系統功能語法教學於記敘文連貫性之影響

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. 系統功能語法教學於記敘文連貫性之影響. The Impact of the Instruction Informed by Systemic Functional Linguistics Theory on the Coherence in Narrative Writings. 指導教授：張珮青 Advisor: Dr. Peichin Chang 研究生：楊介萱. 中華民國一百零六年七月 July 2017.

(2) 摘要本研究旨在探討系統功能語法於記敘文教學的應用，及其對文章連貫性之影響，以增加學生文章連貫性及修辭架構的完整性。修辭架構及主位推進為文章連貫性指標且皆為系統功能語法的重要概念，台灣學生作文常常出現連貫性的問題，然而寫作教學現場仍偏重單字、文法的解說。因此本研究藉由觀察高中生接受系統功能語法教學後的改變，期望能提供不同於往的教學方式。參與學生為北部一所公立高中高二生，男女各十五位，總共三十位。進行十堂課的系統功能語法教學之後探討此教學法對學生看圖寫作文章的總分高低和連貫性的影響。其中連貫性主要透過分析前後測文章在主位選擇、主位推進和修辭架構的差異而定。研究結果顯示: (一)學生看圖寫作的總分於教學前和教學後有顯著進步， (二)多數學生教學後無標記主位的選擇變多元，(三)後測文章中標記主位的連接詞使用增加，(四)不論前測或後測，學生頻繁使用主位同一型推進模式、(五) 教學後修辭架構更趨完整，有助增加文章連貫性。綜上所述，本文認為系統功能語法對台灣高中生記敘文寫作有正面影響，文步及主位推進教學有助增進學生文章的連貫性。然而此教學法在台灣仍尚未普及，因此仍需更多的教學演示、教案分享、師資訓練和此類研究。本文最後提供教學建議和活動學習單，希望能讓第一線教師更明確知道如何將系統教學法應用於寫作教學中並為相關研究作出些許貢獻。關鍵字:系統功能語法、連貫性、主位選擇、主位推進、修辭架構、記敘文. i.

(3) ABSTRACT The aim of the present study was to assess whether the instruction of TCs (Thematic Choices), TPs (Thematic Progressions) and rhetorical structures assisted English as Foreign Language (EFL) senior high school students in producing coherent and well-organized narrative texts. In the present study, TCs refer to the participants in unmarked themes and logical connectors in marked themes. The connections of themes and rhemes across sentences are TPs. Rhetorical structures refer to a particular rhetorical or linguistic pattern and stage conventionally found in similar texts. Rhetorical structures and TPs are indispensable indicators of coherence and important concepts in Systemic Functional Linguistics (SFL) theory. Students in Taiwan often have difficulty producing coherent texts given that most of the teachers lay emphasis on vocabulary and grammatical problems rather than discourse-level ones. The current research investigated the learners’ performance of TCs, TPs and rhetorical structures in picture writings. The participants were thirty 11th graders, composed of fifteen males and fifteen females, from a public senior high school in Northern Taiwan. The intervention consisted of ten classes of instruction. The students’ performances in TCs, TPs, and rhetorical structures were analyzed to understand whether they made progress in coherence after the instruction. A paired-T test was conducted to determine whether the total scores improved in the posttest.. ii.

(4) Pretest and posttest writings of the three learners, whose pretest scores were the lowest, in the middle and the highest, were further analyzed to verify the findings. Several results were reported: (1) learners’ posttest scores significantly increased. (2) Generally, the learners produced coherent writings after the instruction informed by SFL. Most of the students diversified TCs in posttest writings. That is, the number of logical connectors in marked theme positions increased on posttest, which enhanced coherence. (4) The most frequently used TP type was constant TP in both pretest and posttest writings, but more TP types were employed to increase coherence in the posttest writings. (5) Most of the rhetorical moves were identified and well-established after the intervention, which helped create coherence. In conclusion, the present research found that the instruction helped improve coherence in the 11th graders’ picture writings. Nonetheless, the generalizability of these results was subject to certain limitations. For instance, the intervention merely lasted 3 weeks and more time was recommended to allow sufficient practice. To implement this approach, however, extended professional development is needed to familiarize teachers with the instruction informed by SFL theory.. Keywords: Systemic Functional Linguistics, coherence, thematic choices, thematic progressions, rhetorical structures, narratives. iii.

(5) ACKNOWLEDGMENT Were it not for the assistance and support of many respectful people throughout the period of my study, this work would not have been possible. First and foremost, I would like to express my profound gratitude to my advisor, Dr. Peichin Chang. I am extremely grateful for her instruction and guidance both in class as my teacher, and after class as my advisor. She helped me refine my research design when she knew that I could do better before I knew it. She pointed out the directions for improvement when I thought I had done a satisfying job. She provided valuable advice to help me remove the doubts, and challenged me to be a better writer. To her, I owe a great deal. My gratitude also goes to my thesis committee members- Dr. Chiou-lan Chern and Dr. Ming Huei Lin. It was particularly kind of them to patiently read my work and offered words of encouragement. Their unsurpassed knowledge and insightful comments considerably enrich my work. Next, I am indebted to the professors at NTNU, where I have cultivated my lifelong passion for the field and keen interest in learning English since my college years. With guidance and encouragement from my teachers, I have found my teaching and learning journey not merely worthwhile, but also enjoyable and enlightening. As a matter of fact, I am so proud of being able to learn from the professors at NTNU, who have taught me more than I could ever give them credit for here.. iv.

(6) Last but not least, I am thankful for the unconditional love from my family. Without the kindly help from my parents, parents-in-law and sister, I might not be able to further my education and professional development. Also, my son’s company while I worked on my thesis has made the experience especially delightful. Most important of all, I would like to thank my husband, Chi-kai, who has strengthened my belief in myself along the way and savored every moment of our student lives. I am blessed to have such a loving family. I could not have imagined having better teammates during my studies without them.. v.

(7) TABLE OF CONTENTS ABSTRACT (Chinese)………………………………………………………………...i ABSTRACT (English)………………………………………………………………...ii ACKNOWLEDGEMENT……………………………………………………………iv TABLE OF CONTENTS……………………………………………………………..vi LIST OF TABLES………………………………………………………………..……x LIST OF FIGURES……………………………………………………………...…...xii CHAPTER ONE-INTRODUCTION………………………………………………… 1 1.1 Background…………………………………………………………………. 1. 1.1.1. The Discourse Level Writing Problems…………………………….. 2. 1.1.2. Coherence…………….…………………………………………….. 3. 1.1.3. Coherence and SFL…………………………………………………. 3. 1.2 Purpose of the Study……………………………………………………….. 6. 1.3 Significance of the Study…………………………………………………… 8 1.4 Structure of the Thesis……………………………………………………. 9. CHAPTER TWO-LITERATURE REVIEW…………………………………………11 2.1 Theoretical framework: Systemic Functional Linguistics (SFL)………...…..12 2.1.1. Genre Theory from the SFL Perspective……………………………..18 Genre-based Instruction…………………………………………….21. vi.

