中文安慰言語行為之比較研究：以華語為第二語言的英語與日語學習者為例

全文

(1)國立臺灣師範大學英語學系碩. 士. 論. 文. Master’s Thesis Department of English National Taiwan Normal University. 中文安慰言語行為之比較研究：以華語為第二語言的英語與日語學習者為例 A Comparative Study of Chinese Speech Act of Comforting by English and Japanese Learners of Chinese as a Second Language. 指導教授：陳純音博士 Advisor: Dr. Chun-yin Doris Chen 研究生：葉惠真 Student: Hui-chen Stephanie Yeh. 中華民國一百零七年七月 July 2018.

(2) 摘要. 本研究旨在探討以華語為第二語言的英語與日語學習者對中文安慰策略的有效度判斷以及使用情形，並驗證不同操縱變因（如：被安慰者的社會權力高低、事件嚴重程度及情境類型）對安慰策略的影響程度。研究對象共為十二名英語和十七名日語的中級華語學習者以及二十名中文母語者作為對照組，參與本研究的兩項測驗：一為有效度判斷，二為口頭完成情境對話。研究發現如下：一、情境類型的操縱因素對受試者的有效度判斷及使用情形影響最大，其次是事件嚴重程度，而被安慰者的權力高低影響則最小。二、提供支持、建議和鼓勵策略在有效度判斷中較同情及安撫策略有效。然而只有建議策略在口頭情境完成的測驗中仍被大量使用，其他兩種策略被安撫策略所取代。三、除了對提供支持策略的有效度評價不如中文母語人士高之外，英語和日語的華語學習者的表現大致與對照組相同。四、受試者普遍採用組合式的安慰策略，受歡迎的組合通常皆含有建議策略。五、日語學習者在面對更高權力的被安慰者的情況下比英語學習者更加看重安撫策略，並且在嚴重度高的情況下採用更多鼓勵策略。另一方面，英語組的人均安慰策略總數比日語組要來的多。根據上述結果，我們認為三種操縱因素各發揮了不同程度的影響，也證實對於安慰策略的有效度判斷與實際策略使用間確實存在影響差異。而且，跨文化差異仍存在於安慰策略的學習中。. 關鍵詞：第二語言習得、言語行為、安慰策略、中文. i.

(3) ABSTRACT The present study aims to investigate the perception and production of Chinese comforting strategies by English and Japanese learners of Chinese as a second language (CSL) under the three manipulation factors: power, severity, and situation types. A total of 12 English and 19 Japanese intermediate CSL learners together with 20 Chinese controls were recruited to complete two tasks (i.e., an effectiveness evaluation task (EET) and an oral discourse completion task (ODCT)). The major findings are summarized as follows: 1. The factor of situation types was found to have the greatest impact on the perception and production of comforting strategies by both groups, followed by severity, and power had the least influence. 2. In the perception of strategy effectiveness, the strategies of offer of support, advice, and encouragement were considered more effective than the strategies of sympathy and soother. However, only the advice strategy was still used greatly in the production task, and the other two strategies were replaced by the strategy of soother. 3. The English and Japanese CSL learners performed similarly to the Chinese natives except the evaluation of the offer of support strategy: They deemed it as significantly less effective than the Chinese controls. 4. Using combined strategies rather than single strategies was verified as a universal trend. The patterns with favored combined strategies of comforting mostly contained the advice strategy. 5. The Japanese group viewed the soother strategy more effective in the condition of facing comfortees of higher power and used a larger proportion of the encouragement strategy in the severe condition than the English group. On the other hand, the English produced more tokens of strategies per capita than the Japanese. The results indicated that the three moderating factors exerted their influence to different ii.

(4) degrees, and that there were indeed discrepancies between the subjects’ perception and production of comforting strategies. Moreover, it was found that the intercultural variations played a part during the acquisition process of comforting act.. Keywords: second language acquisition, speech act, comforting strategy, Chinese. iii.

(5) 謝辭. 終於到了提筆致謝的這一刻，這不僅僅意味著近一年來撰寫論文所收穫的甜美果實，更同時象徵著兩年半碩士生涯所刻劃的完美休止符。回顧這條充滿挑戰的研究所求學之路，若是沒有師長、好友、同學與家人相伴鼓勵，我想無論如何都無法堅持到這最後一刻。首先，我要感謝我的指導教授陳純音老師。我始終感恩這份始於大學時期而延續至今的師徒情誼，因為若非這段珍貴的緣分，我絕對無法獲得如此的良師益友。在無數次的論文晤談中，純音老師豐富的經驗、精闢的見解與溫暖的話語總是像黑暗中的那盞燈塔，拯救幾度迷失在茫茫大海的我。在純音老師身上，我所學到的不只是做研究、求學問的嚴謹態度與清晰脈絡，更多的是待人處世的謙遜圓融、遭遇難題的轉念智慧與看待人生的正向積極。另外，也要感謝特地撥空來當我的口試委員的陳淑惠教授以及張妙霞教授，謝謝她們從大綱口試到畢業口試所給予的每個寶貴建議與提點，都讓我的論文更臻完整。再者，我要向在研究所階段曾經導過我的所有師長致謝，謝謝李臻儀老師、吳曉虹老師、吳靜蘭老師、林蕙珊老師、張妙霞老師、詹曉蕙老師、甯俐馨老師、謝妙玲老師以及蘇席瑤老師（按姓氏筆畫排列），謝謝老師們用心的教學與耐心的指導，讓我能夠扎扎實實、一步一腳印地根基自己的語言學知識，進而完成這本論文。謝謝所有的研究所同學與好友在這兩年半的時間裡的陪伴讓我在這條路上從不是孤軍奮戰的鎧甲武士。特別是我最親愛的摯友與戰友—阿升（堡升）、阿寬（洪寬）與愛玲（韋伶），不管是當研究助理一起培養的默契、讀書會中的相互切磋砥礪、與任何喜怒哀樂時刻的分享與傾聽，你們一直都是我在同儕圈中最堅強的靠山，最強大的心靈支持。還要謝謝可愛又可靠的 Anita 江妍學姊與強尼（世強）學長，你們的智慧幽默與溫暖相伴都讓我的研究所生活更加精采充實。也要謝謝從大學一直以來的好朋友秉容與冠妤，雖然在不同的研究道路上，我們卻始終記得彼此。. iv.

(6) 最後，要向我最溫暖的避風港—親愛的家人，致上最誠摯的感謝。謝謝爸爸媽媽總是無條件的鼓勵與支持，放手讓我嘗試一切挑戰，卻也不忘在我跌跤時攙扶一把。謝謝弟弟妹妹三不五時捎來的關心問候，總是讓人倍覺溫暖。是因為有家這個永遠堅強的後盾，才能讓我一路走到現在，成就今天的我。僅以此本論文獻給所有關心我以及愛我的人，以表達我的無限感激。. v.

