way interaction of emotion, communication, age and culture suggested different patterns for each cul- tural group. Given its complexity, this interaction was deconstructed by examining how culture and age af- fected children's responses about not communicating anger and shame. (Happy or okay feelings did not factor into this four-way interaction.) All of the means and standard deviations associated with the four-

way interaction are presented in Table 2. Communicating anger. As expected, age influenced

decisions to not communicate anger for Tamang and U.S., but not Brahman, children. As shown in Table 2, Brahman children were more likely to not reveal anger than were Tamang children in both the older groups, t(57) = 5.91, p < .001, and younger groups, t(71) = 2.96, p < .05, and U.S. children in both the older groups, t(71) = 4.40, p < .001, and younger groups, t(86) = 7.51, p < .001. For the Tamang and U.S. children, age influ- enced decisions to reveal anger in different directions: older Tamang children reported showing anger less than did younger Tamang children, p < .05; and older U.S. children reported showing anger more than younger U.S. children, p < .05.

Communicating shame. Regardless of age group, Tamang children believed they would communicate

Table 2 Means and Standard Deviations for Frequencies of Communicating and Not Communicating Each Emotion as a Function of Culture and Age Group

United

Brahman Tamang States Age Group M (SD) M (SD) M (SD)

Revealing anger Younger .77 (.97) 1.65 (1.86) 1.58 (1.50) Older .63 (.87) .76 (1.06) 2.70 (1.31)

Not revealing anger Younger 3.17 (1.58) 1.06 (1.10) 1.58 (1.47) Older 2.44 (1.34) 1.42 (1.41) .68 (.86)

Revealing shame Younger .27 (.52) 1.55 (1.35) 1.26 (1.09) Older .19 (.74) 1.76 (1.44) 1.27 (1.05)

Not revealing shame Younger 1.50 (1.48) 1.03 (1.50) .93 (1.12) Older 2.00 (1.14) 1.15 (1.30) .88 (.90)

Revealing happiness Younger .27 (.45) .07 (.26) .05 (.21) Older .69 (1.12) .24 (.50) .05 (.23)

Not revealing happiness Younger .00 (.00) .00 (.00) .00 (.00) Older .00 (.00) .03 (.17) .02 (.15)

Revealing okay Younger .00 (.00) .62 (.94) .47 (1.10) Older .00 (.00) .52 (1.06) .37 (.68)

Not revealing okay Younger .00 (.00) .00 (.00) .12 (.32) Older .00 (.00) .00 (.00) .02 (.13)

shame, t(57) = 4.85, p < .001, more than did Brahman children, t(63) = 5.52, p < .001. Contrary to predic- tion, U.S. children endorsed showing shame more than Brahman children, a significant difference for both younger, t(71) = 4.60, p < .001, and older age groups, t(86) = 5.12, p < .001. U.S. and Tamang chil- dren did not differ at either age group, both ts < 1. An effect of age was found only when older chil- dren were compared across cultural groups. Older Brahman children did not communicate shame

more than did older Tamang children, p < .01, and older U.S. children, p < .001. In general, then, the age-related findings showed, as predicted, stability among Brahman children but not among U.S. and Tamang children.

Justifications for Communicating or Not Communicating Emotions

United States children were predicted to justify emo- tional responses with a self-assertive, problem-focused point of view, whereas Nepali children were expected

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

Cole, Bruschi, and Tamang 991

Table 3 Means and Standard Deviations by Cultural Group for Justifications

United

Brahman Tamang States Justification Type M (SD) M (SD) M (SD)

Self-evident 2.13 (1.55) 3.16 (1.80) .91 (1.02) Emotion focused 1.52 (1.49) 1.76 (1.49) . 70 (1.05) Problem focused 1.77 (1.71) .92 (.14) 4.20 (1.45)

to rely on internal, emotion-focused reasons. A main ef- fect of justification indicated that children, regardless of cultural group, reported more problem-focused justifi- cations (M = 2.61, SD = 2.07) than emotion-focused (M = 1.87, SD = 1.70) or self-evident (M = 1.22, SD = 1.38) justifications, ps < .001. There was no difference between emotion-focused and self-evident justifica- tions, t < 1. Justifications, however, were also influ- enced by culture, as predicted. They were also pre- dicted by age, but culture and age did not interact (see Table 1). The means and standard deviations for each type of justification for each cultural group are shown in Table 3.