(8) Narrative Genre……………………………………………………..23 2.1.2. Ideational Metafunction.……………………………………………..25 Participants, Processes and Circumstances…………………………25. 2.1.3. Textual Metafunction………………………………………………...29 Theme and Rheme……………………………………………….....29 Thematic Progressions……………………………………………...31. 2.2 Text Analysis: Investigating Thematic Progressions (TPs) and Thematic Choices (TCs) in Student Writing………………………………..…………..36 2.3 Empirical Studies: The Application of SFL Theory in Language Classrooms……………………………………………………..…………….41 2.3.1. The Effect of Instructing TPs and TCs……………………………….42. 2.3.2. The Effect of Instructing Rhetorical Moves………………………….45. 2.3.3. Integrated Approach of Instructing Rhetorical Moves, TPs and TCs..50. 2.4 Summary………………………………………………………..……………52 CHAPTER THREE-METHODOLOGY…………………………………………….57 3.1 Participants……………………………………………………………..…….57 3.2 Procedure……………………………………………………………….…….58 3.2.1. Stage One………………………………………………………...…..61. 3.2.2. Stage Two………………………………………………………...…..62. vii.

(9) 3.2.3. Stage Three…………………………………………………….....…..65. 3.3 The Grading...……………………….………………………………….…….66 3.4 The Analysis...………………………………………………………….…….70 3.4.1. Evaluation of the Rhetorical Structures…..……………………...…..71. 3.4.2. Evaluation of Thematic Progressions (TPs)……………………...…..74. 3.4.3. Evaluation of Thematic Choices (TCs)…...……………………...…..75. CHAPTER FOUR-RESULTS ...................................…….……………...………….79 4.1 The Pretest and Posttest Picture Writing Scores...……………………..…….79 4.2 Coherence in the Learner Writings….…………………............……….…….83 4.2.1. TC in the Learner Writings………….................................……...…..83 Evaluation of Participants in Unmarked Themes.…….................…..84 Evaluation of Logical Connectors in Marked Themes...................…..87. 4.2.2. TP in the Learner Writings…….………………………………….….91. 4.2.3. Rhetorical Moves in the Learner Writings..…………..…………...…95. 4.3 The Learners' Writing Examples...............................................................…...99 4.3.1. Henry.............................................................................................….100. 4.3.2. Natalie................................................................................................109. 4.3.3. Laurence........................................................................................….116. CHAPTER FIVE-DISCUSSION AND CONCLUSION….………………......……129. viii.

(10) 5.1 Discussion...........................…………………………………………..…….129 5.1.1. Learners' Significant Score Gains after the Instruction of TCs, TPs, and Rhetorical Structures…………..........................................……...….131. 5.1.2. The Variety of Participant Types in Unmarked Themes after the Instruction…….………………………………….........................….133. 5.1.3. The Variety of TP Types in the Posttest Writings..........................….135. 5.1.4. Considerable Improvement in Applying the Obligatory Rhetorical Moves after the Instruction............................................................….136. 5.1.5. Similarities and Differences in the Picture Writings of Henry, Natalie and Laurence.................................................................................….138. 5.2 Conclusion………………......................………………………….………...143 5.2.1. Major Findings of the Study………………………………………...143. 5.2.2. Pedagogical Implications…………………………………………...144. 5.2.3. Limitations of the Study………………………………....……...…..145. REFERENCE….……………………………………………………………………149 APPENDIX….……………………………………………………...………………165 A Teaching Material of the Instruction Program………………………..…….165 B The GSAT Scoring Rubrics in 2014 (Chinese Version)..…………...……...191. ix.

(11) LIST OF TABLES. Table 2.1 The Ideational, Interpersonal, and Textual Metafunctions………………...15 Table 2.2 The Contrast between SFL and Traditional Grammar……………………..17 Table 2.3 The Participants, Processes, and Circumstances……………………...…..26 Table 2.4 Constant TP with the Themes in Bold……………………………………..32 Table 2.5 Linear TP with the Themes in Bold and Picked-up Rhemes Underlined….33 Table 2.6 Derived Hyper TP with the Topic in Bold and the Supporting Details Underlined………………………………………………………………...35 Table 3.1 The Weekly Schedule of the Intensive Writing Program………………….60 Table 3.2 The Scoring Rubrics (Adapted from College Entrance Exam Center in 2014)………………………………………………………………………67 Table 3.3 An Example of the TP analysis in a Student Writing……………………...74 Table 4.1 The Results of Paired-T Test on the Pretest and Posttest Picture Writing Tasks Scores……………………………………....................................…81 Table 4.2 Participants in Unmarked Themes of the EFL Senior High School Learners’ Pretest and Posttest Picture Writings……………………………...………84 Table 4.3 Logical Connectors in Marked Themes of the EFL Senior High School Learners’ Pretest and Posttest Picture Writings………...………..………..88 Table 4.4 Constant TP, Linear TP, and Derived Hyper TP in the EFL Senior High School Learners’ Pretest and Posttest Picture Writings……………..…….92 Table 4.5 The Use of Rhetorical Moves in the Participants’ Pretest and Posttest Writings………….....................................................................…………...95 Table 4.6 The TC, TP and Rhetorical Move Analysis of Henry’s Pretest Picture Writing…………………...............………….. ...............…………...…...100 Table 4.7 The TC, TP and Rhetorical Move Analysis of Henry’s Posttest Picture Writing………...............………….. ...............…………...……………...104 Table 4.8 The TC, TP and Rhetorical Move Analysis of Natalie’s Pretest Picture Writing……………...............………….. ...............…………...………...109 Table 4.9 The TC, TP and Rhetorical Move Analysis of Natalie’s Posttest Picture Writing…………...............………….. ...............…………...…………...112 Table 4.10 The TC, TP and Rhetorical Move Analysis of Laurence’s Pretest Picture Writing…………………...............………….. ...............…………...…...117 Table 4.11 The TC, TP and Rhetorical Move Analysis of Laurence’s Posttest Picture Writing……………...............………….. ...............…………...………...121 Table 5.1 The TC, TP and Rhetorical Move Analysis of the Three Learners’ Pretest x.

(12) and Posttest Picture Writings……………...............…………...………...139. xi.

(13) LIST OF FIGURES. Figure 2.1 The Model of Language: Genre, Registers and Metafunctions (Adapted from Bawarshi & Reiff, 2010)……………………………………….…. 14 Figure 2.2 The Four Stages of Genre-based Instruction (Adapted from Feez, 1998)..22 Figure 3.1 The Picture Prompts of the Pretest………………………………………..58 Figure 3.2 The Picture Writing Organization Framework (Adapted from Labov & Waletzky, 1997)…………………………………………………………...64 Figure 3.3 The Picture Prompts of the Posttest (Retrieved from http://tammy0103.pixnet.net/album/set/14414732 )....................................65 Figure 4.1 Participants in Unmarked Themes of the Pretest and Posttest Picture Writings……………………………............................................................86 Figure 4.2 Logical Connectors in Marked Themes of Pretest and Posttest Picture Writings………………...……….................................................................89 Figure 4.3 TPs in the Pretest and Posttest Writings………..............………………...94. xii.