(7) TABLE OF CONTENTS CHINESE ABSTRACT ......................................................................................................................... i ENGLISH ABSTRACT...................................................................................................................... ii CHINESE ACKNOWLEDGEMENTS............................................................................................ iv TABLE OF CONTENTS .................................................................................................................... vi LIST OF TABLES AND FIGURES ................................................................................................... ix. CHAPTER ONE INTRODUCTION .................................................................................................. 1 1.1 Motivation ..................................................................................................................................... 1 1.2 Research Questions ....................................................................................................................... 5 1.3 Significance of the Study .............................................................................................................. 6 1.4 Organization of the Thesis ............................................................................................................ 7. CHAPTER TWO LITERATURE REVIEW ..................................................................................... 8 2.1 Speech Act of Comforting ............................................................................................................ 8 2.1.1 A Psychological Perspective .................................................................................................. 8 2.1.2 A Linguistic Perspective ...................................................................................................... 11 2.2 Cross-cultural Comparisons of the Speech Act of Comforting................................................... 14 2.3 Previous Empirical Studies of Comforting ................................................................................. 17 2.3.1 Burleson & Mortenson (2003) ............................................................................................. 18 2.3.2 Xu (2007) ............................................................................................................................. 21 2.3.3 Burleson (2008) ................................................................................................................... 23 2.3.4 Clark, MacGeorge & Robinson (2008) .............................................................................. 26 2.3.5 Summary ............................................................................................................................ 30 2.4 Summary of Chapter Two ........................................................................................................... 33. CHAPTER THREE RESEARCH DESIGN ................................................................................... 35 3.1 Participants .................................................................................................................................. 35. vi.

(8) 3.2 Methods and Materials ................................................................................................................ 36 3.2.1 Oral Discourse Completion Task ......................................................................................... 39 3.2.2 Effective Evaluation Task .................................................................................................. 44 3.3 Procedures ................................................................................................................................... 46 3.3.1 Pilot study ............................................................................................................................ 46 3.3.2 Formal Study........................................................................................................................ 48 3.3.3 Scoring Policy and Data Analysis........................................................................................ 50 3.4 Summary of Chapter Three ......................................................................................................... 54. CHAPTER FOUR RESULTS AND DISCUSSION......................................................................... 55 4.1 Factors Affecting Subjects’ Perception of Strategy Effectiveness .............................................. 55 4.1.1 Overall Findings: Power .................................................................................................... 55 4.1.2 Overall Findings: Severity ................................................................................................. 57 4.1.3 Overall Findings: Situation Types ..................................................................................... 59 4.1.4 General Discussion ............................................................................................................ 62 4.2 Factors Affecting Subjects’ Production of Strategy Types ....................................................... 66 4.2.1 Overall Findings: Power .................................................................................................... 66 4.2.2 Overall Findings: Severity ................................................................................................. 68 4.2.3 Overall Findings: Situation Types ..................................................................................... 69 4.2.4 General Discussion ............................................................................................................ 71 4.3 Discrepancies in the Influence of Moderating Factors between the Subjects’ Perception and Production ................................................................................................................................. 76 4.3.1 Discrepancies in the Influence of Moderating Factors between the Subjects’ Perception and Production according to Power Levels ............................................................................. 76 4.3.2 Discrepancies in the Influence of Moderating Factors between the Subjects’ Perception and Production according to Severity Levels .......................................................................... 78 4.3.3 Discrepancies in the Influence of Moderating Factors between the Subjects’ Perception and Production according to Situation Types .......................................................................... 79 4.3.4 General Discussion about the Discrepancies in the Influence of Moderating Factors between the Subjects’ Perception and Production ............................................................. 81 4.4 Single and Combined Strategies Used by the Subjects ............................................................... 83. vii.

(9) 4.4.1 Distributions of Single and Combined Strategies Used by the Subjects ............................ 83 4.4.2 General Discussion about Combined Strategy Use ........................................................... 87 4.5 Intercultural Variation ................................................................................................................. 93 4.6 Summary of Chapter Four........................................................................................................... 95. CHAPTER FIVE CONCLUSION .................................................................................................... 96 5.1 Major Findings of the Current Study ........................................................................................ 96 5.2 Pedagogical Implications .......................................................................................................... 98 5.3 Limitations of the Present Study and Suggestions for Further Research .................................. 99. BIBLIOGRAPHY ............................................................................................................................. 101 APPENDIX A TEST ITEMS FOR ODCT & EET ...................................................................... 106 APPENDIX B CONSENT FORM ................................................................................................. 139. viii.

(10) LIST OF TABLES AND FIGURES Table 2-1 Comforting Strategies Varying with Person Centeredness ................................................ 10 Table 2-2 Comforting Categorization for Discourse Strategies .......................................................... 13 Table 2-3 Major Findings and Limitations of the Previous Studies ................................................... 31 Table 3-1 Summary Table of the Participants .................................................................................... 36 Table 3-2 The Structure of the Research Design ................................................................................ 41 Table 3-3 An Example of the ODCT .................................................................................................. 43 Table 3-4 An Example Scenario with Corresponding Strategies of the EET ..................................... 46 Table 3-5 The Coding System Used in the Study ............................................................................... 52 Table 4-1 Manipulation of Power: Ratings of the Perception of Strategy Effectiveness .................... 56 Table 4-2 Manipulation of Severity: Ratings of the Perception of Strategy Effectiveness ................ 58 Table 4-3 Manipulation of Situation Types: Ratings of the Perception of Strategy Effectiveness ..... 60 Table 4-4 Effects of the Three Moderating Factors on Subjects’ Perception of Strategy Effectiveness ............................................................................................................................................ 64 Table 4-5 Manipulation of Power: The Distributions of the Subjects’ Strategy Use. ....................... 67. Table 4-6 Manipulation of Severity: The Distributions of the Subjects’ Strategy Use ...................... 68 Table 4-7 Manipulation of Situation Types: The Distributions of the Subjects’ Strategy Use ........... 70 Table 4-8 Effects of the Three Moderating Factors on the Subjects’ Production of Comforting Strategies ........................................................................................................................... 74 Table 4-9 Rankings of Strategies between Power Levels ..................................................................... 77 Table 4-10 Rankings of Strategies between Severity Levels .............................................................. 78 Table 4-11 Rankings of Strategies among Situation Types .................................................................. 80 Table 4-12 Distributions of the Overall Strategy Use ......................................................................... 83 Table 4-13 Distributions of Combined Strategy Patterns. ................................................................. 85. Figure 3-1 Research Flow of the Two Tasks in Each Scenario .......................................................... 50. ix.

(11) CHAPTER ONE INTRODUCTION. 1.1 Motivation Serving as a universal speech act across languages, comforting is omnipresent in everyday life. It is not only behavior to assuage negative feelings but also a way to maintaining friendships, romances, and work relationships (Burleson 2003). Below is a comforting scenario commonly seen in partnership. (1). Speaker A: Ganggang biaoxian just.now perform. de. tai cha le,. wo juede. DE too bad ASP 1SG feel. hen very. nanguo. sad. ‘I really did an awful job just now. I felt so sad.’ Speaker B: Meiguanxi, women xiaci hui biaoxian de geng it’s okay 1PL next.time will perform DE more hao. good ‘It’s okay. We will do a better job next time.’ Example (1) shows that the two speakers, Speaker A and Speaker B, are group members. When Speaker B heard that Speaker A was disappointed about his performance, he immediately uttered some promising words. In this case, Speaker B’s words did not merely show his concern as a partner but comfort Speaker A by offering a better future. Thus, functioning as emotional support, comforting skills play a crucial role in the interpersonal communication competence and the interpersonal relationship (Samter 2003, Burleson 2008). However, the preference for emotional support differs from person to person and even from 1.

