.. 3 fiwpgun 13$?st any.) «a... fi a, X . o! .. ‘u.)..ufl...uvofln.. Nun H. 5;.” o. A in“ .i . 3.... hut-.5 ii «I I... 61. .35... 7:13... I rt 1:51;. annuyéxiMin ‘ 2 ’ in. .. :3. 5i? 3:. 9...... b1: #5». .5”; 11., a 1...}... 5 .0- 555: . it? A 4.: 3..) 3.; is . THESIS 23,710 LIBRARY Michigan State Unlverslty PLACE IN RHURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 3 o 1} * 000 41 6 2003 1‘: AA; A U'LZ K 3622165 mm m.“ HIGH/LOW-CONTEXT COMMUNICATION: CONCEPTUALIZATION AND SCALE DEVELOPMENT By Rie Ohashi A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 2000 ABSTRACT HIGH/LOW-CONTEXT COMMUNICATION: CONCEPTUALIZATION AND SCALE DEVELOPMENT By Rie Ohashi The concept of high/low-context traditionally has been discussed as a part of the individualism] collectivism dimension, rather than as a separate concept. This study argues that high/low-context is a distinctive construct that needs to be explicated in its own right, despite the trend in current intercultural communication theories to treat high/low-context as part of their conceptualization and not as a separate construct. The data were consistent with the construct and content validity of the proposed high/low- communication scale, as well as the claim that high/low-context communication is a separate construct from other related constructs in intercultural communication such as self-construal and horizontal/vertical-individualism/collectivism. Further discussion on high/low-context communication and directions for future research are also provided. Copyright by RIE OHASHI 2000 TOT.M iv ACKNOWLEDGEMENT First and foremost, I would like to thank my advisor, Professor Mary Jiang Bresnahan. I could never have completed my degree without her tireless help, support, understanding, and patience. Not only did she give me many ideas, suggestions and feedback, she read my manuscript over and over and over again, correcting numerous grammatical mistakes I made, and editing my careless work at the same time. I can only imagine how much work all this was for her, and I cannot find the words to express my appreciation. 715%? i: <9 5 I) ('5?) V) 7% é: 5 3591 ‘i L710 I would also like to thank Professors Frank Boster, Jim Dearing, and Bernard Finifier for serving on my committee, and for their valuable comments, suggestions, and time. Professor Boster helped me with the Confirmatory Factor Analyses, spending time with me despite his extremely busy schedule. The encouragement that Professor Dearing gave me really helped me get through the process of completing my dissertation. Professor Finifier gave me detailed, extensive comments on several earlier versions of this study, and often corresponded with me on e-mail suggesting useful references. Without their help this study could not have been in its present form. Merci beaucoup. I am very grateful for the support I received from the other faculty and staff at the department. I drew ideas and materials from almost every single class I took from the department. In fact, I based this study on an assignment project for the class I took from Professor Kim Witte in spring of 1996-. The materials from Professor Steve McComack’s class gave me a lot of inspiration as well. I also appreciate the encouragement I received from my professors in Japan, especially Professor Oka and Professor Yamanaka. 3% 3510 I would like to thank my cohort and friend, Reiko Nebashi for collecting the Japanese data for this study in her classes, and for all the support she gave me. Thomas Wagner, Aaron Boyson, and Ken Lachlan let me collect my US. data in their classes. I would also like to thank Takuya Minami for back-translating my questionnaire, and for great ideas, comments and support he gave me. Muchas gracias. Many thanks to Maria Lapinski—LaFaive for her encouragement, her feedback on the manuscript, and for serving as one of the judges for scale items. Shelly Campo, Wen- Ying Liu, and Lisa Murry also served as judges for the scale items, as well as gave me their support. Vielen Danke. I was fortunate to have many friends inside and outside the department, and on both sides of the Pacific Ocean, who gave me encouragement and support while I was struggling. Thanks to ALL my friends, cohorts, and colleagues, in the US, in Japan, and in other parts of the world. Special thanks goes to Tom and Tomoe, Mr. and Mrs. Fukushima, Keiko, Mr. Inoue, Mr. Kamata, Esther, Kenzie, Ren, Cheong-yi, Amy, Victoria, Deb, Sachi, and Lifang for their friendship and support. Thanks also to my friends from OPIE, UTWO, 98LIII10D, and the Department of Language and Information Sciences, the CERI crew, and the NOCKERS and friends. Djenkuie. I am also thankful to the Fulbright Foundation for giving me a scholarship and an opportunity to study in the United States, and to complete this degree. I am grateful to the Department of Communication, the Student Services, and The Graduate School of vi Michigan State University for their financial assistance while I was completing my degree. Tack sci mychet. Lastly, I owe thanks to my parents, for supporting my overseas study and my endeavor for getting a Ph.D. degree. Indeed, I owe them a lot. Thank you very much. vii PREFACE A friend once challenged me for doing “cross-cultural comparison” by sweeping across individuals and making generalizations without considering individual differences. I tried to defend myself by saying that that’s just how I have been trained to do research, and it is not the only way to study cultural differences. Indeed, when I read or hear about the experiences of making transitions across different cultures, I am painfully aware that theories aren’t of much help to those who are struggling on a day to day basis in the process of cultural adaptation. Sweeping theories can’t help the individuals who are going through their everyday life in confusion, being perplexed, not knowing what is it that they are doing wrong, but still not being able to communicate with their friends, not being able to fit in, feel safe, feel assured of themselves. Yet, I still believe that there is some utility to theories, even if it was just to make sense out of the experience after one has already gone through it all. I sincerely hope that those who are struggling now will be able to get through it someday soon, and that those who have already gone through the struggle will be able to get something out of it, like I did. viii TABLE OF CONTENTS LIST OF TABLES ........................................................................... INTRODUCTION ........................................................................... CHAPTER 1: CONCEPTUALIZATION OF HIGH/LOW-CONTEXT COMMUNICATION ....................................................................................... . Definition of High/Low-Context Communication ................................ Related Constructs in the Area of Intercultural Communication . . . . .. Individualism/collectivism ................................................... Allocentrism/ideocentrism ................................................... Self-construal .................................................................. Association between allocentrism/ideocentrism and self-construal ..... Horizontal/vertical-individualism/collectivism ............................ Elaborated/restricted code ..................................................................... Group/ grid ...................................................................... Differentiating High/Low-Context Communication from Other Concepts Relationships Between High/Low-Context Communication, Individualism/ Collectivism, and Self-Construal .................................................. Rationale for Creating a High/Low-Context Communication Scale — A Critical Examination of Gudykunst et a1. (1996) CHAPTER 2: METHOD ................................................................... Construction of the High/Low-Context Communication Scale . .. Confirmatory Factor Analyses - A Test of Content Validity . . . . . . .. Construct Validity of the High/Low-Context Communication Scale .......... High/Low-Context Communication as a Distinct Construct .................... Procedure ............................................................................... CHAPTER 3: VALIDATION OF THE HIGH/LOW—CONTEXT COMMUNICATION SCALE .. .......................................................................... Confirmatory Factor Analyses ....................................................... Establishing the Construct Validity of the High/Low—Context Communication Scale ................................................................ Establishing High/Low-Context as a Distinct Construct ......................... CHAPTER 4: DISCUSSION ON HIGH/LOW-CONTEXT COMMUNICATION ..................................................................... High/Low-Context Communication and Confucianism/Rhetoric ............... High/Low-Context Communication and Politeness .............................. Future Research ........................................................................ ix xi \INNN ll 14 15 17 22 24 27 29 32 32 34 36 36 38 42 42 45 47 50 50 53 57 CHAPTER 5: RESULTS AND DISCUSSIONS ON THE SELF-CONSTRUAL SCALES AND THE HORIZONTAL/VERTICAL- INDIVIDUALISM/ COLLECTIVISM SCALES Relationships Between the Scales ..................................................... Differences Between US. and Japanese Participants .............................. CHAPTER 6: CONCLUSION ............................................................ REFERENCES .............................................................................. APPENDICES ............................................................................... 61 61 64 67 69 76 Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 1 1 Table 12 Table 13 LIST OF TABLES Factor Loading of Each Item in Respective Scales ........................ T-test Between US. and Japanese participants on the High/Low- Context Communication Scale .......................................... T-tests Between US. and Japanese participants on the High/Low- Context Communication Scale for Females and Males T-tests Between Females and Males on the High/Low-Context Communication Scale in the US. and Japan Multiple Regression of Extraneous Variables onto the High/Low- Context Communication Scale ........................................... Multiple Regression of Other Scales onto the High/Low-Context Communication Scale ....................................................... Correlations Between the High/Low-Context Communication Scale and Other Scales Correlations Between Detecting Meaning Items and Other Scales Correlations Between All Scales for Combined Data .. Correlations Between All Scales for US. Data .. . Correlations Between All Scales for Japanese Data . T-tests between US. and Japanese Participants on the Self-Construal Scales and the Vertical/Horizontal-Individualism/Collectivism Scales ........................................................................ T-tests Between US. and Japanese Participants on the Self-Construal Scales and the Vertical/Horizontal-Individualism/Collectivism Scales for Females and Males ........................................... xi 77 83 84 86 88 89 92 94 96 97 98 99 101 INTRODUCTION The concept of high/low-context communication, introduced by Hall (1976), is familiar to all intercultural communication scholars. Despite its long history of being acknowledged as one of the key notions in intercultural communication, high/low-context communication has never been systematically conceptualized and measured. Instead, it exists as a general research concept, and various constructs in the field of intercultural communication have incorporated high/low-context communication as a part of their conceptualizations. The first part of this study discusses how other constructs have incorporated high/low-context communication, and then argues for the need for distinguishing high/low-context communication as a distinct concept. The second part of this study proposes a strategy for developing a measure of high/low-context communication. The third part of this study reports the results of the high/low-context communication scale validation, followed by a discussion of the issues behind high/low- context communication and suggestions for future research. Finally, results and discussion regarding other scales used in this study are reported. CHAPTER 1 CONCEPTUALIZATION OF HIGH/LOW-CONTEXT COMMUNICATION Definition of HigmLow-Context Communication Whether a given culture is a high-context communication culture or a low-context communication culture is determined by the extent to which a person living in that culture is expected to derive information that is not stated explicitly in interpersonal communication. Hall (1976) defines: A high-context (HC) communication or message is one in which most of the information is either in the physical context or internalized in the person, while very little is in the coded, explicit, transmitted part of the message. A low-context (LC) communication is just the opposite; i.e., the mass of the information is vested in the explicit code. Although no culture exists exclusively at one end of the scale, some are high while others are low. (p. 91) As Hall (1976) described, high/low-context is not a dichotomous concept, but a continuous dimension where high-context is placed at one end and low-context at the other. In a culture that is closer to the high-context end of the continuum, there is a social expectation that in order to comprehend a message it is necessary to realize the larger implicit suggestions that lie behind the often minimal words that have been spoken. A person ftmctioning in such culture is expected to “read behind the words” to infer the intent of the speaker that is not verbally stated, and the communicator legitimately expects the receiver to do so. At the same time, the communicator is allowed to communicate his or her real intentions by implication as opposed to explicit verbalization. As a message sender, one must understand that the receiver will try to read more into your words than what you verbally stated, and as a message receiver, one must be cognizant of the fact that the sender may mean more than what he or she actually said, and you are expected to understand it. In a culture that is closer to the low-context end of the continuum, it is a social norm to place less emphasis on hidden, suggested meanings that may lie behind the words, and to place greater emphasis on words that have been explicitly communicated'. A communicator fimctioning in such a culture is expected to communicate his or her intentions explicitly. As a result, in a low-context culture, people are not expected to “read behind the words” as much as they would be required to do in a high-context culture. As a message sender, one must be careful to “say what you mean and mean what you say”, and as a message receiver one must keep in mind that it is expected of the message receiver to ask for clarification and explanations if one is not sure what the speaker is trying to communicate. For instance, I was once told by a good American friend of mine on a day before a test that “I’m sure you’ll fail; the odds that you’ll pass that exam are about 100 to 1.” Her ' Verbal communication is not the only means of communicating - much can be communicated nonverbally. Although the importance of nonverbal communication is recognized, the focus of this study will be placed on verbal communication. This study will not include nonverbals in its conceptualization of high/low-context communication; the relationship between high/low-context communication style and the use of nonverbals warrants investigation in a separate study. communication style was very low-context, as she explicitly made it clear what she thought about the issue. At the same time, I did not have to read behind her words to understand the implicit message, and could assume that she had no hidden messages. On the other hand, my closest Japanese friend told me that “Well you’ll just have to try your best, you know, and see what happens.” His speech style was high-context, because he was not making it explicitly clear what he thought about the issue, and had left it up to me to figure out whether he actually thought I would fail or not. Therefore, I was obliged to read behind the words for the real message that he was trying to conveyz. Indeed, Honna (1989) mentions that “[l]anguage plays a limited role in Japanese society” (p. 164) because “[p]eople are expected to understand meanings in view of the context of situation in which they are embedded” (p. 164). For instance, in Japanese, a compliment can sometimes be a disguised sarcastic complaint, and the receiver is expected to make an appropriate interpretation based on the context of the situation. Naotsuka (1980) gives an example; when a next door neighbor comes and tells you that your daughter practices her violin earnestly, you will have to discern whether that remark was made as an honest compliment or as a complaint that the noise of the instrument is too loud. According to her, this “unmistakable” complaint was misunderstood as a 2 In this case, I would try to discern his meaning by deliberately telling him “Oh, I’m sure I’ll fail!” If he denied it right away and strongly, such denial would be a good indication that it was not his real intention to tell me that he thought I would actually fail. If he gave me another vague response, such as “Well, you never know until you try”, I would infer the response to mean that he was at least thinking of the possibility that I might fail the exam but was afraid to convey such a negative message to me, and I would appreciate his consideration for my feelings. sincere compliment by more than half of the non-Japanese respondents’. Naotsuka (1980) explains that since non-Japanese respondents are not used to inferring the underlying intention that is not explicitly stated (i.e., because they are not accustomed to the social norm of a high-context culture), they were not likely to interpret this remark correctly as a complaint, but interpreted it literally as a compliment. On the other hand, a Japanese should be able to notice that this remark is meant as a complaint, because as a functional Japanese one is expected to infer the implicit intent of the speaker (i.e., a Japanese person should be able to fimction successfully in the high-context culture of Japan). Needless to say, though, this interpretation is a matter of degree. Regardless of their cultural background, people make inferences, especially in face-threatening situations where concern for politeness or for the other person’s feelings overrides concern for clarity of communication (Kim et al., 1996), as in the case of conveying negative information. In fact, Brown and Levinson (1987) claim that “the major motivation for being indirect at all is politeness” (p. 139). If the purpose of indirect speech is to be polite, indirect speech should be observed in all societies. There are instances of high-context communication occurring in low-context communication societies; and indeed, even in the US, various forms of indirect speech (euphemism, indirect request, etc.) are utilized (for an in-depth discussion of indirect speech, see Brown & Levinson, 1987). Yet, compared to low-context cultures, in high-context 3 Naotsuka did not report how many Japanese were able to interpret this apparent compliment as a complaint; she only mentioned that to interpret this compliment literally as a compliment is “unthinkable” for Japanese. cultures one is expected to a much stronger degree to make inferences about what is not explicitly mentioned. In Japan, “[t]acit understanding is more important than elaborate speech” (Honna, 1989, p. 164) and “[p]eople who cannot understand speech in its social context are frowned upon” (Honna, 1989, p. 164). Honna (1989) goes so far as to say that “[p]eople who resort to elaborate speech are felt as noise makers” (p. 164), which is an echo of an earlier mentioning by Naotsuka (1980) that “to say everything explicitly is equal to insulting the other person in Japan (p. 161; my translation). This is a sharp contrast to the “problem-oriented, direct, explicit, personal, and informal” American communication style (Stewart & Bennett, 1991 , p. 155). Not every person living in a society will conform to social norms, and there are exceptions to every rule. Specific situations may call for a communication style which deviates from the expected norm, such as the use of low-context communication between close friends in a high-context culture, or the use of high-context communication between a doctor and a terminally-ill patient in a low-context culture". Brown and Levinson (1987) mention that cultural taboos and touchy areas are often avoided or communicated indirectly. These examples illustrate that there are exceptions within the general social norm. It is possible that certain topics (e. g., topics that are considered taboo in all cultures, if there are any) may never be discussed explicitly in any culture. While acknowledging the existence of situational boundaries, this study will conceptualize high/low-context communication at the societal level as the prevalent communication " This idea originated from an article entitled “‘Passive euthanasia’ is the norm in today’s hospitals, doctors say” by Gina Kolata in New York Times (Saturday, June 28‘“, 1997, pp. 1, 10), and was developed following discussion with B. Finifter (personal communication, July 1997). style of a given culture, and argue for establishing high/low-context communication as a distinct construct. Related Con_structs in the Area of Intercultural Communication Individualism/collectivism. Individualism/collectivism is considered as “the major dimension of cultural variability isolated by theorists across disciplines” (Gudykunst & Ting-Toomey, 1988, p. 40). With the other cultural dimensions identified by Hofstede (1980, 1983a, l983b)’, it has been viewed as one of the fundamental cultural variables explaining differences among cultures" (e. g., F rymier, Klopf, & Ishii, 1990; Gudykunst & Nishida, 1984, 1986; Gudykunst & Ting-Toomey, 1988; Hui, 1988; Sueda & Wiseman, 1992; Thompson, Klopf, & Ishii, 1991; Ting-Toomey, 1988a, 1988b). Collectivistic cultures are described as emphasizing group-orientation, placing the group’s goals over individual goals, and making a sharp distinction between the ingroup and the outgroup. Individualistic cultures are described as emphasizing self-orientation, placing personal goals over the group’s goals, and not making a sharp distinction between the ingroup and outgroup. Individualistic communication style is characterized as direct 5 Hofstede (1980) identified four dimensions of cultural variability (uncertainty avoidance, individualism/collectivism, power distance, and masculinity/femininity) through factor-analyzing work values in 40 countries. He expanded his data to 50 countries and 3 regions in 1983. Uncertainty avoidance is the degree to which members of a society feel anxious toward unfamiliar situations (Hofstede, 1980, 1983a: Gudykunst & Ting-Toomey, 1988). The other three dimensions will be discussed later in this paper. 