(14) CHAPTER ONE. Introduction. 1.1 Background Writing plays an indispensable part in communication and reflects writer’s logical thinking. In Taiwan, an English as a Foreign Language (EFL) context, writing has received increasing attention on the high-stake assessments, such as General Scholastic Aptitude Test (GSAT), Department Required Test (DRT), and General English Proficiency Test (GEPT). As the college entrance exam, GSAT is the test almost every Taiwanese senior high school student takes. Scoring high on the GSAT composition section is thus one of the goals students endeavor to reach. On GSAT, the writing section consists of translations and an essay, which account for eight and twenty points respectively. The topics of the essay range from writing a letter to picture description and the latter is the more common one in recent years. Picture description, or picture writing, usually comprises a series of picture prompts or a comic strip. Test-takers are required to create a story from the visual stimulus by describing details in the pictures. According to previous research (Chen, 1.

(15) 2012; Lin, 2006; Wright, 1996), the prompts provided storylines to assist unskilled writers. Familiarizing the senior high students with the frequent topics on GSAT might be the first step to help them excel on the GSAT writing section. Therefore, polishing students’ descriptive skills might be one of the teaching foci in Taiwanese senior high schools.. 1.1.1 The Discourse Level Writing Problems EFL learners, nevertheless, often find writing one of the most difficult skills to develop and sharpen. The problems that learners encounter include word-level problems, such as word usages, sentence-level problems, such as different syntactic structures, and discourse level ones (Chen, 2002). According to Chen (2002), it is the discourse level that challenges a great number of students. For instance, students have to familiarize themselves with different organizations in English writing, unlike traditional Chinese writing structure, i.e. qi, cheng, zhuan, he. Traditional writing pedagogy in EFL contexts lays considerable emphasis on vocabulary and grammar instruction. Also, comments from teachers often revolve around vocabulary and grammatical errors in traditional methods (Albufalasa, 2013; Yasuda, 2015). Coherence and organization might not have received sufficient attention from teachers and learners. As a result, students often produce 2.

(16) grammatically correct sentences but find it difficult to write a coherent and well-organized essay (Liu & Liu, 2013).. 1.1.2 Coherence According to Danes (1974) and Lee (2002), coherence is defined as the use of coherence-creating devices, including thematic progressions (TPs), connecting devices between the sentences and the transition between the paragraphs. To establish coherence, it is of crucial importance for writers to order ideas in a logical sequence and connect themes and rhemes, which reflects the continuity of TPs (Danes, 1974). Moreover, examining the coherence of a text will also investigate the disposition of ideas, i.e. the organization of texts. Coherence and organization are inseparable (Parsons, 1991). Organization is considered the way different parts of a system are arranged and work together. That is, organization refers to how writers present ideas to convey meanings across texts. Researchers have identified the rhetorical actions writers take in well-organized texts and advocated the instruction of rhetorical structures to improve organization and then coherence in student writings (Chen & Su, 2012; Christie & Dreyfus, 2007; Henry & Roseberry, 1998).. 1.1.3 Coherence and SFL 3.

(17) To help learners increase coherence in writings, many researchers have suggested adopting the Systemic Functional Linguistics (SFL) approach as a comparatively new pedagogical method in writing instruction (Henry & Roseberry, 1998; Liu & Liu, 2013; Mellos, 2011; Wang, 2007). They emphasized the social purposes of texts and how language is linked to contexts systematically. The research focus is on the function of language, which is how languages are put together to communicate meanings. According to Halliday and Hasan (1976), coherence can be achieved at two levels, i.e. at the discourse level and at the sentence level. First, at the discourse level, writers can produce coherent texts by including the rhetorical structures. SFL theorists argued that particular patterns of linguistic choices represented a certain type of text (Brown & Marshall, 2012; Cheng, 2008; Christie & Dreyfus, 2007; Henry & Roseberry, 2001; Martin, 2003). According to Hyland (2002), Derewianka (2004), and Crombie and Johnson (2008), the rhetorical actions writers use in similar texts are referred to as genre. Genre-based research has been conducted to describe rhetorical moves in different genres, including narration, commentary, and exposition. In 1997, Labov and Waletzky identified the six rhetorical moves in narratives, which were Abstract, Orientation, Complication, Evaluation, Resolution, and Coda. Picture description, which requires students to write a story, belongs to narrative genre. 4.

(18) Thus, the present study adopted the rhetorical structures and devised a picture writing framework based on Labov and Waletzky (1997). Having identified rhetorical moves in genres, researchers then proposed genre-based instruction with a purpose of raising learners’ awareness to rhetorical structures, grammatical and cohesive choices (Bhatia, 1999; Brown & Marshall, 2012; Christie & Dreyfus, 2007; Derewianka, 2003; Henry & Roseberry, 1998; Pang, 2002). Adopting genre-based approach in writing instruction has been widely advocated in the field of SFL to help learners improve the rhetorical structures and coherence of texts. Coherence can also be created through the use of theme and rheme at the sentence level. To examine coherence in texts, researchers employ theme-rheme as the analytical framework. Observing TCs and TPs gives readers an insight into the coherence of texts and how TCs and TPs function in different genres. Over the years, scholars have applied the theme-rheme analytical framework in examining a variety of speeches and writings, such as academic writings (Bloor & Bloor, 1992; Ebrahimi & Khedri, 2011; Fetzer, 2008; Gao, 2012; Hawes & Thomas, 2012; Jalilifar, 2010; Lores, 2004; Lovejoy, 1998; Martínez, 2003; North, 2005; Wang, 2007). One line of SFL research has been exploring theme-rheme in learner writings. To identify problems of thematisations, which may compromise coherence, researchers have designed comparison studies between different groups of writings, including 5.

(19) non-native learner writings and native-speaker writings, learner writings and proficient writings, together with writings produced by high-proficiency groups and writings of low-proficiency groups (Cheng, 2008; Christie & Dreyfus, 2007; Green, Christopher & Lam, 2000; Hawes & Thomas, 1997). As several scholars have suggested, it is necessary to incorporate SFL into writing pedagogy to increase coherence in learner writings (Green et al., 2000; Hawes & Thomas, 1997; Mellos, 2011; Vande Kopple, 1991).. 1.2 Purpose of the Study The present research examined coherence by analyzing the learners’ use of TPs, TCs, and rhetorical moves in picture writings with a view to extending previous studies in three directions. First, some researchers have sought to uncover the efficacy of SFL instruction mostly in the first language (L1) contexts (Fetzer, 2008; Lovejoy, 1998; North, 2005; Vande Kopple, 1991) and more recently, but still few, in second language (L2) contexts (Cheng, 2008; Liu & Liu, 2013; Siahaan, 2013). Liu and Liu (2013) claimed that incorporating SFL theory into writing pedagogy will help increase coherence in learner writings and students’ writing proficiency. Yet, more empirical evidence is needed to examine the impact of applying SFL theory in English as a Second Language (ESL) and EFL language classrooms (Jing, 2014; Wang, 2007). To 6.