(12) culture to culture (Burleson 2008). Considerable studies have focused on individual differences of merely either perceiving or employing comforting strategies (e.g., Burleson & Samter 1978), but the discrepancies between people’s perception and production of comforting strategies are worth investigation as well. Besides, an individual’s perception of the effectiveness of comforting strategies is closely related to his/her preference (Burleson 1994), but it has not been widely explored. A broader aspect of cultural differences is worth further investigation. Burleson & Mortenson (2003) and Duong (2008) were two of the few studies on the cultural differences between American and Chinese speakers’ evaluation and American and Vietnamese speakers’ preference for comforting strategies. Burleson & Mortenson (2003) found that Chinese speakers evaluated more positively on the messages that were viewed as avoidance by Americans. Duong (2008) compared different preferences for comforting strategies between Vietnamese and Americans, and found that Americans used the least amount of encouragements and Vietnamese seldom expressed sympathy when comforting others. Few have discussed the second language acquisition (SLA) of comforting, especially comforting in Chinese. In addition, the first language (L1) influence in the field of SLA remains a hotly discussed topic (e.g., Su 2010, 2012). Speaking of culture-related theories, Hofstede’s cultural pattern (1980) is a theory to explain the cultural differences of evaluating or employing comforting strategies. Apart from 2.

(13) it, Brown and Levinson’s politeness theory (1987) accounts such discrepancies from different perspectives. From a Western perspective, Brown and Levinson (1987) divide face into negative and positive. Negative face is the need to enjoy the freedom to be independent and not to be forced by others. On the other hand, positive face refers to the need to be liked and accepted as innergroup members. According to them, a large proportion of speech acts are intrinsically facethreatening acts (FTAs). Acts such as advice, invitation, and promises threaten negative face of a hearer (H) by forcing him/her to do something. Acts like complaints, criticism, and disapproval threaten H’s positive face by ignoring of his self-image. As a result, to protect the face of both a speaker (S) and a hearer, people tend to employ strategies consisting of positive and negative politeness to minimize threats. From an Eastern especially Chinese perspective, Brown and Levinson’s politeness theory has been criticized by Gu (1990) for not being suitable for Chinese data, since there is a gap between the concept of negative face in a Chinese view and a Western notion. For example, inviting in Chinese is not considered threatening H’s negative face as shown in the following example. (2) Speaker A: Mingtian tomorrow. lai chi wanfan. com eat. dinner.. ‘Come to have dinner tomorrow.’. Speaker B: Bu not. le,. tai mafan.. ASP too troublesome.. ‘No! It will bother you.’ 3.

(14) Speaker A: Mafanshenma. Trouble.nothing ‘Not at all.’. (Gu 1990: 252). In Example (2), speaker A insisted on inviting Speaker B to dinner even if Speaker B had already explicitly expressed his desire that Speaker A not do it. In this case, Gu argues that a European will feel that Speaker A’s act is intrinsically forcing, but a Chinese will think that Speaker A’s act is intrinsically polite. He claims that Speaker A’s insistence on inviting serves as good evidence of Speaker A’s sincerity so that the Chinese negative face is not threatened in this situation. In addition, to perfectly elaborate Chinese data, Gu (1990) amends Leech’s politeness maxims (1983),1 which not just makes H easier to accept S’s offer, and indicates S’s sincerity in making the offer. In short, polite behavior can be discussed from Eastern and Western perspectives. Thus, when we specifically concentrate on the issue of comforting strategies cross-culturally, different perspectives of politeness as well as the various cultural patterns merit further consideration. Moreover, Brown and Levinson (1987) argue that sociological factors such as social power referring to the power relationship between different interlocutors, and imposition, which is defined as the degree of enforcement on H, influence the strategy use for speech acts.. In Gu (1999: 244), each maxim is further divided into the motivational level and the conversational level. Different from the motivational level, the operation at the conversational level only regulates speech behavior without altering the nature of the cost at the motivational level. 4 1.

(15) These two factors have later been adopted in studies of speech acts like apologizing, requesting, requesting, complaining, (e.g., Blum-Kulka, House & Kasper 1989, Trosborg 1995, Cheng 2001, Chen, Chen & Chang 2010). They have also been used in performing comforting behavior: Xu (2007) attested the effect of social power on producing comforting strategies. Other factors like severity of problems and types of situations have been proved to affect people’s strategy adoption not only in other speech acts (e.g., Lin 2010, Hsu 2016), but also in the speech act of comforting (Burleson 1982, 1984). That is, whether a condition is severe or not has effect on people’s strategy employment, and so do types of situations like academic performances and social rejection. Therefore, the present study thus investigated these factors, namely, social power, severity of the problem, and types of situations. For the subject recruitment, our study as a cross-cultural one thus targeted at two language groups, Japanese and English, as the representatives from Eastern and Western cultures. That is, the present study examined the perception and production of comforting strategies by Japanese- and Englishspeaking leaners of Chinese as a second language (CSL). 1.2 Research Questions Inspired and motivated by the aforementoned niche, the present study aimed to answer the following questions: 1). Will moderating variables such as social power, severity, and situation types influence comforting perception and production of the participants from different cultural 5.

(16) backgrounds? 2). Will the results show discrepancies between CSL learners’ perception and production of comforting strategies?. 3). What are the English- and Japanese-speaking CSL learners’ single and combined strategy use in performing the act of comforting?. 4). To what extent does the intercultural variations between English and Japanese learners of Chinese as a second language influence their comforting strategies? The first question stressed the influence of moderating factors. The second question. concerned the discrepancies of the participants’ perception and actual production of comforting strategies. The third research question pertained to the usage tendencies of single and combined comforting strategy use between the two experimental groups. Last, the fourth question centered on cultural differences of the act of comforting and the impact of intercultural variations on CSL learners’ language acquisition. 1.3 Significance of the Study Acquisition of speech acts has been extensively discussed in the literature (e.g., House and Kasper 1981, Trosborg 1995). However, comforting behavior, due to its complexity, has been relatively understudied compared with other speech acts. Burleson (1982, 1984), a leading researcher of comforting behavior, examined both children’s development and adults’ preferences for comforting strategies in Western cultures, but only a few studies on Chinese 6.

(17) comforting strategy use (Wen 1999, Wang & Li 2003) were conducted. Xu (2007) was one of these few studies on the strategy employment of native Chinese speakers, and the interplay between strategy use and the influence of social factors. To date, none of the aforementioned studies has probed into the second language acquisition of Chinese comforting strategies. Therefore, the present study was in an attempt to fill the gap by exploring the usage differences of Chinese comforting strategies by English and Japanese CSL learners. Besides, the study intended to investigate the interplay and the influence among moderating factors such as social power, severity of the problem, and situation types. Last but not least, the extent of intercultural variations and their influences on SLA and the discrepancies between perception and production of comforting strategies were also the concern. It is hoped that the findings of the present study will shed light on foreign CSL learners’ perception and production of Chinese comforting strategies. 1.4 Organization of the Thesis This proposal is organized as the follows. Chapter Two introduces what comforting is with its definition, framework, and classification. In addition, cultural differences of act of comforting and several experimental studies are reviewed. Chapter Three introduces the research design of the present study. Chapter Four presents the results and discussion. Lastly, the major findings, the pedagogical implications, and the limitations of the study are summarized in Chapter Five. 7.