Self-evident justifications. Tamang children used self- evident justifications more often than did Brahman children, t(122) = 3.44, p < .001, and U.S. children, t(159) = 10.11, p < .001. In turn, Brahman children en- dorsed self-evident justifications more than did U.S. children, t(159) = 6.07, p < .001. Age did not influence self-evident justifications, all ts > 1.

Emotion-focused justifications. Tamang children en- dorsed emotion-focused justifications more often than did U.S. children, t(159) = 5.35, p < .001, but Tamang children did not differ from Brahman chil- dren, t < 1. Age only influenced emotion-focused jus- tifications; regardless of cultural group, older chil- dren reported more emotion-focused justifications (M 1.41, SD = 1.42) than did younger children (M 1.00, SD = 1.29), t(221) = 2.20, p < .05.

Problem-focused justifications. United States children reported more problem-focused justifications than did Brahman children, t(159) = 9.64, p < .001, or Tamang children, t(159) = 14.79, p < .001. Further- more, Brahman children endorsed problem-focused justifications more frequently than did Tamang chil- dren, t(122) = 3.19, p < .005. Age did not influence problem-focused justifications, all ts < 1.

Descriptions of Behavior

United States children were expected to describe efforts to alter situations whereas Nepali children were predicted to accept, rather than act to alter, diffi-

cult situations. A main effect of behavior indicated

that children were more likely to describe accepting difficult situations (M = 4.07, SD = 1.54) than acting to change them (M = 1.82, SD = 1.52). Culture influ- enced the behavior children reported, however. United States children reported acting to change situ- ations (M = 2.49, SD = 1.62) more than did Brahman children (M = 1.21, SD = .98) and Tamang children (M = 1.34, SD = 1.39), both ps < .05. Conversely, Brahman and Tamang children were more likely to accept negative situations (Ms = 4.65, 4.61, SDs 1.03, 1.40, respectively) than were U.S. children (M = 3.38, SD = 1.62), both ps < .05. There were no effects related to age.

DISCUSSION

School children from Tamang and Brahman villages of Nepal and a rural community in the northeastern United States, as predicted, revealed culturally dis- tinct patterns of appraising and acting on emotions in difficult interpersonal situations. The evidence strengthens the position that emotion appraisals, communication, and action are influenced by culture (Mesquita & Frijda, 1992) and that Asian cultural values involving relational goals influence emotion understanding as well as social and moral under- standing (Keller et al., 1998; Miller & Bersoff, 1992). The findings are discussed in terms of each group's distinct cultural model, cross-sectional differences re- lated to children's age, and future directions.

Brahman communities, like all Nepali communi- ties, prize group harmony and respect for authority. They organize their communities around the Hindu caste system, which emphasizes social distance and status differences among people of different castes. To behave according to caste requires a high level of awareness and self-control in social interaction. Brah-

man children's belief that anger should not be com- municated, even among the younger group, reflects a level of self-awareness that may be fostered by a cul- tural niche in which consciousness of self in relation

to other is part of competent behavior (Joshi & MacLean, 1994). This finding, also reported for first- grade Brahmans (Cole & Tamang, 1998), indicates a reliable cultural pattern, even among young school children, and invites more research on the age at which emotional display rules can be articulated by children. Brahman children's justifications and be- havioral choices further point to their understanding of the need for group harmony and respect for au- thority. Consider the story of being slapped for snatching an eraser that was not in use (see the Ap- pendix). The modal response of Brahman children

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

992 Child Development

was to feel angry because this was unjust, but not to show anger. They explained that it was wrong to show anger because "father gave me life" or "friends must cooperate" and that one should just "sit quietly."