(20) respond to the need of more empirical studies conducted aside from L1 contexts, the current study implemented the instruction informed by SFL theory in Taiwan, an EFL context. Second, many studies have focused on college students (Green et al, 2000; Herriman, 2011; Hu, 2008; Liu & Liu, 2013; Ma, 2001; Wang, 2010). Few have examined the incorporation of genre theory and theme theory in EFL senior high school language classrooms. One of the learners’ writing difficulties is the inappropriate use of TCs and TPs, such as overuse of logical connectors and lack of TPs in student writings, which tended to decrease coherence (Green et al., 2000; Liu & Liu, 2013; Vande Kopple, 1991; Wang, 2007). The researchers agree that teaching learners to enhance the quality of TCs and TPs will help students produce coherent writings (Christie & Dreyfus, 2007; Green et al., 2000; Hawes & Thomas, 1997; Mellos, 2011; Vande Kopple, 1991). Thus, the present study aims to explore how the instruction informed by SFL theory improves coherence in EFL senior high school student writings. Although many researchers have proposed that SFL-based instruction could be applied in writing pedagogy, few studies have developed and provided teaching materials to offer concrete information regarding how language teachers can implement theme theory in L2 writing pedagogy (Bonhnacker, 2010; Mellos, 2011). 7.

(21) The current research employed genre-based approach and devised a three-week intensive picture writing instruction in an EFL context. The present study attempts to contribute to SFL research with empirical evidence by employing complete teaching materials in Taiwanese senior high schools. Based on the above-mentioned objectives, the research questions for the current study are as the following: 1. Do the 11th graders’ overall picture writing scores increase after the instruction of TC, TP and rhetorical moves? 2. Does coherence improve in the following three aspects after the instruction of TC, TP and rhetorical moves? 2.1 Thematic Choices (TCs) 2.2 Thematic Progressions (TPs) 2.3 Rhetorical Moves. 1.3 Significance of the Study The present study aims to help senior high school students in Taiwan produce coherent picture writings. The intensive instruction program introduced rhetorical moves in picture writings, TPs, and TCs to produce coherent essays. The study intended to add to the existing knowledge by combining two important concepts, rhetorical structures and theme-rheme analytical framework, in EFL writing pedagogy. 8.

(22) It is hoped that the findings of the study will provide insights into how SFL-based instruction improves coherence, starting from the sentence level and then the discourse level, in EFL senior high school student writings.. 1.4 Structure of the Thesis Chapter 2 consists of two sections. The first one is dedicated to introducing the theoretical foundation of the present research, SFL theory with a specific focus on narrative genres, TCs, and TPs. The second section of the chapter reviews previous research examining TCs and TPs in learner writings, and the efficacy of explicitly teaching TCs, TPs and rhetorical moves. With the hope of helping the EFL learners increase coherence in their picture writings, a three-week intensive instruction was devised for the study. Chapter 3 elaborates on information concerning the participants, procedure with details of the three teaching stages in the intensive instruction program, grading and analysis. The chapter also illustrates the instruments employed in the pretest and posttest, and explains how the pretest and posttest data are graded and analyzed. Chapter 4 is devoted to the report of results and findings. To present a complete picture, quantitative statistical data are provided besides the qualitative analysis of TCs, TPs, and rhetorical structures in the students' picture writings. The chapter ends 9.

(23) with further analysis of three learners’ writings to support and verify the findings. Finally, Chapter 5 is divided into two sections, i.e. Discussion and Conclusion. The first section begins with the discussion of major findings and then proceeds to the comparison of the three learners’ writings. The second section starts with the summary of the major findings, followed by pedagogical implications. The chapter ends with limitations of the study and suggestions for future research. Lastly, complete teaching material of the writing program is included in the appendix.. 10.

(24) CHAPTER TWO. Literature Review. Under the traditional form-focused teaching in Taiwan, which mainly revolves around vocabulary and grammar instruction, the problems frequently found in student writings are lacking coherence and organization (Chen, 2002). To address the issue, the instruction informed by Systemic Functional Linguistics (SFL) approach has risen as solutions since 1960s. Genre theorists advocated familiarizing learners with discourse knowledge, such as rhetorical structures of genres, and linguistic features associated with genres, including TCs and TPs, to improve coherence (Cheng, 2008; Christie & Dreyfus, 2007; Liu & Liu, 2013; Mellos, 2011). The present study incorporated genre-based approach into the instruction program to teach rhetorical moves in narratives, specifically the special genre on GSAT in Taiwan, i.e. picture writings. Aside from drawing learners’ attention to the overall structure of picture writings, linguistic features shared in narrative texts, including theme-rheme and TPs were taught to help students connect sentences in texts. 11.

(25) Chapter 2 consists of four sections. Section 2.1 addresses the theoretical foundation of the study, i.e. SFL theory. Specific emphasis is put on the rhetorical structure of narrative texts, ideational organization, dealing with the word choice in narratives at the word level, and textual organization, focusing on the connection of sentences across texts. Section 2.2 reviews existing research investigating TPs and TCs in student writings. The focus of Section 2.3 is on the previous studies that incorporate SFL approach into writing pedagogy and Section 2.4 is the summary of the literature review.. 2.1 Theoretical framework: Systemic Functional Linguistics (SFL) Halliday, McIntosh and Strevens proposed Systemic Functional Linguistics (SFL) in 1964. According to Halliday (1994), the conceptual framework of SFL is functional instead of formal. The focus of SFL is on functions of languages, which is how language is used to convey different meanings, rather than on language forms to regulate what people can or cannot say. Unlike traditional grammarians who merely examined texts word by word, SFL researchers lay emphasis on how words, sentences and paragraphs are related to each other. Enlarging the scope by investigating the relationship between language and its social functions helps researchers view texts as a whole from a functional-semantic approach rather than a formal-syntactic one. They 12.

(26) also explored how language use varied in different social situations to achieve communication purposes. Thus, different from the form-focused traditional grammar, SFL is meaning-based, emphasizing functions of the language. Some scholars suggested that writers make systematic linguistic choices to construct texts of different genres, so language is not merely “a set of abstract rules” (Thompson, 2004, p.6). Texts are influenced by social functions, like writing to narrate, explain, or argue (Lin, 2017; Schleppegrell, 2004). Two types of contexts, i.e. context of culture (genre) and context of situation (register), play a significant role in text formation to achieve coherence (Albufalasa, 2013). The context of culture influences the rhetorical moves that characterize genres. Typical language is used in each context of situation, and the context-specific language use is termed register. They are field, which explains what takes place, tenor, describing how the subjects are related to each other, and mode, stating what channel is used to communicate (Eggins, 2004). The registers, field, tenor, and mode, which function at the level of context of situation, actually corresponds to what happens at the linguistic level, i.e. the three metafunctions, ideational, interpersonal, and textual, respectively. According to Halliday (1994), the three metafunctions shape and organize all languages. The ideational metafunction construes experiences, interpersonal metafunction explicates 13.