6 Smith, Dugan, and Trompenaars (1996) attempted to replicate cultural variabilities identified in several previous studies, including Hofstede’s (1980) four dimensions. They found dimensions resembling Hofstede’s individualism-collectivism and power distance, but failed to replicate masculinity-femininity and uncertainty avoidance factors. and assertive, while collectivistic communication style tends to be indirect and face- saving. A scale for measuring individualism and collectivism was first developed by Hui for his doctoral dissertation in 1984 (Triandis, Leung, Villareal, & Clack, 1985), using US. American and Hong Kong participants. The scale was subsequently published in 1988. Since then, many measures for individualism/collectivism have been developed, both from etic and emic perspectives (see Triandis and Gelfand, 1998, for a summary of existing measures for individualism/collectivism). In previous research, however, the concept of high/low-context communication has conventionally been regarded as a component of the concept of individualisrn/ collectivism (e. g., Gudykunst & Ting-Toomey, 1988; Triandis, 1994), both conceptually and methodologically. Gudykunst and Ting-Toomey (1988) claim that “the dimensions of low-high-context communication and individualism-collectivism are isomorphic” (p. 44), saying that “low— versus high-context communication and direct versus ambiguous communication are predominant forms of communication in individualistic and collectivistic cultures, respectively” (p. 45). Consequently, previous studies have not generally differentiated high/low-context communication from individualism/ collectivism, and the concept of high/low-context communication has never been explicated clearly. Although Triandis (1993, 1996) advocated the idea of conceptualizing individualism/collectivism as “cultural syndromes’”, recent research on individualism/ collectivism has focused on individual differences within a culture in order to explain why a certain culture is individualistic or collectivistic. This line of research seems to indicate that whether a culture is individualistic or collectivistic is determined by the personality (in terms of ideocentric/allocentric or predominant self-construal; see next section) of the majority of people who are living within that culture. Allocentrism/ideocentrism. Soon after Hofstede (1980, 1983a, 1983b) identified the individualism/collectivism dimension, Triandis and his colleagues embarked on a line of research on the “psychological dimension” of individualism/collectivism, which they named “ideocentrism” and “allocentrism” (e. g., Triandis et al., 1993; Triandis, Chan, Bhawuk, Iwao, & Sinha, 1995; Triandis, Leung, Villareal, & Clack, 1985). Triandis, Leung, Villareal, and Clack (1985) began their paper by declaring that “[t]he focus of this paper is on a psychological dimension to be named allocentn'c vs. ideocentric tendencies” (p. 395, emphasis in the original), whose cultural correspondence is “collectivism vs. individualism” (p. 395). Triandis (1995) explains that“ [i]n every culture there are people who are allocentric, who believe, feel, and act very much like collectivists do around the world. There are also people who are ideocentric, who believe, feel, and act very much like individualists do around the world” (p. 5, emphasis in the original). The concept of allocentrism/ideocentrism has gone through several stages of development 7 Triandis (1993) claims that “[a] cultural syndrome can be identified when shared attitudes, beliefs, norms, roles, values, and other such elements of subjective culture, identified among those who share a language, historic period, and geographical location, (a) are organized around a theme, (b) there is evidence that the within-culture variance of these constructs is small relative to the between-culture variance, and (c) there is a link between these patterns of subjective culture and geography” (p. 155). under Triandis and his colleagues’ research. Triandis, Bontempo, Villareal, Asai, and Lucca (1988) claimed that there are three factors within the structure of ideocentrism in the US. (self-reliance with competition, concern for ingroup as a reverse factor, and distance from ingroup). In Triandis, McCusker and Hui, (1990), only the distance from ingroup and self-reliance factors were considered when determining an individual’s allocentrism/ideocentrisms. Throughout several studies (e.g., Triandis, Bontempo, Villareal, Asai, & Lucca, 1988; Triandis, Chan, Bhawuk, Iwao, & Sinha, 1995; Triandis, Leung, Villareal, & Clack, 1985; Triandis, McCusker, & Hui, 1990), Triandis and his colleagues have attempted to measure allocentrism/ideocentrism by using several versions of the measure they developed themselves and also with pre-existing measures, such as Hui’s Individualism-Collectivism scale (1988) and the modified version of the Twenty- statement test (Kuhn & McPartland, 1954). It should be noted, however, that measure of individualism/ collectivism and measure of allocentrism/ideocentrism were not always conceptually distinguished. Hui’s (1998) Individualism-Collectivism Scale was first developed as a scale for measuring individualism/collectivism, but was subsequently used to measure allocentrism/ideocentrism (e.g., Triandis, Leung, Villareal, & Clack, 1985). Yamaguchi, Kuhlrnan, and Sugimori (1995) added 4 “individualist items” to Yarnaguchi’s (1994) “Collectivism Scale” to measure allocentrism, thereby also not making a clear distinction between individualism/collectivism and allocentrism/ 8 Those who were in the top half of the distribution in distance from ingroups and self- reliance factors were categorized as ideocentric, and those who were in the bottom half of the distribution in the same two factors were categorized as allocentric. 10 ideocentrism’. It should also be noted that throughout the line of research on allocentrism/ideocentrism, communication style has not been subsumed as a part of the construct. Triandis, Chan, Bhawuk, Iwao, and Sinha (1995) recommend using the modified Twenty-statement test and two scales developed in the process of Triandis and his colleagues’ research, none of which includes any item on communication style. Self-construal. Another prevalent perspective in the field of intercultural communication for understanding the relationship between individuals and culture is that of independent/interdependent self-construal. In 1991, Markus and Kitayama identified the construct of independent/interdependent self-construal as a different way of viewing the self, raising a question about “what have been thought to be culture-free aspects of cognition, emotion, and motivation” (Markus & Kitayama, 1991, p. 224). They defined independent self-construal as “a conception of the self as an autonomous, independent person” (p. 226), and interdependent self-construal as “the significant feature of the self [that is] found in the interdependent and thus, in the more public components of the self’ (p. 227). Singelis and Brown (1995) explain the construct of self-construal as “the theoretical nexus of cultural collectivism and a variety of individual-level communication behaviors” (p. 359). In the current literature, it is conceptualized that independent self- construal and interdependent self-construal co-exist within a single person (Singelis, 1994; Gudykunst et al., 1994 July), and either self-construal can be activated, depending 9 However, his case may only be a naming issue; Yamaguchi (1994) describes his Collectivism Scale as “a scale to measure collectivism among individuals” (p. 180). Yamaguchim Kuhlman, and Sugimori (1995) consider “personal collectivism” as equal to “allocentrism’ (p. 659). It is possible that the word “allocentrism” was not available to Yamaguchi when he constructed his Collectivism Scale in Japan. 11 on the situation. However, usually one or the other tends to predominate (Gudykunst et al., 1994 July) as the “default” self-construal; in individualistic cultures people’s self- construals are predominantly independent, whereas in collectivistic cultures people’s self- construals are predominantly interdependent (Gudykunst & Nishida, 1994)”. From such line of logic, it is plausible to assume that self-construal can be conceptualized either as a cognitive state (when studying situational activation of independent or interdependent self-construal) or a trait (when studying the “default” self-construal that is predominant in a given culture). Recent research on self-construal by Gardner, Gabriel and Lee (1999) claim that self-construal can be “primed” (p. 321) in an individual, by having the participants read “an independent or interdependent story” (p. 322). The story “described a dilemma in which a general had to choose a warrior to send to the king. In the independent condition, the general chose the person who was the best individual for the job and considered benefits to himself. In the interdependent condition, the general chose a member of his own family and considered benefits to his family” (Gardner, Gabriel, & Lee, 1999, p. 322). The authors examined how the induced self-construal affected the participants’ values, which was measured by presenting 56 values to the participants, '0 However, this conceptualization may need some modification. A study by Park and Levine (1999) showed that participants scored higher on independence than on interdependence, regardless of whether they were living in the mainland U.S., Korea, or Hawaii (chosen to represent individualistic culture, collectivistic culture, and multicultural environments). The only difference between the three cultures was the degree of differentiation between independence and interdependence (participants in mainland U.S. made the most differentiation between independence and interdependence; Korean participants differentiated the least). The authors claim that “[t]aken together with previous findings, these findings suggest that it may be necessary to rethink the role of self-construals in intercultural research” (p. 199). 12 some of which were “individualist values” (Gardner, Gabriel, & Lee, 1999, p. 322) while others were “collectivist values” (Gardner, Gabriel, & Lee, 1999, p. 322). The participants were asked to indicate “the extent to which each value represented a guiding principle in their lives (Gardner, Gabriel, & Lee, 1999, p. 322). The authors demonstrated that priming the self-construal that is considered inconsistent with the predominant self-construal in a particular culture can shifi the individual’s value considerably toward the value endorsed by the primed self-construal type. In their study, US. American participants who were primed with interdependent self-construal endorsed collectivistic values significantly more than participants who were not primed, and Hong- Kong participants who were primed with independent self-construal endorsed individualistic values significantly more than non-primed participants. In the research domain of self-construal, direct/indirect communication style has often been incorporated into the construct without receiving an independent status. For example, Singelis (1994) claims, in his summary of independent/interdependent self- construal, that “being direct in communication” is one of “the constellation of elements composing an independent self-construa ” (p. 581). Similarly, he claims that one emphasis of interdependent self-construal is “being indirect in communication and 9” ‘reading other’s minds (p. 581). In developing a scale concerning independent/ interdependent self-constuals, Singelis included the item “ I’d rather say ‘No’ directly, than risk being misunderstood” as a measure of independent self-construal (p. 585). Gudykunst et al. (1994 July) argue that this item is actually measuring the concept of high/low-context communication and not interdependent self construal, cautioning l3 against incorporating low-context communication items in the self-construal measureI '. Their argument is compelling, because there are culturally bound rules for how and when to say “no” (Ruben, 1976), suggesting that whether or not one will say “no” directly is at least partially dependent upon the cultural norm under which one is operating, not on a idiosyncratic inclination. Assocfiition between allocentrism/ideocentrism and self-construai Allocentrism/ ideocentrism was conceptualized by translating culture-level variables to the individual level, while the construct of self-construal views culture through individuals’ concept of the self. Although the two constructs seem to be clearly related, no effort has been made to integrate the two, even though the two lines of research have co-existed for the past decade. Based on the descriptions of the two constructs, it seems that Triandis’ “allocentric” can be considered as someone whose self-construal is predominantly interdependent, and “ideocentric” as someone whose self-construal is predominantly independent. Therefore, for example, Japan is a collectivistic culture because the majority of Japanese are more allocentric than ideocentric. This also implies that the majority of J apanese’s self-construal is predominantly interdependent rather than independent. Similarly, because the majority of the US. Americans are ideocentric, meaning that their self-construals are predominantly independent rather than interdependent, the US. is characterized as an individualistic culture (Gudykunst & Matsumoto, 1996; Gudykunst et al., 1996). ” There are several attempts by different researchers to measure self-construals; not all of them incorporate communication style items in their measure (e. g., Cross, 1995). 14 Horizontal/vertical-individualism/collectivism. Triandis (1995) claims that besides the independent and interdependent self (as was advocated by Markus and Kitayama in 1991), there are two other dimensions of self, referred to as “same” and “different”. He later rewords his own explanation by saying that the vertical/horizontal axis corresponds to “power and achievement” and “benevolence and universalism” (Triandis, 1996, p. 410). He explains that horizontal collectivism (HC) is adding benevolence to collectivistic values, horizontal individualism (H1) is adding universalism (which includes “equality”) to individualistic values, vertical collectivism (V C) is adding power to collectivistic values, and vertical individualism (V I) is adding achievement to individualistic values. In Triandis and Gelfand (1998), the construct is further refined, and described as follows: [I]n HI, people want to be unique and distinct from groups, are likely to say “I want to do my own thing,” and are highly self reliant, but they are not especially interested in becoming distinguished or in having high status. In VI, people often want to become distinguished and acquire status, and they do this in individual competition with others. They are likely to say “I want to be the best.” In HC, people see themselves as being similar to others (e. g., one person, one vote) and emphasize common goals with others, interdependence, and sociability, but they do not submit easily to authority. In VC, people emphasize the integrity of the in- group, are willing to sacrifice their personal goals for the sake of in-group goals, and support competition of their in—group with out-groups. If in-group authorities want them to act in ways that benefit the in-group but are extremely distasteful to them, they submit to the will of these authorities. (p. 119) 15 Triandis (1995) claims that “vertical collectivism and horizontal individualism are the “typical” patterns around the world” (p. 46), while granting that “some individualist cultures are more horizontal than others” (p. 46). He conceptualizes horizontal/ vertical-individualism/collectivism as situationally dependent and individually unique at the same time. He claims that different cultures should have different profiles regarding the four dimensions, and that each person also has a unique profile of horizontal/vertical- individualism/collectivism, because people will act differently in different situations (Triandis, 1996, March). In line with the concept of self-construal, the concept of horizontal/vertical- individualism/collectivism also recognizes individual differences within a culture, although the conceptualization of horizontal/vertical-individualism/collectivism stresses the situational factor much more strongly than the conceptualization of self-construal. In fact, Triandis, Chen, and Chan (1998) presented 29 situations to their respondents, and asked them to rank order the four choices representing horizontal individualism, vertical individualism, horizontal collectivism, or vertical collectivism. In addition, the concept of horizontal/vertical-individualism/col1ectivism acknowledges that people are not always rigidly individualistic or collectivistic, but instead will behave flexibly according to what is appropriate under various circumstances (Triandis, 1995). Horizontal/vertical-individualism/collectivism is not exceptional in incorporating direct/indirect communication style as a part of its general concept. Triandis’ (1995) scale for measuring horizontal individualism includes the item, “I prefer to be direct and forthright when I talk with people”, alongside other items such as “One should live one’s life independently from others” and “I am a unique individual” (p. 206, 207). By asking 16 one’s preference for being “direct and forthright”, this item measures communication style as a part of horizontal individualism. The claim made by Gudykunst et a1. (1994 July) towards Singelis’ (1994) operationalization of independent self-construal (i.e., that the item is measuring high/low-context communication and not horizontal individualism) can also apply here. This is another illustration of how direct/indirect communication style has been confounded with other constructs. Elgborated/restricted code. In the sociology field, Bernstein (1964) developed the concept of elaborated code and restricted code by examining the linguistic variations within the UK. 12: If it is difficult to predict the syntactic options or alternatives a speaker uses to organize his meanings over a representative range of speech, this system of speech will be called an elaborate code. In the case of an elaborated code, the speaker will select from a wide range of syntactic alternatives and so it will not be easy to make an accurate assessment of the organizing elements he uses at any one time. However, with a restricted code, the range of alternatives, syntactic alternatives, is considerably reduced and so it is much more likely that prediction is possible. In the case of restricted code, the vocabulary will be drawn from a narrow range but because the vocabulary is drawn from a narrow range, this in itself is no ‘2 Bemstein’s primary interest seems to have been in how status difference is encoded in the speech pattern. The first characteristic of the restricted code is listed as “The status aspect of the social relation is salient” (p. 60). Bernstein explicitly mentions later that “[r]estricted codes can be considered status-oriented speech systems” (p. 63) . l7 indication that the code is a restricted one. (Bernstein, 1964, p. 57)'3 According to Bemstein’s (1964) conceptualization, the restricted code “symbolizes and reinforces the form of social relations by restricting the verbal signaling of differences [between the interactants]” (p. 60), because restricted code “presupposes a local cultural identity which reduces the need for the speakers to elaborate their intent verbally and to make it explicit” (p. 60). In other words, the verbal component of the speech is considered to be used primarily for the purpose of encoding what is assumed to be already shared, such as social status differences. Consequently, “[n]ew information is made available through extraverbal channels and these channels will become objects of special perceptual activity” (p. 60), and “[d]iscrete intent can only be transmitted through variations in the extraverbal signals” Q). 60). As a result, “the verbal component of the message is highly predictable” (p. 58). Bernstein (1964) argues that because of the amount of presumably shared information and the fact that the speaker does not need time to think deeply about the verbal message (because the verbal message is not the focus of the attention for obtaining new information), restricted codes can be observed as “fast, fluent, with reduced articulatory clues, the meaning might be discontinuous, dislocated, condensed and local, but the quantity of speech might not be affected” (p. 62), with “a low level of vocabulary and syntactic selection” (p. 62). In addition, because language is not considered as a tool for expressing individual thought and/or experiences, “mm unique meaning of the person would tend to be implicit” (Bernstein, 1964, p. 62; emphasis in the original). '3 It should be noted that Bernstein explicitly states that the elaborate/restricted code is “not defined in terms of vocabulary” (p. 57; emphasis in the original). 18 In contrast, the elaborated code assumes that “the discrete intent of the other person may n_ot be taken for granted” (p. 63, emphasis in the original). Therefore, “the listener is dependent upon the verbal elaboration of meaning” (p. 63, emphasis in the original), and the verbal channel, not the extraverbal channels, “become[s] the objects of special perceptual activity (p. 63). “The orientation of the speaker is based upon the expectation of psychological differences - his own and that of others” (p. 64); therefore, it is “punctuated by relatively frequent pauses and longer hesitations” (p. 65), because it takes longer time to produce a verbal message due to the careful word choice and much self-editing. Bernstein (1964) also argues that as a result of explicitly verbalizing one’s thoughts and experiences, an elaborate code user will “perceive language as a set of theoretical possibilities available for the transmission of unique experience” (p. 64). Bemstien (1964) argues that because speech style is reinforced through family communication, the study’s findings that students from lower working class families have lower verbal IQs than students from middle-upper class families may be the result of the differences in speech codes between lower working class families and middle-upper class families. Bemstein’s argument is that students from lower working class families are only familiar with the restricted code, which is not used for the verbal expression of thoughts. As a result, they are not accustomed to the use of language in the education system, and are thus disadvantaged. Students from middle-upper class families, on the other hand, are familiar with both elaborated code and restricted code, therefore they are at an advantage in the education system merely because they are accustomed to using language as a tool for logical thought and its expression. There are similarities between the construct of elaborated/restricted codes and 19 high/low-context communication styles. The parallel between explicit communication style in low-context communication style and the emphasis on verbal channel in the elaborate code is apparent. Elaborated code is seemingly related to verbal explicitness, and restricted code seems to be related to verbally implicit forms of communication; therefore, the two concepts may seem to have some convergence, in that the idea of predictability in the restricted code is also relevant to the irnplicitness of the messages in high-context communication. Furthermore, it is most interesting that Bemstien (1964) mentions that “the ‘how’ rather than the ’what’ of the communication would be important” (p. 62) in the restricted code. This description sounds very close to the conceptualization of high-context communication style. However, there are several differences between the concept of elaborated] restricted code and that of high/low-context that must be noted. First, the most striking difference is the level of analysis; elaborated/restricted code is conceptualized in terms of the linguistic variation within a culture (intracultural variability), while high/low-context is conceptualized in terms of cross-cultural variability. Bernstein (1964) concludes by saying, “[e]laborated and restricted codes and their variants should be found in any social structure where their origination conditions exist” (p. 67), arguing that those who can coMunicate using the elaborated codes are also capable of communicating using the restricted code, depending on the social context the individual happens to be in. Therefore, it is conceivable that subcultures which operate in elaborated code and restricted code may co-exist at the same level in a given culture. It is also conceivable that the same person may use either an elaborate code or restricted code depending on the situation. However, high/low-context communication is a perceived social norm in a 20 given culture and each culture falls into different places on the continuum of high/low- context communication style. The norm of high/low-context communication is conceptualized to be shared at the cultural level; deviation within a culture at the subculture level is subsidiary to the cultural norm. If one is to communicate with someone outside of the micro culture, he or she must conform to the norm of the larger culture”. Moreover, there is a limit to the extent in which one can communicate in low- context within a high-context culture, because individuals living within a culture must conform to the cultural norm of communicating, and those who do not will be socially deviant. Another difference between elaborated/restricted code and high/low-context communication style is their theoretical orientation. Elaborated/restricted code was conceptualized in terms of the lexical and structural predictability of the individual’s speech, while high/low-context communication is conceptualized in terms of the perceived social norm regarding communication style. In the former example of compliment and criticism, the communication style would be described as very high- context style if the remark was meant as a criticism (and the fact that the listener must consider such possibility is the result of the shared social norm of high-context communication style). On the other hand, if the remark was in fact an honest compliment, it is not a high-context communication, because the word explicitly stated conveys the actual meaning. However, the remark cannot be said to have been made in a ’4 There is a limit to making these distinctions. The definition of a culture is still being debated by different scholars, and the distinction between a culture and a subculture is never clear-cut. 21 restricted code, regardless of whether it is interpreted as a compliment or criticism, because it does not bear any of the characteristics of the restricted code. To summarize the above discussion, it can be said that both elaborated and restricted code may be used in any culture, regardless of whether the predominant social norm in a given culture is high-context communication or a low-context communication. The judgment of whether a speaker is using a high-context or low-context communication style depends on how to interpret the message; the judgment of whether a speaker is using an elaborate code or restricted code depends on the characteristics of the verbal message itself. Group/ grid. The concept of group/grid in the anthropological literature was developed independently of the intercultural communication field. Getting some hints from Bemstein’s elaborated/restricted code, Douglas (1970, 1978) conceptualized group/grid as a framework for systematic explanation of the belief systems held by different tribes in understudied comers of the world such as Afi’ica and Micronesia. According to Gross and Raynor (1985), the group coordinate “represents the extent to which people are restricted in thought and action by their commitment to a social unit larger than the individual” (p. 5), whereas the grid coordinate “is the complementary bundle of constraints on social interaction, a composite index of the extent to which people’s behavior is constrained by role differentiation, whether within or without membership of a group” (p. 6). Gross and Raynor (1985) explain that in “high group” situations, “individuals are expected to act on behalf of the collective whole, and the corporate body is expected to act in the normative interest of its members” (p. 5); in “low group” situations “people negotiate their way through life on their own behalf as 22 individuals, neither constrained by, nor reliant upon, a single group of others” (pp. 5-6). The explanation for grid given by Gross and Raynor (1985) is that “high grid” situations are such that “roles are distributed on the basis of explicit public social classification, such as sex, color, position in a hierarchy, holding a bureaucratic office” (p. 6), et cetera, whereas “low grid” situations are such that “access to roles depends upon personal abilities to compete or negotiate for them, or even of formal regulations for taking equal turns” (p. 6), in other words, “where access to roles is not dependent on any ascribed characteristics of rank or birth” (p. 6). Gross and Raynor (1985) develop their discussion further into how a society progresses from low-group-low-grid condition to low-group- high-grid, high-group-high-grid, high-group—low-grid, and back to low—group—low-grid. The group coordinate sounds very similar to the prototypical individualism/ collectivism, “high group” being collectivistic and “low group” being individualistic. The grid coordinate sounds somewhat similar to a combination of Hofstede’s (1980) power distance and masculinity/femininity dimensions. Power distance is the degree to which members of a society accept unequal distribution of power, whereas masculinity/ femininity is defined as the relative importance or unimportance placed on such things as sex role differentiation, material success, personal achievement (Hofstede, 1980, 1983a; Gudykunst & Ting-Toomey, 1988). High grid society corresponds to high power distance (where people accept the ascribed role difference) and masculine (strong sex-role differentiation is part of the definition for high masculinity) society. Low grid society corresponds to a society that is low in power distance (where people believe they stand on equal ground regardless of their role ascribed according to their sex or birth) but still high in masculinity (valuing personal achievement is still a part of definition for masculinity). 