The Brahman respect for group harmony and au- thority, however, did not deter Brahman children from appraising situations in ways that made them expect to feel anger. It may be that Brahman children are proud of their high social status-a caste from a heritage of priestly scholars-and come to expect being treated ac- cordingly. A sense of relative privilege may explain why Brahman and U.S. children appeared to appraise situations similarly, although they had quite different opinions about communicating and acting on anger. Although there are differences in the communication of anger, Brahman and U.S. children may feel entitled to anger in the sense that they have pride in their pasts and their present achievements (e.g., school achieve- ment). Such an interpretation points to the complex in- fluences of religion and sociopolitical factors in culture. Sociopolitical factors such as majority and minority sta- tus influence social behavior (Ogbu, 1981).

Tamang communities also cherish interpersonal harmony and authority but achieve social unity through a more communal, egalitarian approach. Their form of sociality is supported by their Tibetan Buddhism phi- losophy, which guides them away from a focus on temporary individual differences and self-concern, and encourages them to be selfless and free of emo- tional wants. Tamang school children appraised situ- ations such that they did not feel anger, and instead endorsed two alternative emotions. First, as shown in previous research, Tamang children had a unique ten- dency to report feeling thiken (okay) in difficult situa- tions (Cole & Tamang, 1998), a state that is ideal from the Tamang Buddhist perspective. The inclusion of shame as a response choice revealed that Tamang were more likely than others to endorse shame. When probed why they would not be angry, they replied, "Tilda bomo khaba?" (Why be angry?), explaining that the situation had already occurred and questioning the value of being angry. Thus, the cultural influences of Tamang society may promote a pattern of emotion regulation that involves modifying how one appraises situations. This approach is consistent with Buddhist teaching that one make one's mind calm and without strong feeling at all times. Another possibility is that Tamang children experience some shame when they fail to meet the state of being thiken that is most con- sistent with their community's Buddhist values. Both Brahman and Tamang children, then, accommodate their behavior to the values ascribed to their own, and

more broadly, Eastern cultures, but they achieve this be- havioral result through different emotional pathways.

The tendency for Tamang children to endorse shame more than did Brahman and U.S. children may also reflect an internalization of a submissive role that

fits their lower caste and disadvantaged minority group status. For example, in the slapped hand story, Tamang children typically reported that they would be ashamed, and communicated that it was their fault (e.g., "I snatched the eraser.") and that like Brahman children, they thought they should sit quietly in the face of such interpersonal difficulties. The association between sociopolitical status and a child's sense of appropriate emotion or control orientation requires further study.

Even when Tamang children felt negatively, they believed emotions could be communicated. In their

minds, this emotional communication was a natural consequence-self-evident-perhaps indicating that they had not yet understood or adopted a value of hiding emotions. Older Tamang children referred to not communicating anger more than did younger Tamang, something they may have learned as they ex- perienced the school setting, where they were taught by Brahman teachers. Thus, to understand children's understanding of appropriate emotion, the socializa- tion experiences associated with the school context, in- cluding those that involve intergroup relations, must be considered, in addition to the community's cultural values. The Tamang view that emotions are usually communicated may reflect an openness and tolerance in Tamang culture that is captured in the Nepali stereo- type that the Tamang are sojho. This Nepali word con- notes the quality of being frank and without artifice.