(27) relations, while the textual metafunction enables the first two to connect coherently. Genre, registers and metafunctions realize and relate to one another. In together, the three construct the model of language, as diagrammed in Figure 2.1, adapted from Bawarshi and Reiff (2010).. marks Context of culture. genre. is characterized by Context of situation (register) field. tenor. mode. interpersonal. textual. carries Language (metafunctions). ideational. Figure 2.1 The Model of Language: Genre, Registers and Metafunctions (Adapted from Bawarshi & Reiff, 2010, p.31). Figure 2.1 indicates that languages are influenced by contexts, such as different cultures and situations. Context of culture marks the features of genres while registers, i.e. field, tenor and mode, characterize context of situation. Registers contextualize language, which carries three metafunctions, i.e. ideational, interpersonal, and textual. And rhetorical moves of genres express the three meanings or metafunctions in language (Hyland, 2002). Specifically, the ideational metafunction of languages 14.

(28) describes writers’ experiences of the “real world” represented in languages. As Halliday (1985) contended, it reflected “our experience of the world that lies about us, and also inside us, the world of our imagination.” The ideational meaning is related to the field of the text and can be analyzed through transitivity patterns. Next, the interpersonal metafunction reflects relationships. According to Halliday (1985), it conveys “meaning as a form of action: the speaker or writer doing something to the listener or reader by means of language” (p.53). The interpersonal metafunction is related to the tenor of the text and can be identified through mood analysis. Third, the textual metafunction refers to the organization of the text and carries the meaning of “the relevance to the context: both the proceeding (and following) text, and context of situation” (Halliday, 1985). The textual metafunction is associated with the mode of communication. Furthermore, SFL researchers utilized the theme patterns to identify the textual metafunction. The SFL systematic analytical framework enables people to better understand the nature of languages (Christie, 2004; Eggins, 2004). Table 2.1 summarizes the three metafunctions.. Table 2.1 The Ideational, Interpersonal, and Textual Metafunctions Metafunction. Aspect of the Text. Analytical Tool. Ideational. Field. transitivity patterns 15.

(29) Interpersonal. Tenor. mood analysis. Textual. Mode. theme patterns. One distinct difference between traditional form-focused grammar and SFL is that the former examines how words are organized in a sentence while the latter focuses on how language is put together to communicate meanings. For instance, every word in a sentence is analyzed and divided into different parts of speech in traditional grammar. Yet, in SFL, the English words in a sentence are not examined one by one, but grouped into four major categories, i.e. nominal, verbal, adjective and prepositional groups, and labeled based on the three metafunctions (Feng, 2013). The nominal group, usually the main characters of the sentence, is referred to as the participants, belonging to the ideational metafunction. It is also the theme, belonging to the textual metafunction. The rest of the sentence with the verb is the rheme, based on textual metafunctions. The verbal group that describes actions is the process in the ideational metafunction. The prepositional phrase is the circumstance, which provides further information of time, location, etc. Table 2.2 displays the contrast between the form-focused traditional grammar and SFL.. 16.

(30) Table 2.2 The Contrast between SFL and Traditional Grammar SFL. Example. Nominal group. Verbal group. Prep. group. (theme/participant). (rheme/process). (circumstance). Candace and. Traditional Noun. her. brothers were. Conj. Pron. Noun. BeV. chatting. in. the. park.. V-ing. Prep. Det. Noun.. Grammar. Table 2.2 shows that the linguistic units, i.e. words, chunks, and sentences, operate at different levels and form larger units in the example sentence. The word, “Candace”, forms the nominal group with “and”, “her”, and “brothers”. In traditional grammar, each word is analyzed and labeled according to its part of speech whereas SFL researchers group the words, which are referred to as chunks, or word groups, based on its functions in the sentence. In this case, the nominal group is the subject, which is the participant or theme of the sentence in SFL. The four words, “Candace”, “and”, “her”, “brothers”, in the nominal group are analyzed as a unit rather than separately. The rest of the sentence after the theme is the rheme. The word, “were” together with “chatting”, constitutes the verbal group, also termed as the process. Next, the prepositional group consists of three words, “in”, “the”, and “park”, which is the 17.

(31) circumstance of the sentence. Finally, the three chunks, “Candace and her brothers”, “were chatting”, “in the park”, constitute the sentence. Instead of going to the details to examine the nine words as traditional grammarians do, SFL researchers use the three chunks as the analytical units. The labeling of the units in a sentence is based on the communication functions rather than the parts of speech. According to Chandler (2007), genres are shaped by different contexts to serve different purposes. Writers use a series of linguistic choices, such as participants, processes, and circumstances, to build the context of situation. As a result, the participant and process types might vary in different genres (Correa & Domínguez, 2014). Furthermore, how themes and rhemes are connected across sentences, i.e. TPs, might also differ based on the genre type (Ebrahimi & Ebrahimi, 2012a; Jalilifar, 2010; Rafiei & Modirkhamene, 2012; Zhang & Li, 2009). The following sub-section elaborates on the generic structure before advancing to the thematic structure in SFL theory.. 2.1.1 Genre Theory from SFL Perspective Genre refers to different types of written texts (Johns, 1997). Within the context of studies on speaking and writing, some researchers make distinction between “genre” and “text type”, while others use the two terms interchangeably (Stubbs, 1996). Even 18.

(32) when researchers distinguish, “genre” from “text type”, the definitions vary in different studies. For instance, “genre” is defined as the socially constructed varieties of texts, such as poems and lectures, in Biber’s study (1988). However, in SFL, “genres” refer to the rhetorical actions that people use in similar texts, such as explaining and arguing (Crombie & Johnson, 2008; Derewianka, 2004; Hyland, 2002). Following the mainstream SFL researchers, the present study adopts the definition of the latter researchers and does not distinguish the two terms. As a result, in the current research, “genre” is defined as different rhetorical functional uses of sentences and grammatical features in written texts, inclusive of “description, narration, commentary, exposition,. exegesis,. explanation,. demonstration,. instruction,. argumentation,. persuasion” (Council of Europe, 2001, p.126). Thus, particular patterns of linguistic choices are mobilized to create a certain type of text. Analyzing genres by recognizing the patterns will help readers infer the genre of texts. Researchers lay different emphasis when analyzing texts, including focusing on the social role of texts and how texts are organized to achieve communicative purposes. Three broad schools are commonly identified in the field of genre theory. The first is the New Rhetoric group, which employs ethnographic methodologies to explore how genres imply writers’ attitudes, beliefs and values (Hyland, 2002). 19.