(18) CHAPTER TWO LITERATURE REVIEW. Chapter Two provides both theoretical and empirical studies on the speech act of comforting. Section 2.1 defines the act of comforting with examples and reviews the frameworks of comforting from two perspectives: psychological and linguistic. Section 2.2 puts emphasis on the relationship between cultural differences and the speech act of comforting. Section 2.3 reviews four empirical studies on speakers’ perception and production of comforting strategies. Last, Section 2.4 summarizes the whole chapter. 2.1 Speech Act of Comforting The speech act of comforting is discussed in two sections: To begin with, a psychological perspective introduces what comforting is, and how people define effective comforting strategies. In addition, a linguistic perspective presents linguistic features and a categorization of comforting strategies. 2.1.1 A Psychological Perspective Comforting originally refers to the activity which counselors and psychotherapists regularly engage in. It is then proposed as a popular approach by famous psychologists, Rogers (1957, 1975) and his associates (Aspy 1975, Carkhuff & Berenson 1977). The Rogerian approach claims that only if messages can express EMPATHY, RESPECT, and. 8.

(19) GENUINENESS 1 to recipients can they be regarded as successful comforting strategies. However, it was subjected to criticisms of being too general and vague even being widely used in therapeutic conditions, and thus was replaced by constructivism. Applegate (1978, 1980), one of the representatives of constructivism, proposes that comforting messages could be evaluated for the extent to which the feeling and perspective of a distressed other are explicitly acknowledged, elaborated, and granted legitimacy. He then develops a coding system composed of nine hierarchies: Three major levels with three sublevels within each major one; higher levels refer to more sophisticated and advanced messages. 2 Because of its concrete features, his constructive coding system gradually replaced the Rogerian approach. Later on, Burleson (1985, 1994) defines comforting strategies as the messages having the goal of alleviating or lessening others’ emotional distressed experiences. The act of comforting is also seen as an activity directed at managing or modifying the physical states or feelings of others, rather than the treatment of physical or material conditions. Example (1) shows that the speaker expresses his/her understanding to the situation (“I really understood…”); he/she in turn provides similar experiences of being rejected (“I haven’t been invited to…”), and he/she. The three special terms are defines as the following in Rogerian approach. (i) EMPATHY refers to the extent to which therapists are successful in communicating their awareness and understanding of clients. (ii) RESPECT is the degree to which therapists communicate positive regard for clients. (iii) GENUINENESS is the extent to which therapists are real and authentic in communicating with clients. 2 Applegate’s coding system (1978, 1980) is divided according to the extent that messages acknowledge other’s distressed feelings. (i) The lowest major level (Levels 1~3) refers to either the explicit or implicit denial. (ii) The second major level (Levels 4~6) is the implicit recognition. (iii) The highest major level (Levels 7~9) is the explicit acknowledgement or recognition. 9 1.

(20) finally offers a possible reason (“maybe Jean really wanted to…”) to alleviate the disappointment of the distressed others. (1) An example of Burleson’s comforting messages “Well, I’d tell her (the distressed one) that I really understood how she feels, that I haven’t been invited to special party sometimes and I know it hurts—you can feel rejected. But I’d say maybe Jean really wanted to have you but her parents wouldn’t let her invite everybody. So it doesn’t mean that Jean doesn’t like you or anything.” (Burleson 1982: 1581) In addition, based on the coding system of Applegate, Burleson (1985, 1994) specifically associates it to the concept, PERSON CENTEREDNESS: “In comforting contexts, person centeredness reflects the extent to which messages explicitly acknowledge, elaborate, and contextualize the distressed others feeling and perspectives (Burleson 1994, 2008).” Messages are thus categorized into three levels of person centeredness: low person centeredness (LPC), moderate person centeredness (MPC) and high person centeredness (HPC). Table 2-1 summarizes Burleson’s statements, and demonstrates comforting strategies varying with person centeredness with definitions and examples.. Table 2-1. Comforting Strategies Varying with Person Centeredness (Burleson 1994: 141) Person Centeredness. Definition. Example. Low. Denial of individual. “I’d tell her she had no reason to feel this. perspectivity. way.”. Moderate. Implicit recognition of individual perspectivity. “I’d ask her if she wants to come to my party to play…”. High. Explicit recognition and elaboration of individual perspectivity. “The movie was scary. It also scared me…” 10.

(21) From the above table, some implications can be inferred (Burleson 1994): Firstly, messages of lower person centeredness (LPC) are more speaker-centered by interpreting the situation mainly from the speaker’s own perspective, and advising how others should think or act. Conversely, messages of higher person centeredness are more listener centeredness by recognizing other’s feeling. Secondly, moderate person centeredness (MPC) or high person centeredness (HPC) comforting messages are more neutral than LPC messages because they generally describe the feelings and situations. Therefore, among the three levels of comforting strategies, higher person centeredness are usually preferred because of the greater degree of involvement of problems, the neutral evaluation, and the cognitive-oriented explanation of the feelings (Burleson 1994). In short, the psychological perspective of comforting describes its development from the vague definition given by Rogerian researchers to Applegate’s concrete coding system (1978, 1980) under constructivism. Finally, Burleson (1985, 1994) links Applegate’s nine hierarchies with the concept of person centeredness within three levels. The new comforting framework simplifies the complexity but remains the specific characteristics, and thus it has been prevailing until now. 2.1.2 A Linguistic Perspective Although the psychological framework has been extensively adopted (e.g., Samter, Burleson & Murphy 1989, Kunkel 2002, Xu 2007), the implications that it can provide for 11.

(22) pedagogy are still limited. Under this situation, Suzuki (2010) defines the act of comforting from a linguistic perspective, proposing that the speech act of comforting should be a faceenhancing act (FEA). Speakers use the FEA, which is supposed to benefit hearers, to convey sympathy, offer encouragement or support, and give suggestions (Suzuki 2010). In addition, according to Suzuki (2010:84), “Comforting is assumed to belong chiefly to Searle’s EXPRESSIVE (Searle 1979), and Leech’s CONVIVIAL3 (Leech 1983) because of its FEA nature.” Example (2) displays some features including conveying sympathy (“I’m really sorry…”) and offering support (“if you need…”). (2) An example of Suzuki’s comforting messages “Hey! I’m really sorry to hear about your sister. If you need anything let me know!” (Suzuki 2010:96). In terms of the categorization, Suzuki (2010) analyzes comforting strategies from single strategies to combined strategies. First, for single strategies, Suzuki (2010) divides them into five main types of strategies, each of which is carefully elaborated through its functions, features, formulae, and examples. Table 2-2 summarizes his categorization as follows.. According to Searle (1979), EXPRESSIVE is one of the five illocutionary acts which express a speaker’s attitude toward a proposition or statement. EXPRESSIVE is later categorized as COVIVIAL by Leech (1983:104) according to its relationship with the Politeness Principle. 12 3.