Finally, U.S. society also values group harmony and respect for authority but, more than most other societies, it values self-expression and the assertion of individual rights (Baer et al., 1996; Harkness et al., 2000; Markus & Kitayama, 1991; Rothbaum et al., 2000; Triandis, 1994). Consistent with this value system, U.S. children revealed a third unique description of emotion communication and action. They believed anger could be communicated in difficult situations. This tendency differentiated U.S. children from Brah- man children and was greater among older than younger U.S. children. United States children ap- peared to have acquired a sense that feeling and com- municating anger can be acceptable and effective, consistent with their society's value on self-reliance and self-assertion, a value that perhaps strengthened over the course of elementary school. This interpreta- tion is strengthened by U.S. children's action and jus- tification responses, which dealt directly with prob- lem situations and implied a sense of control and ability to effect situations. For example, with regard to the story about being slapped for taking the

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

Cole, Bruschi, and Tamang 993

unused eraser, a typical U.S. child's reply was to com- municate anger "... so he would realize it was only a mistake," and to tell the person "you can have the eraser, but you don't need to hit me." Thus, U.S. chil- dren, in comparison with Nepali children, provided evidence of a higher tolerance and appreciation for anger and an orientation toward primary control (cf. Essau & Trommsdorff, 1996; McCarty et al., 1999).

Children's age, as indexed by their grade, affected children's responses to a degree. First, understanding that one can feel happy in a difficult situation, and that one can manage emotion through how one thinks about a situation occurred more frequently among older children, regardless of cultural group member- ship. The ability to reflect on one's emotions and to manipulate them consciously emerges during middle childhood (Brown, Covell, & Abramovitch, 1991; Meerum Terwogt & Stegge, 1995; Saarni, 1979). Possi- bly, the response that one felt happy in a difficult sit- uation reflects the ability of older children to feel an "opposite" emotion, understanding that one can make oneself take a positive view despite the situation.

Second, age influenced children's responses that involved decisions about communicating anger. Brah- man children as young as second grade understood that one can feel differently than one reveals (see also Cole & Tamang, 1998), a pattern that may reflect Hindu socialization pressure (Joshi & MacLean, 1994). For Tamang and U.S. children, however, age differ- ences were found in opposite directions. Older Tamang children were twice as likely as younger Tamang to state that anger should not be revealed, although as a group they still reported this less than did Brahman children. In contrast, U.S. children were the least likely to describe not communicating anger, and older U.S. children reported this half as much as did younger U.S. children. Thus, the data suggest that cultural factors interact with age with regard to a child's sense of what is appropriate anger, perhaps reflecting changes in understanding of self-control and self-efficacy (cf. Essau & Trommsdorff, 1996). Therefore, both culture and age must be considered when studying the development of emotion under- standing, because culture is the developmental niche in which beliefs about communicating emotions are cultivated.

The paucity of cross-cultural research on emo- tional development limits the understanding of child development. Cultural comparisons, particularly those that go beyond contrasts between complex nations to examine within-nation variations in systematic ways, will contribute to knowledge on the complex social factors that influence emotional lives and on the uni-

versal and culturally variant aspects of socialization

of emotion. The present findings suggest paths for fu- ture research. The present study, in particular, was limited by a method that precluded analysis of how situational context (parent versus peer, public versus private context) influences children's descriptions of appropriate emotional behavior, and the lack of for- mal assessment on the degree to which each culture, and the individuals within them, adhered to prin- ciples such as collectivism and individualism or rela- tion- versus self-oriented goals. Finally, there is a need for research that studies the developmental processes through which such cultural understand- ing is acquired.

ACKNOWLEDGMENTS

The authors acknowledge the enthusiastic willing- ness of the participating villagers. In addition, they appreciate the interviewing assistance of Ishwori Lama, additional translation by Tirtha Dong, artwork by George Leary, and coding by Kara Sontag and Jil- lian Driscoll. This project was supported by a grant from the National Science Foundation (SBR-9711519) to the first author.

ADDRESSES AND AFFILIATIONS

Corresponding author: Pamela M. Cole, Department of Psychology, Pennsylvania State University, 417 Moore Building, University Park, PA 16802; e-mail: [email protected]. Carole J. Bruschi is also at Pennsylva- nia State University; Babu L. Tamang is at the Sanman Prabhi School, Tekanpur, Kavre, Nepal.