23 Despite the parallel between group/grid and some of the most popular constructs in intercultural communication, the group/grid concept has never gained attention in the intercultural communication field. Since no previous studies have examined the relationship between group/grid dimensions and high/low-context communication, power distance and high/low-context communication, or masculinity/ femininity and high/low- context communication, no further discussion on this concept can be presented in this study. Differentiating High/Low-context Communication from Other ConceDt§ According to Triandis (1994), “in high-context cultures people value the unspoken, the implicit [C]ommunicators focus on the perceiver of the communication and train each other to understand the implicit” (p. 185). This claim suggests that there is social pressure in high-context cultures to understand implicit messages. Consequently, the characterization of a culture as high-context comes from the social norms of such a culture, not from whether the majority of the people have predominantly independent] interdependent self-construal, or whether they are ideocentric or allocentric. Although there are differences in the degree to which an individual is socialized to conform to the norms of a particular culture, high/low-context communication is a cultural characteristic which overrides individual preferences. Think of an international student from the US. who is studying in Japan. An American is likely to be more ideocentric rather than allocentric, and her independent self-construal is likely to predominate over her interdependent self construal. It will not be surprising if she felt the need for autonomy, just as she did back home. However, in Japan, need for autonomy is not highly recognized, due to the collectivistic nature of 24 Japanese culture. What can the American student do to fulfill her needs? She cannot just say “Please leave me alone because I need some time of my own”; she must communicate her need for autonomy in a very indirect manner if she wants to function effectively in Japan. In other words, she should be able to fulfill her need for autonomy if she knows how to communicate in a high—context manner. However, according to the current conceptualization of high/low-context communication, utilizing a high-context communication style means that she is acting according to her interdependent self- construal. On the other hand, in this case, her need is based on her independent self- construal, not on her interdependent self-construal. Which self-construal is in fact predominant here? Such an uncomfortable experience of a sojoumer is the result of the discrepancy between inner self (i.e., predominant self-construal or allocentric/ideocentric personality) and the social norms to which one must conform in order to function effectively in a culture. Such issues can be studied fruitfully only when the construct of high/low-context communication is differentiated from other constructs, and is recognized as an independent construct. Concepts in the intercultural communication field under their current constructions cannot provide a satisfactory explanation for this example, precisely because high/low-context communication is confounded with other concepts. Individualism/collectivism was originally conceptualized as cultural variability (Gudykunst & Ting-Toomey, 1988). However, as more studies accumulated, researchers found that some people within collectivistic cultures are rather individualistic, and there are also some collectivistic people in individualistic cultures (Triandis, 1995). In order to explain these “exceptions” within a culture, subsequent researchers began focusing on 25 individual differences within cultures. The emergence of the independent/interdependent self-construal construct gave a powerfirl tool for explaining such individual differences within culture. Despite Triandis’ call (1993, 1996) to conceptualize individualism/ collectivism as cultural syndromes, the field of intercultural communication is developing in the direction of focusing on individual differences within culture. Even Triandis (1996) himself resorted back to the psychological level in his discussion of cultural syndromes. As a result, the current trend in the intercultural communication field is to conceptualize cultural individualism/collectivism as being derived from the personality traits of individuals (i.e., predominance of independent or interdependent self-construal of the individuals living in that culture). When we consider the fact that many cross-cultural researchers are trained as psychologists, such direction in the field seems only natural, because psychologists are, in general, interested in intra-psychic dimensions. However, as described by Miller and Steinberg (1975) and suggested by Leung, (1989), it is also possible to study the cultural level apart from the personal level. Rather than trying to define separate categories, this study proposes an alternative approach by conceptualizing high/low-context communication as characterized by the perceived shared norm15 in the culture regarding its predominant communication style, a norm which competent members of a given culture are aware that one could be reprimanded if he or she deviated from it. In essence, I am suggesting that “culture” be put back into “intercultural” communication, based on the idea that cultural norms regarding a behavior (i.e., how to '5 Morris (1956, p. 610) defines norms as “generally accepted, sanctioned prescriptions for, or prohibitions against, other’s behavior, belief, or feeling”. 26 talk, in this case) supersede individual preferences for that behavior. In this study, a high- context culture is conceptualized as a culture where high-context communication is practiced as the predominant, perceived social norm. Where a specific culture falls along the high/low context continuum depends on the social norms of each culture regarding whether and how one should speak in a given situation, and people learn these norms through various socialization processes”. Relationships Between High/Low—Context Communication, Individualism/Collectivism and Self-Construall7 The reason that the concept of high/low-context communication has not gained an independent status in the field of intercultural communication lies in the widespread assumption that most individualistic cultures have low-context communication as their social norm, and that most cultures with a high-context communication norm are collectivistic. Thus, while the two are conceptually distinct, they may be highly correlated. Singelis and Brown (1995) argue that “culture affects the development of an individual’s psychological makeup, which, in turn, affects communication behavior” (p. 355). Gudykunst et al. (1996) propose a more specific model which states that individualism/ collectivism “has both a direct effect on communication behavior and an '6 The intention here is to probe the perceived social norm itself, rather than the degree of conformity to the social norm in different cultures (Davidson, Jaccard, Triandis, Morales, & Diaz-Guerrero, 1976; Triandis & Gelfand, 1998). The notion of tight/loose culture (Chan, Gelfand, Triandis, & Tzeng, 1996; Triandis, 1994) includes the degree of conformity to social norm as a part of its conceptualization. '7 Due to the lack of previous research in the intercultural communication discipline regarding elaborated/restricted code and group/ grid, the relationship between these constructs and high/low-context communication, individualism/collectivism, and self- construal could not be discussed here. It is a very intriguing agenda for a future research. 27 indirect effect on communication behavior that is mediated through individual-level factors such as self construals and values” (p. 511)”. It is reasonable to think that individualism/collectivism will affect how an individual will communicate. For instance, if a person is socialized to be an individualist, that person will retain that individualistic orientation to a certain extent as a part of his or her own personality, even when that person is in a collectivistic culture". A predominantly independent person in a high- context culture might feel frustrated in having always to attend to the possibility that interactants expect inferences of the implicit intent behind what is said. A person might in fact prefer to be direct, but knows that under the circumstance that is not the most appropriate behavior. The construct of high/low-context communication points to major differences between cultures regarding predominant communication styles, and therefore needs to be conceptualized more clearly. Unclear explication of the concept of high/low-context communication has led to the premature neglect of the construct. The conceptualization of high/low-context communication as a social norm holds promise to advance future cross-cultural research, because it will permit examination of the impact of social norms about speech on communication behavior. '8 For a diagram of their claim, see Figure 1 (p. 512) in Gudykunst et a1. (1996). ‘9 The degree of acculturation will moderate the extent to which his/her original cultural trait is retained (Kim, 1995). 28 MIC for Creating a High/Low-Context Communication Scale - A Critical Examination of Gudykunst et al. (1996) There has been only one prior attempt to measure high/low-context communication independently of other constructs. Gudykunst et a1. (1996) studied the relationship between individualism/collectivism, self-construals, individual values, and high/low-context communication, theorizing that cultural individualism/collectivism has both direct and indirect influences on communication behavior. The authors attempted to determine whether cultural-level variables (individualistic/collectivistic culture) or individual-level variables (self-construals and value orientation) were better predictors of high/low-context communication style. They developed a questionnaire to assess high/low-context communication style, independent/interdependent self-construals, and individualistic/collectivistic values, using “members of individualistic cultures” (represented by US. American and Australian college students) and “members of collectivistic cultures” (represented by Korean and Japanese college students) as respondents. In order to measure high/low-context communication style, the authors generated 156 items (some were drawn from various scales used in past research, and others were specially generated for the purpose of their study), and subjected them to principal component factor analysis with equimax rotation. Eight factors composed of 80 items were derived from the solution, identified as (1) “respondents’ perceptions of their ability to infer others’ meanings”, (2) “respondents’ tendencies to use direct/ambiguous communication”, (3) “respondents’ interpersonal sensitivity”, (4) “speakers’ tendencies to use dramatic communication”, (5) “the tendency of respondents to use feelings to guide 29 their behavior”, (6) “openness in and initiation of communication with others”, (7) “respondents’ preciseness in communication”, and (8) “respondents’ positive perception of silence in communication” (Gudykunst et al., 1996, pp. 521-523). Their study was a significant departure from previous studies on individualism/ collectivism and high/low-context communication in the sense that the two-concepts were finally being differentiated, instead of high/low-context communication being measured as a part of individualism/collectivism. However, Gudykunst et al. (1996) claim that factors (1), (4), (5), (6), and (7) are related to low-context communication, whereas factors (2), (3), and (8) are related to high-context communication”. This claim indicates that the authors are considering high-context communication and low-context communication as separate dimensions, rather than as opposing poles of a single continuous dimension. Consequently, their communication measure departs from Hall’s (1976) conceptualization of high/low-context communication, in which high/low—context communication was thought of as a continuous single dimension, not as two independent dimensions. I agree with Hall’s (1976) conceptualization of high/low-context communication, because, as I mentioned earlier, indirect and/or implicit communication is utilized in every culture to a certain extent (Brown & Levinson, 1987). The difference between a 2° On face value, the first factor (respondents’ perception of their ability to infer others’ meanings) may seem to be related to high-context communication. However, the authors claim that it is related to low-context communication, because it “emphasizes the listener’s abilities to infer speaker’s meanings, not the extent to which they actually infer meanings” (Gudykunst et al., 1996, pp. 523, 525; emphasis in the original). The authors explain firrther that such ability will not become an issue in a high-context culture, because inferring meanings is taken for granted in high-context culture; therefore, this factor is related to low-context communication, instead of high-context communication. 30 high-context communication culture and a low-context communication culture is a matter of degree regarding how much indirect and/or implicit communication is used by people living in that culture. In addition, Gudykunst et a1. (1996) conclude that “individual level factors (i.e., self construals and values) are better predictors of low- and high-context communication styles across cultures than cultural individualism-collectivism” (p. 510). This claim goes against my conceptual definition of high/low-context communication style as a perceived social norm superceding individual-level characteristics”. Consequently, I argue that what Gudykunst et a1. (1996) measured in their study is not the same construct as I am conceptualizing in this study. Therefore, I believe it is worthwhile to create a new scale for measuring high/low- context communication style which more closely reflects Hall’s (1976) conceptualization of high/low-context communication, as an alternative tool for studying high/low-context communication style across cultures. 2' I would argue that this finding is the result of designating “cultural individualism/ collectivism” a priori by the participants’ nationality, and how high/low-context communication style is conceptualized in their study. The authors do recognize that their sample may not have been representative of their cultural individualism/collectivism. 31 CHAPTER 2 METHOD CgLstruction of the High/Low-Context Communication Scale High/low-context communication style is defined in this study as the normative belief for explicit/implicit forms of communication and whether a message can be taken literally, or whether there is some unstated implicit message behind the words (referred to as “literality” of the message). Direct/indirect aspects of a message were considered as an alternative indicator for explicit/implicit aspects of a message. The normative belief for explicit/implicit communication style was measured by the perceived social norm regarding how one’s message should be conveyed - in a clear and direct manner or in ambiguous and indirect manner. It was also measured as the perceived social norm regarding the degree to which a speaker expects a listener to infer meaning from what he or she did not state explicitly by words. The “literality” of a message was measured as the perceived social norm about the degree to which a listener should try to read behind the words actually being stated explicitly, or the degree to which a listener can believe that the words he or she hears represent what the speaker means (no hidden meanings). Consistent with the conceptualization of high/low-context communication as a social 32 norm, the construct was not measured by asking the respondents for their individual traits and behaviors”. A total of 93 items was generated according to an earlier conceptualization of the construct discussed above. Most of the items were constructed based on the conceptual description of high/low-context communication in previous research (e. g., Gudykunst & Ting-Toomey, 1988; Hall, 1976; Triandis, 1994, etc)”. Some additional items were generated by consulting other students in my department who study intercultural communication. The items were then evaluated for face validity (whether an item looked like it was measuring what it purports to measure) by three judges“, including myself. Items with most agreement between the judges were subjected to a small scale pre-pilot study, whose participants were US. American undergraduate students and Japanese undergraduate students who came to the US. for a short-term English learning program. After revising and rewording the items from the pre-pilot test, a total of 20 items for the scale were included in the final questionnaire. All items were Likert-type items on a 7- point scale, from “strongly disagree” to “strongly agree”. 22 Although such format was not used in this study, the distinction between personal preferences and social norm can be highlighted by asking the participants’ own personal response a_n_d what they perceive the others’ response to be (e. g., Kluckhohn & Strodtbeck, 1961). 2’ Gudykunst et a1. (1996) was not consulted when generating the items, because the original 93 items were generated before the article’s publication. 2" Ph.D. students in the Department of Communication at Michigan State University who were studying intercultural communication served as judges for choosing the appropriate items. They were asked to read my conceptualization of high/low-context communication, and assess each item on its content (that is, they read the items and judged whether an item looked like it was consistent with the conceptualization of high/low-context communication). 33 Confirmatorv Factor Analyses - A Test of Content Validity Confirmatory factor analysis was conducted in order to establish the content validity of the high/low-context communication scale, and also to test the unidimensionality of the scale (Hunter & Gerbing, 1982). Confirmatory factor analysis specifies the measurement model by theoretical reasoning, instead of analyzing the items blindly without any underlying theory about which item is measuring what construct (as in exploratory factor analysis). A priori theoretical specification of the measurement model enables confirmatory factor analysis to test the content validity and unidimensionality of a scale. Hunter and Gerbing (1982) claim that “[t]he construct validation of a variable operationalized with multiple indicators is accomplished by the analysis of the measurement model” (p. 269). A measurement model is the specification of “the causal relations between the theoretically defined variables of interest and the responses to the observed variables, which are presumed to be determined by the theoretical variables” (Hunter & Gerbing, 1982, p. 270). In essence, it is a theoretical designation of which item is supposed to measure which construct. In the present study, I am constructing a measurement model for my scale by claiming that all the items for my high/low-context communication scale are measuring the construct of high/low-context communication. There are two separate tests that are conducted to establish the content validity of a scale. The test of internal consistency examines whether the items are in fact related to each other. “If all the items in a cluster measure the same factor, then, the correlations between the items will satisfy a ‘product rule for internal consistency’. That is, the correlation between two items in the same cluster should be the product of their 34 correlations with the underlying trait” (Hunter & Gerbing, 1982, p. 277). The test of parallelism examines how the items that are supposed to be measuring the same construct relate to an outside variable. “The general statement of parallelism is that the items in a unidimensional cluster have similar patterns of correlations with (1) items in other clusters or (2) other traits” (Hunter & Gerbing, 1982, p. 279). The tests of internal consistency and parallelism examine the deviation of the actual data from this theoretical model. In practical terms, the test of internal consistency and test of parallelism examine the difference (i.e. error) between the predicted correlation and actual correlation obtained from the data. In the test for internal consistency, the correlation between any two items belonging to the same factor is theoretically predicted to be the product of the factor loadings of the two items. In the test for parallelism, the correlation of any two items belonging to different factors is theoretically predicted to be the product of each item’s factor loading and the correlation between the two factors. The errors are calculated by subtracting the predicted correlation from the obtained correlation. The smaller the error, the better fit the data are to the theoretically specified model. Inappropriate items which are inconsistent with the factor structure specified in the test of internal consistency and/or parallelism were deleted from the final scale. When the data from a given factor are consistent with the unidimensional structure specified in the test of internal consistency, it can be claimed that all items are in fact measuring the same construct. When the data from all scales are consistent with the factor structure specified in the test of parallelism, it can be claimed that different scales are measuring different constructs. Once the data are consistent with the factor structure specified in the 35 test for internal consistency and the test for parallelism, it can finally be claimed that the content validity of all scales is confirmed. Consguct Validity of the High/Low-Context Communication Scale A valid high/low-context communication scale should be able to distinguish the difference between a high-context culture from a low-context culture. In order to test whether the new high/low-context communication scale developed in the current study is able to serve this purpose, Japan was chosen as an example of a high-context culture, and US. was chosen as an example of a low-context culture. These two countries were chosen because their different communication styles are widely documented (e.g., Gudykunst & Ting-Toomey, 1988; Hall, 1976, 1983; Okabe, 1983), suggesting that Japan is a high-context culture and US. is a low-context culture. If the scale can measure Japan to be a more high-context communication culture than the US, the construct validity of the scale can be demonstrated. Hi ow—Context Communication as a Distinct Construct In order to establish high/low-context communication as a distinct construct, it must be shown that high/low—context communication is different from cognate constructs. Therefore, it must be demonstrated that the high/low-context communication scale and the self-construal scales do not correlate so highly as to indicate that the two scales are measuring the same construct. Similarly, it must also be demonstrated that the high/low- context communication scale and the scales measuring horizontal/vertical individualism/ collectivism do not correlate so highly as to indicate that the two scales are measuring the same construct. 