(23) Table 2-2. Comforting Categorization for Discourse Strategies Comforting strategies. Function. Feature or Formula. Examples. Soother. Relieving H’s hurt or sad feeling. used as a head act “It’s okay.” at the beginning (Suzuki 2010:91). Encouragement. Offering H a better future. using future tense. “You will do better.” (Suzuki 2010:92). Sympathy. Representing S’s understanding. subj. + cognitive verb + fact / object’s feeling. “I know how you feel.”. (Suzuki 2010:92) Advice. Offering something beneficial for H. in an imperative mood. “Don’t worry about it.” (Suzuki 2010:93). Offer of support. Another strategy to encourage hearers. in a declarative or “I am here for you.” interrogative mood (Suzuki 2010:94). Moreover, Suzuki (2010) discovers that strategies incorporate with each other to constitute a combination for the achievement of a speech act. It has been found that while people in some cases use one single strategy, they utilize more elaborated combinations of utterances or formulae to show their intention and consideration for others, as shown below: (3) “Hey! I’m really sorry to hear about your sister. If you need help, please let me know. (Suzuki 2010:96) (4) Take medicine and lie down for a bit. It will make you feel better. (Suzuki 2010:97) Example (3) shows a combination of the two strategies, sympathy and offer of support to show his understanding of the situation, and then makes offer to help. In addition, as can be seen in Example (4), the strategy of advice combines with the strategy encouragement to form a pattern. Such a pattern can be a powerful support for the hearer in the sense that the clear 13.

(24) guidelines for solving problems and encouragement for the better future. These patterns suggest the fact that people usually use combined strategies to reach a more powerful and complete comfort. All in all, a psychological perspective and a linguistic perspective of comforting have been reviewed. The psychological perspective of comforting strategies discusses the necessary features and functions of an effective comforting message from the concept of person centeredness. However, without further analysis of the form and content, it was hard to categorize comforting strategies. Therefore, the present study adopted Suzuki’s categorization to form a concrete and unambiguous standard in research design and data coding. 2.2 Cross-cultural Comparisons of the Speech Act of Comforting That people from different countries tend to communicate differently is acknowledged as a general sense (Nishimura, Nevgi & Tella 2008). In fact, countries belonging to different cultural patterns demonstrate different propensities in interpersonal communication. Previous research has endeavored to propose some cultural taxonomies to explain such differences in Western and Eastern cultures (Hall 1976, Hofstede 1980, 1991). These taxonomies can be viewed as implications to help predict cultural preferences for comforting strategies. To begin with, Hall’s (1976) high- and low-context cultural patterns emphasize the role of context in communication. He proposes that cultures differ on a continuum that ranges from high to low context. People of high context (HC) cultures including Chinese, Japanese, and 14.

(25) Korean prefer to use covert and implicit messages because of their clear distinction of in- and out-groups. In other words, the meaning of a message is either implied by the physical setting or is presumed to be part of the shared knowledge within a group, and messages can even be expressed nonverbally. Conversely, people of low-context (LC) cultures such as German, and English cultures are inclined to use messages in a rather explicit code. For the LC cultures, only when messages are plain with meanings specifically stated are they able to clearly understand the words and continue the communication. On the other hand, Hofstede (1980, 1991) explains various cultural patterns from different dimensions. He argues that the interaction between the power distance index (PDI) and the individualism index (IDV)4 appears to be negatively correlated to each other: That the higher a culture’s PDI is, the lower its IDV score will be. The power distance index (PDI) examines how a culture values the appropriateness and importance of status differences and social hierarchies. People of high PDI cultures stress a power distance, but those of low PDI cultures believe everyone is born equal. In terms of individualism index (IDV), it refers to the degree to which a culture depends on to the self and to the group. People of high IDV cultures view independence, privacy, and self as vital concepts. On the other extreme, people of low IDV cultures prefer a collectivist orientation which center on the strong interpersonal bonds, the. Hofstede (1980, 1991) discusses cultural patterns from four dimensions, power distance, uncertainty avoidance, individualism-collectivism, and masculinity- femininity. The dimensions are respectively indexed as the power distance index (PDI), uncertainty avoid index (UAI), individualism index (IDV), and masculinity index (MAS). Concerning the degree of relevance, the present study only focuses on the interaction between the PDI and IDV. 15 4.

(26) pursuit of group profit, and the maintenance of group harmony. Eastern cultures including Japan and Taiwan are collectivist cultures which believe that the concept of “we” instead of “I” should be the priority and that each person has a rightful place in the social order. Therefore, they are the cultures of a high PDI but a low IDV. Children raised in Eastern cultures are taught to obey their parents without challenging or questioning them. Similarly, students in such cultures are expected to conform the requests from teachers. In addition, the high PDI feature makes parents of Eastern cultures not only take children’s academic performances seriously but also have high expectation on them, causing high parental involvement in children’s academic achievements (Chua 2011, Kung 2015). This situation in turn makes Eastern children care much about their academic performances. Conversely, Western cultures such as USA and Canada prefer small power distances and minimizing hierarchies, and emphasizes independence of oneself. They belong to the cultural pattern of a low PDI and a high IDV. Children raised in Western cultures put less value on obedience and are taught to seek reasons or justifications for their parents’ actions. Students in this cultural pattern value high on their independence, and are taught to ask questions and challenge authorities bravely. On top of that, parents of Western cultures respect children’s individual differences and encourage them to pursue their own interests, showing a relatively loose parental involvement in academic performances (Kung 2015). As a result, various cultural patterns result in the differences of belief, value, thinking, and 16.

(27) even ways to speaking in Eastern and Western cultures. These differences may extend to different tendencies of perception and employment of comforting strategies. First, following the cultural variations, it is expected that Americans would value more positively on HPC comforting strategies, and less positively on LPC comforting strategies than people of Eastern cultures. People of Western cultures would endorse recognition and acknowledgement of emotional state such as HPC strategies (Burleson 1994). However, people oriented to collectivism sometimes avoid directly focusing on others’ distress to minimize the chance of threatening faces. Moreover, Eastern cultures value high on maintaining the harmony due to the high PDI feature (Hofstede 1980), so the possibility of the participants using advice may be lower than those of Western cultures when they address comfortees of higher power. The reason is that advice is always in imperative mood, which thus threatens the face and disrupts harmony. In short, people from Eastern and Western cultures might vary a lot in belief, value, and ways of communication. The different tendencies are discussed through the discrepancies between their oriented cultural patterns. Last but not least, these differences can serve as the implications of the perception and production of the act of comforting. 2.3 Previous Empirical Studies of Comforting In Section 2.3, four empirical studies concerning the comforting behavior are reviewed, three of which centered on adults, and one of which focused on children and adolescents. 17.