APPENDIX

INTERVIEW VIGNETTES

Stories about Injustice or Thwarted Goals

1. You are doing your homework. Your father/friend is sitting next to you. You want to use the eraser and you see it by his/her hand. You take the eraser and your father/ friend slaps your hand and says, "Don't grab things-wait until I'm finished!"

2. You are sitting in your house. Your father/friend is there with you and he/she is drinking some iced/hot tea. You say to him/her, "Look at my homework." He/she reaches for your book and spills the tea all over the table. Quickly, your father/ friend takes the homework and cleans up the tea on the table. Now the homework is ruined and you must do it all again.

3. You have a piece of candy. You are taking the paper off so that you can eat this great candy. Your mother/friend is with you and he / she snatches the candy away from you.

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

994 Child Development

You say, "Give it back!" but by that time your mother/ friend has eaten it.

Stories about Public Mistakes or Mishaps

4. Your teacher put a very large red X on a page of your schoolwork. After school, your family/friends are all out- side your house. Your father/friend takes your book, sees the red X the teacher made, and shows everyone. He/she says "How could you have done this badly, didn't you try?" Everyone is looking at you.

5. Your mother / friend can't find her /his money. She /he and some of her /his friends are looking everywhere for it. Suddenly your mother /friend points toward you and says, "There's the money! It's right next to you! I can't believe you would steal my money! How could my son/ daughter/ friend do that?" Everyone is looking at you.

6. You're outside your house / school, playing with ev- eryone. Suddenly, everyone is shouting and running to see something. You run after them but then you fall down and get mud all over you. Your mother / friend and all the others laugh at you.

Stories about Pleasant Events or Interactions

7. It is a beautiful day. You are sitting outside in the sun with your mother/friend. You are talking and laughing. Your mother/friend smiles at you and you think to your- self, "My mother/friend likes to do things with me."

8. It is afternoon. You must rake the leaves/cut grass. There is so much work to do! Your father/friend helps you rake the leaves/ cut grass. After that, you and your father/ friend have a nice time together, eating a snack together.

9. You are eating your lunch with your mother/friend. You finished eating your lunch but you are still hungry. You tell your mother/friend as she/he gives you some of her/ his food to eat.

REFERENCES

Baer, D., Curtis, J., Grabb, E., & Johnston, W. (1996). What values do people prefer in their children? A comparative analysis of survey evidence from fifteen countries. In C. Seligman, J. M. Olson, & M. P. Zanna (Eds.), The psychol- ogy of values: The Ontario symposium, Vol 8. The Ontario symposium on personality and social psychology (pp. 299- 328). Mahwah, NJ: Erlbaum.

Barrett, K. C. (1995). A functionalist approach to shame and guilt. In J. P. Tangney & K. W. Filcher (Eds.), Self-conscious emotions: The psychology of shame, guilt, embarrassment, and pride (pp. 25-63). New York: Guilford Press.

Barrett, K. C., & Campos, J. (1987). Perspectives on emo- tional development II: A functionalist approach to emotions. In J. Osofsky (Ed.), Handbook of infant develop- ment (2nd ed., pp. 555-578). New York: Wiley.

Barrett, K. C., Zahn-Waxler, C., & Cole, P. M. (1993). Avoid- ers vs. amenders: Implications for the investigation of guilt and shame during toddlerhood? Emotion and Cog- nition, 7, 481-505.

Bennett, L. (1983). Dangerous wives and sacred sisters: Social and symbolic roles of high-caste women in Nepal. New York: Columbia University Press.

Berreman, G. D. (1997). Hindus of the Himalayas: Ethnography and change (2nd ed.). Delhi, India: Oxford University Press.

Billings, D. K. (1989). Individualism and group orientation. In D. M. Keats, K. Munro, & L. Mann (Eds.), Heterogene- ity in cross-cultural psychology (pp. 92-103). Lisse, The Netherlands: Swets & Zeitlinger.

Brown, K., Covell, K., & Abramovitch, R. (1991). Time course and control of emotion: Age differences in under- standing and recognition. Merrill-Palmer Quarterly, 37, 273-287.