36 Of several scales that purport to measure self-construal, (Cross, 1995; Gudykunst et al., 1994 July; Kim & Leung, 1997; Singelis, 1994), Kim and Leung’s (1997) self- construal scale was used to examine the relationship between high/low-context communication and self-construal. Their scale was chosen over the other scales, because when creating their scale Kim and Leung (1997) took the good items from Gudykunst et a1. (1994 July) and Singelis (1994); therefore, their scale is considered to be an improvement over the other two scales. The scale created by Singelis, Triandis, Bhawuk, and Gelfand (1995) was used to examine the relationship between high/low-context communication and horizontal/vertical-individualism/collectivism. This scale is almost identical to the measure in Triandis (1995), and there are no other validated scales measuring horizontal/vertical-individualism/ collectivism. Using the horizontal/vertical- individualism/collectivism scale was deemed more appropriate for the purpose of this study than using an individualism/collectivism scale (e. g., Hui, 1988), because horizontal/vertical-individualism/collectivism is a further refinement of individualism/collectivism (Singelis, Triandis, Bhawuk, & Gelfand, 1995). Additional evidence of discriminant validity can be provided by a scale which correlates differently with my high/low-context communication scale and the other two scales (F. Boster, personal communication, March, 1999). That is, if the same scale correlates highly with the high/low-context communication scale and not with the self- construal and the horizontal/vertical-individualism/collectivism scale (or vice versa), it is an additional proof that the high/low-context communication scale and the other two scales are measuring different constructs. The “detecting meaning” factor from the 37 conversational sensitivity scale (Daly, Vangelisti, & Daughton, 1987) was chosen with the expectation that it would serve this purpose. Conversational sensitivity is defined as “the propensity of people to attend to and interpret what occurs during conversation” (Daly, Vangelisti, & Daughton, 1987, p. 168). The authors propose a 36-item scale with 8 factors, the first of which is the “perceptiveness in finding deeper and often multiple meanings in what others say” (Daly, Vangelisti, & Daughton, 1987, p. 171). This description of the factor fits comfortably within my definition of high-context communication. On the other hand, this description of the factor does not seem related to the construct of self-construal or horizontal/vertical-individualism/collectivism. Therefore, I expect that the items that load on this factor will correlate rather highly with my high/low-context communication scale, while they will correlate only moderately with the other scales. Procedure A questionnaire containing the proposed high/low-context communication scale (20 items), independent self-construal scale (15 items), interdependent self-construal scale (14 items), vertical individualism scale (8 items), vertical collectivism scale (8 items), horizontal individualism scale (8 items), horizontal collectivism scale (8 items), and items in the detection of meaning factor from the conversational sensitivity scale (8 items) was prepared. The items for self-construal scale, horizontal/vertical- individualism/collectivism, and items in the detection of meaning factor from the conversational sensitivity scale were intermingled in order to avoid the carryover effect fi'om one scale to another. The questionnaire was translated into Japanese and back translated into English for conceptual equivalence (Brislin, 1986). After being piloted, 38 the English version of the questionnaire was distributed to undergraduate students in Michigan State University, and the Japanese version of the questionnaire was distributed to undergraduate students at Rikkyo University. The data were collected between May and October of 1999, including the pilot test. The questionnaire was collected from 230 participants in the US. (146 females - 63.5%, 84 males - 36.5%) and 223 participants in Japan (98 females - 43.9%, 120 males - 53.8%, 5 unclassified) from various majors. The mean age for US. participants was 19.26 (SD = 1.38), while the mean age for the Japanese participants was 18.92 (SD = 1.43). The mean year in college was 2.04 for the US. participants (SD = 1.08), whereas the mean year in college for the Japanese participants was 1.08 (SD = 0.45). Among the US. participants, 77.8% identified themselves as Caucasian American, and 13.5% identified themselves as African American. The other 6.5% identified themselves as either Asian American, Native American, or Other (most in this category were Mixed Races). Four participants (1.7%) identified themselves as international students, while one participant (0.4%) did not indicate their ethnicity. Among the Japanese participants, 95.5% identified themselves as Japanese, and 0.9% identified themselves as Korean Japanese. One participant (0.4%) identified self as an international student, while 3.1% did not indicate their ethnicity. Those who did not indicate their ethnicity and those who identified themselves as international students were eliminated from all subsequent analyses. The test of internal consistency and the test of parallelism were performed on all 8 scales. To test internal consistency, the predicted correlation for all pairs of items within a scale was calculated. Then, each predicted correlation was subtracted from the 39 corresponding obtained correlation to calculate the error between the theoretically predicted model and the actual data. Items with large errors were deleted fiom the scale, and the same procedure was repeated” until an acceptable solution was reached. When all errors were reasonably small and the occurrence of a large error was within the number predicted by chance“, the scale was considered internally consistent. The test for internal consistency was performed separately on all scales that were used in this study. Once all scales were deemed internally consistent, parallelism between the scales was tested. Again, the predicted correlation for every pair of items was calculated, and was subtracted from every corresponding obtained correlation to calculate the error. Items which produced large errors with other items in other scales were deleted, and the process was repeated until an acceptable solution was reached. The solution was considered acceptable when all errors were reasonably small and within the number expected by chance. The final solution was again tested for internal consistency to reconfirm the unidimensional structure of each scale. A preliminary solution for internal consistency of each scale was obtained by using the data from all participants (referred to as “combined data” ). Then, the preliminary solution was checked separately against the data obtained from US. participants and Japanese participants. Items which did not pass the test for internal consistency in either the US. data or the Japanese data were deleted, and only items 2’ Factor loading of an item changes every time a variable is added to or deleted from the factor. 26 At the conventional 95% probability level, one in twenty correlations can have large error by chance alone. 40 which passed the test of internal consistency in all three data sets (combined data, data obtained from the US. participants, and data obtained from the Japanese participants) were subjected to the test for parallelism. Once again, the preliminary solution for the test of parallelism was obtained by conducting the test on the combined data, and then checked separately against the US. and Japanese data respectively. Items which failed the parallelism test in any of the three data sets were eliminated from the final solution. The internal consistency of each scale in the final solution was reconfirmed in all three data sets after all scales passed the test of parallelism. The score for each scale was computed for each participant by summing the items in each scale determined through the procedure described above. These scale scores were used in the analyses. 41 CHAPTER 3 VALIDATION OF THE HIGH/LOW-CONTEXT COMMUNICATION SCALE Confirm_atorv Factor Analyses27 Seven items from the high/low-context communication scale were consistent with the factor structure specified in the test of internal consistency and parallelism for the three data sets (combined, US. only, and Japanese only). Although there were several reverse items in the questionnaire, all of these items were eliminated during the confirmatory factor analysis process. The reliability of the seven items was .70 in the combined data, .64 in the US. data, and .71 in the Japanese data. The mean for US. American participants was 30.08 (N = 224) with a standard deviation of 5.49, while the mean for Japanese participants was 34.21 (N = 211) with a standard deviation of 4.90 (the minimum possible score = 7, the maximum possible score = 49). Five items from the independent self-construal scale (Kim & Leung, 1997) were consistent with the factor structure specified in the test of internal consistency and parallelism for the three data sets. The reliability of the five items was .70 in the combined data, .61 in the US. data, and .67 in the Japanese data. The mean for US. American participants was 24.42 (N = 223) with a standard deviation of 3.74, while the 27 Items and the factor loading of each item is presented in Table 1. 42 mean for Japanese participants was 24.82 (N = 211) with a standard deviation of 4.10 (the minimum possible score = 5, the maximum possible score = 35). Three items from the Interdependent self-construal scale (Kim & Leung, 1997) were consistent with the factor structure specified in the test of internal consistency and parallelism for the three data sets”. The reliability of the three items was .56 in the combined data, .53 in the US. data, and .62 in the Japanese data. The mean for US. American participants was 13.15 (N = 225) with a standard deviation of 3.23, while the mean for Japanese participants was 12.54 (N = 213) with a standard deviation of 3.12 (the minimum possible score = 3, the maximum possible score = 21). Four items from the Vertical Individualism scale (Singelis et al., 1995) were consistent with the factor structure specified in the test of internal consistency and parallelism for the three data sets. The reliability of the four items was .74 in the combined data, .73 in the US. data, and .76 in the Japanese data. The mean for US. American participants was 16.86 (N = 224) with a standard deviation of 5.01 , while the mean for Japanese participants was 16.64 (N = 211) with a standard deviation of 4.41 (the minimum possible score = 4, the maximum possible score = 28). Four items from the Vertical Collectivism Scale (Singelis et al., 1995) were consistent with the factor structure specified in the test of internal consistency and parallelism for the three data sets. The reliability of the four items was .57 for the combined data, .52 for the US. data, and .67 for the Japanese data. The mean for US. American participants was 17.26 (N = 225) with a standard deviation of 4.28, while the 2” The scale was internally consistent by definition because there were only three items left in the scale afier eliminating the bad items. 43 mean for Japanese participants was 16.04 (N = 210) with a standard deviation of 4.09 (the minimum possible score = 4, the maximum possible score = 28). Six items from the Horizontal Individualism Scale (Singelis et al., 1995) were consistent with the factor structure specified in the test of internal consistency and parallelism for the three data sets. The reliability of the six items was .62 in the combined data, .63 in the US. data, and .57 in the Japanese data. The mean for US. American participants was 34.13 (N = 224) with a standard deviation of 4.17, while the mean for Japanese participants was 30.67 (N = 209) with a standard deviation of 4.31 (the minimum possible score = 6, the maximum possible score = 42). Four items from the Horizontal Collectivism Scale (Singelis et al., 1995) were consistent with the factor structure specified in the test of internal consistency and parallelism for the three data sets. The reliability of the five items was .72 in the combined data, .59 in the US. data, and .70 in the Japanese data. The mean for US. American participants was 22.92 (N = 225) with a standard deviation of 2.74, while the mean for Japanese participants was 19.70 (N = 210) with a standard deviation of 3.66 (the minimum possible score = 4, the maximum possible score = 28). Seven items from the Detecting Meaning items of the Conversational Sensitivity Scale (Daly, Vangelisti, & Daughton, 1987) were consistent with the factor structure specified in the test of internal consistency and parallelism for the three data sets. The reliability of the seven items was .86 in the combined data, .79 in the US. data, and .87 in the Japanese data. The mean for US. American participants was 36.60 (N = 224) with a standard deviation of 5.38, while the mean for Japanese participants was 31.55 (N = 207) 44 with a standard deviation of 6.81 (the minimum possible score = 7, the maximum possible score = 49). Establishing the Construct Validig of the High/Low-Context Communication Scale According to the current literature in the field, Japan should score higher than US. on a high/low-context communication scale. A t-test between US. and Japanese participants on the high/low—context communication scale shows that Japanese participants scored significantly higher than US. participants on the high/low-context communication scale (Table 2). The mean score for the US. participants was 30.08 (SD = 5.49, N = 224), while the mean score for the Japanese participants was 34.21 (SD = 4.90, N = 211). The treatment correlation was r = .37. There were no significant differences (t = 0.821, df = 207, p = .412) between Caucasian Americans (mean = 30.19, SD = 5.56, N = 178) and Afiican Americans (mean = 29.32, SD = 4.62, N = 31). No difference was found (t = 0.616, df= 222, p = .538) when all non-Caucasians were grouped together (mean = 29.63, SD = 5.26, N = 46) and compared against Caucasians. Because there was only a small number of participants in other ethnic categories, no further comparison between the ethnic groups in the US. was performed. Due to the difference in the male-female ratio between US. participants and Japanese participants, the same analysis was conducted separately for males and females to rule out the possibility that sex differences were a potential confound for the differences between US. and Japanese participants. The mean score for US. females was 29.98 (SD = 5.20, N = 143), while the mean score for Japanese females was 34.60 (SD = 5.42, N = 95). A t-test between the two female groups showed that Japanese females 45 scored significantly higher than US. females, with the treatment correlation of r = .39 (Table 3). The mean score for US. males was 30.25 (SD = 6.00, N = 81), while the mean score for Japanese males was 33.89 (SD = 4.42, N = 116). A t-test between the two male groups showed that Japanese males scored significantly higher than US. males, with a treatment correlation of r = .33 (Table 3). Since the result of the overall analysis was replicated in both females and males, the possibility that the male-female difference was driving the difference between US. and Japan can be ruled out. In addition, no sex difference on the high/low-context communication scale was found within the US. data as well as within the Japanese data (Table 4). In order to control for the differences in age and year in college between US. and Japanese participants, multiple regression analysis was conducted to see whether age and year in college would have any impact on the high/low-context communication scale score. Table 5 shows that the beta weight of age and year in college is nearly zero. Therefore, the differences between US. and Japanese participants on the high/low- context communication scale was not driven by the differences in age and year in college between the US. and Japanese participants. The present data are consistent with the literature in the field that Japan is more high-context than the US. Therefore, these results suggest that the high/low-context communication scale has construct validity. Table 6 shows the result of the multiple regression of all other scales onto the high/Iow-context communication scale. The results show that the detecting meaning scale was a significant predictor of high/low-context communication style in all three data sets. Vertical collectivism was a significant predictor of high/low-context 46 communication style in the combined data and the US. data, and horizontal collectivism was also a significant predictor in the combined data. However, the detecting meaning scale was the only consistent predictor of high/low-context communication style across the three data sets. These results give additional support to the construct validity of the high/low-context communication scale. Establishing High/Low-Context Communication as a Distinct Construct If the construct of high/low-context communication were to be independent of individualism/collectivism and independence/interdependence, the correlations between the high/low-context communication scale and other scales, after being corrected for measurement error, should not correlate so highly as to suggest that they are measuring the same construct. Table 7 shows the uncorrected and corrected correlations between the high/low-context communication scale and the other scales. It should be noted that none of the correlations are close to 1.00, which indicates that high/low-context communication is indeed a distinct construct fi'om individualism/collectivism and independent/interdependent self-construal. These results are consistent with the claim that high/low-context communication is a distinct construct from other related constructs. However, the correlations between the high/low-context communication scale and other scales were much lower than was predicted, especially in the Japanese data, although moderate correlations were expected between all scales. For example, the correlation between the high/low-context communication scale and interdependence self-construal scale is nearly zero in the combined data and the Japanese data (Table 7). In the Japanese data, the high/low- context communication scale does not correlate with vertical collectivism scale, nor with 47 horizontal collectivism scale. Considering the discussion in the existing literature, these minimal correlations were not expected. It was predicted that detecting meaning items from the conversational sensitivity scale would provide further evidence for discriminant validity of the high/low-context communication scale by correlating positively with the high/low-context communication scale and correlating to a lesser degree with other scales. The data were not consistent with this prediction. As shown in Tables 8, the detecting meaning scale correlated more strongly with the independent self-construal scale and the horizontal individualism scale than other scales in all three data sets. The detecting meaning scale did not correlate with the high/low—context communication scale in the combined data, but did correlate weakly to moderately in the US. data and the Japanese data. In the combined data and the US. data, detecting meaning items correlated moderately with the horizontal collectivism scale and weakly with the interdependent self-construal, vertical individualism, and vertical collectivism scales, but the corresponding correlations were minimal in the Japanese data. These results indicate that detecting meaning items from the conversational sensitivity scale did not provide additional support for the discriminant validity of the high/low-context communication scale at zero-order correlation level. The reason why the detecting meaning scale did not correlate strongly with the high/low-context communication scale cannot be determined from the present study. One plausible explanation is that the detecting meaning scale was also tapping into the individual dimension, which is at a different level from the social norm itself that the high/low-context communication scale was intended to measure. Therefore, even though the content of the detecting meaning items looked similar to that of high/low-context 48 communication scale, it seems plausible that the similarity in face validity was not supported by the data when scrutinized from their construct validity. The same argument can be made for explaining the unexpectedly low correlation between the high/low- context communication scale and other scales. Because all other scales were measuring individual level attitudes, they may not have correlated with the high/low-context communication scale, which was measuring the social norm, which is not necessarily congruent with the individual psychological makeup. It is interesting that the item “I prefer to be direct and forthright when discussing with people” passed the test of internal consistency and parallelism as a horizontal- individualism measure; the factor loading of this item to the horizontal-individualism scale was .44 in the combined data, .48 in the US. data, and .51 in the Japanese data, whereas the factor loading of this item to the high/low-context communication scale was .12 in the combined data, .17 in the US. data, and .12 in the Japanese data. Although this item is asking about one’s communication style, it is asking for an individual dimension, just as the other items in horizontal individualism scale were. Because the operationalization of high/low-context communication was strictly restricted to the social norm itself, it is understandable that this item did not correlate with the high/low-context communication scale to the extent that it did with the horizontal individualism scale. These results suggest further support for my claim that the social norm should be investigated independent of the individual level analyses. In hindsight, a scale which measures another social norm may have provided better evidence of discriminant validity, rather than a scale measuring an individual dimension. 49 CHAPTER 4 DISCUSSION ON HIGH/LOW-CONTEXT COMMUNICATION The current scale was designed to measure the general social norms regarding any society’s communication style in terms of direct - explicitness or indirect - implicitness. The results confirm that the Japanese participants in this study are a more high-context communication culture than the US. American participants. The following section will focus on some possible reasons for these results. High/Low-Context Communication and Confucianism/Rhetoric Culture is not created overnight - it is important to recognize that what we are observing as “culture” today is affected by some factor from the past, be it history, customs, or traditions. I suggest that there must be some historical or traditional factors that have influenced why Japan became a high-context culture, and why the US. became a low-context culture. Although many explanations are possible, numerous scholars have described the role of Confucianism in shaping Japanese culture (Barnlund, 1989; Yum, 1994). Therefore, it would be informative to examine how Confircian ideals and value preferences are reflected in speaking. As for a Western counterpart to what Confucius is to Japan, I chose Aristotle, based on the notion that “[t]he dominant American communication function is persuasion” (Stewart & Bennett, 1991, emphasis in the 50 original), and Aristotle’s indisputable influence on persuasive communication. The two philosophers share several common features. Both Confircius and Aristotle were real people (as opposed to some imaginary figure); Confucius was born in 552 BC. (Huang, 1997), and Aristotle was born on 384 BC. (Kennedy, 1991). They were both considered as the leading scholar of their time, and many people came to seek their teaching. Neither of them was a religious leader - although Confucianism is sometimes considered a religion (e.g., Confucianism is one of the entries in The Illustrated Encyclopefif World Religions, edited by C. Richards), it is more of a philosophy than a religion. Yum (1994) clearly states that “Confucianism is a philosophy of human nature that considers proper human relationships as the basis of society” (p. 77): Confucius mentions very specifically about the conduct of speech in the Analects: The Master said: “Sweet words and pleasing countenance have indeed little humanity29 in them!” (1.3, 17.16). The Master said: “The reason that the ancients30 would not rashly utter words is that they deemed it shameful not to live up to them” (4.22). The Master said: “The gentleman31 deems it shameful if his speech exceeds his action” (14.27; a very similar thought is expressed in 2.13 and 4.24). Master Kong said: “Three types of fiiends are beneficial; three types of fiiends are 2’ The word “humanity” is used to mean “the supreme virtue and the sum of all virtues” (Huang, 1997, p.16) 3° “The ancients” refers to the ideal kings of the ancient times. 3' “The gentleman” refers to an ideal man with talent and virtue; “man of humanity” (Huang, 1997, p. 34). 51 harmful. To befriend the upright, befriend the truthful, and befiiend the erudite is beneficial; to befriend the excessively respectful, befriend the obsequious, and befi'iend the glib-tongued is harmful” (16.4). It is obvious throughout the m that Confucius did not regard speech very highly. He mentions that action is preferred to words, and suggests refraining from speech when the action cannot be delivered. In Confucius’ mind, a virtuous person should not deliver empty words. Considering that Japan was strongly influenced by this philosophy, it is not surprising that speech has not been regarded highly since the ancient times. Such attitude toward speech would naturally shape a high-context culture. Aristotle, on the other hand, did not provide concrete guidelines about how a virtuous person should speak. In Aristotle’s mind, speech was only a tool for persuading others, and what mattered was whether one is able to convince others of what one is advocating. Reflecting such orientation toward speech, Aristotle’s Rhetoric is entirely about how to persuade people, or how to make a persuasive speech: Persuasion occurs through the arguments when we show the truth or the apparent truth from whatever is persuasive in each case (1.2.6). Let the matters just discussed be regarded as understood, and let the virtue of style be defined as “to be clear” (speech is a kind of sign, so if it does not make clear it will not perform its fimction ) (3.2.1). ... speaking well is not a matter of rapidity or conciseness but moderation, and that means saying just as much as will make the thing clear or as much as will make [the audience] suppose that something has happened or (3.16.4). 