(28) Although one of the studies was to investigate children’s perception of the act of comforting, it discovered some findings that were different from others. Firstly, Burleson & Mortenson (2003) investigated cultural differences between American and Chinese speakers’ perception of comforting strategies. Secondly, Xu (2007) examined the comforting strategies used by native Chinese college students. Thirdly, Burleson (2008) analyzed how individual differences affect people’s perception of comforting strategies. Finally, Clark, MacGeorge & Robinson (2008) explored children and adolescents’ evaluation of comforting strategy from the perspective of comfortees. 2.3.1 Burleson & Mortenson (2003) Burleson and Mortenson (2003) investigated whether there were indeed cultural differences (American vs. Chinese) in preferred comforting strategies. Their participants included 203 college students from Euro-American and Chinese cultural groups, 59 of whom were American males, 39 were American females, 44 were Chinese males, and 61 were Chinese females. The average age for the two experimental groups were 23 and 26, respectively. In the study, the participants were asked to finish four questionnaires by rating items on a 7-point scale. The first questionnaire was a value orientation measure including self-reporting scores of collectivism and individualism and the Communication Functions Questionnaire (CFQ) for value placing on comforting skills. The participants rated how important it was for a close, same-gender friend to be able to skillfully comfort the distressed. The second 18.

(29) questionnaire required the participants to evaluate interaction goals such as avoidance, problem management, and emotion management. Two comforting scenarios were given, each of which was accompanying with a list of nine statements describing the comforter’s coping behavior. The participants were asked to rate each statement on its importance. The third questionnaire concerned evaluations of the quality of comforting messages varying on person centeredness. They needed to evaluate nine comforting strategies corresponding to each scenario on their sensitivity5 and effectiveness. As for the task in the fourth questionnaire, the participants were asked to evaluate the appropriateness of interactive coping behavior like solve and solace behavior6 in three given situations: having conflict with friends, getting fired from a job, and breaking up with important others a romantic relationship. The results concerning people’s evaluation of comforting strategies and interactive coping behavior indicated that although Chinese speakers considered comforting strategies more sensitive than American speakers, Americans viewed comforting strategies of high person centeredness more positively and LPC comforting strategies less positively than Chinese. This finding was consistent with the prediction that Chinese would discriminate less than Americans in evaluating comforting strategies. Moreover, as predicted, the findings showed that Chinese According to Burleson (1982), sensitivity refers to the degree to which a comforting message recognizes and understands the causes of the other’s distressed emotional state. 6 Barbee & Cunningham’s typology (1995) of interactive coping behavior proposes that the interaction between the two theoretical dimensions (approach-based vs. avoidance-based and problem-focused vs. emotion-focused) results in a four- strategy typology of supporting strategies: (i) Solace behavior is approach- and emotion-based. (ii) Solve behavior is approach- and problem-based. (iii) Escape behavior is avoidance- and emotion-based. (iv) Dismiss behavior is avoidance- and problem-based. 19 5.

(30) participants viewed escape and dismiss behavior significantly more positively than Americans. However, it was contradictory that Chinese evaluated the strategy of solace more positively than Americans, which was against to the hypothesis that Americans would rate it more positivity than Chinese. On the other hand, they found that both value and goal orientation significantly played a part in the evaluation of comforting messages, and that interaction goals were indeed a stronger mediating factor affecting the evaluation. Burleson and Mortenson concluded several similarities and differences between American and Chinese speakers, and tried to account for the discrepancies between the finding and the prediction. For the similarities, both groups viewed HPC strategies as the most sensitive strategies, and believed that MPC strategies were more sensitive than LPC strategies. It showed that there was a transcultural similarity in perceiving comforting strategies, and proved that hierarchies of comforting strategies were valid. As for the differences, the finding that Chinese discriminated less in their evaluations of supporting behavior and further evaluated avoidant behavior more positively than Americans could be explained by different cultural patterns. Oriented to a high- context culture, Chinese rely less on the specific content of supporting behavior to infer others’ intentions and concerns than Americans. In addition, due to the nature of collectivist culture, Chinese people view expressing distress as a way to disturbing group harmony. It might be the reason that the Chinese participants in the study endorsed escape and dismiss behavior. The surprising finding that Chinese viewed solace behavior more appropriate 20.

(31) than Americans could be inferred by the distinction between individualism and collectivism. Being collectivists, Chinese are more responsible for the welfare of their in-group friends so that they are motivated to provide solace. Taken together, the results suggest a model connecting culture, interaction goals, and evaluation of supporting behavior. 2.3.2 Xu (2007) Xu (2007) conducted a study on the speech act of comforting in hope of filling the gap that little research on the issue was done in Chinese context. Based on the theoretical framework of the Speech Act Theory (Brown & Levinson 1987) and the Politeness Principle (Leech 1983), she aimed to investigate the strategy use, the influence of social factors on the selection of strategies, and the realization patterns of the comforting strategies collected from Chinese native speakers. By reviewing the theoretical framework and previous studies, Xu (2007) concluded that the act of comforting in Chinese was indirect, and it belonged to EXPRESSIVE under Searle’s categorization (Searle 1979). When it comes to the methodology of the study, Xu (2007) collected 172 (39 males and 133 females) valid questionnaires of DCTs from 180 subjects to reach the reliability and validity. The participants were native speakers of Chinese as well as college students in Guangdong University. The questionnaire of DCTs consisted of 10 situations, varying according to relative power, and contextual of specific distressed emotions. Adopting the categorization of Burleson (1994), Xu coded comforting strategies into two main 21.

(32) groups: HPC and LPC. To be more specific, three strategies (concern, sympathy, and approval of the distressed emotion) were divided from HPC, and five strategies (reason, problem solving, different perspective, denial, and weakening) fell under the category of LPC. Within the eight strategies, 19 sub-strategies were in turn created for a more careful inspection. Xu (2007) found that among the responses elicited from the questionnaires, LPC comforting strategies accounted for a considerably large proportion, 94.78%, showing that Chinese preferred focusing directly on the event or denying the emotion rather than approving negative emotions. In contrast, HPC comforting strategies only touched sparsity, indicating the dislike for Chinese to encourage a distressed emotion. Specifically speaking, offering different perspectives, denying the value of negative emotions, and weakening the extent of the distressed emotion were the strategies that the participants mostly used to comfort the addressees under the category of LPC. In terms of the influence of social factors, relative power and gender both played crucial roles. Although relative power did not have an apparent effect on the total frequency of the two main comforting strategies, it strongly affected the strategy use in the three situations, upper power [+P], equal power [P], and lower power [-P], respectively. HPC strategies were used more frequently when the participants comforted a person with equal status rather than with higher and lower status. In addition, there were clear differences in employing specific strategies when the participants faced comfortees with different power status. That is, none of 22.

(33) the three situations had the same order as the total frequency. For example, offering different perspectives ranked first when facing upper-power addressees, however, it only ranked second in the [P] situation, and ranked last in the [-P] situation. Similarly, the influence of gender performed in such a way: Despite the fact that there was no clear difference in choice of the total number of strategy use, the respective strategy preferences differed a lot from males to females. Men were reported to use denial, and weakening more frequently; nonetheless, women tended to use approval of the feeling, different perspectives, and reasons. Furthermore, as for the realization patterns of the comforting behavior, Chinese were found to usually use a conventional phrase like negative adverbs + negative emotions, like “buyao shangshin” ‘don’t be sad’. In short, social factors like relative power and gender exercised great impact on strategy use. In conclusion, based on the theoretical framework of Speech Act Theory (Brown & Levinson 1987) and the Politeness Principle (Leech 1983), the instrument, DCT questionnaires, and the adoption of Burleson’s categorization (1994), Xu (2007) found Chinese speakers’ preferences for LPC categories rather than HPC ones. In addition, she discovered that relative power and gender could exert their impacts on the employment of comforting strategies. 2.3.3 Burleson (2008) Burleson (2008) reported three studies concerning how certain psychological and situational factors affect people’s responses to various comforting messages.. 23.