Cameron, M. M. (1998). On the edge of the auspicious: Gender and caste in Nepal. Chicago: University of Chicago Press.

Central Bureau of Statistics. (1994). Statistical pocket book Nepal 1994. Kathmandu, Nepal: His Majesty's Govern- ment.

Chao, R. K. (1995). Chinese and European American cul- tural models of the self reflected in mothers' childrearing beliefs. Ethos, 23, 328-354.

Chung, T., & Asher, S. R. (1996). Children's goals and strat- egies in peer conflict situations. Merrill-Palmer Quarterly, 42, 125-147.

Clark, R., Anderson, N. B., Clark, V. R., & Williams, D. R. (1999). Racism as a stressor for African-Americans: A biopsychosocial model. American Psychologist, 54, 805- 816.

Cole, P. M., & Tamang, B. L. (1998). Nepali children's ideas about emotional displays in hypothetical challenges. De- velopmental Psychology, 34, 640-646.

Eisenberg, N., Fabes, R. A., Minore, D., Mathy, R., Hanish, L., & Brown, T. (1994). Children's enacted interpersonal strategies: Their relations to social behavior and nega- tive emotionality. Merrill-Palmer Quarterly, 40, 212-232.

Ekman, P. (1994). Strong evidence for universals in facial ex- pressions: A reply to Russell's mistaken critique. Psycho- logical Bulletin, 115, 268-287.

Ekman, P., & Friesen, W. V. (1978). Facial action coding system manual. Palo Alto, CA: Consulting Psychologists Press.

Essau, C. A., & Trommsdorff, G. (1996). Coping with uni- versity-related problems: A cross-cultural comparison. Journal of Cross-Cultural Psychology, 27, 315-328.

Ferguson, T. J., Stegge, H., Miller, E. R., & Olsen, M. E. (1999). Guilt, shame, and symptoms in children. Develop- mental Psychology, 35, 347-357.

Fiske, A. P. (1992). The four elementary forms of sociality: Framework for a unified theory of social relations. Psy- chological Review, 99, 689-723.

Fricke, T. E. (1986). Himalayan households: Tamang demogra- phy and domestic processes. Ann Arbor: University of Michigan Research Press.

Fridja, N. H. (1986). The emotions. Cambridge, U.K.: Cam- bridge University Press.

Fuchs, D., & Thelen, M. H. (1988). Children's expected in- terpersonal consequences of communicating their affec- tive state and reported likelihood of expression. Child Development, 59, 1314-1322.

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

Cole, Bruschi, and Tamang 995

Gautam, R., & Thapa-Magar, A. (1994). Tribal ethnography of Nepal (Vols. I and II). Delhi, India: Book Faith India.

Gnepp, J., & Hess, D. L. R. (1986). Children's understanding of verbal and facial display rules. Developmental Psychol- ogy, 22, 103-108.

Gray, J. N. (1995). The householder's world: Purity, power, and dominance in a Nepali village. Delhi, India: Oxford Univer- sity Press.

Halberstadt, A. G., Denham, S. A., & Dunsmore, J. C. (2000). Affective social competence. Social Development, 10, 79- 119.

Han, G., & Park, B. (1995). Children's choice in conflict: Ap- plication of the theory of individualism-collectivism. Journal of Cross-Cultural Psychology, 26, 298-313.

Harkness, S., Super, C. M., & van Tijen, N. (2000). Individu- alism and the "Western mind" reconsidered: American

and Dutch parents' ethnotheories of the child. In S. Harkness, C. Raeff, & C. M. Super (Eds.), Variability in the social construction of the child: No. 87. New directions for child and adolescent development (pp. 23-39). San Fran- cisco: Jossey-Bass.

Heckhausen, J., & Schulz, R. (1995). A life-span theory of control. Psychological Review, 102, 284-304.

Hochschild, A. R. (1983). The managed heart: The commercial- ization ofhuman feeling. Berkeley: University of California Press.