(33) The second perspective is based on Michael Halliday’s (1994) SFL theory, which emphasized the social purposes of genres, and how languages are linked to contexts systematically. Research in this area seeks to describe how rhetorical structures are developed and how lexical, grammatical together with cohesive choices are made to construct the functions of genres. Based on the theory, discourse structures are identifiable. The final orientation is the English for Specific Purposes (ESP) approach, which combines the two views. Heavily drawing on the SFL theory, it is also regarded as the application of SFL and the ESP scholars strive to extend research findings to pedagogical practices (Bloor, 1998; Hyland, 2002). Research of the three schools have mainly focused on academic genre, such as sections of research articles (Basturkmen, 2012; Brown & Marshall, 2012; Loi, 2010; Martin, 2003), and business genre, including business letters (dos Santos, 2002; Henry & Roseberry, 2001; Lai & Tseng, 2012). In particular, move analysis adopting the SFL approach has also identified a variety of macro genres, including narrative, recount, argument, and report. Researchers have promoted explicit genre-based instruction (Brown & Marshall, 2012; Cheng, 2008; Christie & Dreyfus, 2007). According to Cheng (2008), raising learners’ awareness to rhetorical structures of genres helps learners significantly improve in overall writing. The current study employed the SFL approach to teach the rhetorical moves and linguistic features 20.

(34) associated with genres. Further information regarding the genre-based instruction is provided in the following section.. Genre-based Instruction According to SFL genre theorists, incorporating genre-based instruction into writing pedagogy generally involves four stages (Feez, 1998; Gibbons, 2009; Kalan, 2013). As demonstrated in Figure 2.2, the focus of the first stage is building the field, which draws learners’ attention to the focused genre. The second stage is modeling the genre. Instructors present model writings of the genre that learners are going to produce in language classrooms and teach generic structure together with lexicogrammatical features of the focused genre. During the first two stages, learners deconstruct the texts, which might help them, especially EFL learners, build knowledge of the unfamiliar genre. The third phase is joint text construction, in which students discuss the purposes, rhetorical structures and lexicogrammatical features of the texts, such as the participant types commonly found in the genre. Then, they organize a text of the focused genre together by using the rhetorical structures and linguistic features. The last stage is individual construction, in which students have to practice their writing skills by composing an essay to demonstrate their understanding of the genre.. 21.

(35) building the field. modeling the genre. joint text construction. individual construction. Figure 2.2 The Four Stages of Genre-based Instruction (Adapted from Feez, 1998). Applying a genre-based approach in writing instruction helps learners analyze the organization of texts, and understand the relations, contexts, apart from functions of the rhetorical moves in texts (Lemke, 1994). As a result, researchers have advocated explicit instruction of genre in teaching writing (Bhatia, 1999; Derewianka, 2003; Hyland, 2002; Hyon, 1996; Paltridge, 2001; Pang, 2002). However, promising as the instruction seems to be, there is a lack of empirical evidence to support the effectiveness of genre-based instruction, in which learners are guided to use the rhetorical moves so as to produce organized texts. The existing research on employing genre-based approach in writing pedagogy attempted to investigate the effect of the instruction through text analysis and yielded inconclusive results. The current study employed the SFL approach to explore the rhetorical moves and lexicogrammatical features associated with genres, specifically, the narrative genre. Further information regarding the narrative genre is provided in the following section. 22.

(36) Narrative Genre According to Polkinghorne (1988), it is defined as the linking of human actions and events in a comprehensible composite. Picture writing, a frequent test type on GSAT, requires students to narrate a story based on the picture prompts. Thus, narrative is the target genre of the present research. Narratives are one of the most ubiquitous genres (Barthes, 1977). Narrators usually describe the changing events over time from a particular perspective. It is a representation of reality. Nevertheless, narrative is not merely outlines of events. Instead, narrative is subjective and will vary depending on how narrators select and sequence events (Tappan & Brown, 1991). With a purpose in mind, writers only include the event which is deemed necessary or important to the narrators, regardless of what actually happens in reality. Narrators reconfigure reality based on how they make sense of the experience and place incidents in a certain order to give them particular meanings (Bruner, 1986). A line of research has been devoted to investigating the narrative genre, such as writing personal stories and other peoples’ stories (Kormos, 2011). Social purposes and the contexts under which texts are produced determine text types (Schleppegrell, 2004). The overall purpose of narratives is “to entertain…to teach or inform, to embody the writer’s reflections on experience, and—perhaps most important—to 23.

(37) nourish, and to extend the reader’s imagination” (Derewianka, 2004, p.40). The structure of narratives might be shaped by different contexts, such as cultures, and narratives usually consist of beginning, middle and ending stages (Chandler, 2007). Studies within the SFL framework often include the six rhetorical moves, i.e. Abstract, Orientation, Complication, Evaluation, Resolution, and Coda, in fully formed narratives (Labov & Waletzky, 1997). According to Labov and Waletzky (1997), each stage has different functions. First, Abstract begins the narrative, and informs readers of what the text is about and why it is told. Orientation sets the scene and provides information regarding the main characters, location, and time. The main body is Complication, introducing a series of events or problems that characters have to solve. It creates tension or sparks readers’ interest. Next, Evaluation highlights “the relative importance of some narrative units compared to others” (Labov & Waletzky, 1997, p.32). Resolution follows evaluation and describes the outcome or how the issues are resolved. Coda returns “the verbal perspective to the present moment” (ibid, p.35), which states the changes of characters and what they have learned from the experience. Rhetorical structures and linguistic features vary from genre to genre. In addition to specifying rhetorical moves, SFL theorists identify particular language features and connecting devices, mainly theme and rheme, to help writers reconstruct the events 24.

(38) (Cheng, 2008). The next section introduces ideational organization, which focuses on the participants in unmarked theme positions.. 2.1.2 Ideational Metafunction In traditional grammar, transitivity refers to whether a verb takes an object, which is used to distinguish a transitive or intransitive verb. Researchers from the field of SFL enlarge the scope of transitivity, and employ the system of transitivity to examine the whole sentence. According to Halliday (1985), the system belongs to the Ideational, or experiential metafunction. It is through the system of transitivity that communicators share experiences of the external and internal world. Different linguistic choices, such as different words and sentence patterns, are made to convey the same meaning (Fowler, 1996). Examining the lexicogrammatical options in texts will reflect writers’ social stance and ideologies. The focus of SFL research is thus on the content of message rather than the grammatical forms.. Participants, Processes and Circumstances The researchers utilize the system of transitivity to analyze the linguistic units, which represent different experiences, i.e. “the actions, events, processes of consciousness and relations” (Halliday, 1985). To be more specific, the representation of reality in the SFL model consists of processes, 25.