(34) Study 1 was designed to explore whether, and to what extent, responses to different levels of comforting strategies vary as a function of the value people place on emotional support skills. One hundred and eighty-four participants (89 men and 95 women) were recruited to rate the importance of a skillful comfort on a 5-points scale, and then to rate the effectiveness and quality of three levels of messages within comforting situations. The results indicated that people valuing high on support skills rated HPC comforting skills more positively, and LPC messages less positively than those valuing low. However, all the participants—regardless of their support values—rated HPC messages as significantly more positively than MPC and LPC messages. In the discussion, a practical implication was proposed: Although people valuing high on comforting messages pay more attention to the details of comforting messages than the others, HPC messages invariably did the best job for recipients regardless of the support values. Study 2 concerned the influence of people’s self-concept about the communicative behavior. Self-concept, the way we think about ourselves, can be divided into two aspects, expressive and instrumental. People who see themselves as expressive believe themselves to be kind, warm, and sensitive to others’ feelings. In a contemporary American society, an expressive orientation is connected to femininity, and it is thus expected that high expressives would evaluate HPC messages more positively. In contrast, people belonging to instrumental orientation trust themselves to be independent, decisive, and practical. Associated with masculinity, instrumental-oriented people are expected to evaluate MPC messages (problem24.

(35) solving strategies), and, perhaps even LPC messages more favorably. Collecting the data and analyzing the answers from 387 college students (190 men and 197 women), Burleson found that people with high expressivity considered HPC comforting messages to be the best strategies. In addition, even though people with high instrumentality viewed MPC messages more helpful than those with high expressivity, they still rated HPC strategies as the most helpful strategies. These findings indicated that HPC comforting messages were the most helpful messages to all the participants, which was highly similar to the findings of Study 1. Different from Studies 1 and 2, which focused on personal traits, Study 3 examined how aspects of the communicative situation influence responses to different levels of comforting strategies. In this study, a total of 131 participants (59 men & 72 women) were recruited, and two aspects of communicative situations were scrutinized: the severity of problem experienced by message recipients and the gender of the helper. It was expected that the gender of the helper was more likely to exert influence on mild upset situations than more intense upsets. But when coping with a more intense upset, it was expected that message person centeredness would be the main impact. Although the manipulation of problem severity was successful, the effect was neither significant for problem severity nor the gender of the helper. Only message person centeredness had a significant effect. However, a near significant interaction between problem severity and the helper’s gender was discovered, indicating that female helpers were perceived as using more helpful messages than male helpers in the mild upset but not in the more intense 25.

(36) upset. In short, that message person centeredness still appeared to be the strongest influence on message evaluations—regardless of problem severity and helper’s gender, HPC messages were rated as the most effective strategies. Burleson (2008) showed that although certain psychological and situational factors had some effect on responses to comforting strategies, their impacts were relatively minor compared to the influence of message person centeredness. This indicates that people generally care much more about the features of comforting strategies than recipient characteristics or situational factors. In addition, it is apparent that HPC messages are always the most effective comforting strategies. Hence, some instructions were provided at the end of the review like creating a comfortable conversational environment for the addressees, assisting the distressed person in telling a story about the upsetting problem, and avoid evaluating, denying, or ignoring others’ feelings. These tips would help to provide effective emotional supports. 2.3.4 Clark, MacGeorge & Robinson (2008) Compared to the studies on adults’ perception and production of comforting strategies , little is known about how children and adolescents respond to the comforting strategies offered by peers. Given the lack of such kind of research, Clark, MacGeorge & Robinson (2008) conducted an experiment about children and adolescents’ evaluation of and preference for comforting strategies to fill the gap and help understand their conceptualization of effective comforts. Previous research has shown that children from early to middle childhood (2-7 years old) 26.

(37) tend to use nonverbal behaviors, simple verbal expressions, and recruit adults. On the other hand, adolescents in late adolescence and early adulthood (17-22 years old) are capable of employing sophisticated verbal comforting strategies, i.e., strategies with high-person centeredness (cf. Burleson 1982). Therefore, Clark, MacGeorge & Robinson (2008) selected “Tweens (8-14 years old)” as the target because of the substantial growth in the variety of verbal comforting strategies children employ with their peers and the increasingly important role of peer relationships. In terms of the strategies commonly employed by tweens, there were six main types: sympathy, advice, optimism, companionship, account, and minimization. Within the psychological framework of constructivism, sympathy was rated as the most positively due to its high-person-centered characteristics, followed by account, companionship, optimism, advice, and minimization. However, taking the significance of companionship at the stage of tweens into consideration, they thus proposed a revised ranking of these strategies: It was supposed that companionship would replace sympathy, and it would be the most favored strategy, followed by sympathy, account, advice, optimism, and minimization. For the variables that might moderate the results, Clark, MacGeorge & Robinson (2008) included three variables, age, gender, and types of situations, and proposed corresponding hypotheses for the results according to previous literature. Firstly, for the variable of age, it was hypothesized that tweens’ evaluations of sympathy and accounts would increase, whereas 27.

(38) evaluations of optimism and minimization would decrease. The reason was that adolescents’ friendship is more talk-based than childhood friendship, and tweens have gained greater knowledge of the real world. Secondly, for the gender variable, the hypothesis is that girls would evaluate companionship and sympathy more positively, whereas boys would evaluate optimism and minimization more positively. Lastly, in terms of the hypothesis on types of situations, it was proposed that the companionship strategy would be evaluated more positively in the social rejection situation than in the academic difficulty situation. A total of 292 participants (142 males and 149 females) including 107 fifth graders, 98 seventh graders, and 87 ninth graders from two different communities were recruited. All of the participants were asked to do a message evaluation task, in which there were two hypothetical situations: academic difficulty, and social rejection. They were firstly asked to imagine that they encountered hypothetical distressed situations. Later on, within each situation, they were required to evaluate six messages corresponding to different strategies on two 5point scales of affect improvement and caring. In addition, one community of the participants were asked to further complete a strategy preference task to select the preferred strategies. Firstly, for the hypothesis of the revised rankings of strategies, Clark, MacGeorge & Robinson (2008) found that there were significant effects for strategies in the results of both affect improvement and caring. It indicated that there were various score distributions among six strategies: Companionship was evaluated the most positively, and account was rated the 28.

(39) least positively. They also found that companionship appeared to be the mostly preferred strategy in the strategy preference task by accounting for over a half of the percentage. In short, the companionship strategy, as predicted, was evaluated the most positively; however, the account strategy appearing to be the strategy with the lowest score surprisingly violated the expectation. The plausible explanation was because tweens could not recognize or appreciate they were being given esteem support as adults. However, it was odd that 9th graders, who should have found it easier to recognize the esteem support, did not rate the account more positively than 7th graders. Due to the situation, there was another possible explanation that the designed messages of the account strategy were viewed as inaccurate. Secondly, for the age variable, the results showed there were significant effects for affect improvement and caring and indicated the 5th graders evaluated all strategies more positively than other age groups. It was against to the hypothesis that higher graders would evaluate sympathy and account more positively, and rate optimism and minimization less positively. It was argued that it was possible that higher graders were stringent critics of the comforting strategies or because of the designed content which was not as typical as the theoretical one. Thirdly, for gender variable, a significant effect for the caring indicated that girls evaluated all strategies more positively than boys. It indicated that the hypothesis that girls only evaluated some strategies more positively than boys was not supported. The main effect might be explained by assuming that girls tended to imagine “friends” delivering comfort as another girl, 29.