H6fer, A. (1979). The caste hierarchy and the state in Nepal: a study of the Muluki Ain of 1854. Innsbruck: Univer- sitditsverlag Wagner.

Hofstede, G. (1980). Culture's consequences. Beverly Hills, CA: Sage.

Hoftun, M., Raeper, W., & Whelpton, J. (1999). People, poli- tics, and ideology: Democracy and social change in Nepal. Kathmandu, India: Mandala Book Point.

Holmberg, D. H. (1989). Order in paradox: Myth, ritual, and exchange among Nepal's Tamang. Ithaca, NY: Cornell Uni- versity Press.

Hubbard, J. A., & Coie, J. D. (1994). Emotional correlates of social competence in children's peer relationships. Mer- rill-Palmer Quarterly, 40, 1-20.

Izard, C. E. (1979). The maximally discriminative facial move- ment coding system (MAX). Newark: University of Dela- ware, Instructional Resource Center.

Jones, D. C., Abbey, B. B., & Cumberland, A. (1998). The de- velopment of display rule knowledge: Linkages with family expressiveness and social competence. Child De- velopment, 69, 1209-1222.

Joshi, M. S., & MacLean, M. (1994). Indian and English chil- dren's understanding of distinction between real and apparent emotion. Child Development, 65, 1372-1384.

Kagitcibasi, C., & Poortinga, Y. H. (2000). Cross-cultural psychology: Issues and overarching themes. Journal of Cross-Cultural Psychology, 31, 129-147.

Keller, M., Edelstein, W., Schmid, C., Fang, F., & Fang, G. (1998). Reasoning about responsibilities and obligations in close relationships: A comparison across two cultures. Developmental Psychology, 34, 731-741.

Kitayama, S., & Markus, H. R. (1995). Culture and self: Im- plications for internal psychology. In N. R. Goldberger &

J. B. Veroff (Eds.), The culture of psychology reader (pp. 366- 383). New York: New York University Press.

Lazarus, R. S. (1991). Emotion and adaptation. New York: Ox- ford University Press.

Lebra, T. S. (1983). Shame and guilt: A psychocultural view of the Japanese self. Ethos, 11, 192-209.

March, K. S. (1986). Women's informal associations in develop- ing countries: Catalysts for change? Boulder, CO: Westview Press.

Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, motivation, and emotion. Psychological Review, 98, 224-253.

McCarty, C. A., Weisz, J. R., Wanitromanee, K., Eastman, K. L., Suwanlert, S., Chaiyasit, W., & Band, E. B. (1999). Cul- ture, coping, and context: Primary and secondary con- trol among Thai and American youth. Journal of Child Psy- chology and Psychiatry and Allied Disciplines, 40, 809-818.

Meerum Terwogt, M., & Stegge, H. (1995). Children's un- derstanding of the strategic control of negative emo- tions. In J. A. Russell & J. M. Fernandez-Dols (Eds.), Every- day conceptions of emotion: An introduction to the psychology,

anthropology, and linguistics of emotion (pp. 373-390). Dor- drecht, The Netherlands: Kluwer Academic.

Mesquita, B., & Frijda, N. H. (1992). Cultural variations in emotions: A review. Psychological Bulletin, 112, 179-204.

Miller, J. G., & Bersoff, D. M. (1992). Culture and moral judgment: How are conflicts between justice and inter- personal responsibilities resolved? Journal of Personality and Social Psychology, 62, 541-554.

Miller, P. J. (1996). Instantiating culture through discourse practices: Some personal reflections on socialization and how to study it. In R. Jessor, A. Colby, & R. Shweder (Eds.), Ethnography and human development: Context and meaning in social inquiry (pp. 183-204). Chicago: Univer- sity of Chicago Press.

Miyake, K., & Yamazaki, K. (1995). Self-conscious emotions, child rearing, and child psychopathology in Japanese culture. In J. P. Tangney & K. W. Filcher (Eds.), Self-con- scious emotions: The psychology of shame, guilt, embarrass- ment, and pride (pp. 25-63). New York: Guilford Press.