52 There is a demand for clarity in Aristotle’s discussion of style (Kennedy, 1991), and although he mentions moderation, his major concern is on making sure that the audience understands what is being advocated. This is a stark contrast to Confucius’ thinking, in which he said, “[a]s long as speech conveys the idea, it suffices” (15.41). Although I recognize the different context in which Confircius and Aristotle made these comments (in Confucius’ case, the comments were made as part of his teaching on everyday conduct; in Aristotle’s case, the context was ostensibly persuasive rhetoric of epideictic, judicial, or deliberative speech), the underlying differences in their attitude toward speech is clear. Both Confucius and Aristotle’s philosophy continue to have currency in describing important, long-standing distinctive values about communication in the Eastern and Western culture today. Their philosophies have been very influential in reflecting the different development of cultures in East and West. Confucius’ little importance placed on speech can easily be linked to the high-context communication style in East Asia, including Japan; Aristotle’s emphasis on persuasion by clarity can be considered as the basis for the low-context communication style in the Western cultures, including the US. Such historical perspective is outside the area of investigation in the communication field, yet it brings an interesting insight into the issue of communication style differences. High/Low-Context Communication and Politeness Assuming, as Brown and Levinson (1987) claim, that people utilize indirect communication to be polite, if Japan is a high-context culture where indirect communication style is valued, does that mean that Japanese society is more polite? 53 Maynard (1997) disagrees - “Japanese people are usually characterized as polite. Although the Japanese language has a built-in system of politeness strategies that requires a choice of appropriate politeness levels, this does not mean that the Japanese are always polite” (p. 56), and goes on to discuss the situations in which they are allowed to be impolite. Maynard explains that the Japanese seem more polite to observers, because the observer is usually an outgroup member. Japanese regard an outgroup member to be someone to whom they must maintain their social and psychological distance, and this distance requires maintaining an appropriate level of politeness. However, the “observer paradox” does not explain why the Japanese participants were more high-context than the US. American participants in this study, because the participants’ responses in this study were self-report answers about the respondents’ own culture. In other words, the observers, in this case, were ingroup members. The results of this study suggest that Japanese are “polite” to ingroup members also. An alternative explanation can be made based on the conceptualization of high/low-context communication in this study; namely, that high/low-context communication style is a perceived social norm in each culture. If being high-context is the social norm in Japan, one would see more instances of high-context communication, because the instances where one is allowed to deviate from the social norm are more limited than the instances where one must follow the social norm. On the other hand, if the social norm of the US. is to be low-context, we would naturally see more of low- context communication. Another explanation can be made, based on the premise that “[d]istribution of politeness (who has to be polite to whom) is socially controlled” (Brown & Levinson, 54 1987). If there are more situations in Japan than in the US in which one is expected to maintain social distance, then we might expect to see more polite speech in Japan than in the US. For example, in situations where social distance between people is closer in the US. than in Japan, we might see a low-context communication style in the US, whereas we would expect to see a high-context communication style in Japan. If there are fewer instances in the US. to be “socially distant” where indirect communication is required, it may seem like the US. is a more low-context society than Japan, if only because of the sheer frequency of observing such interaction. The conclusion, then, would be that Japan is a more high-context culture than the US, because people maintain more social distance in Japan than in the US, hence having to be more polite more often. Alternatively, the differences in “how” to be polite may be a factor. “Politeness is required in every society, and all languages provide for different styles and variations to accommodate it” (Maynard, 1997). In fact, Lakoff (1975, p. 65) proposes “formality (keep aloof), deference (give options), and camaraderie (show sympathy)” as three basic rules of politeness, and suggests that different societies have different preference orders among the rules”. She discusses that in first encounters Japanese emphasize deference over other rules while Americans emphasize camaraderie, and as the relationship progresses Japanese will slowly start incorporating camaraderie and end up being friendly ’2 Brown (1966) claims that “status norm” and “solidarity norm” exist in every society. He goes on to argue that in Western culture solidarity norm has become stronger than the status norm, drawing evidence from the shift in second person pronouns throughout the course of European history. 55 and deferential”. My point here is that to Americans being friendly is the utmost politeness, while to the Japanese being fiiendly is not being polite Instead, being deferent is considered to be the proper form of politeness, and that is how they will start the relationship. To them, being fi'iendly is being less formal, hence less polite, and should only happen in a more advanced stage of the relationship. In essence, how “politeness” may be manifested is different from one society to another. In a situation where one is expected to be polite, one will fulfil the expectation by behaving in a fiiendly manner in one culture, by showing deference in another culture, and by being formal in a third culture. If being fi'iendly is the utmost form of politeness in the US, we would expect a “friendly” communication style to be the “polite” form. And because “being friendly” entails being frank and straightforward in the US, being polite would mean speaking straightforwardly; i.e., in a low-context manner. On the other hand, if being deferential is the utmost form of politeness in Japan, a respectfirl communication style would be the “polite form”. As I mentioned earlier, being overly explicit is an insult to the other person in Japan (Honna, 1989; Naotsuka, 1980), because it implies that the other person is insensitive. Therefore, if one wishes to be polite to the other person, implicit communication style would be expected. When low-context communication is utilized in Japan, it is clearly recognized as a violation of the social norm. Therefore, there is a significant meaning to applying low- 3’ Unfortunately, Lakoff does not discuss how Americans would move as the relationship progresses. However, based on the analyses by Brown (1966), I predict that Americans will continue to reinforce comradeship. 56 context communication - it signifies disregard for formalities, which indicate intimacy, which is an “unusual” circumstance within Japanese society. That is why it is only allowed between family members or very close friends. In the US, on the other hand, there is “no higher ideal than that of equality, followed by insistence on independence and a distaste for formality. The first tends to reduce social barriers, the second to favor spontaneous relationships, the third to encourage frankness in communicating with others” (Barnlund, 1989, p. 60). Therefore, intimacy is a goal that the relationship should strive for, not an “unusual circumstance”. This difference in orientation toward fiiendliness and intimacy may possibly be an underlying factor driving the differences in high-context communication culture and low- context communication culture. In turn, then, where does the differences in this “friendliness - intimacy drive” come from? Bamlund (1989) points out, after describing an American ideal person as “[a]n independent, self-realizing person, faithful to his or her own inner truth” (p. 37), that “[i]mplicit in this focus on an independent human being is insistence upon equality of treatment” (p. 37). Maynard (1997) also mentions the American’s “ethic of ‘equality’”. The “camaraderie” rule posited by Lakoff (1975) may be an intervening variable between individualism/collectivism and high/Iow-context communication style. A further examination of this idea seems warranted. Future Research In future research, the social norms of communication that pertain to more specific domains of every day life need to be examined. Scenarios were used in Triandis, Chen, and Chan (1998) and also in Triandis and Gelfand (1998) to measure horizontal/ vertical-individualism/collectivism. Such methods may be useful for this line of 57 investigation. As discussed in the previous section, situations that require being polite would be interesting to investigate. In situations where “being polite” matters, we might see differences, whereas in situations where being polite is not an issue, we may not see as much difference. How politeness is encoded in the communication style would also be worthy of investigation. In addition, Brown and Levinson (1987) mention that culturally taboo areas are often left implicit. This is another area that would be of interest to investigate”. For instance, the level of face threat involved in any situation is said to be a large determiner of the communication style, especially when making a request (Brown & Levinson, 1987). The larger the face threat, the more polite one is expected to be, and there are many strategies with varying degrees of politeness that one can choose from in order to make a request (Brown & Levinson, 1987; Lim, 1990). It would be useful to examine whether there are any differences between different societies in terms of high or low-context communication style when given the same situation. For instance, it is often talked about among the Japanese staying in the US. that US. Americans are more explicit and direct than Japanese when making the same request, such as asking to borrow class note from someone who is not your good friend. This could be because the same request is seen as more face threatening in Japan than in the US, and if it is more face threatening, it would be communicated more politely. Alternatively, it may be because of the differences in the “default” social norm between US. and Japan; regardless of the circumstance, one might be positively evaluated to be explicit in the US, and one might 3" Leaving culturally taboo areas implicit may be a part of being polite; I believe that part of being polite is to be in accord with social norms. 58 be able to avoid being evaluated negatively by erring in the direction of being implicit in Japan. Another factor in determining the level of politeness in the commmiication would be the social distance between the communicators (Brown & Levinson, 1987). It is intuitively natural that we speak differently to our closest friends compared to how we speak to strangers. One way to look at this phenomenon is to pay attention to the differences in the social distance between the speakers. The social distance between us and our closest friend is different from the social distance between us and a stranger. If social distance is a determining factor of the level of politeness that is expected to be conveyed in the interaction, then it is natural that we adopt a different communication style when we are talking to our best fiiend and when we are talking to a stranger, because of the differences in social distance between the two situations. But just how different, or different in what way, is a separate question involving the social norm regarding how to speak to a closest friend and how to speak to a stranger. The distance may be seen as smaller in the US. and larger in Japan, hence reflecting on their respective communication styles. These are just speculations, and it would be interesting to test them, if there is a way. As we delve deeper into the issue, we are being forced to recognize the difficulty of intercultural studies, in that it is almost impossible to vary one and only one factor. It is a challenge to present a situation where social distance, psychological distance, and the perceived degree of face threat are the same in two cultures. 59 Another interesting issue that can be investigated in future research is how much of communication style differences are influenced, if at all, by the language itself”. For example, it would be interesting to see whether we can still detect the differences in communication styles between different societies where English is spoken as a native language, for example, between US, England, Ireland, Australia, and India. Among these countries, India is the only country which is listed as a “high context culture” in Gudykunst and Ting-Toomey (1988), and there is some anecdotal evidence suggesting that England is a little more high-context culture than the US. (Naotsuka, 1980). If we can detect the differences between these societies, we can have more confidence in the content validity of the scale, in that it is measuring the communication style used in that society and not some unidentified effect inherent in translating from one language to another. On the other hand, if we cannot detect differences between different societies using the same language, then we may have to accept that conversation style in a particular society is inseparable from the language spoken in that society. Finally, a more qualitative approach, such as the analyses of the actual conversation or content analyses of media, would also make a great contribution to the ' study of communication style. A qualitative study would help shed light on the actual social norm itself, whereas the questionnaire survey method could only uncover the individuals’ perception of the social norm, which would be essential for further explication of the construct of high/low-context communication. 3’ It is often mentioned that it is easier to leave the speaker’s intention implicit in Japanese because the predicate is placed at the end of the sentence, whereas the verb is placed right after the subject in English. 60 CHAPTER 5 RESULTS AND DISCUSSIONS ON THE SELF-CONSTRUAL SCALES AND THE HORIZONTAL/VERTICAL-INDIVIDUALISM/COLLECTIVISM SCALES Relationships Between the Scales The current literature in the field suggests that independent/interdependent self- construals are the personal variables driving cultural individualism/collectivism. Therefore, it seems reasonable to expect that the independence scale, vertical individualism scale, and horizontal individualism scale would correlate positively with each other. Similarly, it seems reasonable to expect that the interdependence scale, vertical collectivism scale, and horizontal collectivism scale would correlate positively with each other. On the other hand, the results obtained by Singelis, Triandis, Bhawuk, and Gelfand (1995) contradict these expectations. Their results showed a moderate positive correlation between independent self-construal and horizontal individualism (.36 for US. American students, .43 for Hawaii students, .45 for the combined sample), but vertical individualism was not correlated with independent self-construal (.04 in the US American students, -.10 in the Hawaii students, -.01 in the combined sample). They found a moderate correlation between interdependent self-construal and horizontal collectivism (.45 in the US. American students, .41 in the Hawaii students, .43 in the combined sample), and also between vertical collectivism and interdependent self- 61 construal (.40 in the US. American students, .54 in the Hawaii students, .50 in the combined sample). The results in the present study replicated the findings by Singelis, Triandis, Bhawuk, and Gelfand (1995) to a certain extent (Tables 9-11). In accordance with the results by Singelis, Triandis, Bhawuk, and Gelfand (1995), all three data sets in the present study showed that the independent self-construal scale correlated strongly with the horizontal individualism scale (r’ = 1.00 in all data)”, and that the correlation between the independent self-construal scale and vertical individualism scale was weak at best (r’ = .17 for combined data, r’ = .29 in the US. data, and r’ = .07 in the Japanese data). In all three data sets in the present study, the interdependent self-construal scale correlated moderately or strongly with the horizontal collectivism scale (r’ = .70 for combined data, r’ = .55 in the US. data, and r’ = .88 in the Japanese data), and the correlation between the interdependent self-construal scale and vertical collectivism was moderate in all three data sets (r’ = .46 in the combined data, r’ = .35 in the US. data, and r’ = .51 in the Japanese data). In the present study, vertical individualism and horizontal individualism correlated moderately in all three data sets (r’ = .34 in the combined data, r’ = .40 in the US. data, and r’ = .32 in the Japanese data); vertical collectivism and horizontal collectivism also correlated moderately in all three data sets (r’ = .41 in the combined data, r’ = .32 in the US. data, r’ = .39 in the Japanese data). This is in line with the results reported in Triandis and Gelfand (1998); they found a correlation of .30 between 3" The r’ is used to represent a corrected correlation. 62 horizontal and vertical individualism, and .50 between horizontal and vertical collectivism”. However, this is inconsistent with Singelis, Triandis, Bhawuk, and Gelfand (1995), in which they found the correlation between horizontal and vertical collectivism scale to be -.00 (-.03 for US. American students, .04 for Hawaii students), although they also found a .39 correlation between horizontal and vertical collectivism (.33 for US. American students, .45 for Hawaii students). It is also expected from theoretical reasoning that vertical individualism and vertical collectivism should correlate moderately (because they share the “vertical” dimension), and that horizontal individualism and horizontal collectivism should also correlate moderately (because they share the “horizontal” dimension). In Singelis, Triandis, Bhawukl and Gelfand (1995), “the two horizontal dimensions and the two vertical dimensions were slightly, but significantly, positively related” (p. 258). Their data showed that the correlation between horizontal individualism and collectivism was .20 (.08 for US. American students, .22 for Hawaii students), and the correlation between vertical individualism and collectivism was .14 (.19 for US. American students, .18 for Hawaii students). In the present study, the two horizontal dimensions correlated moderately in the combined data (r’ = .38) and US. data (r’ = .47), but did not correlate in the Japanese data (r’ = -.07). For the two vertical dimensions, the correlations in the present study was lower than that of Singelis, Triandis, Bhawuk, and Gelfand (1995) (r’ = -.04 in the combined data, r’ = -.03 in the US. data, r’ = -.06 in the Japanese data). ’7 The participants in their study were all US. Americans. 63 These results suggest that self-construal has a stronger relationship to the horizontal dimensions than the vertical dimensions. The results also suggest that the shared feature of individualism/collectivism is stronger than the shared feature of vertical/horizontal. In other words, the relationship between the two individualism dimensions and the two collectivism dimensions (i.e., vertical individualism - horizontal individualism, and vertical collectivism - horizontal collectivism) is stronger than the relationship between the two vertical dimensions and the two horizontal dimensions (i.e., vertical individualism - vertical collectivism, and horizontal individualism - horizontal collectivism). In addition, the results indicate that the two horizontal dimensions are more strongly related to each other than the two vertical dimensions. While the main goal of the present study was not to validate the horizontal/vertical-individualism/ collectivism scale, it is clear from these results that further research investigating the relationship between these constructs is warranted. Differences Between U.S.fiand Japanese ParticipapLs The current literature in the field suggests that US. Americans should score higher than Japanese on independence, vertical individualism, and horizontal individualism, while Japanese should score higher on interdependence, vertical collectivism, and horizontal collectivism. Table 12 shows the results of t-tests between the US. and Japanese participants on the self-construal and vertical/horizontal- individualism/collectivism scales. When both sexes were combined, U.S. participants scored significantly higher on all scales except for vertical individualism. The fact that US. participants scored higher than Japanese participants on vertical collectivism and horizontal collectivism was contrary to what was expected from the existing literature in 64 intercultural communication field, which describe Japan as a more collectivistic society than the US. Table 13 shows the results of the same analyses conducted separately for females and males. US. and Japanese were not significantly different in interdependent self- construal for both men and women. US. women were similar to Japanese women in vertical individualism, whereas US. men scored higher than Japanese men in vertical individualism. In addition, there were no U.S.-Japanese differences on vertical collectivism for women, but the difference was significant for men. The fact that US. men turned out to be more vertically collectivistic than Japanese men was contrary to the expectation. U.S. Americans scored higher than Japanese on horizontal collectivism for both men and women, which also unexpected from the existing literature which describe Japanese as more collectivistic than the US. Americans. The reason for these unexpected results cannot be determined from this study. It may seem natural to suspect that some unidentified variable is moderator or mediating the relationship. However, while these results contradict the expectations made from the existing literature, there have been several studies in the past that found similar unexpected results. Kim et al. (1996) and Oetzel (1998) both reported that US. Americans were significantly more independent than Japanese, but there were no differences between the two groups in interdependence. Even Gudykunst et a1. (1996) reported in their study that the results for the mean scores of self-construal and individualistic/collectivistic values “do n_ot reflect the general cultural tendencies usually associated” with US. and Japan (p. 530, emphasis in the original). It is interesting to note that these results were obtained from studies using different scales for measuring 65 self-construal, hence making the possibility of systematic measurement error less plausible”. In light of these findings, it is important to note that the justification for assuming cultural characteristics a priori from previous studies is getting weaker than ever before. 3’ Kim et al. used an earlier version of the self-construal scale used in the present study. Gudykunst et al. (1996) constructed a measure for assessing self-construal and individualistic/collectivistic values in their study by conducting exploratory factor analyses of the collection of items from various existing scales. Oetzel (1998) used the self-construal measure used by Gudykunst et a1. (1996). 