(39) which is a semantic verb or the state that is being described; participants, i.e. the animate or inanimate noun phrase involved; and circumstances, usually the adverbial or prepositional phrases that specify where, when, how, and why the experience takes place. Transitivity is employed to describe the relationship between participants and processes as who (or what) does what to whom (or what). The functions and frequent types of processes, participants, and circumstances are presented in Table 2.3, adapted from Humphrey and Droga (2002).. Table 2.3 The Participants, Processes, and Circumstances SFL. Traditional. Function. Types. to help track the. people, objects,. Grammar 1) Participants. the repeating. nouns in a passage key elements. dummy, referent, intangible entity…. 2) Processes. the event that is. to show the action. material, mental,. described,. relational, verbal,. including the verb. behavioral, existential. 3) Circumstances the adverbial, or. to provide relevant 26. location, manner,.

(40) prepositional. facts of the event. cause, extent, destination…. phrases. As Table 2.3 shows, the participants are the repeating nouns and the themes of the sentences, which help tracking the key elements. There are many types of participants, including people, objects, time, place, dummy, referent, and intangible entity. Participant types characterize genre (Correa & Domínguez, 2014). In narratives, participants are often characters with distinct features (Cheng, 2008; Correa & Domínguez, 2014; Derewianka, 2004). Writers often use modifying clauses, adjectives and adverbs to describe participants (Fang, Schleppegrell, & Cox, 2006). The two types frequently appear in narratives are people participants, such as the male and the female as in “Andy and Elva felt astonished” and object participants, such as the antique in “The vessel looked very old.” Another participant type is the dummy, like “it” in “It was dangerous to leave children alone at home.” The circumstances are usually the adverbial or prepositional phrases in sentences, which provides further information of the event. For instance, in the sentence, “Tina and her boyfriend went to a chinaware exhibition at National Palace Museum,” the circumstance, “at National Palace Museum”, signifies the location of the date. Other types are manner, such as “politely”, cause, as in “due to 27.

(41) the typhoon”, destination, such as “to the airport”. The processes refer to the event being described, including the verb. The purpose is to show the action and characterize participants, such as what the participants do, what they think and how they feel. There are six main types of processes, i.e. material, mental, relational, verbal, behavioral and existential processes. The material process embodies a verb of doing or happening, which refers to the physical action of an entity or concrete depiction of the outer world. On the other hand, the mental process is concerned with the inner thoughts, encoded by verbs, such as “like”, “feel”, and “think”. A relational process, often includes verbs of being, becoming or having, and depicts abstract relation, such as possession, equivalence, attributes, and so on. Next, a verbal process is related to communication, linguistic expressions of human beings, or “any kind of symbolic exchange of meaning” (Halliday, 1985). It is signaled by verbs, like “suggest”, “ask”, and “nominate”. A behavioral process refers to the physiological or psychological performance while an existential process consists of “there” and a “be” verb, which describes the existence of an entity, action or event. Similar to participants, processes and circumstances in ideational metafunction, themes and rhemes also hold distinctive traits depending on genre types (Ebrahimi & Ebrahimi, 2012a; Jalilifar, 2010; Rafiei & Modirkhamene, 2012; Wu, 2003; Zhang & Li, 2009). The next sub-section introduces theme and rheme in textual metafunction. 28.

(42) Specific emphasis is laid on the three common ways to connect themes and rhemes in narratives, which are constant TP, linear TP and derived hyper TP.. 2.1.3 Textual Metafunction The textual metafunction manages the flow and the internal organization of texts. It helps serve the communicative purpose of texts and organize message both within and between clauses. The textual metafunction is related to mode and encompasses the theme and rheme analytical framework to create coherence in texts under the context of a certain situations.. Theme and Rheme To examine and improve coherence in texts, the analytical tool of the textual metafunction, i.e. the theme and rheme, should be first identified. Analyzing theme and rheme can provide a systematic way to investigate textual metafunction and the logical relation between a sentence and the larger context (Eggins, 2004). Theme is defined as the start of the message, before the main verb (Halliday, 1985). Theme takes the initial position, which reveals the prominence of the information. It serves as a point of departure for the development of a message and influences the readers’ interpretation of the text. Therefore, theme dictates the relations of the sentences and represents the ‘psychological subject’ or the ‘subject of 29.

(43) discourse’ in texts (Brown & Miller, 1991). Themes typically contain given information, which is mentioned earlier or frequently appears in the text. The rest of the sentence with the main verb is the rheme, which develops the topic and follows the point of departure to make a text coherent (Berry, 1989). Rhemes usually carry new information. For example, in the sentence, “I went home cheerfully,” “I” is the subject and the main character. It is also the theme, realized by a noun. The rest of the sentence, after the verb, “went”, is the rheme. It elaborates on the subject. There are two types of themes according to its markedness. If the theme is the grammatical subject, like the theme, “I”, in the previous example, it is unmarked; unmarked theme is often the norm. On the other hand, marked themes are often phrases that foreground unfamiliar information, as in the following sentence, “One day after class, I went home cheerfully.” The theme of the sentence is “I”, followed by the rheme, “went home cheerfully”. To be more specific, “I” is the grammatical subject, so it is the unmarked theme. The marked theme of the sentence is “One day after class”, which is not the grammatical subject. Rather, it creates the setting and specifies when the incident takes place in this case. Usually, marked themes can be an adverbial group, prepositional phrases and dependent clauses. According to Vande Kopple (1991), marked themes are more 30.

(44) unusual, compared with unmarked themes. Nevertheless, one of the reasons why writers choose marked themes at the beginning of a sentence is to set the stage for the succeeding texts. Another reason is to draw reader’s attention with its starting position. Also, marked themes can be transitional devices, or logical connectors, which make texts coherent. For instance, writers use “Besides”, “Moreover”, and “Furthermore”, to provide further information and elaboration (Hawes, & Thomas, 1997).. Thematic Progressions In texts, themes are connected to the themes and rhemes of previous sentences in numerous ways with a view to restating given concepts, continuing the development of ideas or providing new information. The connections or successions of themes across sentences form patterns of thematic progression, or TP (Danes, 1974). According to Halliday (1985), TP plays a vital role in the internal relation between sentences in texts and establishes coherence by informing readers of what the text is about and where the text is going. In narratives, writers usually make use of TP to connect sentences in texts and dispose events in a time sequence. There are three common TPs, i.e. constant TP, linear TP and derived hyper TP. First, when a writer uses constant TP, the themes of the sentences are the same, as exhibited in Table 2.4. The pattern, with the repeated predominant themes, can also be referred to as theme reiteration. 31.

(45) Table 2.4 Constant TP with the Themes in Bold Marked Theme. Theme. Rheme. (new information). (the subject and. (elaboration of the subject). main character) 1) 2). was mad that he yelled at the boy.. The driver. opened the door and took off the boy’s. He. earphones. 3). hit the boy angrily.. The driver. In the example, the theme, “The driver”, in the first sentence, is realized as the pronoun, “He”, in the second sentence. Then, the same theme appears again in the third sentence. Frequently found in narrative texts, constant TP draws readers’ attention to the repeating theme, which is usually the key concept (Wu, 2003). Nonetheless, it may also create a sense of monotony if the theme repeats continuously (Mellos, 2011). Without further development of the theme, the static text seems to be a stream of disconnected ideas (Enkvist, 1974). Another common flow is linear TP, or zig-zag TP, which provides a more dynamic effect. It refers to the type of flow in which the rheme of the first sentence is 32.