(40) whereas boys imagine other boys as comforters. Traditionally speaking, female support providers are often considered more caring and empathetic. Furthermore, as for the variable of situation types, a significant effect affect improvement showed that test situations were evaluated more positively than social rejection situations. The result was in consistent with the hypothesis, showing that the companionship strategy was indeed a better match to the social rejection situation. Previous research has not offered any explanation for the result; therefore, Clark, MacGeorge & Robinson (2008) claimed that it was because of the compensatory feature of the companionship strategy that the lost acceptance and inclusion were restored. Although acknowledging several limitations such as the lack of pretest, the seemingly unreal outcome, the study still offered practical and theoretical conclusions. For the theoretical side, the study found adults might enhance adolescents and children ability to provide comforts by the strategies of sympathy and advice, and that they should let kids know why some strategies might not work. The theoretical conclusion lay in the different conceptualizations of effective comforting strategies between children and adults. 2.3.5 Summary Table 2-3 presents the major findings and limitations of the four empirical studies reviewed in this section.. 30.

(41) Table 2-3. Major Findings and Limitations of the Previous Studies Major Findings. Limitations. Burleson & 1. Favored types of comforting 1. Strategy categorization: Only a Mortenson strategies: HPC and solace psychological framework (2003) 2. Factors affecting strategy 2. Tasks: Evaluation tasks on evaluation: Culture and Value & comforting and cultural goal orientations differences 3. Factors not included: Social power and severity of problems Xu (2007). 1. Favored type of comforting strategies: LPC 2. Factors affecting strategy employment: Power and gender. 1. Strategy categorization: Only a psychological framework 2. Task: Only a written DCT 3. Factors not included: Severity of problems. Burleson (2008). 1. Favored type of comforting strategies: HPC 2. Factors affecting strategy evaluation: Person centeredness of comforting strategies7. 1. Strategy categorization: Only a psychological framework 2. Task: Only on evaluation of comforting strategies 3. Factors not included: Social power and types of situations. Clark, MacGeorge, & Robinson (2008). 1. Favored type of comforting strategies: Companionship 2. Factors affecting strategy evaluation: Age, gender, and types of situation. 1. Strategy categorization: Only a psychological framework 2. Task: Only a message evaluation task 3. Factors not included: Social power and severity of problems. Generally speaking, comforting strategies varying from moderate to high personcenteredness (MPC & HPC strategies) and solace behavior were favored, and comforting strategies of low person centeredness were less favored (Burleson & Mortenson 2003, Burleson. Burleson (2008) in fact investigated other factors like people’s perception of emotional supporting skills, selfconcept, the helper’s gender, and severity of problems; however, they have not been proven to be the main factors that affected people’s evaluation of comforting strategies. 31 7.

(42) 2008, Clark, MacGeorge & Robinson 2008), indicating that strategies explicitly acknowledging and recognizing addressees’ feelings were more appropriate and preferred. However, this universality did not appear to work for Chinese culture: Burleson & Mortenson (2003) discovered that Chinese have positive evaluations on both solace and avoidant strategies, and Xu (2007) found that LPC strategies were mostly used among native Chinese speakers. These differences proved that cultural and value orientations influenced the perception and production of comforting strategies. As for other factors affecting employment or evaluation of comforting strategies, age was a factor that influenced perception of comforting (Clark, MacGeorge & Robinson 2008). Gender differences appeared to be another influential factor for both strategy employment and strategy evaluation, showing that females evaluated strategies more positively than males, and that they preferred to use different strategies (Xu 2007, Clark, MacGeorge & Robinson 2008). Moreover, the higher frequency of using HPC strategies when facing comfortees of higher power indicated that social power was a critical factor (Xu 2007). Lastly, types of situations would cause different preferences for strategy evaluation (Clark, MacGeorge & Robinson 2008). As for the limitations, firstly, these studies simply categorized comforting strategies according within a psychological framework under constructivism, suggesting the lack of analysis from a linguistic perspective. That is, these studies only focused on the function of strategies from a psychological perspective but ignored the form of comforting strategies from 32.

(43) a linguistic perspective. Therefore, in order to discuss both the form and function of comforting strategies, the present study analyzed comforting strategies from both the psychological and linguistic aspects. In addition, although a written DCT (Xu 2007) is a widely used tool to elicit data, it has often been criticized for lacking authenticity. To improve naturalness, the present study adopted a spoken DCT. Moreover, these studies merely looked into either perception or production tasks. To avoid the task bias, the present study included both tasks. Lastly, in terms of the factors affecting strategy evaluation and employment, although most of the empirical studies did not include the factor of severity of problems, its significance should never be ignored. It was found that research on speech act acquisition has often included severity of problems as one of the affecting factors (cf. Lin 2010, Hsu 2016). As a result, the present study included types of situations and severity of problems as the moderating factors. 2.4 Summary of Chapter Two In this chapter, we firstly defined the act of comforting from two perspectives: psychological and linguistic. From a psychological one, we introduced a three-leveled fundamental framework under constructivism; from a linguistic one, we presented a classification based on the linguistic features of strategies. This can serve as the reference in the following data analysis. Section 2.2 discussed cultural differences between Western and Eastern cultures through Hall’s (1976) and Hofstede’s (1980, 1991) cultural taxonomies. Lastly, in Section 2.3 we reviewed several empirical studies, and found that some factors including 33.

(44) types of situations and social power indeed exerted profound influence on the strategy employment or evaluation, indicating that these factors could also be taken into consideration in the research design. In the following chapter, the research design of the study that based on the research questions stated in Chapter One will be introduced.. 34.

(45) CHAPTER THREE RESEARCH DESIGN. Chapter Three presents the research design of the present study examining the perception and production of Chinese comforting strategies by English- and Japanese-speaking learners of Chinese as a second language. Section 3.1 introduces the basic information of the participants. Section 3.2 describes the methods and the materials, and Section 3.3 reports the experimental procedures of the study. Finally, a brief summary of this chapter is given in Section 3.4. 3.1 Participants A total of 49 participants were recruited to participate in the present study. They were further divided into two experimental groups, and a control group. The two experimental groups consisted of 29 learners of Chinese as a second language (CSL), whose native languages are English and Japanese respectively. Among the 29 subjects, 12 were English and 17 were Japanese native speakers. The control group comprised of 20 native Chinese speakers from Taiwan. The participants for the experimental groups were recruited from the Mandarin Training Center (MTC) in National Taiwan Normal University (NTNU). It is currently the largest Chinese language center in Taiwan with around 1,700 students enrolled each academic year. Classes at the MTC are divided into 9 levels, from the beginner levels to the advanced levels. 35.