66 CHAPTER 6 CONCLUSION The most important contribution of this study was to construct a high/low-context communication scale, if only in a preliminary form. The results of the internal consistency test and parallelism show that the scale has construct validity, and the correlations with other scales of related constructs indicate that high/low-context communication is a distinct construct. The differences in the scale scores between the US. and Japanese participants suggest the content validity of the scale. A limitation of the current study is that it could not provide direct additional support for the discriminant validity of the high/low-context communication scale, because the detecting meaning items did not correlate highly with the high/low—context communication scale at the zero- order level. The next step in this line of research is to increase the reliability of the scale by adding more items to the scale, and to obtain additional support for demonstrating the discriminant validity of the scale. It is also important to note that other scales used in this study were also subjected to rigorous measurement tests, which is not often the case in many studies. Yet, they did not behave in line with the predictions made from current literature in the field - only the results fi'om independence scale and horizontal individualism scale were consistent with the theoretical predictions. Contrary to the prediction that Japanese participants would score higher on the interdependent self-construal and horizontal/vertical collectivism 67 scales, U.S. participants scored higher on all these scales. Although U.S. participants were expected to score higher than Japanese participants on the vertical individualism scale, the result was consistent with this prediction only for men, and there were no differences between US. and Japanese participants for women and when both sexes were combined. More research is necessary before any decisive interpretation of these results can be made. However, we can no longer deny the possibility that Japanese individuals are starting to depart from the description they were traditionally associated with (e. g., Miyanaga, 1991). On the other hand, the present study also showed that the social norm of high-context communication is still very much alive in Japan. The results in the current study suggest an increasing discrepancy between what each individual feels, believes, or thinks privately, and how they actually interact and communicate with others. This discrepancy is especially noticeable among the Japanese participants. These results suggest that tapping into an individual’s psychological aspect, such as his or her self- construal orientation, is not enough to predict that person’s communication behavior. Markus and Kitayama (1998) mention that people “organize their actions” (p. 66) based on the framework provided by the society. Communication is a social act, which is inevitably influenced by social norms. We must begin to pay more attention to the existence of social norms which affect the relationship between the individual’s psychological makeup and their actual behavior. The scale proposed in this study is a step in this direction. 68 REFERENCES Bamlund, D. C. (1989). Communication sgles of Japanese and Americans: Images and realities. Belmont, CA: Wadsworth Publishing Company. Bernstein, B. (1964). Elaborated and restricted codes: Their social origin and some consequences. In J. J. Gumpertx & D. Hymes (Eds), The ethnography of communication. American Anthropologists, 66 (special edition), 55-69. Brislin, R. W. (1986). The wording and translation of research instruments. In W. J. Lonner & J. W. Berry (Eds), Field methods in cross-cultugal research (pp. 137- 164). Beverly Hills, CA: Sage. Brown, R. (1966). Social psychology. New York: The Free Press. Brown, R, & Levinson, S. C. (1987). Politeness: Some universals in la_ngu_ag§ usage. Cambridge, Great Britain: Cambridge University Press. Chan, D. K.-S., Gelfand, M. J ., Triandis, H. C., & Tzeng, O. (1996). Tightness- looseness revisited: Some preliminary analyses in Japan and the United States. International Journ_al of Psychology. 31. 1-12. Cross, S. (1995). Self-construals, coping, and stress in cross-cultural adaptation. Journal of Cross-Cultural Psychology, 26, 673-697. Daly, J. A., Vangelisti, A. L., & Daughton, S. M. (1987). The nature and correlates of conversational sensitivity. Human Communication Research. 14. 167-202. Davidson, A. R., Jaccard, J. J ., Triandis, H. C., Morales, M. L., & Diaz-Guerrero, R. (1976). Cross-cultural model testing: Toward a solution of the etic-emic dilemma. Intgnational Journ_al of Psychology. 11, 1-13. Douglas, M. (1970). Natural symbols: Explorations in cosmology. London: The Cresset Press. Douglas, M. (1978). Cultural bias. London: Royal Anthropological Institute of Great Britain and Ireland. Frymier, A. B., Klopf, D. W., & Ishii, S. (1990). Affect orientation: Japanese compared to Americans. Communication Research Reports, 7, 63-66. 69 Gardner, W. L., Gabriel, S., & Lee, A. Y. (1999). “I” value freedom, but “we” value relationships: Self-construal priming mirrors cultural differences in judgment. Psychological Science. 10. 321-326. Gross, J ., & Raynor, S. (1985). Measurinflilture: A paradigm for the analysis of social organization. New York: Columbia University Press. Gudykunst, W. B., & Matsumoto, Y. (1996). Cross-cultural variability of communication in personal relationships. In W. B. Gudykunst, S. Ting-Toomey, & T. Nishida (Eds.), Communication in personal relationships across cultures (pp. 19-56). Thousand Oaks, CA: Sage. Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., Kim, K., & Heyman, S. (1994, July). Measuring self construals across cultures. Paper presented at the International Communication Association Convention, Sydney, Australia. Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., Kim, K., & Heyman, S. (1996). The influence of cultural individualism-collectivism, self construals, and individual values on communication styles across cultures. Human Communication Research. 22. 510-543. Gudykunst, W. B., & Nishida, T. (1984). Individual and cultural influence on uncertainty reduction. Communication Monographs, 51, 23-36. Gudykunst, W. B., & Nishida, T. (1986). Attributional confidence in low-and high-context culture. Human Communication Research. 12. 525-549. Gudykunst, W. B., & Nishida, T. (1994). Bridging Japanese/North American differences. Thousand Oaks, CA: Sage. Gudykunst, W. B., & Ting-Toomey, S. (1988). Culture and interpersonal communication. Newbury Park, CA: Sage. Hall, E. T. (1976). Beyond culture. New York: Doubleday. Hall, E. T. (1983). The dance of life: The other dimengion of time. New York: Doubleday. Hofstede, G. (1980). Motivation, leadership, and organization: Do American theories apply abroad? Organizational Dynamics, 9, 309-330. Hofstede, G. (1983a). National cultures revisited. Behavior Science Research. 18, 285-305. 70 Hofstede, G. (1983b). Dimensions of national cultures in fifty countries and three regions. In J .B. Deregowski, et al. (Eds), Explications in cross-cultural psychology (pp. 335-386). Lisse, Netherlands: Swets & Zeitinger. Honna, N. (1989). Social function of language. In N. Honna & B. Hoffer (Eds), An English Dictionary of Japanese way of thinking (pp. 164 - 167). Tokyu: Yuhikaku. Huang, C. (1997). The a_grlects of Confucius (Lun Yu). Oxford: Oxford University Press. Hui, C. H. (1988). Measurement of individualism-collectivism. Journal of Regarch in Personality. 22. 17-36. Hunter, J. E., & Gerbing, D. W. (1982). Unidimensional measurement, second order factor analysis, and causal models. In B. M. Staw & L. L. Cunnings (Eds), Research in organizational beflvior. Vol. 4 (pp. 267-320). Greenwich, CT: JAI Press. Kennedy, G. A. (1991) Aristotle. on rhetoric: A theory of civic discourse. Oxford: Oxford University Press. Kim, M.-S., Hunter, J. E., Miyahara, A., Horvath, A.-M., Bresnahan, M. J., & Yoon, H.-J. (1996). Individual - vs. culture-level dimensions of individualism and collectivism: Effects on preferred conversational styles. Communication Monogyaphs, §_3_, 29-49. Kim, M.-S., & Leung, T. (1997). A revised self-construal scale. Unpublished manuscript, University of Hawaii at Manoa. Kim, Y. Y. (1995). Cross-cultural Adaptation: An integrative theory. In R. L. Wiseman (Ed.), Intercultural communication theog (pp. 170-193). Thousand Oaks, CA: Sage. Kluckhohn, F., & Strodtbeck, F. (1961). Variation_s in value orientLions. Evanston, IL: Row, Peterson. Kuhn, M. H., & McPartland, T. S. (1954). An empirical investigation of self- attitudes. American Sociological Review. 19. 68-76. Lakoff, R. (1975). Language and woman’s place. New York: Harper Colophon. Leung, K. (1989). Cross-cultural differences: Individual-level vs. culture-level analysis. Intern_ation_al Journ_al of Psychology. 24. 703-719. 71 Lim, T.-S. (1990). Politeness behavior in social influence situations. In J. P. Dillard (Ed.), Seeking compliance: The production of interpersonal influence messages (pp. 75-86). Scottsdale, AZ: Gorsuch. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review. 98. 224-253. Markus, H. & Kitayama, S. (1998). The cultural psychology of personality. Joml of Cross-Cultural Psychology, 29. 63-87. Maynard, S. (1997). Japanese communication: Language and thought in context. Honolulu: University of Hawai’i Press. Miller, G. R., & Steinberg, M. (1975). Between people: A new analysis of interpersonal Communication. Chicago, IL: Science Research Associates. Miyanaga, K. (1991). The creative edge: Emerging individualism in Japan. New Brunswick, NJ: Transaction Publishers. Morris, R. T. (1956). A typology of norms. American Sociologi_C_al Review. 21, 610-613. Naotsuka, R. (1980). Oubeij in ga chinmoku suru toki [When Westerners Stop Talking]. Tokyo: Taishukan. Oetzel, J. G. (1998). Explaining individual communication process in homogeneous and heterogeneous groups through individualism-collectivism and self- construal. Human Communication Research. 25, 202-224. Okabe, R. (1983). Cultural assumptions of East and West: Japan and the United States. In W. B. Gudykunst (Ed.), Intercultural communication theog (pp. 21-44). Beverly Hills, CA: Sage. Richards, C. (Ed.). (1997). The illustrated encyclopedia of world religions. Rockport, MA: Element Books. Park, H. S., & Levine, T. R. (1999). The theory of reasoned action and self- construal: Evidence from three cultures. Communication Monoggaphs, 66, 199-218. Ruben, J. (1976). How to tell when someone is saying “no”. Topics in Cultural Learning, 4, 61-65. Singelis, T. (1994). The measurement of independent and interdependent self- construals. Personality and Social Psychology Bulletin. 20. 580-591. 72 Singelis, T. M., & Brown, W. J. (1995). Culture, self, and collectivist communication: Linking culture to individual behavior. Human Communication Research. 21. 354-389. Singelis, T. M., Triandis, H. C., Bhawuk, D. P. S., & Gelfand, M. J. (1995). Horizontal and vertical dimensions of individualism and collectivism: A theoretical and measurement refinement. Cross-Cultural Research, 29. 240-275. Smith, P. B., Dugan, S., & Trompenaars, F. (1996). National culture and the values of organizational employees: A dimensional analysis across 43 nations. Journal of Cross-Cultural Psychology. 27, 231-264. Stewart, E. C., & Bennett, M. J. (1991). AmeriLan cultural patterns A cross- culm perspective. Yarmouth, ME: Intercultural Press. Sueda, K., & Wiseman, R. L. (1992). Embarrassment remediation in Japan and the United States. International Journal of Intercultural Relations 16 159-173. Thompson, C. A., Klopf, D. W., & Ishii, S. (1991). A comparison of social style between Japanese and Americans. Communication Research Reports. 8. 165-172. Ting-Toomey, S. (1988a). Rhetorical sensitivity style in three cultures: France, Japan, and the United States. Central State Speech Journal, 39, 28-36. Ting-Toomey, S. (1988b). Intercultural conflict styles: A face-negotiation theory. In Y. Y. Kim & W. B. Gudykunst (Eds), Lh_eorie§ in intercultural communication (pp. 213-238). Newbury Park, CA: Sage. Triandis, H. C. (1993). Collectivism and individualism as cultural syndromes. Cross-cultural Research. 27. 155-180. Triandis, H. C. (1994). Culture and Social Beh_avior. New York, NY: McGraw- Hill. Triandis, H. C. (1995). mividualisnflnd collectivianl Boulder, CO: Westview Press. Triandis, H. C. (1996, March). Individualism-collectivism; Theog, method, and implications for communication. Speech presented at the plenary session of Conference on Theory and Research on Culture and Communication, Fullerton, CA. Triandis, H. C. (1996). The psychological measurement of cultural syndromes. Amerim Psychologist. 51. 407-415. 73 Triandis, H. C., Bontempo, R., Villareal, M. J ., Asai, M., & Lucca, N. (1988). Individualism and collectivism: Cross-cultural perspectives on self-ingroup relationships. Journal of Personality and Social Psychology. 54. 323-338. Triandis, H. C., Chan, D. K.-S., Bhawuk, D. P. S., Iwao, S., & Sinha, J. B. P. (1995). Multimethod probes of allocentrism and ideocentrism. International Journal of Psychology. 30, 461-480. Triandis, H. C., Chen, X. P., & Chan, D. K.-S. (1998). Scenarios for the measurement of collectivism and individualism. Joumflf Cross-Cultural Psychology, 22, 275-289. Triandis, H. C., & Gelfand, M. J. (1998). Converging measurement of horizontal and vertical individualism and collectivism. Journal of Personality and Social Psychology. 74. 118-128. Triandis, H. C., Leung, K., Villareal, M. J., & Clack, F. L. (1985). Allocentric versus ideocentric tendencies: Convergent and discriminant validation. Journal of Research in Personality. 19. 395-415. Triandis, H. C., McCusker, C., Betancourt, H., Iwao, S., Leung, K., Salazar, J. M., Setiadi, B., Sinha, J. B. P., Touzard, H., & Zaleski, Z. (1993). An etic-emic analysis of individualism and collectivism. Jougal of Cross-Cultural Psychology. 24. 366-383. Triandis, H. C., McCusker, C., & Hui, C. H. (1990). Multimethod probes of individualism and collectivism. Journal of Personality and Social PsychologY. 59. 1006- 1020. Yamaguchi, S. (1994). Collectivism among the Japanese: A perspective from the self. In U. Kim, H. C. Triandis, C. Kagitcibasi, S.-C. Choi, & G. Yoon (Eds)., Individualism and collectivisrm: Theory. method. and applications (pp. 175-188). Thousand Oaks, CA: Sage. Yamaguchi, S., Kuhlman, D., & Sugimori, S. (1995). Personality correlates of allocentric tendencies in individualist and collectivist cultures. Joumal of Cross-Cultural Psychology, 26, 658-672. Yum, J. O. (1994). The impact of Confucianism on interpersonal relationships and communication patterns in East Asia. In L. A. Samovar & R. E. Porter (Eds), Intercultural communication: A reader (pp. 75 - 86). Belmont, CA: Wadsworth. 74 APPENDICES 75 APPENDIX A TABLES 76 .888.» 88:88.. 85 E 288—. :w. 3 8388 83 2886 83:3 85 5.. 8:88 2F .. :88 09 238% xmt 8 .85 888 @5888 8: 888 9 88a am. mm. 54. mm x 85 88288 b888w mm x 88% 8:83 85 E 85 88:3 H cw 88:8 :88: 85 8.5 05 80¢ 88on 85 mo “:85 85 98888:: 9 8888 8 888: M3. cm. 0.8. 8 85 88288 b888w mm z 885 :oHED 85 E 8:: 88:3 H Amv .888 3820 Ho: 8 z a 88 888 .382 888% 85 8:3 So ofiwm 9 38 up 8 888: 8 88:8 :8 888% 8. 2. a. a a: 86228 £888 a a 88m 88: a: a a: 26:3 H 3 8808888 9 8:38 28 8 on 85 33388 .98 8: 88 888% 85 b 88 >8 2 may 8 888% 8 8:? 88888:: 3 28 on 3:08 888: cc. NV. cm. 8 85 88288 388:8 mm x 4.88% 82:: 85 E 85 28:8 ~ CV 28m 888930685 88D 88:88.. 88G .m.D 88: 8:588 wfiwmoq 888m :8: gum 96883“ :m :8: 88m mo :88; 888m .: 23 77 mm. mm. 8. .8888 85 .8 885 58, 388.88 8 8888 >8 8:88 8.88 H AC no. Hm. on. .888 8888858 :8 88 88 8 88 :8 888.888 38> 8 8H Cu 8m. Hm. 8m. .8888 :28 >8 .8.“ 388888: 88: H HQ 8 8 3.. .228 8% 888% ea 8.88 _ 8 no. em. no. .8850 808 88888 .888 88:: 8 88 88 H A 3 88am 8:88o0..8m 8888888 .8888 85 .8 8888 85 :85 88.888 888 388 8 8828888 8 885888 mm. Hv. mm. 38H 85 88288 388:8 8 8 88m 8883 85 885 88:8 H AC .3888 888 8 8 888: 8058, 808888 on :88 888 8.888% mm. 8m. 3. 85 85 88288 388:8 8 8 88m 88:: 85 8 85 88:8 H GV .3888 :85 38:8 885 .88 8 88888 888 38:8 m8. mm. mm. 8 8 85 88288 388:8 8 8 888 88:3 85 8 85 88:8 H AB 883 8888888 88D .m.D 885 8:888 88804 888 88H 8888 H 28H. 78 # n. we. no. A80: 08965 805 .8 28 8: 8L $8883 083.380 0303 080m Cb 8. 8. we. waste: 8 mafia; 6 ch. mm. mm. .3308 woow a PE: 3 03688 8: fl 8 825388 805;? AS mm. we. 5. .288 .8 32 05 mm 8253800 A: 28m 883.6385 13:85 88388 av. 3. me. 9538 283 m8? :2: How 28 £805 820 2:88 9 88888 mm : Amv me. we. we. 95$ >8 8 805 mo $08qu: 05 so 883% $0838 m2 ANV 808528888 78083 8. av. mm. >8 55 88898 808 98 anew >8 8 8050 53, 3380838 >2 CV Egoufiom 808833.58— SmQ 8833. Sam .m.D 83 38388 madame; 88mm 80: 3:888 _ 2an 79 mm. 3. mm. mm. em. am. wc. om. «m. wv. ov. mm. 2. ”gm. mm. mm. mm. 3. me. wv. mm. 5». mm. mm. @808 :26 >8 m_ 08 8 83%: “an? Sc Bacon 5? @5886 80:? 83888 98 888 on 8 6.8.8 _ AS .528 E as H av ...88 55 E 8.. 5% H 8 2.3 858223 3:88: .080: 8 m: 55> .3888 $8? So 38— Esogm 0.3 $3 .855 >58 88... >=8£ >8 no 83808 “88 53» 8388 _ .88 8.88 a mac—.8 880m Amv >868 35 @8868 H t :05 .388 >8 883 2:95 8:3 ow 2:95 ~ Amv .: .8 308% 8: 28 >=8£ >8 .t 858 >83 >080 ~ 85 33:3 8“ 8588 2:03 _ CV 2.8m 88380200 1838 > Sum 8289:. 38 .ms 9% 8888 wfivmoq 88mm 80: 3:888 _ 2an 80 Hm. mm. NV. oo. 8. R. om. om. 80D 000883. Hm. wm. mm. 3. cm. on. on. om. 98 0.: 5% 8888 85004 880$ 2‘. oo. om. cc. we. mm. 3. av. 002006 05 803 5.800 08000 505 8050858 H0 35 05: 805080800 805 a: x08 H .0085 >802 Amv 0805080800 8 m8>0m 000 0800a :23 H0 £00m 80 00098 05 8500005 .HH0m>8 H85 8050 H CH 0H00m 88002 w850000o .8050 53» 085 858000 8 080008 .08 0 H. a; .8050 53, 05800000 H 8055 000w H00H H AmV 5:08 H05 2:03 H .088 0 80m 80503.00 0 .HH ANV .08 00 808008 8 0503.00 >8 H0 8:00-203 08% CH 0H00m 80:58:00 380800: .m>03 >808 8 8050 805 8080.55 H80 008:: mafia >080 H g 005250 >8 H0 0020000 >58? 8 5H 5000030 H 8055 A3 80: 0038800 H 030;. 81 .mfiomummuo>fioo R. R. mm. 5 wcHom can 0303 05:3 $888 :3 H 35 03% a 26: auto H AD mm. 3. we. .mcoummbéoo 5 $5538 03:8 250: cote H GV .mcosmmbéoo ow. ow. Hm. E wfizam 2a 030a 8:3 5 mwcaaofi 5qu BE auto H HS .3 93m 0% 8 2H 2&3 :05 >3 9 wfiow mH on. Ho. 3.. ~8th 8523 3:3 860a auto 9 2% on 9 Baum H mcoHEmSZSU E a; .252: 35 wow fact mu. om. mu. who—to swag: :96 @3888 2mm 0:888 33 @536qu :oao :8 H HO Sun 80:33. Sac .m.D Saw @8388 wfiumoq 88mm 88H 325:3 H 2an 82 .250 8338:8800 83:00:32 0:08 3:388 :38 3:an < .80 Z womd VmH H.m 33$ onHHu coo. mm: momw 0:0 3%: :3 332 4WD - :30: 8280.88 3:28:00 8:80th :38 :8 80803: H0888 3:08.800 $3 00:03HHHH0 :38 AH H: : mummd H Hm vwmwé mwomém :32. $3.0 vmm 236 33.3. .m.D mm 7H Qm :32 23m 3838:8800 8880045042 HHH 05 :0 08mmHoHtmnH 03:0:de H03 .m.D :330m 3.8% .N 0331 83 03% 333808800 80800-:an 0:08 000388 :38 :0:an < 0:02 8:3 2 : $03. 38.2 5%: 88.0 5 ~83 $38 0.: 00:08 83.0 3 8:“: 88.x 8%: 23.0 M: :33 83.8 0.: 00380.: 00 z :m :82 0202 0:: m0H080HH :00 23m 8038:8800 8080045042 HHH 0:: :0 0808280: 000825. 08:: .m.D :00E0m 300:? .m 033. 84 ommo voofim omCN 309m ooo. m2 ooo:V 208 momo 33% vamd Smoé ooo. 0mm mood 0.080.: 0:0 :25: 0:0 :032 Add - 5:03 30:00-83 :000_0::00 00:0:0tm0 :008 :8 80880:: _0>:0:8 00:00:.800 £03 00:0:0bm0 :008 : o0 : 0008800 m 030B 85 .0090 :0000808800 80800-:3: 0:08 00:00:08 :008 :0:w_: < 0:0 Z woSmo 2 : wvmvé 33.3 00:02 oommo 3 00:06 ooocém 00—080:— :0:0_. onooo S Nmood oovmom 00.02 Sumvo m3 526 oonodm 00—080,: .m.D mm 2 Gm :002 alga: 90 0.0 2: a 280 80888800 0080-332 8 20 8 802 Ea was: 8233 as: v 2%.: 86 mnoo ommoo oovoN- :NKo mom. oom :mo: :0:0_. mmoo 3:: 00mm. 7 onomo- 5mm. Nmm ommo .m.D 0:0 :08: 0:0 :038 3:08 - 2080b 30:80:53 800—0800 00:88.80 :008 :8 80880:: 0:208: 00:00:.800 o\°mo 00:88.80 :008 q 80 : 0008800 v 030% 87 .m0_:0::0> 800:0:008 00 00:00:: 0:03 003089» :0::0 =< 0:000 8088:8800 8080045228: 0:: 0: 03005» 800:0:00 8% 0:0 Z mtg. wKo 00o. 0w0 mom. _om.o NNo. 0w0=00 8 :00» mum. Smo- omor x00 ooo. oSH :33. 80:9. :0 .m.DV 88:00 3000 a : 08208000 0080:0030 0:00m 8088:8800 808004508: 5 0:: 080 00508.5» 800:0:8m .8 :0:000 08 0_ 2:2 m 038 88 .mO—£NE> ”EDUGDQDUGM mm @3605 0.53 mOEQm .550 =< .2000 8088:8800 8080045228: 0:: 0: 03083 800:0:00 0: P 0:02 So. 83 :3. 05:82 0:08:00 so. 33- 02.- 8236280 8:88: 30. $20- 5 5.- 8038305 08800: :5. ~50 mi. 880230 0200:: am. $3 $0. 880388: 35> 08. $20 08. 08:88.80 80058235 0: :. 02. :- 0%.. 02:88.80 0.008305 000 085800 02000 a : 080688000 0080:0080: 0:00m 8088:8800 8080045042 E 0:: 0::0 00:00m :0::O .8 :02000: 08 0: 2:2 0 030.2. 89 00305:, E03032: 00 00:00:: 0:03 00:08 :0::0 =< 0:000 3300:5888 :x0::00-30<:w:: 0:: w: 030.90» 8000000: 0::. .80 Z So. 8; RN. 05502 058300 mg. 80.0- 08.- 80:08:30 @8580 $0. 5.0- 03.- 8003238: 3808: 000. 82 :0. 8238230 35> an £2 08. 80:22:05 35> SN. 02.: ms. 33:88.:om 05053225 : _ _. 80.: :3. 3388.000 05200005 .000 0.0 00000 0 : 03065000 0363056 “0005300 0 030B 90 0050:0000 0000000000: 00 00:00:: 0:03 02000 00:00 =< .2000 0000300000000 :x0:000-30_\:wE 0:: 0: 030E0> :00000000 0: H .0:0 Z So. 0: m.m gm. w0:0002 90:00:20 now. om _ . : 5. 803000—60 3:000:00: m _ m . m 3. ooo. 800002300: 3:000:00: 03. ~00. mmo. 803002—00 :00:::0> 0mm. wmc. 000. 00000000000: :00:::0> $0. 00 _ . _- 000.- 02008000 08058235 30. $0. :- E _ .- 02.88000 0505805 0:0Q 0000000.. 00000 0 : 30065000 000000080 00000000 0 030... 91 0:00m w0:000:>: w0000:0n: u :00:0Q .0:00m 000:>::00::0U 3:000:00: n :00 00:: .0:00m 00000000000: :0:00~:00:: u 00: 00:: .0:00m 0003000009 :00000> n :00 00> .0:00m 00000000000: :00000> u 00: 00> .0:00m :000:0000..::0m :0000000000:0: n 00:0: .0:00m :000:0000.0:0m :000000000: n 0000: 0:0 Z 000 00000 .000 0030: 000200000 00:000000 00.: om. .wm. 0m. .8. 0m. 6:. 0m. .3. mm. .3. mm. .no. 00. .oN. :0>00:0: 000000000 003 0m. 0:. mm. m0. 0:. on. mm. 0000:00000 00:000000 0000:00000 00m. oo :. EV :. mom. 00o. m: :. :0 :. 00:00000000 0:00 .m.D 000 00000 .000 0030: 000200000 00:000000 00.: 0:. .:o.- ::.- .mmr oo. .m:.. cm. .wo. 0:. .:o. o:. .wo... mo: .mmr :0>00:0:000000000 o\om0 mo. om: 0o.- 0 :. o:. :o. 0: .- 00::0:00000 00:00000 00020000 So. 03.- moo... no :. who. woo. 00o; 00:00000000 0:00 0000:0000 :00:0Q :00 00:: 00: 000: :00 00> 00: 00> 00:0: 0000: 00:00m 00:5 000 0:00m 00000000000000 00000000506: :0: 0:: 0003:0mJ00000:00000 0 0::00. 92 0:00m w0:000:>: w0000:0Q n :00:0Q .0:00m 000:>::00::0U :0:00N:00~: u :00 003 .0:00m 00000000000: :0:00N:00:: u 00: 00:: .0:00m 0:0:>::00::0U :00::00> n :00 00> .0:00m 00000000000: :00000> H 00: 00 > .0:00m :0000000..::0m :0000000000:0: u 00:0: .0:00m :000:0000.::0m :000000000: n 0000: 0:0 Z 000 00000 .000 0030: 0000:00000 00:000000 00-: am. .m:. no. .0:.- w:. .wo.- mm. .vor mm. .8: 0o. .omr mo. .mmr 0.0000000000000003 R. 00.- mo. 8. S. S.- 2- 802208 3:808 N 0 N. 0.0.- 08. So. 08. 03.- 000.- 802208 03808.5 0% 805.00 :00:0Q :00 00:: 00: 00:: :00 00 > 00: 00 > 00:0: 0000: 00000000 :- 030% 93 0:00m 000:>::00::0U 300000: n :00 000: .0:00m 00000000000: :0:00~:00:: n 00: 003 .0:00m 000:>::00::0U :00000> n :00 00 > .0:00m 0000002300: :00000> u 00: 00 > .0:00m :00000000..::0m :0000000000:0: u 00:0: .0:00m :000:0000-0:0m 0000000000: n 0000: .0:00m 0000000000000 :00:000-30.:\:w::: n 30::w::.: 0:07: 000 00000 .000 0030: 00020000 00:000000 000 00. .00. 00. .0... X. .8.- 00. .00. a. .8. 00. .30 00. .00. 0225 850.008 :00 00.. 00. :. 0.0. 0:. 00. am. 802208 8.8008 0:0. 09.. NS. 00:. 0::. Em. 000. 802208 02808.5 0% 0.0 000 00000 .000 0030: 0000:00000 00:000000 00.: m0. .0m. 00. .mm. mm. .0o. mm. .vo. 0m. .wo. oo. .3. 0:. .:o... :0>00:0:00000m:000 o\om0 mm. mm. 0:. m :. 0:. 0m. mo. 00::0:00000 00:00000 owm. 3.0. N: :. No:. 0: :. 0 :0. moo. 00::0:00000 0800000000 0:00 0000:0000 :00 00:: 00: 00:: :00 00 > 00: 00 > 00:0: 0000: 32:30: w0|:00m 00:5 000 080:: w0:000:>: w0000:0Q 0003:0m: 0000300000 0 2.00 94 0:00m 0000300330 :00000:00:.: u :00 00:: .0:00m 00000302300: 00000000: n 00: 003 .0:00m 0000300230 :00000> n :00 00 > .0:00m 00020003000: :000000> u 00: 00 > .0:00m :00000000..0:0m 0000000000000: .I. 0000: .0:00m :00000000..0:0m 0000000000: n 0000: .0:00m 00000000000000 00000009306030: u BoEwE .0007: 000 00000 .000 0030: 0000:0008 000000000 00.0 m:. .::.- 00. Km. on. do: 0:. .O:.- mm. .05.. 00. .NN. 0m. .m:. 02250000000000.0003 No. wm. no. 8. o :. 0m. 0N. 00:00:00000 000000000 m :o. mom. bmo. mmo. m3. mom. N:N. 00:00:00000 00000000000 0000 00000000 :00 00:: 00: 00:: :00 00 > 00: 00 > 0000: 0000: 050in 0000:0000 m 0300. 95 Table 9 Correlations Between All Scales for Combined Data Highlow Indep Inter Ver Ver Hor Hor Detect Ind Col Ind Col Highlow Indep Inter Ver Ind Ver Col Hor Ind Hor Col Detect N_otg The numbers in the lower triangle (darker shade) are the uncorrected correlations. The numbers in the upper triangle (lighter shade) are the correlations corrected for attenuation due to error of measurement. The standard error for corrected correlations is indicated in parentheses. The numbers in the diagonal are reliabilities of the scale used when correcting for attenuation. Some corrected correlations are 1.00 because of the algorithm for calculating the corrected correlation; [obtained correlation] + [(«l reliability)x(‘/ reliability)]. High Low = High/Low-context Communication Scale, Indep = Independent Self-construal Scale, Inter = Interdependent Self-construal Scale, Ver Ind = Vertical Individualism Scale, Ver Col = Vertical Collectivism Scale, Hor Ind = Horizontal Individualism Scale, Hor Col = Horizontal Collectivism Scale, Detect = Detecting Meaning Scale 96 Table 10 Correlations Between All Scales for US. Data Highlow Indep Inter Ver Ver Hor Hor Detect Ind Col Ind Col Highlow Indep Inter Ver Ind Ver Col Hor Ind Hor Col Detect 1191;, The numbers in the lower triangle (darker shade) are the uncorrected correlations. The numbers in the upper triangle (lighter shade) are the correlations corrected for attenuation due to error of measurement. The standard error for corrected correlations is indicated in parentheses. The numbers in the diagonal are reliabilities of the scale used when correcting for attenuation. Some correlations are 1.00 because of the algorithm for calculating the corrected correlation; [obtained correlation] + [(x/ reliability)x(\/ reliability)]. Highlow = High/Low-context Communication Scale, Indep = Independent Self-construal Scale, Inter = Interdependent Self-construal Scale, Ver Ind = Vertical Individualism Scale, Ver Col = Vertical Collectivism Scale, Hor Ind = Horizontal Individualism Scale, Hor Col = Horizontal Collectivism Scale, Detect = Detecting Meaning Scale 97 Table 11 Correlations Between All Scales for Japanese Data Highlow Indep Inter Ver Ver Hor Hor Detect Ind Col Ind C01 (.068) (.0683) (.068) (.068) ((31058 " " .67 3:0, .07, _ 1.311;, -_ Indep Inter Ver Ind Ver Col Hor Ind Hor Col Detect 13% The numbers in the lower triangle (darker shade) are the uncorrected correlations. The numbers in the upper triangle (lighter shade) are the correlations corrected for attenuation due to error of measurement. The standard error for corrected correlations is indicated in parentheses. The numbers in the diagonal are reliabilities of the scale used when correcting for attenuation. Some correlations are 1.00 because of the algorithm for calculating the corrected correlation; [obtained correlation] + [(‘J reliability)x(‘/ reliability)]. Highlow = High/Low—context Communication Scale, Indep = Independent Self-construal Scale, Inter = Interdependent Self-construal Scale, Ver Ind = Vertical Individualism Scale, Ver Col = Vertical Collectivism Scale, Hor Ind = Horizontal Individualism Scale, Hor Col = Horizontal Collectivism Scale, Detect = Detecting Meaning Scale 98 owomd 0cm owomé 00006». 