(46) taken up as the theme of the second sentence, as demonstrated in Table 2.5.. Table 2.5 Linear TP with the Themes in Bold and Picked-up Rhemes Underlined Marked Theme. Theme. Rheme. (new information). (the subject and. (elaboration of the subject). main character) 1) However,. the accidents. did not lead the brother and his sister to despair and sorrow.. They. 2). then found their interest in bees and earphones.. 3) Years later,. their interest. motivated them to publish the book, Bee-ctionary.. 4). Jack and Jill. won themselves the Nobel Prize in 2020.. The theme of the first example sentence in Table 2.5 is “the accidents”, succeeded by the characters, “the brother and his sister,” in the rheme. “The brother and his sister” then becomes the theme in the subsequent sentence, which causes an effect on “their interest” in the rheme. Next, the marked theme, “Years later”, 33.

(47) signifies the passing time, during which “their interest” influences the characters “them”, the rheme of the third sentence. In the last sentence, “them”, which refers to the characters, i.e. the siblings, or “Jack and Jill”, becomes the theme. It is followed by the successful results of “the accidents” in the first sentence as a happy ending. Linear TP, in which new information is further elaborated, discloses the details in a cumulative way. With more cross-referential links, the text appears rather academic as an idea is expanded from the rheme to the theme in successive sentences. The rheme is picked up in the following sentence, which presents more details as the text proceeds. Linear TP informs readers of the origin of the information and where it is going, which then enhances coherence of the text (Wang, 2007). As the example in Table 2.5 presents, linear TP makes the four sentences more connected, and the passage coherent. The third TP frequently appeared in narratives is derived hyper TP, proposed by Belmonte and McCabe-Hidalgo (1998) based on Danes (1974)’s original model. Usually, the scene description or the topic, is announced in the first sentence of derived hyper TP. Developed from the same overriding theme, the succeeding sentences provide further information. The supporting details help develop the account as the text unfolds. Table 2.6 provides an example of the derived hyper TP.. 34.

(48) Table 2.6 Derived Hyper TP with the Topic in Bold and the Supporting Details Underlined Marked Theme. Theme. Rheme. (new information). (the subject and. (elaboration of the subject). main character) 1). I. woke up at a tranquil night.. 2). The room. was cool and quiet, with both windows open.. 3). The moon. was high in the sky.. 4). Its soft light. fell on my bedroom floor.. In Table 2.6, the major setting, “a tranquil night”, is introduced in the first sentence. The following three sentences depict the night from three different perspectives. In the second sentence, the adjectives, “cool” and “quiet”, are used to describe the room at the serene night, with the moon hanging high up, as illustrated in the next sentence. Lastly, the focus shifts from the moon back to the floor of the room with the “soft” moonlight. In the passage, the writer employs derived hyper TP, introducing the topic in the first sentence and using the supporting details in subsequent sentences to present a more comprehensive picture of the peaceful night. 35.

(49) What the author tries to convey becomes clear with derived hyper TP, and thus the passage creates a clear image of the tranquil night and makes it easier for readers to follow. Thematic structure plays an indispensable role in conveying messages effectively and coherently (Halliday, 1985). Analyzing TPs and TCs in learner writings might help locating learners’ difficulties, which could assist learners to solve their writing problems and to produce coherent texts. The next section reviews previous research that conducted text analysis to examine TPs and TCs in student writings.. 2.2 Text Analysis: Investigating Thematic Progressions (TPs) and Thematic Choices (TCs) in Student Writing Since Halliday proposed the concept of theme and rheme in the 1960s, researchers have examined thematic progression (TP), or the flow of information from theme and rheme across sentences, and thematic choices (TC), or how writers position themes in discourses and articles, to analyze coherence in texts (Cheng, 2002; Ma, 2001; Wang, 2010; Zhang, 2004). The results indicated that TP and TC are crucial in coherent texts (Eggins, 2004; Fetzer, 2008; Green et al, 2000; Halliday, 1985; Hawes & Thomas, 1997; Vande Kopple, 1991). Another focus of SFL research is on the investigation of TPs and TCs in student writings to unveil learners’ writing difficulties 36.

(50) and then to improve coherence in texts (Ebrahimi & Ebrahimi, 2012b; Green et al, 2000; Hu, 2008; Lu, 2013; Wei, 2013). Analyzing TCs and TPs in student writing, Vande Kopple (1991) pointed out the significance of themes in enabling readers to identify genres of texts, and detecting writers’ concerns along with important aspects of the context. Most important of all, he stressed that coherence of a text is closely related to the continuity of themes. Connections between themes across texts, such as linear TP, make a text coherent. In this way, readers can track or store information in groups. On the other hand, incoherent texts lack consistent themes, linear TP, and logical connectors, like “moreover”. Therefore, readers might find it difficult to follow the progression of ideas without sufficient and clear signposts directing the readers. Unlike Vande Kopple (1991), who conducted the research in L1 settings, Wang (2007) examined TCs and TPs to explore students’ writing problems in EFL contexts. The author emphasized the importance of theme theory in writing pedagogy because students were usually poor at producing coherent texts by connecting themes and rhemes. The analysis of the TCs and TPs in a Chinese university student writing revealed three common problems stemming from misusing themes and rhemes. First, the student writer used constant TP most of the time and overused “There” apart from personal pronouns in themes. Next, similar to the problems in other studies (Bloor & 37.

(51) Bloor, 1992; Mauranen, 1996; Mellos, 2011), new information was put in theme position and brand new themes abruptly occurred in sentences frequently. The themes were not introduced earlier in texts, which not only impeded readers’ comprehension, but also disrupted the continuity of themes. The last problem of misusing themes and rhemes in learner writings was that there were numerous empty rhemes which appeared once without followed-up information. Thus, the ideas are disconnected. In line with Vande Kopple (1991) and Wang (2007), Green et al (2000) also reported students’ misuse of themes. They compared Hong Kong freshman students’ writings from a non-native speaker corpus with texts from three native speaker corpora. Examining TCs in student writings, they discovered that non-native speakers tended to misuse and overuse topic-fronting devices together with logical connectors in theme position. First, Chinese learners used more topic-fronting devices, such as “For” and “Concerning”, in theme position than native speakers. Second, there were also more logical connectors, like “Moreover”, “Furthermore”, and “Besides”. Different from the findings in Vande Kopple (1991), it was not a lack of using logical connectors, but overusing and misusing the logical connectors as marked themes in sentences that were judged to be inappropriate by the researchers. For example, the Chinese students tended to use “Besides” to present arguments from different perspectives instead of elaborating on the same point. The frequent thematisation of 38.