Ogbu, J. (1981). The origins of human competence: A cul- tural-ethological perspective. Child Development, 52, 413- 429.

Poortinga, Y. H. (1989). Equivalence of cross-cultural data: An overview of basic issues. International Journal of Psy- chology, 24, 737-756.

Rothbaum, F., Pott, M., Azuma, H., Miyake, K., & Weisz, J. (2000). The development of close relationships in Japan and the United States: Paths of symbiotic harmony and generative tension. Child Development, 71, 1121-1142.

Rubin, K. H., Bukowski, W., & Parker, J. G. (1998). Peer in- teractions, relationships, and groups. In N. Eisenberg (Ed.), W. Damon (Series Ed.), Handbook of child psychol- ogy: Vol 3. Social, emotional, and personality development (5th ed., pp. 619-709). New York: Wiley.

Russell, J. A. (1994). Is there universal recognition of emo- tion from facial expressions? A review of the cross- cultural studies. Pscyhological Bulletin, 115, 102-141.

Saarni, C. (1979). Children's understanding of display rules

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms

996 Child Development

for expressive behavior. Developmental Psychology, 15, 424-429.

Saarni, C. (1997). Coping with aversive feelings. Motivation and Emotion, 21, 45-63.

Saarni, C. (1999). The development of emotional competence. New York: Guilford Press.

Selman, R. L., & Demorest, A. P. (1984). Observing troubled children's interpersonal negotiation strategies: Implica- tions of and for a developmental model. Child Develop- ment, 55, 288-304.

Skinner, D., Pach, A., & Holland, D. C. (1998). Selves in time and place: Identities, experience, and history in Nepal. Lan- ham, MD: Rowan & Littlefield.

Stearns, C. Z., & Stearns, P. N. (1986). Anger: The struggle for emotional control in America's history. Chicago: University of Chicago Press.

Super, C. M., & Harkness, S. (1993). The developmental niche: A conceptualization at the interface of child and culture. In R. A. Pierce & M. A. Black (Eds.), Life-span de- velopment: A diversity reader (pp. 61-77). Dubuque, IA: Kendall/Hunt.

Tamang, P. R. (V.S. 2051; in Western calendar, 1994). Tamang jati [The Tamang people]. Kathmandu: Nepal Rajkiya Pragya Pratisthan.

Tangney, J. P. (1990). Assessing individual differences in

proneness to shame and guilt: Development of the self- conscious affect and attribution inventory. Journal of Per- sonality and Social Psychology, 59, 102-111.

Triandis, H. C. (1994). Culture and social behavior. Cham- paign, IL: McGraw-Hill.

Wierzbicka, A. (1994). Emotion, language, and cultural scripts. In S. Kitayama & H. R. Markus (Eds.), Emotion and culture: Empirical studies of mutual influence (pp. 133-196). Washington, DC: American Psychological Association.

Zahn-Waxler, C., Friedman, R. J., Cole, P. M., Mizuta, I., & Hiruma, N. (1996). Japanese and United States preschool children's responses to conflict and distress. Child Devel- opment, 67, 2462-2477.

Zelko, F. A., Duncan, S. W., Barden, R. C., Garber, J., & Mas- ters, J. C. (1986). Adults' expectancies about children's emotional responsiveness: Implications for the develop- ment of implicit theories of affect. Developmental Psychol- ogy, 22, 109-114.

Zeman, J., & Garber, J. (1996). Display rules for anger, sad- ness, and pain: It depends on who is watching. Child De- velopment, 67, 957-973.

Zeman, J., & Shipman, K. (1996). Children's expression of negative affect: Reasons and methods. Developmental Psychology, 32, 842-849.

This content downloaded from �������������152.13.18.10 on Tue, 06 Oct 2020 14:30:24 UTC��������������

All use subject to https://about.jstor.org/terms