0000:. med 0mm 08:0 32.0w .m.D 0002002300: :0:00~:00:: 0000.0 000 0000.... 0000.0: 0.0000 0000.0 000 000.0 00.00.: 0.0 80:08:00 080005 0000.0 :0 033. 0000.00 0000.. 0000.0 0.00 0000.0 0000.00 0.0 80000003050020; 0000.0 000 00000 030.00 00000 mm:m.o mmm :wmm.m $3.0“: .m.D :00000000..::0m 0000000000000: 0000.0 :0 0000.0. 000.00 00000 0000.0 000 003.0 0:000 0.0 03000800000805.0005 00 z 00 08: 000 00000002300000::000:>:00:-:m0000:00::\:00:000> 00:0 000 00:3m :0000000.0:0m 00:0 00 0000060000 00000000 000 .m.D 0003000:i00000-.:. .N: 0:000. 99 0000:0000: 05 008,: 00:02:00 0003 00008 :0 0000000 00:: 005 00000::00: 0000:0000 00000000 0>00w00 0::. .0002 :06. 3%. m good om :N. m ooo. m m:0 N : 0.0: 000300230 0300003.: wnmd 83.0 5ch 0035 ooo. :90 m9; 80:00:0300: 300000: mid- $0on mmmvd aim: moo. mmv vmod 000300330 :00::0> mmod 30: :.: mowed- w : mmd 0mm. m m0 306 0002002300: :00:t0> vmod- 30:: $86 :Nood wvo. 090 $0.: :0000000..::0m 0000000000000: 0:06 mammé mmmwd vmmmd ooo. N30 9090 6000000.::0m 0000000000: :000 0000:: :000 0030: 00000:. - .mdv 80:00-03: 0000:0000 000000030 0008 00,: 000000.:m::0 00000000 :03000: 00003000 003 0020 0: :0 0 mNmNd Sm ommoa 080.0: 0000:. 0mm : .o mmm 0305N om : QNN .m.D 0000600200 3000000: mm 2 mm 000:): 0000:0000 m: 030:0 100 0000.0 0: 0000.0 000000 00000 00000 00 0000.0 0000.00 0.: 0202 0000.0 00 0000.0 000.00 00000 0000.0 0: 0000.0 0000.00 0.00 0200000 000008.0000000000000200 0000.0 0: 0000.0 000.00 00000 0000.0 00 0000.0 0000.00 0.00 0202 00 :00 00 0000.0 000.00 00000 000.0 0: 200.0 0000.00 0.: 0200000 00000000000000.00000000 00 00 000: 0200300 000% 000300239000::000:>::00:-:m0000:00::\:00:t0> 05 :000 00:00m :0000000.0:0m 00:0 00 0000000000: 00000000 000 .m.D 0003000: 00000-00 .m : 030:. 101 0000.0 00 0 3000 000000 00000 0000.0 00 0000.0 0000.: 0.00 0202 0000.0 00 0000.0 0000.00 00000 00000 0: 0000.0 0000.: 0.0 0200000 000308000 000E0> 0000.0 0: 0000.0 0000.: 00000 0000.0 00 0000.0 0000.00 0.0 0202 0000.0 00 0000.0 000.00 00000 000.0 0: 0000.0 0000.00 0.: 00000000 0000000000000 0000005 00 00 000: 0000008 00 2000 102 mmnmd o: 300.00 mm 5.2 50900. Nammd 0w wmmoxm meNN .m.D 02002 GE md 3 ammo. m mnomdm 08900. wmomd 3: voovN mmmm.mm .m.D 02008000 803082—00 100085003 wm 0 To m: 0 m N010 vavdm 5005. omomd ow $90.00 mmSém .m.D 00:02 named 00 33.00. 0003.00. S0900. wummd 3L ommmd cemndm .m.D 02089.0 80203230051000.3083 mm Om S002 0000008 00 0300. 103 0530:0800 05 80000 800230 0.53 008200 00 0000000000000 05 005 08365 00202208 2025020 gamma: 20H .80 Z :md- mmmo.m 090900 $va moo. m2 vmod 0032 ~36. co 0 m .0 336. m 0 30d omv. cmm Bad 02080000 800382—00 080:“; mvmd comma mmmmd ovnmd So. 02 Sm.m 0202 owed- mmmfio o 0 mm . 0- womvd- @000. 2.0. 0 _ 5o- 0208000 802323000: €30.50, Sod- momm; mmmmd- Romd 00m. .30 2.0.0 0202 coed- 3:1 mm. 0 .o- Emod m3. 0mm 20m.— 0208000 _wE00=oo..:om 0500050030085 306 0900.0 mvwod 0m $.m ooo. 30 $06 0202 wovd N330 NmeN who?” ooo. 3.0. 03.0 0208000 Rabmcoofium 0500500205 25 0003: 25 0032 AS00000. - .mdv 8209-028 8:20.006 08:28.08 00on 00.0 3285 88.5.0006 00025300 8:030:00 003 00on a .20 0 335:8 2 030.0. 104 080860.000 05 80¢ 800203 82> 0:008 .00 0020806 05 0.05 08865 08322.08 0580020 9503: 2C. .30 Z 00.0.6- 330 mm 5.0 m0mo.m ooo. m3 806 0202 0506- maid 0 _ mmd mmmo.m ooo. omm 00m.» 000—0800 80038230 30800.82 mmvd 336 good 0.0030 ooo. m3 omnd 0202 80¢ 50va 0900.0 030:0. ooo. 0mm cmm.m 0208000 802006365 3000008: 25 .8qu use 832 900000. I .m.Dv Gozfiégv 800800.020 nous—oboe 538 80 BEBE 02080ch 0505800 85000.38 003 00.88 a :0 0 3:50:00 m 0 2an 105 APPENDIX B QUESTIONNAIRE AND RESEARCH CONSENT FORM (IRB# 99293, approved May 18, 1999) The seven-point anchor was inserted afier each item in the questionnaire. The abbreviation after each item indicate the following: [HIGHLOW #] — High/Low-context Communication Scale item # (passed CFA tests in this study) (HL) — item thought to be measuring high/low-context communication, but failed CFA test(s) in this study [INDEP #] — Independent Self-construal Scale item # (passed CFA tests in this study) (D) — Kim et al.’s independent self-construal scale item, but failed CFA test(s) in this study [INTER #] — Interdependent Self-construal Scale item # (passed CFA tests in this study) (T) — Kim et al.’s interdependent self-construal scale item, but failed CF A test(s) in this study [VERIND #] — Vertical Individualism Scale item # (passed CFA tests in this study) (V I) — Singelis et al.’s vertical individualism scale item, but failed CFA test(s) in this study [VERCOL #] — Vertical Collectivism Scale item # (passed CFA tests in this study) (V C) — Singelis et al.’s vertical collectivism scale item, but failed CFA test(s) in this study [HORIND #] — Horizontal Individualism Scale item # (passed CFA tests in this study) (HI) — Singelis et al.’s horizontal individualism scale item, but failed CFA test(s) in this study [HORCOL #] — Horizontal Collectivism Scale item # (passed CFA tests in this study) (HC) — Singelis et al.’s horizontal collectivism scale item, but failed CF A test(s) in this study [DETECT #] — Detecting Meaning Scale item # (passed CFA tests in this study) (DM) — Daly et al.’s detecting meaning scale item, but failed CFA test(s) in this study 106 Please indicate how much you agree or disagree with the following statements by circling the appropriate number on the scale. Example Children’s cartoon programs are too violent. 1 2 3 4 {5) 6 7 strongly moderately slightly neither agree slightly moderately strongly disagree disagree disagree nor disagree agree agree agree The circle on 5 indicates that you slightly agree that children ’s cartoon programs are too violent. Please do not skip any questions even if some of them seem redundant. Thank you very much for your cooperation. A. In the following section, you will find a set of statements asking your belief about daily conversation in the US. Please respond according to what you believe to be the general practice among typical people that you meet and observe in this society. <1> In the US. it is generally considered that a listener should be able to understand what a speaker is trying to express even if the speaker does not say everything that he or she intends to communicate. [HIGHLOW 1] l 2 3 4 5 6 7 strongly moderately slightly neither agree slightly moderately strongly disagree disagree disagree nor disagree agree agree agree <2> In the US. it is generally considered that what a person really means can often be different from what the person actually expresses in words. (HL) <3> In the US. it is generally considered that how a person says something is usually more important than what she or he says. (HL) <4> In the US. it is generally considered that the content of the message is usually more important than how it is communicated. (HL) 107 <5> In the US. it is generally considered that some things that one wants to express are better left unsaid. (HL) <6> In the US. it is generally considered that a speaker usually can expect a listener to be able to figure out what the speaker really means even if it is not clearly stated. [HIGHLOW 2] <7> In the US. it is generally considered that the speaker’s intent carmot really be understood unless they are stated clearly. (HL) <8> In the US. it is generally considered that if a listener misunderstands what a speaker is saying, it is because the speaker had not adequately explained what he or she meant. (HL) <9> In the US. it is generally considered that everything that one wants to communicate should be expressed fully by words. (HL) <10> In the US. it is generally considered that a listener is expected to understand the intent of the speaker from the way the person talks. [HIGHLOW 3] <11> In the US. it is generally considered that there are some things that cannot be expressed by words. (HL) <12> In the US. it is generally considered that a listener usually cannot be expected to be able to figure out what the speaker really means unless it is clearly stated. (HL) <1 3> In the US. it is generally considered that it is better to risk not speaking enough than to risk speaking too much. [HIGHLOW 4] In the US. it is generally considered that it is usually more important to say things politely than clearly. [HIGHLOW 5] <15> In the US. it is generally considered that the speaker’s intent can be understood without having to be stated clearly. [HIGHLOW 6] <16> In the US. it is generally considered that a speaker should express everything clearly for the listener in order for the listener to understand what a speaker intends to communicate. (HL) <17> In the US. it is generally considered that how something is communicated is usually more important than the content of the message. [HIGHLOW 7] <18> In the US. it is generally considered that it is usually more important to say things clearly than politely. (HL) 108 <19> <20) B In the US. it is generally considered that it is better to risk speaking too much than to risk not speaking enough. (HL) In the US. it is generally considered that if a listener misunderstands what a speaker is saying, it is the listener's problem, not the speaker’s. (HL) The following section asks about some of your personal preferences. We are interested in what you think and how you behave. Please respond according to what you personally think, or how you yourself would behave. <21> I don’t like depending upon others. [INDEP 2] 1 2 3 4 5 6 7 strongly moderately slightly neither agree slightly moderately strongly disagree disagree disagree nor disagree agree agree agree <22> I often find myself detecting the purpose or goals of what people are saying in <23> <24> <25> <26> <27> <28> <29> <30> <31> <3 2> conversations. [DETECT 1] It annoys me when other people perform better than I do. (VI) I enjoy being admired for my unique qualities. (D) The well-being of my co-workers is important to me. [HORCOL 1] I feel uncomfortable disagreeing with my group. (T1) I often “do my own thing”. [HORIND 1] Understanding myself is a major goal in my life. (D) I would sacrifice an activity that I enjoy very much if my family did not approve of it. [VERCOL 1] I should be judged on my own merit. (VC) Competition is the law of nature. [V ERIND 1] I conceal my negative emotions so I won’t cause unhappiness in my group. (T) 109 <33> <34> <35> <36> <3 7> <3 8> <39> <40> <41> <42> <43> <44> <45> <46> <47> <48> <49> <51> One should live one’s life independently of others. (HI) Many times, I pick up from conversation little bits of information that people don’t want to disclose. [DETECT 2] If a co-worker gets a prize, I would feel proud. [HORCOL 2] I try to meet demands of my group, even if it means controlling my own desires. (T) I would do what would please my family, even if I detested that activity. [VERCOL 2] The security of being an accepted member of a group is very important to me. (T) If a relative were in a financial difficulty, I would help within my means. (HC) My relationships with others in my group are more important than my personal accomplishment. [INTER 1] When another person does better than I do, I get tense and aroused. (VI) Having a lively imagination is important to me. (D) Before taking a major trip, I consult with most members of my family and many fi‘iends. [VERCOL 3] I can ofien understand why someone said something even though others don’t see that intent. [DETECT 3] I like my privacy. [HORIND 2] I voice my opinions in group discussions. (D) I usually sacrifice my self-interest for the benefit of my group. (VC) Speaking up in a work/task group is not a problem for me. (D) I prefer to be direct and forthright when discussing with people. [HORIND 3] My happiness depends on the happiness of those in my group. [INTER 2] It is important to me to maintain harmony within my group. (HC) 110 <52> <53> <54> <55) <56> <57> <58> <59> <60> <61> <62> <63> <64> <65> <66> <67> <68> <69> <70> <71> It is important to consult close friends and get their ideas before making decisions. [INTER 3] Without competition, it is not possible to have a good society. [VERIND 2] In conversations I seem to be able to often predict what another person is going to say even before he or she says it. [DETECT 4] Winning is everything. [V ERIND 3] I often consider howl can be helpful to specific others in my group. (T) I like sharing little things with my neighbors. (HC) I ofien hear things in what people are saying that others don’t seem to even notice. (CS) I am a unique individual. (HI) I don’t change my opinions in conformity with those of the majority. [INDEP 5] Children should be taught to place duty before pleasure. (V C) My personal identity, independent of others, is very important to me. (D) What happens to me is my own doing. [HORIND 4] I am careful to maintain harmony in my group. (T) I hate to disagree with others in my group. O] C) I enjoy being unique and different from others. (D) I often find hidden meanings in what people are saying in conversations. [DETECT 5] I feel good when I cooperate with others. [HORCOL 3] I have an opinion about most things; I know what I like, and I know what I don’t like. (D) It is important that I do my job better than others. (VI) If my brother or sister fails, I feel responsible. (T) 111 <72> <73> <74> <7 5> <76> <77> <78> <79> <80> <81> <82> <83> <84> <85> <86> <87> <88> <89> C I often notice double meanings in conversations. [DETECT 6] My happiness depends very much on the happiness of those around me. (HC) When I’m with my group, I watch my words so I won’t offend anyone. (T) We should keep our aging parents with us at home. [VERCOL 4] It is very important for me to act as an independent person. [INDEP 4] I enjoy working in situations involving competitions with others. (VI) I should take into consideration my parents’ advice when making education and career plans. (T) When I succeed, it is usually because of my abilities. [HORIND 5] I prefer to be self-reliant rather than dependent on others. (D) To me, pleasure is spending time with others. [HORCOL 4] I would sacrifice my self-interest for the benefit of my group. (T) I enjoy being unique and different fi'om others in many ways. [HORIND 6] I take responsibility for my own actions. [IN DEP 3] Children should feel honored if their parents receive a distinguished award. (V C) I act as a unique person, separate from others. [INDEP 1] Some people emphasize winning; I’m not one of them. [V ERIND 4; reverse item] I often have a sense that I can forecast where people are going in conversations. [DETECT 7] I act as fellow group members would prefer me to. (T) Please circle the appropriate choice, or fill in the blank. What is your sex? How old are you? male female 112 What is your ethnicity? African American Asian American Caucasian American Hispanic American Native American international student Other (please specify) What is your major? What is your year in college? Was any part of this questionnaire confusing to you? Yes No If Yes, which question? This questionnaire is printed on both sides of the paper. Please make sure that you filled out all pages of the queStionnaire. THANK YOU VERY MUCH FOR YOUR PARTICIPATION! 113 RESEARCH CONSENT FORM The purpose of this study is to develop a new scale. Some of your attitudes and behaviors will be asked. It will take approximately 30 minutes of your time. 0 Apart from your participation in this study, your actual performance in this study will in no way affect your evaluation in this course, or in any other courses. a Any credit you may earn via participating in this research is not transferable to another class or another semester. 0 Your participation in this study does not guarantee any beneficial results to you. 0 You have the right to withdraw from this study at any time without penalty. 0 You have the right to have this study explained to you to your satisfaction. 0 The results of this study will be treated with strict confidence with regard to the data of any given participation. With this restriction, the results will be made available to you at your request. 0 The data you provide to the researcher as a result of your participation in this study may be used by other scientists for secondary analysis. Again, data will be treated with strict confidence. 0 Should you have any questions, problems, complaints, or if you desire fitrther information, you have the right to contact the following person: Rie Ohashi Department of Communication Michigan State University East Lansing, MI ************************************************************************ Given these understandings, I voluntarily agree to participate in this research as described. SIGNED DATE NAME (print) STUDENT # COURSE INSTRUCTOR ************************************************************************ 114 aoaxutmnrmbwsmwc.hatueoaamn.:tmanczr m?%I®T®R§K#mnrm5&$®¢c.fitfiruiba$QOLrT 0m. W $flfl8®7=itfi§h9~yfiéf§bo 1 2 3 4 1“ 6 7 W é<%5li 7:51: 00%511 8156?“ 4‘15 7:5 MN) .0000 Efiflm\ Ebfiw saw J€WE5 B5 %5E5 5iconic»\Tu\6®l1.¥fltmlr®7:x tui£7w~yn§3fa°6 J: b \ 5:0Kbatfiwbmfifbcteibifo [5‘] DJ: 5 it: & EBEDVCD‘E) J: 5 llfixibfiflfiifibo‘t‘bfil‘fémvfié’twi FnfilCQZiT'Féh‘o :‘fifiléL/(IBEWLiTo A uTwfifin.B$cwafiéfi®—&fiwmwfbificfoS<—&m&a $A®B$c®¥fiwafififiuomf.—&muafifiianfwétfimi Tm?batwawait—fififimomTEELTTéw. vau—fimm.Eitm:t&$fl§§KL&vau—&WK.Aolfifififimaicifiéné:ttufifi: kib§l< battzennxb. (HL) <3>axrn—flmm.éiwaxvfnéafi‘fimwfifikWfiktz EIL‘CW6O (HL) 115 <4> uteri—Habit. 5161850373110 6008557150“ him fitiien‘tmb. (HL) <5> vau—flmlc. {Eitwtfiorbgmztfléttmfimwx:k1b buzzennxa. (HL) <6> atria—£99K. Lao’szalailtmoa D numerack'cb. Bfi< 61001001301120 Lr< #16186 5 azzennxe. [HIGHLOW 2] <7> Basra—mm. 1:039 kaxvaaumna. 1,000on Idiomatic«kai‘awtiienrwb. (HL) Etta—RWK.w atria—12391:. Eituxaameraxvaarb&afieaza h‘Cméo (HL) <10> atria—191391;. Baa$mac$wc0mmseeoxonléa amecaamaténmxe. [HIGHLOW/3] < 11 > Bili'C‘ti-‘fi‘fl‘llt. axvaavsammm Us: 521111,er (HL) <12> atria—£39112. LJP’Sélflllbi‘lioé '9 mcmémxctm. aa Eaten—Rama. 31065336 x 0 113103. 9 mammmsa 2;: 6 nine. [HIGHLOW 4] < 14> Risen—191391;. kttfiwifié. 110% D 3: 1605'? 5 J: MLKEM aajimktmsezzennxé. [HIGHLOWS] <15> atria—”#31:. 110% D tarmac?“ 'C‘b Ma’séwwfil Nomazwfiewaazennxb. [HIGHLOW6] <16> Batman—£391; Lam‘zawofileafiéflcfificrbe51:201.: u.Eitmckéérmoaot§§KTb&afiafiienrwé. (HL) 116 <17> Bili‘Cli—&3‘JIC. 10008502000750. 507101000000; Dbkflfikfiil‘on‘cwbo [HIGHLOW/7] <18> Brahma—flame. 7:160:50}. #LfiIElA 50235 ..t 01103 D t aamsxwmtzennxa. (HL) <19> atria—whiz. emu 72:1»; 9 ciamiflgbjitmfit‘ttia n’cwéo (HL) <2o> Baird—$891; BfiiCfiGT§iTT$Wo <21 > fittMcfibwrtfimt‘, [INDEP 2] 1 2 3 4 5 6 --------------- 7 éci<7c5li 7:51: 9:197:51: 8’66? 421, 765 753729 .5?ka £50be \ $.11?th \ t) a» \ 7C 5 PE 5 .133 5 7C 5 E 5 <22> $330 LTméIfifi¥®§wtwc 0:93:05 1: 120% :. 3: that» L‘CD\Z>:&12§§W\. [DETECT 1] <23> {maestro 52 gammaaaabnamwét‘. (D) <2s> 35001311100000 :. 4:11. a 53‘]: a ortfim: A: page. [HORCOL 1] <26> Wflflwfifilcfixfiémigmwfib\. (r) 117 <27> <28> <29> <30) <31> <32> <33> <34> <35> <36> <37) <38) <39> <40> <4l> <42> 3%.afioctw:terb.nmmmn] afiaagaxraema:amaawxio¢vxaaaav00. m) Efiflfiétltfb.bLifiWBfifiéhtBbéBbéo wmmmn] aaaflmtaae.¢arwmcréb$&otflflit§. fawAlzliEif£< T‘béfilflibmé :tns‘zw, [DETECTB] afim7fi4fiv—é%owfiflafi.mmmwn] fw—7cfibéwéfbfimu.Bfimfifiétbewa<§5°w) kfi.wwwfinokuaaafiofl§2%fimfb.wq mmwmrafiogaeaacaummamuaw.mu) Aaamfbaammauboeawtw.mmmnn] 555$ 80) < Vomifi‘mli E 5503151311023 80) < Eb\$'li‘z)>l: J: 60 [mnmn @mwmeao:amafiuaorxwazacbé.mo <52) film’zfitfi‘i‘éafilcti. fibthcfiLrfifieflk :. tibi‘jct‘ll‘cb 40[mumn <53> fiémmmiawttémca m1. [VERINDZ] <54) game.miamaeaosoaom.m$0§5mmerwvza : a 20%» ans—n). [DETECT 4] <55) 3’): kflé‘C‘Cbéo [VBRINDB] <56) <57) <58) afiu;<.WM®¢ofii®AK%LT.Eérotetoxo&u 32167365000716. ('1) BlahLtbwéffifikfibfibfbwfifléfiom0 nmmmmgeafi.wwxaadmawxaactcsafieid< .2 2: 0010. (DM) 119 <59> afiiitafimflflcabé. (HI) <60> éwn‘mfifilcmcrafimfifiéfizé:tummy [INDEPS] <61> Him. afiwmwwgaarw‘astrb0gmmcantata N‘é'fihé. (VC) <62> mxmxwmc. amwmarmmvmns (EETJ'T‘V-F‘J ’7‘4) 11. BfillkoTKWT‘béo (D) <63) 99203135 fromiiafiwfimvtfi'cbb. [HORIND 4] <64> {wwwfu’i'fi‘coiflziéfifi‘cwbo (T) <6S> (mania/01424000311075. (VC) <66> (1111(0)!) zafioAflfi'cwémi‘fiart“. (D) <67> 03311111; 00600011100. fiéntfltmcfid< 12011331». [DETECTS] <68) (manna—5009mm. [HORCOL 3] <69> 85311110IVEUJIfialcxtL‘CfiRérfio'cmbL. it}? divisii 03 9 LTV\60 (D) <7o> (13AM 5$< afloarweacafigarbé. (VI) <71> bbafiwfifitikafitbtfifi £92066. (T) <72> gramme; Illa¥o>§orwbzklcEfibbcklcivx $12831». [DETECT6] <73> 05300000530310 '9 l:b\6AtB®$‘li‘l:J:6 J: r. macaw. (HC) <74> WMt—filcwélfiii. Awfifiefiéttw:biznwfiafilcfiefi 116. (I) 120 <7s> fizmtauafiwicfififélsét‘. [VERCOL4] <76> (EA): infill-res:amafimorxmazapbé. [INDEP4] <77> flgAkwfiétmét:6'cti32‘9‘60311fi315‘. (VI) <78> fi$$fitmlCBflféfitfiéféfiéltmfiwé5 I: 3: EHXDAi’LbNé '63)!» (T) <79> 85505231111760 afiofiemcib. [HORINDS] <80> flQAlCfiélb wag—msg#01301» (D) <8l> afilckoroafiLaad. (01100134003): 0ch1. [HORCOL4] <82> (wanmalnwtmme00041100315136. (I) <83) § t) Lans‘mmermtéan Lt. 6. $901414): 3: eaticfi 5 «sir (we. (VC) <86> itflAtliB'llC. —A®ifiA& L’Cfi‘fll‘l‘éo [INDEP l] <87> aozaafixtmnmraxsmmx E5511%5'ciifrb\o [VERIND 4; reverse item] <88> £336: 1,1106 3: . aaomnérmrar can?“ : 5: 20331». [DETECT 7] <89> aficiimunfisnnxuautism-z). (r) C. fifl’fifiREK‘fl’Efilfi zillérflbbf'l‘ 3 17 \o 121 1&3” 11M???» ? Eli #ifi (61365177): ? fi (51 Act»: ? 5212A EBHIA EB=PIA 7431A E$$ «tr/3111.094; El ”IA (Emma) gaumerm? MEET?» ? :oiF-fimfimn'c. 13:21:49sz i: < h‘gfifibi‘h D E Ltbl? bolt. @5107: ‘b Lbotfié‘. 80351555555195) 9 l: < motor»: ‘? :oiflmfimfimfilmmaonxxt T’srox—sjofiflfimaimw 5 10:. “101153100 3373‘ filibbfik 5 Cé’hi Lit! 122 flfilfilfifi :orya—bu.fiLwR&(xa—w)ewfifbtwobwcf.é%c sofififickboifo batfizwfliuwhfbm85m.&0.batfi:®uiua5§ibmu x01.:ogfirwofiiobatomfififiiéné:tmwfimbvit».:oa amaze:aurorbatumma0maflnwbaeamv0000.ranwtfi nnfléfiofifiufiacrmtfiwtztuaoif.6L!%$bnd.batmmfi ofi HOHH HmHH HHHH HH H HH H HH H HH H HH H HN H HO H HO H H0 H H0 H HO H H0 H ON vH OH O b O O 0 0H O N1 O1 m1 O N1 mo> HOHH HHNH HOHH HH H Hm H HH H HH H HN H H0 H Hm H Hm H HO H HO H Hm H 00 H HH OH OH O1 N v O O OH O 0H H NH O OH NU> HNHH HOHH HmHH HH H HH H HH H H0 H HH H Hm H HO H HO H HO H HO H HO H HO H O vH OH OH1 N H1 O O OH H1 O1 MH1 O1 N1 O HU> HN1H HmuH HNIH HO H Hm H H0 H HO H Hm H HN H HO H HO H H0 H 00 H HO H H0 H O1 O1 OH1 H OH O O b v m1 O «I O1 O1 O1 vH> HN1H HmuH HNIH HO H Hm H 00 H HO H Hm H HN H HO H HO H H0 H H0 H HO H H0 H 01 N NH1 O O O vH b O 0 O N h H m1 mH> HNIH HNIH HNIH HO H H0 H HO H HO H 00 H HN H HO H HO H H0 H H0 H HO H 00 H OH HH m1 m1 HH 0H O O NH O O O O O OH NH> HN1H HN1H HN1H HO H Hm H HO H HO H HHHH HN H HO H HO H H0 H H0 H HO H H0 H N O1 O1 O OH O OH N O O NH v OH O v HH> HHmH HONH HO1H HwnH HO1H H01H H01H Ho H HH H HH H Ho H HO H HH H Ho H ON Hm ON O1 O1 O VH1 O1 O m1 v O N H v OH vaH HOuH 001H HmnH HOuH HmnH Ho H HH H HH H HH H HH H HH H HH H 0 H0 vm O O O 01 m1 O H1 O1 O1 O O O NH HOnH HmnH HO1H HO1H H01H HO H HH H HH H Ho H HO H HH H HH H o 0 ON vHI OH1 O1 O1 NNI m m v H1 n O H He m9 NB HE OQ «a mo No Ho hHm OHz OHm va mHm NHI HHm 126 HnHH HNNH HOHH HO H HNHH HOHH 00 H HHHH HmnH HOIH HOIH H01H H01H HOnH HOIH vm HO OH O O «H H1 NH H1 OH1 O1 NH1 HH1 OH1 O1 vow HONH HONH HNNH HOHH HOHH HNHH HO H HOHH HmuH H01H HPIH HOnH HO1H H01H HOIH HO ON ON O OH NN O O O O1 N O1 m HH1 H mom HONH HomH HONH HHHH HOHH HmHH HOHH HOHH H01H HOIH H01H HOIH HOIH HOuH HO1H HN NN OH OH OH ON OH «H v vHI >1 OH1 >1 OH1 m1 Num HONH H0mH HONH HNHH HOHH HOHH HHHH HwHH H01H HOIH HO1H H01H H01H HOIH H01H HN Hm ON O HH ON O b O O1 O OH1 O1 N1 H1 Hum HO1H H01H HOuH HNmH HO0H HmmH HomH HO0H HH1H HmuH HmuH HN1H HNIH HmnH HOIH H1 O OH1 Ov Ow Om NN NO N OH1 O1 O1 OH1 OH1 O1 OHm H01H HOIH H01H HONH H0mH HomH HONH HOmH HHIH HmnH HNIH HN1H HN1H HN1H HN1H O O OH1 HO OO «O ON HO OH HH1 OH1 OH1 VH1 O1 N OHm HmnH H01H HmnH HnHH HONH HHNH HOHH HONH HHIH HNIH HNIH HHIH HHnH HN1H HHIH «1 H >1 OH Om ON OH ON N O OH H O N1 m VHm H01H HOIH H01H HmmH H0mH HONH HHNH HNmH HHIH HNIH HNIH HN1H HN1H HNIH HmuH v N1 OH1 ON mm ON OH NN O v NH O OH O N MHz H01H HO1H H01H HNNH HmmH HONH HHNH HHmH 0H1H HNIH HN1H HNIH HNIH HN1H HN1H H O vHI «N «O ON NN mm O O1 HH1 OH1 OH1 O1 O1 NHI H01H HO1H H01H HONH H0mH HomH HmNH HOOH HHIH HmnH HNIH HNIH HNIH HN1H HNIH v O1 OH1 MN MO HO O HO b v1 O O H1 H1 O1 HHm OH NH HE O0 «0 mo NO HO hHm OHm OHE sz mHm NHm HH: 127 NH OH O1 NO ON OO ON OO OH O O1 O1 H1 O HH ED OO OO OO OH NN OO NH OH O OH1 O1 ONI O1 OH1 O1 um O O HO1 OO OO OO OO OO NH OH1 H1 O1 O1 NH1 O1 H: NN ON ON OH1 O O O H ON HH O O1 N O HH U> O N OH1 O OH OH OH O OH O OH N O N O H> OO OO OO OH1 OH1 H OH1 OH1 O O1 N O1 O H O H OH1 H1 ONI OO OO OO OO OO O OH1 O1 NH1 O1 OH1 N D N O1 O OH1 O O O1 O1 ON OO OO OO OO OO OO HO HO H HO H HO H HONH HONH HONH HOHH HONH HN H HO H Hm H HO H HO H HO H HO H NH O O1 HO ON ON OH ON HH O1 O H O O OH O20 HO H HO H HO H HOHH HONH HHNH HOHH HONH HH H HO H HO H Hm H HN H HO H HN H O NH O1 OH NN HN OH OH HH O O H1 O OH O OED HO H HO H HO H HHNH HHOH HONH HOHH HONH HN H HO H HO H Hm H HO H Hm H Hm H O OH O1 ON ON OO OH ON O O O1 O1 O1 O O OED HO H HO H HO H HOHH HONH HONH HOHH HONH HH H Hm H HO H HN H HN H Hm H HO H O NH O1 HN ON OH OH ON O H1 O1 N1 O1 O O OED HO H HO H HO H HOHH HONH HNNH HOHH HONH HH H HO H HO H HN H HN H HO H HO H NH OH H1 NN ON ON NN HN OH N1 O1 O1 H O1 O OED HO H HO H HO H HNHH HOHH HOHH HHHH HOHH HH H HN H HN H Hm H HN H HN H HN H O OH O NH NH OH OH OH OH O1 OH1 O1 N1 O O NED HO H HO H HO H HOHH HONH HOHH HOHH HNNH HH H HO H HN H HN H HN H Hm H HN H O OH O NN OH NN ON ON O O O N1 O O HH HEQ OH NH HO OQ OD OD ND Ho OHm OHm OHm OHm OHz NHm HHm 128 m+ O+ M1 N+ N+ N1 O1 H1 O1 HH1 MI O1 ml NI OD H1 N1 m+ HI H+ VI H+ O+ o H+ N+ H1 M+ m+ O0 0 V+ M+ N+ N1 H+ O+ m+ M+ N1 v1 H1 m+ M+ OQ m1 H1 OI MI H+ VH1 M1 H1 v+ N+ o HH+ MI NH+ ND bH+ m+ N+ OH1 N+ V1 m1 M+ M+ H1 H1 H1 V1 MI HQ M+ HH+ M+ O+ 0H+ m+ °H+ NH+ VH+ MH+ N+ M+ OH+ V+ OH: VH1 01 m+ O+ O1 N1 HH+ n1 G OI m1 N1 v+ Q OHm N1 MHI NH+ VH+ m1 O+ O+ OI O+ mHI m1 H+ M+ O+ OHm b1 VH1 N+ O+ NH1 m+ °H+ OH1 MI NH1 m1 N1 H1 0 OH: m1 NH1 QH+ MH+ ml H+ N+ O1 O+ MH1 MI M+ H+ HH+ OH: MHI O1 o m+ OI H+ m+ m1 N1 m1 HH1 VI H+ Q NHm NI v+ V+ v+ M1 O1 V+ b1 m+ v+ at O1 VH+ o HHm OHm OHm OHm OHm NHm HHm OU> OU> NU> HU> OH> OH> NH> HH> 129 00.00.0000. :-H :-H 0- 01 0+ 0+ 0- 0 00H. 0a.. 0 0 00> .... 00.00. :-H 0: 00+ 0+ 0+ 0- 00+ 0- 0000 0 0- 00> HO0H 001H HNIH 0+ 0- 0 01 0+ 0- 00 0 0 00> _ :H :-H 0- 0+ 0- 00. 00+ 00- ,1, 00 00- 0- 00> 000Hm000H 0- 0+ 01 0+ 0- 0- 0- 0+ 0+ 01 m0,.0mm. 00> 0000 0000 01 0+ 0- 0+ 0- 0- 0+ 0- 0+ 0- 00 .00._ 00> 0000_ 0- 0. 0+ 0+ 01 0- 0+ 0+ 0+ 00- .00 ,00 00> 0 0 00+ 00+ 0- 0+ 0+ 0- 0+ 0- .01 0. .0+ 00 00> 0+ 0+ 0- 0+ 0+ 0+ 0- 00+ 0- 0- 0- 0- 00+ 0+ 00 00+ 00+ 0+ 0+ 0+ 01 0- 0+ 0- 0- 0- 0+ 00+ 0- 00 0- 00- 0- 01 00- 00- 0+ 0- 0- 0+ 00- 00- 0- 0- 00 000 000 000 000 00: 00: 00> 00> 00> 00> 00> 00> 00> 00> 130 000H 00 H 00 H 00 H 00 H 00 H 00 H 00 H 000H 000H 001H 001H H0-H H0-H 00 00 0- 0 0 0 0- 00 0 0 0- 0- 0 0- 0o: 000H 000H 00 H 000H 000H H00H 00 H 000H H00H 000H 001H 00-H 001H H01H 00 00 0 0 00 0 0 00 00 00 00- 00- 0 0 000 000H 000H 00 H 000H H00H 000H 00 H 000H 000H 000H 001H 001H 001H H01H 00 00 0 0 00 0 0 00 00 00 00- 00- 0- 0 000 w 000H 000H 000H 000H,000Hw 00 H 00 H 00 H 00 H 000H 000H 000H 000H 00 00 00 00 00 00 . 0- 00 0 0- 00 00 00 00 000 000H HNNH 000H 000H H0 H 00 H 00 H 00 H 000H 000H H00H 000H 0+ 00 00 00 00 00 0- 0 0 0 00 00 0 00 000 000H 000H HO0H. 00 H 00 H 00 H H0 H 00 H 00 H 00 H H0 H 0- 0- 00 00 00 00 . 0 0 0 0- 0 0 00 00 000 000H H00H 00 H 00 H 00 H 00 H 000H 000H 000H 00 H 0 00- 0+ 00 00 00 0 0- 0- 00- 00 00 00 00 00: H00H 00 H 00 H 00 H 00 H 000H 000H 00 H H0 H 0 00+ 0- w- 00 00 0- 00 0 00 0 0 0 0 000 00 H 00 H 00 H 00 H 000H 000H 000H 000H 01+, 0- 0+ 0+ 01 00 0 0- 0- 00- 0 0 0 00 000 000 000 000 000 000 000 00> 00> 00> 00> 00> 00> 00> 00> 131 OO OO HH NN HO OH O NH O O N OH O OH 20 ON OO O NH ON O O OO ON NN OH1 HN1 N N om OO OO OO OO OO OO O OH H1 O1 HN OH OH ON HO O O H OH1 OH O1 ON OO OO OO O1 O O O1 U> OH OH OH ON O NH O O O NH1 OO OO NO HO H> O1 O1 O1 O1 O1 OH1 O HO ON NN OH1 O1 OH O1 B OO OO OO OO OO OO O1 O O O O OH O OH O OH1 OH1 O NH OH1 N ON O ON O1 O1 O OH OH HO HONH HNNH HOHH HONH HOHH HHNH HO H HO H HO H HO H HO H HO H HO H HO H OO ON NH NN ON OH H O O O H1 O O OH O20 HONH HOHH HOHH HOHH HOHH HOHH HO H HO H HO H HO H HO H HO H HO H HO H ON OH O NH OH OH O O O O O O O O OED HONH HONH HOHH HONH HONH HNNH HO H HO H HOHH HO H HO H HO H HO H HO H OO HO O OH ON OH OH O O H H1 NH O OH OED HONH HHNH HOHH HOHH HOHH HONH HO H HO H HO H HO H HO H HO H HO H HO H ON ON HH OH ON NH O OH O O O OH N O OED HONH HONH HOHH HOHH HOHH HONH HO H HO H HO H HO H HO H HO H HO H HO H ON NO O OH ON OH O OH O O H1 O H1 OH OED HOHH HOHH HO H HNHH HNHH HOHH HN H HO H HO H HO H HO H HO H HO H HO H OH ON O O NN OH N1 O O OH N O O O NED HNNH HOHH HHHH HOHH HOHH HOHH HO H HO H HO H HO H HO H HO H HO H HO H HN OH O NH OH OH O O O O N O H O HEQ OHO OHm OHm OHm NHm HHS OU> OU> NU> HU> OH> OH> NH> HH> 132 HH+ o m+ N+ M+ O O+ HI MI m+ M1 OD bl V1 O1 M1 m1 OI OI VI M+ N+ OI OD N+ o m+ OI V+ V+ V+ V+ 0H+ O+ m+ OD 01 N1 N1 H1 m+ m+ m+ m1 MI m+ N1 No H+ O1 V1 M1 ml 0 H+ H+ O1 H1 OH1 HO m+ OH+ w+ m+ MH+ MH+ O+ N+ O+ m+ O+ qu O1 o H1 V1 MI m1 o m1 N+ OI M+ OHm MI m+ OI HH1 ml NH1 H+ O m+ o m+ qu N1 M1 O1 V1 m1 m1 V1 Ni V1 HH1 OI qu MI N+ OI OI HI VI NI OI m+ HI H+ qu m+ O+ O+ m+ O1 H+ M+ m1 V1 MI 0+ N4: O+ m+ m+ M+ O V+ m+ VI O+ M+ b+ Ham OED OED OED O20 O20 N20 HZQ Oom Oom Nom Hum 133 NI V+ O+ O H+ VI H+ OHI O OI OI OU> H1 H1 H+ V+ OH+ V+ H+ O+ VH+ OH+ O+ OU> MI OI O1 O1 N1 NI N+ NH1 O1 O1 OI NU> OI OI OI HI N+ O+ O N1 H1 O+ HI HU> O1 O O1 O1 O1 N1 OI O OI OI VI OH> N1 O1 O+ O+ NI V+ N+ VI OI OI OH1 OH> H+ H+ O V1 O1 N+ VI V+ VH+ V+ O+ NH> O+ V+ O+ H1 V+ O+ H+ HI OH+ O+ O+ HH> O+ HI NI O+ O+ O+ N+ OH+ HH+ NI VI OH O O+ V+ V+ O+ HH+ O+ O+ HI OI O1 NH OH1 NH1 NH1 HHI OI V1 H1 VI NI OI OI HE OED OED OED OED OED NED HED Oom Oom Nom Hum 134 N1 N1 HI O+ OH+ O+ H+ Oom V+ VI m+ H1 OH+ H+ V+ O0: VI OI O1 N1 O+ N+ OI No: OHI OH1 O1 O1 O+ NI N1 Hum O+ N+ O+ HI O N1 H1 O+ m+ N+ OI OHE N+ OI O+ O+ NH+ OH+ N+ OH+ O+ O+ V+ OHm O1 O1 O1 OI O1 o O1 O1 N+ O1 O1 OH: N+ OI N1 N1 H1 O1 O1 H1 N1 O1 O1 OHI V+ OI O+ OH+ OH+ QH+ O H1 OH+ V+ OI NHE VI NI O1 O1 O1 O1 O1 V1 N1 OH1 O1 HH: OED OED OED OED OED NED HED O0: mom NU: HQ: 135 00 00 00 00 00 00 00 00 00 00 00 mo 00 00 00 00 00 00 00 00 00 00 00 um 00 00 00 00 00 00 00 00 00 00 00 00 0 0 00 00 00 00 00 00 00 00 00 o> 00 0 00 0 0 00 0 0- 0- 00- 00- 0> 00 0 0 00 00 00 00 00 00 00 00 0 N0 00 00 00 0m 00 00 00 00 00 00 0 0 00 0 0 0 0 0 00- 0- 00- 0- 00 000H 000H 000H 000H 000H 000H. 000H 000H 000H 000H 00 00 00 00 00 00 00 00 00 00 00 020 000H 000H 000H 000H 000H 000H 000H HO0H HO0H 0+ 00 00 00 00 00 00 00 00 0 0 020 000H 000H 000H HO0H 000H 000H 000H HO0H 0+ 0+ 00 00 00 00 00 00 00 00 00 020 000H 000H 000H 000H H00H 000H HO0H 0+ 0- 01 00 00 00 00 00 00 00 00 020 000H 000H 000H 000H 000H 000H 01 0- 01 0+ 00 00 00 w 00 00 00 00 020 000H 00 H 000H 000H 000H 0- 01 01 0+ 0+ 00 00 00 00 00 00 020 000H 000H 000H 000H 0 0+ 0 w 0. 0+ 01 00 00 00 0 00 020 020 020 020 020 020 020 020 000 00: No: 000 136 OOH OO OO OH OH OH OO O ED OO OOH OO OO OH1 OO OO OH1 om OO OO OOH O OO OH1 OOH O1 HO OH OO O OOH O1 OO O OH U> OH OH1 OO O1 OOH O1 OH OH H> OH OO OH1 OO O1 OOH HN1 N O OO OO OOH O OH HN1 OOH OH1 D O OH1 O1 OH OH N OHI OOH am ED om Hm U> H> B D 43 000B Ego—Ema 08 H0003 000200—0030 080$ 137 01 0- 0- HI NH+ V1 «0+ 00+ NI 0 VI 01 m+ . O+ 0+ O1 0- 0- 0+ 0- 0+ 00+ 0+ 0+ 00- 0+ 01 0+ 01 00- 0 00- 0+ 0+ 0+ 0- 0- 0+ 0+ 0+ 00- 0- 0+ 0+ 00+ 0+ 0+ 0+ 0+ 0+ 0+ 0+ 0+ 0+ 0+ 0 00- 0- 0+ 0+ 0+ 0+ 0- 0- 0- 0+ 0 OH N9 09 mo 00 mg mg Hm H 00 H H1 O 000H 000H ON OH 00 H 00 H 00 N- 00 H H0 H 0 0 D0 H 000H OH HI .. H0H 00H 00H 00H 0 0 00- 0- 00 000H 00 H 000H 000H 00 00 0 0 00 SH SH 00H 00H OH O O OH OD 00 H 00 H 00 H 00 H 00 H 00 H 00 H 00 H V VI VI HQ M000. ”QM”... 01H: 0.... .. OHI .0 ., . 000 0 000 000 000 00: 00000 0.: 00. .00 00.02 0000008 138 HN H HH H HN H Hm H Hm H Hm H OI MI 0 HI 0 HH v NH mH OH MH VH H OH O 00> Hm H Hm H HO H HH H H0 H Hm H HN H HN H HO H Hm H Hm H Hm H Hm H H0 H Hm H NN m v ml v 0 NI O OH OH NH O m O O mo> HNHH HmHH HVHH Hm H HOHH HO H Hm H H» H HmHH HmHH HHNH HVHH HmHH HOHH HmHH OH OH NH ml v 0 O OH mH m 0H O MH O OH NU> Hm H HmHH HOHH HN H H0 H H0 H H0 H Hm H HOHH HvHH HmHH HOHH Hm H HHHH HHHH O O OH OHI O m O H 0 NH O vI OI w mH HU> Ho H Ho H Ho H Hm H HOHH HOHH Hm H HHHH Hm H Hm H Ho H H0 H H0 H Hm H Hm H HHI MI OHI MI MH NI 0 v m O H v v HHI NI vH> Ho H Ho H Ho H Hm H HnHH HHHH Hm H HNHH Hm H HO H H0 H Hm H H0 H Hm H Hm H N O 0| HI MH HH HH O m N O OH MH H MI mH> HO H Ho H HO H Hv H HvHH Hm H H0 H HOHH H0 H Hm H Hm H H0 H Hm H H0 H H0 H mH OH N PI OH HN HI mH 0H NH O 0 m HI OH NH> Ho H HO H Ho H Hm H HOHH HOHH Hm H HHHH H0 H HO H HO H H0 H H0 H Hm H Hm H H ml mI 0 ON ON Hm HH O m OH O «H v m HH> HmNH HNNH HmIH HNHIHHOIH HOIH HmIH H0 H Hm H Hm H Hm H Hm H H0 H H0 H ON ON HN OHI mHI v OHI vI mI OI m v O O a me HNmH HmIH HOHIHHHHIHHmIHHNHIH Hm H H0 H Hm H Hm H Hm H HO H Hm H o mw Nm 0H OI HI OHI mI O O OH ml O OH 0 NB HvIH HmHIHHmIH HOIH HmIH H0 H Hm H HO H H0 H H0 H H0 H H0 H HI 0 0N mHI OHI OHI MI bHI m m h H w v m HE m9 NH HE mo 00 mo NO HQ 04: qu mHm «Hm mHm NHm HHm 139 HO H HmHH HO H Hm H HOHH HO H HO H HO H HN H Hm H Hm H HN H HN H HN H HN H ON NN O MI O OH OI O HI OI OH HI O M O vow HHHH HOHH HNHH H0 H HNHH HO H HO H HO H Hm H H0 H H0 H Hm H HN H Hm H Hm H ON OH MH OI MN ON O OH NI H HH MI O H O Mom HNHH HOHH HOHH H0 H HVHH HO H HO H HH H Hm H H0 H H0 H Hm H Hm H Hm H Hm H O M HH O OH HN MH HH O OI OH NI NI O M N03 HOHH HONH HHNH HO H HNNH HVHH HHHH HOHH HO H HO H HO H HO H Hv H HO H HO H NH NN HN vI O OH M O 0 N OH «I OI OH O Hum HOIH HOIH HOIH HvHH HOOH HNmH HVNH HOmH HO H HO H HO H HO H H0 H HO H HO H OI OI OHI OM Ow OM OH OO OH OI OH HH M OI HI OHm HOIH HOIH HOIH HOHH HOVH HONH HNNH HHmH HO H HO H HO H H0 H H0 H HO H HO H O HI vHI NH Hv OM NH OM OH O O O N N O OHm HvIH HOIH HvIH HO H HHmH HOHH HOHH HHNH Hm H H0 H H0 H Hm H Hm H Hm H Hm H OI O H M MM ON OH OH MH O O M OH N O va HOIH HOIH HOIH HNHH HmvH HONH HHNH HomH H0 H HO H HO H H0 H H0 H HO H H0 H OI OHI OHI OH OM ON NN OH MH M HH O OH M HI MHm HVIH HOIH HOIH HOHH HOmH HONH HOHH HONH H0 H HO H HO H Hm H Hm H H0 H H0 H OH OI HHI MI Om OH OH OH O O O O NI OI OI NHm HOIH HOIH HOIH HvHH HOvH HomH HmNH HmmH HO H HO H HO H HO H H0 H HO H HO H HI OHI HNI ON HM MM OH NM O OI OH OH O H OI HHm MB NB HE OD 00 Mo NO HQ qu OHm qu vqm qu NHm Ham 140 NN O NI OH OM OM OH ON OH HH NN HH OH ON NH 2O HM HM ON MI ON OM O OH O OI ON OI N MH O om HI NHI ONI OM HO OO OM OO NN HI OH OH HH O O HO OH OH OH MHI O MH O OH ON ON ON OH O OH MN U> N O OI NI ON ON OH OH MH O HH O MH MI O H> OO OO OO MNI MNI OI OHI OHI O O OH O OH OH NH O MHI OHI ONI MN HO HO OM OO ON OH ON HN OH N O O O OH O N OM ON OH O HO OO OO OM OM OO NO HO HO H HO H HO H HOHH HOMH HHNH HOHH HMNH HNHH HOHH HOHH HHHH HOHH HMHH HNHH OH O HI OH OH ON O OH O MI O O OH OH O OZO HO H HO H HO H HO H HONH HOHH HOHH HONH HOHH HmHH HOHH HO H HO H HHHH HOHH N O OI MH NN OH NH OH HH OH ON M OH OH O OZO HO H HO H HO H HO H HomH HOHH HOHH HHNH HOHH HOHH HOHH HOHH HO H HHHH HHHH OH MI OI OH NN NM NH NN O O O M M OH HH OEO HO H HO H HO H HO H HONH HOHH HmHH HOHH HO H HNHH HOHH HO H HO H HOHH HO H OH O O O ON NH NH MH H O O O OH OH MI OZO HO H HO H HO H HO H HONH HOHH HNHH HOHH HO H HHHH HNHH HO H HO H HO H HO H OH OI OI O OH ON MH OH OH NH OH HH HN OH O MzO HO H HO H HO H HO H HNNH HOHH HHHH HOHH HO H HOHH HHHH HO H HO H HO H HO H OH MH OI O HN HM O O NH O O MH O NN OH NEO Hm H HO H Hm H HO H HOHH HoHH HO H HHHH HO H HO H HO H HO H HO H HO H HO H OH NH NH MH OH OH ON OH OH NH MN M OH O HH H2O MB NB HO OO OO MO NO HO OHm OHm OHm OHO MHm NHI HHm 141 OH+ HI OI O+ OHI HH+ NI OHI OI NHI OI OI HHI H+ OO VI VI N+ VI O+ OHI O+ Q OI OI OI VI N+ NH+ OO O+ m+ OH+ N+ OI O+ QH+ V+ NI HI NHI o NH+ OH+ MO OI OHI O+ H+ N+ OI O+ VI H+ V+ VI O+ OI ON+ NO QO+ HI NI HHI OI HI OH+ V+ HH+ VI OI VI O+ O HO O+ O+ OH+ O+ O+ H+ NH+ V+ 0 MI NI V+ OH+ N+ OHm OHI OI V+ OI OI HHI HH+ HH+ HHI NI OI VI O+ OI OHm O+ H+ O+ O+ o O+ O+ O+ VI OI OI HI 0 m+ OHO O+ N+ o O+ O+ OH+ HH+ N+ VHI VHI O O+ o N+ OHm HI NI O+ OH+ OI o NI V+ o OHI o m+ O OH+ MHZ OHI OI HI NI OI VI O+ N+ OI OI OHI VI OI HI NHm OI O+ N+ OI ml HHI V+ O H+ V+ NHI OI VH+ Q Ham OHE OHm OHm MHm NHm HHm OU> MU> NU> HU> OH> MH> NH> HH> 142 S H S H S H S H 0+ 0+ 0+ 00+ 0+ 0+ 0- 00 0 0 00> ., , 8 H 8 H S H 8 H 0+ 0+ 0+ 0- 0+ 0- . a . w m- 0- m w- 00> :-H :-H :-H :-H 0+ 0+ 0- 00- 0- 0- g ._H. m 00 m o 00> 0.. _ HHIH :-H :H :-H 0- 0- 0- 00- 0- 0- 0g . 0+ 0+ 00 0-. 0- 00- 0- 00> .0; 000H ”00H HNOH 0- 00- 0- 00+ 00- 0- 0- 0- 0+ 0- m0 om. 00 00 00> «gr __ HO0H HO0H 0- 0- 0: 00+ 0: 0- 00+ 0- 00+ 0- 0+“ .00 00 00 00> H,. , HO0H 0+ 0- 00+ 0 0n 0- 0+ 0+ 0+ 00- 0- 0 mm 00 00> 00+ 0+ 00+ 00+ 0- 0+ 0+ 0- 0+ 0- o 0: 0+ 00 00> 0 00+ 0- 0 00+ 0+ 0- 00+ 0+ 0- 00- 0+ 00+ 0+ 00 0+ 0+ 00+ 0- 0 0- 0- 0+ 0+ 0- 0- 0+ 00+ 0- 00 0| 0- 0+ 0- 0- 00- 0+ 0- 0- 0+ 00- 0- 0+ 0- 00 000 000 000 000 000 000 00> 00> 00> 00> 00> 00> 00> 00> 143 rwiilll I .l. u . HO H HO H HO H HO H HO H HNIH HNIH 0 om 0 0 0- 0 0- 00 00 N 0- 00- 0 0- 0mm H00H HOHH H0 H HOHH Hm H 000H Hm H 00 H H00H Hm H HMIH HmIH HmIH HmIH 00 mm om w 00 00 o 00 0 0- 00- 0- 00 0 000 HmHH HNHH Hm H OHHH 00 H _NHH Hm H O0 H HmHH 00 H HmIH HmIH OmIH OmIH 00 00 0 H 0 00 0 00 00 00 0- 00- 0- o 000 OQNH _00H HNHH HOHH HO0H 000H mm H Hm H HONH HO0H H0IH _0IH HmuH O0IH m 00 o 0 0 00 m 00 00 0 0- 00- 0 0 000 HmNH HO0H HONH HNNH HomH. H0 H Hm H H0 H 00 H O00H HO0H HNHH 000H mm 00 00 mm 00 mm _ 0 m 00 0- 00 00 00 mm 000 H00H OONH HomH HONH OH H 00 H H0 H H0 H HmHH OmHH HHHH HNHH 0+ 00 00 mm 00 mm m 0 00 NI 0 0 m 00 000 HOHH H00H HmHH. H0 H Hm H H0 H Hm H H0 H H0 H Om H Hm H 0+ 0+ 00 00 00 00 m m m o 0 o 00 00 000 000H HO0H OH H Hm H O0 H H0 H HNHH OMHH 000H ONHH 0- 0- 0+ mm 00 00 00 m- 0. 0- mm mm 00 00 00: HHNH H0 H H0 H H0 H H0 H HOHH H00H 00 H OOHH 0- 0- 0- 0+ 00 00 m 0 0- 0 0- m o m 000 O0 H Hm H H0 H .0 H OmHH O00H _00H HMHH 0+ 01 0I 0+ . 0+ 00.. N 0- N. 0- 0 00 0 00 000 000 000 00: 000 000 000 00> 00> 00> 00> 00> 00> 00> 00> 144 . “II rill—0.0“ OO MN NN OM O OM OH OH O NI O O ON MM 2O ON OO NH O OH HN N ON HN O OHI HNI OH M um OO OO OM OO OO MO OH O O NI MH ON NN OO HO OH OH O NI O NI OH OM NO OO OI O HI OI U> ON NH OH OM N OH O OI O NHI OO OO OO OO H> OHI OI H OHI OI ONI MI NN OM OH OHI O HN OI O OO MO HO HO OM OO OH M HH H O OH OH OM O O NH OH OH O O ON MN ON NH NI NH OH OH HO HONH HHNH HOHH HONH HOHH HNNH HN H Hm H HO H HO H HNHH HOHH HHHH HNHH HM OH OH ON O HN O O H OI H O OH ON OEO HONH HOHH HNHH HOHH HOHH HOHH HH H HN H HO H HO H HOHH HHHH HO H HOHH ON OH OH OH OI NN OH O H N H O O OH OZO HHNH HOHH HmHH HOHH HOHH HONH HH H Hm H HO H HO H HHHH HHHH HO H HHHH OM HN O MN HI HN OH O NI HI HI OH OH ON OZO HOHH HOHH HHHH HOHH HOHH HOHH HH H HN H HO H HO H HOHH HoHH HO H HO H NH OH OH OH NH NN O O OI NI O OH OH OH OEO HOHH HOHH HOHH HOHH HNHH HOHH HH H HN H HO H HO H HO H HO H HO H HO H OH ON OH ON O ON O OH O OI OI HI O ON OZO HOHH HOHH HOHH HOHH HHHH HOHH HH H HN H HO H Hm H HO H HO H HO H HO H HH OH HN HN O ON O O MI MI O O OH OH NZO HNHH HOHH HO H HOHH HO H HHHH HH H HH H Hm H HN H HO H HO H HO H HO H OH O O NH HI OH O NI O O O O NH OH H2O OH: OHO OHO MHm NHm HHm OU> MU> NU> HU> OH> MH> NH> HH> 145 O+ O+ OI OI HI O O+ OI HHI V+ OHI OO OHI OI OI OI OI HI 9 OI HH+ O VHI OO O+ H+ OH+ OI OH+ OH+ O+ OH+ OH+ NH+ N+ OO HHI NI OI HI H+ OI OH+ HHI H+ O+ OI NO OI VI H+ OI OI OI O+ HI O+ OH+ OI HO OI H+ OI OI O+ V+ V+ OI OI V+ HI OHE OHI OI OI OI H+ OI V+ OI OI NHI VI OHI OHI O+ OI OI O+ VI OH+ HH+ O+ NH+ O+ OH: OI OI OI HI O+ O+ NI OI OI OI OI OHm O O+ OI N+ VH+ H+ O+ V+ O+ OI HHI OHm O+ O+ O+ O+ O+ VH+ N+ H+ NI O+ OH+ Nam OI OI O NHI NI N+ O+ O+ O+ O OI HHm OZO OZO OZO OEO MZO NZO H2O Oum Mum Nom Hum 146 O+ OH+ NH+ V+ O+ O+ V+ OI OI V+ NI OU> N+ V+ O+ O+ VH+ NI OI O+ O+ HH+ OH+ MU> OI OI OI OI OI OI O+ O+ OI V+ OHI No> OHI NI OI OI OHI OI O+ VI OHI O+ OI HU> HHI OI NHI OI VHI OI OI OI OI VI NI OH> OI HHI V+ O OHI OI N+ HHI VI OI OI MH> O+ OI O+ N+ O HH+ O+ O+ ON+ N+ O+ NH> VH+ O+ VH+ OH+ HH+ OH+ O+ OI OH+ O+ O+ HH> VH+ OI O+ OH+ O+ VH+ HH+ NH+ OH+ OI OI MB OI O OHI OI VHI O+ O+ O+ H+ OHI OI NO OI OI HHI OI NHI OI OH+ VI H+ NI O HO OZO OZO OZO OEO MZO NZO H2O Oom Mo: Nom Hum 147 0- 0+ 0- 0+ 00+ 0+ 0+ 00: 0 0: 0+ 0+ 00+ 0+ 00+ 000 00: 0- 0: 0- 0- 0+ 0+ 000 00- 0- 0- 0- 0- 0- 0+ 00: 0+ 0- 00+ 0- 0- 0- 0+ 0 0+ 0+ 00- 000 0- 0- 0: 0- 0+ 0 0- 00+ 00+ 0+ 00+ 000 0 0- 0. 0+ 0+ 00+ 0 0 00+ 0- 00' 000 0+ 0+ 0+ 0- 00+ 0+ 0+ 0- 0- 00- 00- 00: 00: 00- 00- 0- 0: 0- 0- 00- 00+ 0- 00- 00: 0- 0+ 0+ 0+ 00+ 0+ 0+ 0- 0+ 0+ 0- 000 020 020 020 020 020 020 0:0 00: 000 000 000 148 00 00 00 00 00 00 mm 00 00 00 00 mo 00 0m 0m 00 mm mm om 00 00 om 00 0: mm om mm mm O0 00 00 00 00 ON 00 00 m 00 00 0 0 0- 0 OH 0 00 00 o> ON 0 00 00 0 00 00 0- N 0- 0- 0> 00 0 0 00 0- 00 mm om 0m 00 mm 0 00 mm 00 0m 00 00 mm O 0m 0m 00 a 00 00 00 00 00 00 00 0 0 0 0 00 HomH HmmH‘H00H‘H00H HmmH HmmH HNHH 000H O00H H0NH 00 00 00 00 mm 00 om 0 00 0 0 020 HO0H HQOH H0mH HmmH HmmH HOHH _00H HO0H OmmH 0- 00 00 00 00 mm 00 00 00 00 00 020 HN0H HmmH OmmH HONH‘ H00H OmHH 000H H0NH 0+ 0+ 00 mm 00 mm 00 0 mm 0 00 020 H0mH HHmH HmmH H0 H _NHH HmHH HHNH 0+ 0- 0- 00 00 mm 00 00 00 00 00 020 HmmH HHNH H0 H HHHH HNHH HO0H 0- 0- 0- 0+ mm om 00 ON 00 00 00 020 HmHH H0 H OO0H HHHH HO0H 0- 0 0 0+ 0+ 00 00 00 00 00 00 020 H0 H H0 H H0 H Hm0H 0+ 0+ 0+ 0. 0+ 0- 00 00 00 00 00 020 020 020 020 020 020 020 020 000 000 000 000 149 OOH OO OO HH ON OH OO OM 2O OO OOH OO HM OHI OO OM OH oz OO OO OOH OH OM ONI HHH NN Hm HH HO OH OOH NI OM OH OO U> ON OHI OO NI OOH H HO OH H> OH OO ONI OM H OOH OMI ON O OO OM HHH OH HO OOI OOH NM O OM OH NN OO OH ON NM OOH Hm 2O 0m Hm U> H> O O HO 30,—. amzflgm 08 @003 mcouflobou 88g 150 .....ch HNIH HHIH 0.00 0.5., .000. 0.00 .00: 151 HOHH HOHH HOHH HOIH HOIH HOIH HOIH HOIH HN H HO H HN H HN H HO H HO H HO H OH OH ON H O OH OI OHI O H OI OI OI O NI Oo> HOHH HOHH HOHH HOIH HoHIHHOIH HOIHHOHIH Hm H HO H Hm H HN H Hm H HO H HO H OM HH MN OHI O OI OHI OHI OH OI O OI OI O OI MU> HOHH HHNH HNNH HOIH HNHIHHOHIHHOIHHNHIH HN H HO H HO H Hm H HO H HO H HO H O OH NN OI NI M N OHI HN OH HH O HH N N NU> HNHH HOHH HOHH HOIH HOIH HOIH HOIH HOIH HN H HO H HN H HN H HN H Hm H Hm H H OH NN OMI OHI HNI OI HNI MN MI OI OI O O OH HU> HOIH HOIH HOIH HN H Hm H HN H HH H Hm H Hm H HO H Hm H Hm H HO H HO H HO H OI NHI NMI MI N M HI O M O O OI NHI O O OH> HOIH HOIH HOIH Hm H HO H Hm H HH H Hm H HN H HO H Hm H Hm H Hm H HO H HO H OHI NI ONI H N H OH O O O O OI HI N NI MH> HOIH HOIH HOIH HN H HO H HN H HH H Hm H HO H HO H Hm H HO H HO H HO H HO H HN O OHI M O O O OI HH O O HI O OH OH NH> HOIH HOIH HOIH HN H Hm H HN H HH H HN H HN H HO H Hm H Hm H Hm H HO H HO H O HI OI N O NI N OI O O O MI OH O N HH> HOOH HNOH HOIH HOIH HOIH HOIH HOIH HHIH HNIH HHIH HHIH HHIH HNIH HNIH ON OM MOI OHI OI M ONI NH OH OI O O N H H OH HHOH HOIH HNHIHHOIH HOIHHNHIH HHIH HmIH HNIH HHIH HNIH HNIH HNIH O OM HO . H H N NI NHI MI OI MHI H NI OI OI NO HOHIHHNHIHHOHIHHOIHHNHIH HHIH HmIH HNIH HNIH HNIH HNIH HNIH H+ O OO MHI OHI OI OI ONI M MI OI OHI O OI MI HO MB NB HO OO OO MO NO HO OHz OHO OHm OHm OHm NHO HHm 152 HOHH HmNH HONH HHIH HHIH HHIH HHIH HHIH Ho H HHIH HHIH Ho H HHIH HHIH HHIH OO OM OM HHI OI NI NNI OI OI OI N O OHI OHI OHI Oom HONH HmmH HOOH HHIH HNIH HHIH HHIH HNIH HHIH HHIH HHIH HHIH HHIH HHIH HHIH HM ON OM OI N O OHI OHI O O OH O OH OI O Mom HONH HOOH HOmH HHIH HNIH HNIH HHIH HNIH HHIH HHIH HHIH HHIH HHIH HHIH HHIH ON OM OO O OH O O OI OI MI OI OHI O OHI H No: HmmH HNOH HOOH HNIH HNIH HNIH HHIH HNIH HHIH HNIH HHIH HHIH HHIH HHIH HHIH ON MM HO O O OH O OI MI HI M OI OH OI O Hum HOIH HOIH HOIH HONH HOOH HONH HOHH HOOH HH H HN H HH H HH H HH H HN H HN H OI O HHI MM HO OM OH OO MHI OHI O OI OI OI O OHm HOIH HOIH HOIH HOHH HONH HOHH HO H HONH HH H HH H HH H HH H HH H HH H HH H OI OI MNI ON OH OH O HO O O H NI OI OH OH OHm HOIH HOIH HOIH HONH HOOH HONH HOHH HOOH HH H HN H HH H HH H HH H HN H HN H H OI ONI ON OM NN O OO NHI M O OI O HHI HI OHm HOIH HOIH HOIH HONH HOOH HONH HOHH HOOH HH H HN H HH H HH H HH H Hm H HN H OH O NHI OM ON OM OH ON OI O OH O OH O O MHO HOIH HOIH HOIH HNHH HOHH HNHH HO H HOHH Ho H HH H HH H HH H HH H HH H HH H OHI OHI OHI OH OH OH O ON O O O OI O OH O NHm HOIH HOIH HOIH HONH HOOH HOOH HOHH HOmH HH H HN H HH H HH H Hm H HN H HN H O HI OHI NN OM ON O HO O H O NI OI HI OHI HHm MB NB HO OO OO MO NO HO OHm OHm OHm OHm MHm NHm HHm 153 NI ON H NN OH ON OH NN NH OH O HH O OH ON 2O MO OO MO HI O HH MHI MHI OI NI O OI O OHI MI om NI OI OMI HO OO OO OH OO OI O OH OI M O O HO HN ON OM NNI OI OI OHI ONI OM H NI OI O O N U> O OI ONI H O O O NI O O HH OI O HH O H> OO NO OO OHI MHI O ONI ONI O OI OI MI N OI OI H NNI OI ONI NO OO OO HM OO OHI OI O OHI HI OI HH O O OHI OI NHI MI O OHI M MM NO MO OM OO OO OO HO HO H HO H HO H HOHH HOHH HOHH HO H HOHH HO H HOHH HO H HO H HO H HNHH HHHH O O MI OM HN OH N NM HH NH NH OH H NH NN OEO HO H HO H HO H HOHH HOHH HOHH HO H HOHH HO H HNHH HO H HO H HO H HHHH HOHH O OH OI NH OH OH O NH O O M O H HH NH OZO HO H HO H HO H HOHH HONH HOHH HO H HONH HO H HOHH HO H HO H HoHH HOHH HOHH OI NN O MN OH ON HH OH O OH O O O OH OH OZO HO H HO H HO H HOHH HOHH HOHH HO H HOHH HO H HmHH HO H HO H HO H HHHH HHHH OI OH OI ON MH HH O ON O OH N NH MI OH ON OEO HO H HO H HO H HOHH HOHH HOHH HO H HOHH HO H HmHH HO H HO H HO H HHHH HHHH O ON OH OH O OH OH O O OH O O O O NH MZO HN H Hm H HN H HO H HO H HO H Hm H HO H HN H HO H Hm H Hm H HO H HO H HO H OI O OH OI MHI OI NH O OH OH HI HI O O OH NZO HO H HO H HO H HmHH HOHH HmHH HO H HOHH HO H HOHH HO H HO H HO H HHHH HHHH OI O NI ON OH OH OH OH O O MI O HI OH OH H2O MB NB HO OO OO MO NO HO OHO OHm OHO OHm MHO NHO HHm 154 O+ V+ H+ O+ O+ OI O+ NI O+ ONI OI HI H+ O OO O+ HI O+ OHI V+ NI NH+ OH+ OH+ HI HI HI O+ NI OO N+ H+ OI H+ O+ HI OH+ N+ OH+ VHI H+ HI O+ VI MO OI NI HHI OI OI OHI HI OI O+ VI NI OH+ V+ H+ NO OH+ HH+ V+ OI OH+ OI OHI OI N+ NHI N+ N+ OI OHI HO VHI HI OHI OI O+ O+ O+ VH+ OH+ HN+ O NI O+ V+ OHm NHI O+ H+ O+ O+ HI OI OHI O+ OI OI O+ OI N+ OHS V+ O O+ OH+ V+ O+ OI OI O+ OHI O+ O+ O+ V+ OHm OI OI OI HI OI OI OHI OI H+ OI HHI NHI VI OI OHm OI OI V+ OH+ O+ OI OI OI O+ O+ OHI VI O+ OH+ MHm HHI NH+ OHI H+ VH+ OI H+ H+ NI O+ O+ OI O+ H+ NHS NI OH+ OI V+ O+ NHI OI OI NI O+ OI OI NH+ NI HHO OHm OHI OHm MHm NHI HHm OU> MU> NU> HU> OH> MH> NH> HH> 155 HmmH HOmH HONH HNIH HNIH HNIH HNIH 0 0+ 00+ 0: 00+ 0+ 0m- .0m 00. 00 m- 0- 0. MI 00> . . a HO0H HomH HmIH HNIH ONIH HNIH 00+ 00- 0+ 0+ 0+ 0+ 0+, 00 mm mm o 0- 00 m 00> 0.. . . H0mH HmIH HmIH HMIH HNIH 0+ 0- 0+ 0+ 00+ 0+ 0+ -0} 00 00 o 0- 0 0 00> . H , HNIH HNIH ONIH HNIH 00- 0: 0- 00- 0+ 00- 0I 0: 0+ 00 00- 0-, 0- 00- 00> .H HO0H ”00H ”00H 0- 00+ 0- 0- 0+ 0- 0- 0+ 0+ 00: on 00 00 00 00> . . HO0H ”mmH 0- 00+ 0. 0: 00+ 00- 0- 0- 0- 0- 00+ 00 mm 00 00> HO0H 0- 0+ 0- 00+ 0- 0- 0- 00+ 0+ 0- 0n 0: 00 00 00> 0- 0+ 0| 0 0+ 0- 0- 0+ 0+ 0- 0: 0: 00+ 00 00> 0+ 0- 0+ 00+ 00- 00+ 0- 00+ 00- 00- 0- 0- 00+ 0+ 00 0+ 0 0+ 00+ 0- 0+ 0 0: 0- 0 0- 0+ 00+ 0+ 00 0- 00- 00- 0- 00- 0- 0+ 0+ 0 0+ 00- 00- 0- 0+ 00 000 000 00: 000 000 00: 00> 00> 00> 00> 00> 00> 00> 00> 156 Ho H HHIH HO H HHHH HO H HOIH HOIH HOIH HOIH m o NI 0 OHI 0 00 0 0I 0I O0I 0I o 0- 000 HHIH OHIH HHIH HHIH HHIH HHIH HNHH HmHH H00H HHHH HmIH OmIH HmIH HmIH 0 00I N 0 NI m 0m 00 00 00 mNI om- m 00 000 OHIH OHIH HHIH HHIH OHIH OHIH 000H HmHH HOHH HmHH OO0IHHOHIHHOHIH HmIH 0 OHI 0 00 0I mI 00 00 00 0 00- NNI N 0 000 HNIH HHIH HNIH HNIH HHIH HNIH H00H HOHH HomH H00H ”NHIHHHHIHHNHIHHOHIH 0- HHI 0 0 00- m- 00 00 00 00 mNI mm- 0I 0 00: HO0H.HONH HONH HHOH HONHe HOIH HmIH HmIH H0IH HHHH HQHH HO0H H0 H mm 00 00 mm 00 om 0- 0 NI mNI 0 0 0 o 000 HO0H HO0H H0 H HO0H H0IH H0IH HmIH H0IH H0 H 00 H _0 H O0 H 0+ 0 mm 0 N 00 NI 00I NHI NHI 00 00 00 00 000 HONH HHHH HmmH H0IH HmIH HmIH HOIH OHHH HHHH HHHH H0 H 0I 0+ 00 om 00 00 0 0 m- 00- 0 0 m 0 000 000H HONH HOIH HmIH HmIH H0IH HHHH OOHH OHHH H0 H 0+ 0I 0I 00 00 00 0- 0I 0- 0m. 0 0 00 m 000 HNHH HmIH HmIH H0IH HmIH Hm H H0 H 00 H H0 H 0 0I 0+ 0+ 0 0 m0 NI 0 0- 0 00 0 0 000 . HOIH HmIH HmIH HOIH ”NHH HHHH ONHH HOHH 0+. .0+ 0 0+ 0I 00 . 0 0- OI 00I 0 NI 0 0 000 000 000 000 000 000 000 00> 00> 00> 00> 00> 00> 00> 00> 157 ON OH O OH HN O OH OI OH OI OI HH MI H 2O O OHI O OH OHI MI ON ON ON HH OMI ONI O O om OO ON NO HO NN MO H OI OHI OOI MN HN MN OH Hm MHI OHI OI ONI O OHI OO OO OO OO OI OHI M MI U> O ON O OH OH O OI O N OHI HO OO OO OO H> OI ONI OHI O ONI OI ON OM ON NN ONI OHI O HI 9 HO OM OO OO NO MO MI OHI OI OMI N OH O NI O HHI O OI NH NH NI NI NI OH O HI N OH NH Hm HOHH HO H HOHH HOHH HO H HOHH Hm H HO H HO H Hm H HH H HH H HH H HH H ON OH OH OH NN NH O OI O OI OI H MI N OZO HmHH HO H HOHH HOHH HO H HOHH HO H Hm H HO H Hm H HH H HH H HH H HH H ON O O O O HH O OI O MHI O O O M OEO HOHH HOHH HOHH HOHH HO H HOHH HO H HO H HO H HO H HH H HH H HH H HH H ON ON HH MH ON O OH OI NH NHI OI O HI O OEO HOHH HO H HOHH HOHH HO H HOHH Hm H HO H HO H Hm H HH H HH H HH H HH H HN OH OH NN OH O NH MI O MI MI OH OI OI OZO HOHH HO H HOHH HOHH HO H HOHH HO H HO H HO H Hm H HH H HH H HH H HH H NH O O NH OH O ON O OH MI OI O OI M OZO HO H Hm H HO H HO H HO H HO H HH H HH H HN H HH H Ho H Ho H Ho H Ho H OI OH N OI O OI O MI MH HH OI OH HI H NZO HOHH HO H HOHH HOHH HO H HOHH Hm H Hm H HO H Hm H HH H HH H HH H HH H OH OH O OH NH O OH O NH NHI HI O OI OI H2O OHO OH: OHm MHm NHO HHO OU> Mo> NU> HU> OH> MH> NH> HH> 158 OH+ HI O+ O+ OI OHI O+ OHI VI O+ O+ OO O+ H+ HI VI OI ONI OI OI V+ NH+ O+ OO V+ O+ HH+ OI O VHI O+ HI OH+ O+ OH+ MO OI O N+ HI N+ O+ O+ HNI OHI O+ H+ NO VH+ VI VI O+ HHI OI H+ VI VHI HI OI HO O+ O+ O O O+ HH+ NI OI O+ OI NI OHm HI O+ O H+ OI O+ OI OI O+ NI H+ OHO V+ VI OI OI OI VI HHI O+ OH+ OI V+ OHm OH+ HI NI O+ O VI H+ V+ O+ OI OI OHO OI OI OHI NHI OI O+ OI OI OH+ O+ HH+ MHO O O N+ O+ OI V+ O+ OI OI OI OI NHO HH+ N+ H+ OH+ H+ OH+ O+ VHI V+ N+ O+ HHO OZO OZO OZO OZO MZO NZO H2O OUO Mom Nu: Hum 159 N+ H+ OH+ O+ OH+ O+ O+ H+ NH+ OI O+ OU> HHI OHI OI OI O VI H+ VI OH+ O+ OI MU> O+ O+ O+ O+ HH+ HH+ O+ OHI OH+ O V+ NU> NHI OHI OHI OI OI OH+ OHI VHI O+ OI OI HU> OI O+ OI VI OHI OI NI OI VHI OHI HHI OH> O O+ O+ O+ O+ OH+ O+ O HHI NHI NHI MH> VI O+ NI OI OI HI OHI O+ NH+ NH+ O+ NH> H+ N+ O+ OI N+ H+ OI H+ OH+ O+ HH+ HH> VI O OI OI O+ OI OI ON+ O+ H+ OI MB O+ O+ OH+ O+ ON+ O+ V+ OH+ OI OI OI NB OI OHI OI HHI O+ O+ OI O+ HI VI VHI HO OEO OZO OZO OZO MZO NEO H2O Oom Mom Nom Hum 160 . HmmngmmH Hme 0+ 0+ 0I 00 .mm. .mm 00 000 0. .. HO0H HO0H 00+ 0 0+ 0+ 00 .00 00 000 .. O. . HO0H 0I 0+ 0I ..0I. HOO.¢¢O0;..00 mom 0I 0- 0+ 0 -OO ,O+ 00 000 0+ 0+ 0+ 00+ 0+ 0+ 0+ 000 00+ 0+ 0+ 0 00I 0I 00I 000 0I 0I 00I 0I 0+ 0+ 0+ 000 0I 0+ 0I 00+ 0+ 00+ 0+ 000 00+ 00+ 0+ 00I 0I 0I 00. 000 00I 00I 0I 0+ 0+ 0I 0I 000 020 020 020 000 00: 000 000 161 00 N0 00 00 00 00 00 N 0 0 HI mu 0 0I 0 N 0N 00I 0 N0 00 00 00 00 N0 mN 00 mm 00 0 0N 0 HI NI 0- 00 HI 0. 0 0 00 N0 0 0 O0 0N 0N u> NI 00 N HI NI 0 0. 0I 00I 00I 00I H> m 0 00 m 0N 0 N N0 N0 mm 00 0 00 0N mm 0N 00 0I 00 00. 00. O0 0 o 0N 00 00 ON 00 00 00 00I N0 OI o 00 HmmH HO0H HmmH HmmH HONH HO0H 00 H HN H HN H HN H 00 00 00 00 N0 0N 00 0 N0 0 o 020 H00H H00H H00H HNNH _00H 00 H HN H ON H HN H 0+ N0 N0 00 00 00 00 0I m 0- 0- 020 000H H00H OmNH 000H 00 H HN H ON H Hm H 0I 0+ 00 00 00 0N m0 0 N0 0 NI 020 HOmH H0NH HO0H H0 H ON H ON H ON H 0+ 00I 0I 00 00 ON 00 0 N m m- 020 HmNH HmmH .0 H HN H HN H HN H 0I 0+ 0I 0+ 00 ON em 00 NN 0 0 020 HmNH H0 H H0 H H0 H H0 H 0+ 0I 0I 0+ 0+ 0 NN mI 00. 0- OI Nzo H0 H HN H _N H HN H 0+ 0+ 0 0+ 0I 0I 00 0I 0 0I m 020 020 020 020 020 020 N20 020 000 000 N00 00: 162 OOH O OM O N HH OM ON mo O OOH OI OM NNI OO OI MI um OM OI OOH ONI HM ONI OOH O HO O OO ONI OOH OI OO ONI OH U> N NNI HM OI OOH OHI O HH H> HH OO ONI OO OHI OOH ONI OI B OM OI OOH ONI O ONI OOH OI O ON MI O OH HH OI OI OOH Hm 2O om Hm U> H> O O HO 00000043030000 .22... 163 II\ 3’ -.. ‘.1P‘\ 0} A,“ l‘ I I"||i“ll {Ill} I I . I.' II. ‘I' I'III III II Ill-o 111‘ I. ll‘lI‘lII ‘1" I'll 111’. IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII MWWMWW